STAT6-Arg1 signaling is essential for M2 microglia/macrophage polarization and neurological recovery after ischemic stroke
W. Cai1,2, Q. Ye1,3, S. Hassan1,3, J. Xu1, J. Zhao1, Y. Shi1, J. Chen1,3 and X. Hu1,3
1University of Pittsburgh School of Medicine, Pittsburgh Institute of Brain Disorders and Recovery and Department of Neurology, Pittsburgh, United States
2The Third Affiliated Hospital of Sun Yat-sen University, Neurology, Guangzhou, China
3Veterans Affairs Pittsburgh Health Care System, Geriatric Research, Educational and Clinical Center, Pittsburgh, United States
Abstract
Objectives: The “alternative activated” M2 phenotype of microglia/macrophage is beneficial for brain repair and functional recovery after stroke, whereas the signaling mechanism underlying this regulation is poorly understood. This study investigated whether the signal transducer and activator of transcription 6 (STAT6)-Arginase1 (Arg1) signaling pathway contributed to microglia/macrophage M2 polarization and stroke recovery in a mouse model.
Methods: Adult male C57/BL6 mice (Stat6+/+ or Stat6−,/,−) were randomly assigned to sham surgery or 60 min transient middle cerebral artery occlusion (tMCAO). Infarct volume, microglial/macrophage polarization markers, and sensorimotor and cognitive functions were assessed up to 35d after stroke. Mechanistic studies were performed using primary microglia and neuron co-cultures derived from Stat6+/+ and Stat6−,/,− mice.
Results: Robustly increased expression of phosphorylated STAT6 (pSTAT6), an indicator for activation, was detected predominantly in Iba1+ cells at 1–5d after tMCAO. As compared to Stat6+/+ mice, Stat6−/− post-stroke mice exhibited exacerbated sensorimotor (foot-fault test, cylinder test, and adhesive-removal test) and cognitive (Morris water maze test) deficits, and significantly increased infarct volume (3d, n = 8/group, p < 0.01; 35d, n = 6–8/group, p < 0.01). Moreover, the number of CD206+Iba1+ or Arg1+/Iba1+ cells were significantly decreased in stat6−,/,− mice than Stat6+/+ mice at 3 days after tMCAO (n = 5/group, p < 0.01). In cultures, interlukin-4 induced the elevation of M2 markers (Arg1, IL-10, CD206) and phagocytic activity in Stat6+/+ microglia, but these effects were greatly diminished in Stat6−,/,− microglia. Furthermore, after inhibiting Arg1 activity, which is a direct downstream target of STAT6, viability and efficiency of phagocytosis decreased in microglia, while NO production was increased.
Conclusions: The STAT6-Arg1 signaling pathway is activated after cerebral ischemia and essential for microglia/macrophage polarization into the inflammation-resolving, tissue-repairing M2 phenotype, which thereby promotes long-term neurological recovery after ischemic stroke.
NLA-2
The Niels Lassen Award Session
Optical control of blood flow in naive animals
R.L. Rungta1, B. Osmanski1, D. Boido1, M. Tanter2 and S. Charpak1
1INSERM U1128, Laboratory of Neurophysiology and New Microscopy, Paris, France
2INSERM U979, Wave Physics for Medicine Lab, Paris, France
Abstract
Over the past decade the use of optogenetics to drive genetically distinct populations of brain cells has profoundly increased our understanding of neural circuitry and brain function in health and disease. Optogenetics is now regularly integrated with functional imaging techniques such as BOLD-fMRI or CBV-fMRI, that rely on neurovascular coupling. However, despite the fact that visible light has been shown to dilate peripheral vessels, the effects of light on cerebral blood flow have not been thoroughly investigated. Here we tested whether light stimulation protocols similar to those commonly used in opto-fMRI or to study neurovascular coupling modulate blood flow in mice that do not express any light sensitive proteins.
Combining two-photon laser scanning microscopy and ultrafast functional ultrasound imaging (fUS), approaches that measure blood flow at the microscopic and macroscopic levels, we report that light per se causes a pronounced pseudo functional hyperemia, of similar magnitude to a sensory stimulation, in the neocortex and the olfactory bulb. Pulses or trains of light (473 to 638 nm), shone on the mouse brain across a chronic cranial window, triggered a reversible and reproducible dilation of arterioles. The effect was energy-dependent appearing at a threshold of ∼0.5mW-1mW. Two-photon imaging of GCaMP6f in transgenic mouse lines revealed that light caused a decrease of vascular smooth muscle cell calcium that preceded the onset of dilation and occurred in the absence of neuronal or astrocyte excitation.
These results impose careful consideration on the use of photo-activation in studies involving blood flow regulation, as well as in studies requiring prolonged and repetitive stimulations to correct cellular defects in pathological models. They also suggest that light could be used to locally increase blood flow in a controlled fashion.
NLA-3
The Niels Lassen Award Session
Acute gliovascular phenotype depends on primary injury severity in a new juvenile Closed Head Injury with Long-term Disorders (CHILD) model
B. Rodriguez-Grande1, A. Obenaus2,3,4, A. Ichkova1, T. Bessy1, J. Aussudre1, E. Barse1, B. Hibba1, G. Catheline1, G. Barriere1 and J. Badaut1,2
1INCIA UMR5287 CNRS, Bordeaux, France
2Loma Linda University School of Medicine, Department of Pediatrics, Loma Linda, United States
3Loma Linda University School of Medicine, Department of Basic Science, Loma Linda, United States
4University of California, Irvine, Department of Pediatrics, Irvine, United States
Abstract
Objectives: Traumatic brain injury (TBI) is the leading cause of pediatric death and disability. Assumptions about the underlying mechanisms of damage are often based on the nature of the injury (penetrating or blunt, rotating or fixed, etc). We hypothesized that, even in the same juvenile TBI (jTBI) model, graded severity could result in different pathomechanisms. We aimed to identify severity-associated differences in a) long-term outcomes (behavior and damage to the corpus callosum), and b) early mechanisms involving gliovascular changes in the white matter (WM).
Methods: In a new jTBI mouse model, Closed Head Injury with Long-term Disorders (CHILD), high-speed cameras characterized the kinematics of mild or moderate (1 or 3 mm depth, 2 or 3 m/s speed respectively) TBI induced with an electromagnetic impactor at postnatal day 17. Magnetic resonance diffusion tensor imaging (DTI) and T2 weighted imaging (T2WI) were used to study WM and edema, respectively. Behavior was tested using open field. Astrocytic activation and alterations in the water channel AQP4 were assessed immunohistologically.
Results: Moderate jTBI produced increased and longer-lasting head acceleration and displacement than mild jTBI. One day post-injury moderate jTBI induced significant edema (increased T2, 3.70% of sham). On the contrary, decreased T2 (−3.95% of sham) was observed in mild jTBI mice, which displayed higher AQP4 expression than the sham and moderate cohorts. One month after jTBI both cohorts showed anxious behavior and lower DTI-derived fractional anisotropy compared to shams, but only the mild group exhibited increased axial diffusivity.
Conclusions: Different gliovascular phenotypes, rather than only graded levels of the same pathology, are observed acutely after different severities of jTBI, which then converge in similar long-term deficits. Since AQP4 is distinctly regulated, it could be a key target for acute treatments based on primary injury severity.
NLA-4
The Niels Lassen Award Session
Selective non-nuclear estrogen receptor activation decreases stroke severity and promotes functional recovery after stroke in mice
U.M. Selvaraj1, E. Plautz2, K. Chambliss3, X. Kong2, S. Rovinsky2, S. Zhang4, C. Mineo3, B. Katzenellenbogen5, J. Katzenellenbogen5, S.H. Kim5, P. Shaul3 and A. Stowe2
1UTSW Medical Center, Immunology, Dallas, United States
2UTSW Medical Center, Neurology and Neurotherapeutics, Dallas, United States
3UTSW Medical Center, Pediatrics, Dallas, United States
4UTSW Medical Center, Advance Imaging Research Center, Dallas, United States
5University of Illinois at Urbana-Champaign, Departments of Molecular and Integrative Physiology, and Chemistry, Urbana, United States
Abstract
Introduction: Estrogens protect the brain from experimental stroke and, correspondingly, stroke incidence is higher in men compared to women, a protection lost after menopause. However, estrogens have undesirable reproductive tract and cancer-related actions. In mice, the selective activation of non-nuclear estrogen receptors has beneficial vascular and metabolic impact without these adverse effects. We hypothesized that selectively activating non-nuclear estrogen receptors (selective estrogen, SE; also known as PaPE-1, pathway-preferential estrogen) lessens stroke severity and promotes post-stroke recovery in mice.
Methods: Female 8 week-old C57BL/6 mice were ovariectomized prior to receiving subcutaneous pellet implants containing vehicle, estradiol, or SE (randomized, n = 9–10/group). In one protocol, mice received rotorod training prior to 45-min transient middle cerebral artery occlusion (tMCAo), with MRI and rotorod performance assessed through 2 weeks post-tMCAo. In a second protocol leukocytes subpopulations in the brain, spleen, and blood were quantified 3d post-stroke.
Results: Compared with vehicle, both estradiol and SE reduced infarct volumes at 3 d post-tMCAo, by 35.8% (p = 0.04) and 65.7% (p = 0.0005), respectively. This was associated with marked decreases in the number of leukocytes infiltrated into the brain 3 days post-stroke of 63% (p = 0.019) and 93% (p = 0.002), respectively. Compared with vehicle, estradiol caused an increase in uterine wet weight whereas SE had no effect. Mice experienced a 53% decline in rotorod performance from baseline at 2 d post-tMCAo. SE improved functional recovery to 85% (p = 0.005) and 82% (p = 0.001) of baseline at 6 d and 13 d post-tMCAo, respectively, while estradiol treatment improved function only at d13 post-stroke, to 76.3% of baseline (p = 0.005).
Conclusion: The selective activation of non-nuclear estrogen receptors decreases stroke severity and improves functional recovery by early regulation of immune cells entering the brain post-stroke. Future studies will determine the efficacy of this treatment when administered only post-stroke, and further identify the basis for the observed structural and functional benefits.
Oral Sessions
BS01-1
Neuroinflammation in cerebral ischemia 1
Triggering receptor expressed on myeloid cells 2 (TREM2) controls subacute injury resolution and myeloid cell reactivity after experimental stroke
A. Alfieri1 and B. McColl1
1University of Edinburgh, Roslin Institute, Midlothian, United Kingdom
Abstract
Neuroinflammation is a major event in the pathophysiology of ischemic stroke, affecting injury and repair after ischemia, being critically orchestrated by resident and invading myeloid cells. TREM2 is a cell surface receptor expressed by resident microglia, monocyte-derived macrophages and neutrophils. Although TREM2 affects myeloid cell responses and survival, its role in subacute complications, injury resolution and repair after brain ischemia is unclear.
Our study aimed to determine the temporal profile and cellular sources of TREM2 expression during the subacute phase after experimental stroke in the brain, alongside markers of brain injury (neuronal loss, edema and hemorrhagic transformation), neuroinflammation and brain repair in wild type and TREM2 deficient mice 1 to 14 days after middle cerebral occlusion (MCAO).
Transcription and protein expression of TREM2 increased significantly after MCAO, peaking at later time-points, mainly produced by microglia. Brain damage was comparable acutely but was greater in the absence of TREM2 14d after MCAO, suggesting a role for TREM2 primarily in the subacute phase. At the same time-point, peri-infarct microglia/macrophage reactivity and neutrophil infiltration were significantly reduced in the peri-infarct region of TREM2 deficient mice. Altered levels of TREM1, IL-1beta, IL-10 and arginase-1 were also observed in TREM2 knockout animals after MCAO indicating a regulatory effect of TREM2 on the neuroinflammatory milieu.
Our findings demonstrate the importance of TREM2 in regulating myeloid cell reactivity and accumulation in the subacute phase of experimental stroke, affecting the neuroimmune environment and ultimately brain damage. In conclusion, we propose that TREM2 may be a valid therapeutic target to direct neuroinflammation towards tissue protective and restorative activity following ischemic stroke.
Funded by the Medical Research Council, Biotechnology and Biological Sciences Research Council.
BS01-2
Neuroinflammation in cerebral ischemia 1
Microglial control of post-ischemic neurogenesis
R. Wakasaki1 and I.P. Koerner1
1OHSU, Portland, United States
Abstract
Background: Adult neurogenesis increases after brain injury, including cardiac arrest (CA). However, even though increased neurogenesis offers the potential for regeneration of lost neurons, few cells survive long-term. The cellular mechanisms behind this remain unclear. We used a novel mouse that allows targeted ablation of microglia to investigate their role for neurogenesis after CA.
Methods: We used the CX3CR1Cre−ER/DTR mouse first described by Parkhurst. Diphtheria toxin receptor (DTR) expression is induced by tamoxifen in CX3CR1-positive microglia and macrophages. A 30 days waiting period allows DTR-negative cells to replace the short-lived macrophages, while long-lived microglia maintain DTR expression and are readily ablated by diphtheria toxin. We ablated microglia before CA and determined neuronal death in the hippocampal CA1 region 3 days after CA. We injected mice with BrdU to label cells born on days 3–6 after CA. We quantified newborn neurons in the dentate gyrus after double-labeling for doublecortin (DCX).
Results: Death of CA1 neurons was dramatically reduced when microglia were ablated before CA (427 ± 333 dead cells/mm2 (control) vs 132 ± 105 (MG ablation), P = 0.02; A). Cell proliferation measured by BrdU incorporation increased in the first week after CA (168 ± 61 cells (CA) vs 69 ± 4 (baseline), P < 0.05; B). To our surprise, this increase was fully blocked when microglia were absent (83 ± 32). Similarly, the increase of DCX-positive young neurons after CA also required the presence of microglia (164 ± 21 (baseline), 349 ± 104 (CA), 63 ± 25 (CA after MG ablation), P < 0.05; C).
Conclusion: While microglia drive neuronal death after CA, they are also required for restorative neurogenesis. This adds another facet of the diverse microglial response to ischemic brain injury, and creates a paradigm shift. Future therapeutic interventions will need to carefully consider these two opposing capacities, and aim to suppress one without interfering with the other.
BS01-3
Neuroinflammation in cerebral ischemia 1
Double negative T lymphocytes from gld mice drive an M2-microglial polarization through FasL/TIAL1/IL-4 pathway in experimental stroke
H. Meng1, L. Weng1, D. He1, Y. Chen1, L. Han1 and Y. Xu1
1Drum Tower Hospital, Medical School of Nanjing University, Nanjing, China
Abstract
Objectives: The crosstalk between T lymphocytes and microglia plays important roles in modulating neuroinflammation after ischemic stroke. We have demonstrated FasL mutant (gld) mice attenuated brain inflammation after ischemic injury, with a dramatic accumulation of double negative (DN) T lymphocytes in peripheral blood and brain. However, the effects of DN T lymphocytes from gld mice on modulating pro-inflammatory (M1) or anti-inflammatory (M2) microglial polarization and the underlying mechanisms are not fully understood.
Methods: The subtypes of T lymphocytes and microglial polarizations in ischemic brain were measured at 24 h and 72 h after middle cerebral artery occlusion (MCAO) in gld and C57BL/6J mice. DN T lymphocytes from gld and C57BL/6J mice after MCAO were cocultured with primary microglia for 24 h, followed by quantification of inflammatory cytokines in the supernatant and evaluation of M1/M2 phenotypes. Neurotoxicity and cell viability were assessed after the coculture supernatant was applied to oxygen glucose deprivation (OGD)-induced neurons for 24 h. Lentivirus-mediated silencing or overexpressed T-cell-restricted-intracellular-antigen-like1 (TIAL1) gene was transfected into DN T lymphocytes.
Results: DN T lymphocytes significantly accumulated in ischemic cerebral hemisphere in gld mice after MCAO, with an increased M2-microglial polarization. Microglia underwent M2 polarization when cocultured with DN T lymphocytes of gld mice. In addition, coculture supernatant from gld mice contained more anti-inflammatory cytokines, and consistently attenuated OGD-induced neuronal death. Furthermore, DN T lymphocytes from gld mice exhibited a strong downregulation of FasL and TIAL1 in protein levels, with an enhanced protein level of IL-4. In contrast, transfection of TIAL1-overexpressed lentivirus abolished the advanced effects of DN T lymphocytes from gld mice in producing IL-4, and simultaneously reduced the M2-microlgial polarization.
Conclusions: DN T lymphocytes with FasL mutation drive M2-microglial polarization, attenuate neuroinflammation and protect neuron from ischemic injury in experimental stroke, which may involve the abolishment of FasL/TIAL1 pathway in inhibiting IL-4 expression.
BS01-4
Neuroinflammation in cerebral ischemia 1
Interleukin-1 contributes to neuronal injury via cell-specific IL-1 type 1 receptor-mediated actions after cerebral ischemia
N. Lénárt1, R. Wong2, B. Martinecz1, G. Coutts2, A. Waisman3, W. Muller2, S. Allen2, E. Pinteaux2 and A. Denes1,2
1Hungarian Academy of Sciences, Institute of Experimental Medicine, Molecular and Developmental Neurobiology, Laboratory of Neuroimmunology, Budapest, Hungary
2University of Manchester, Faculty of Biology, Medicine and Health, Manchester, United Kingdom
3Institute for Molecular Medicine, Mainz, Germany
Abstract
Objectives: The proinflammatory cytokine interleukin-1 (IL-1) is a major contributor to brain injury. Although blockade of IL-1 signalling by IL-1 receptor antagonist is protective in experimental models of cerebral ischemia, the precise mechanisms by which IL-1 triggers neuronal injury remain unknown. The aim of this study was to identify the cell-specific mechanisms of actions of IL-1 responsible for the detrimental IL-1 actions triggered by IL-1R1 after brain injury induced by cerebral ischemia.
Methods: Using a recently generated IL-1R1 floxed (fl/fl) mouse line, we performed conditional deletion of IL-1R1 in myeloid cells, platelets, brain endothelial cells or neurons. Mice were subjected to transient focal ischemia (30 or 45 min middle cerebral artery occlusion, MCAo) to investigate the effect of conditional IL-1R1 deletion on infarct size, neurological outcome and inflammation in the brain.
Results: We show that selective elimination of IL-1R1 from myeloid cells or platelets did not have a significant effect on infarct size compared to control (IL-1R1fl/fl) mice. Conditional deletion of IL-1R1 in neurons and brain endothelial cells using Nestin-CreERT2 or Slco1c1-CreERT mice respectively resulted in significantly decreased infarct size and better neurological outcome compared to control mice. IL-1R1 neuronal deletion was associated with increased microglial CD45 expression, whilst the number of Iba1+ microglia and CD45+ leukocytes were not different in the ipsilateral hemisphere.
Conclusions: Neuronal and brain endothelial IL-1R1 is a key contributor to detrimental IL-1 actions in the brain. Selective targeting of IL-1R1 in the brain could have major therapeutic benefit in cerebrovascular diseases.
BS01-5
Neuroinflammation in cerebral ischemia 1
CCL23: a new chemokine present in brain after cerebral ischemia might play a role as a blood biomarker for brain damage and stroke outcome
A. Simats1, T. García-Berrocoso1, G. Llovera2, A. Penalba1, D. Giralt1, A. Bustamante1, E. Martinez-Saez3, A. Rosell1, A. Liesz2 and J. Montaner1
1Vall d'Hebrón Institute of Research (VHIR), Neurovascular Research Lab, Barcelona, Spain
2Institute for Stroke and Dementia Research, Klinikum der Universität München, München, Germany
3Universitat Autònoma de Barcelona, Neuropathology Unit, Department of Pathology, Barcelona, Spain
Abstract
CCL23 plays a fundamental role in the modulation of the immune response. Circulating CCL23 levels have been found elevated in several inflammatory diseases, but nothing is known about the role of this chemokine in the stroke-associated inflammatory process. Our aim was to evaluate the potential of this chemokine for stroke diagnosis and prognosis and complement it with a molecular study in experimental stroke models.
CCL23 blood levels were evaluated in a total of 369 individuals, including ischemic stroke and stroke-mimicking patients within 4.5 h from symptoms onset. Both disease groups presented higher CCL23 levels than healthy volunteers (p < 0.001 for both comparisons). Although strokes could not be differentiated from mimics (p = 0.173), CCL23 levels identified any type of cerebral injury (p = 0.006) and showed an increase in blood levels at 24 h after stroke (p < 0.001). Moreover, in stroke patients, baseline level of circulating CCL23 resulted an independent predictor of mortality after 3 months of stroke onset (ORadj: 26.969[3.251–223.729], p = 0.002) and disability at hospital discharge (ORadj: 29.073[1.657–510.02], p = 0.021).
In vivo, CCL23 levels were increased 2 h after stroke in cerebrospinal fluid, but not in plasma, from rats submitted to cerebral ischemia (p = 0.006 and p = 0.851, respectively, compared to sham-operated animals). The expression state of the CCL23 rodents' counterparts, CCL6 and CCL9, resulted to be up-regulated in brains from ischemic rats 2 h after stroke (p = 0.095 and p = 0.063, respectively) and remained increased in mouse brains 24 h after the challenge (p = 0.068 and p = 0.038, respectively), corroborating the brain provenance of this chemokine at early time points after cerebral ischemia. Conversely, 5 days after stroke, the presence of CCL23 in human brains was only seen on granulocytes infiltrated into the brain parenchyma by that time.
Thus, CCL23 might have a dual role in the inflammatory processes associated to cerebral injury and its baseline levels might be good predictors of stroke outcome.
BS01-6
Neuroinflammation in cerebral ischemia 1
Perivascular macrophages attract neutrophils to the brain after ischemia
J. Pedragosa1,2, A. Salas-Pérdomo1, M. Gallizioli1, R. Cugota1, F. Briansó3, F. Pérez-Asensio1, A. Gieryng4, B. Kaminska4, F. Miró-Mur2 and A.M. Planas1,2
1Institut d’Investigacions Biomèdiques de Barcelona, Consejo Superior de Investigaciones Científicas, Barcelona, Spain
2Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain
3Vall d'Hebron Research Institute (VHIR), Statistics and Bioinformatics Unit, Barcelona, Spain
4Nencki Institute, Laboratory of Molecular Neurobiology, Warsaw, Poland
Abstract
Objective: Brain perivascular and meningeal macrophages (PVMM) have been attributed functions of immunological surveillance. The aim of this study was to investigate the acute response of this particular subset of brain resident macrophages to ischemic stroke.
Methods: One-hour middle cerebral artery occlusion (MCAo) was induced with the intraluminal technique in adult male Sprague-Dawley rats. PVMM were detected by immunohistochemistry and flow cytometry with antibodies against the scavenger receptor CD163. CD163+ cells were sorted by FACS from the brain of controls and 15 h after MCAo. RNA was extracted from the isolated cells to study gene expression (Affymetrix microarray) (n = 3 per group). The expression of selected genes was validated by qRT-PCR in CD163+ cells sorted from independent groups of rats (n = 6 per group). Depletion of PVMM was induced by intracerebroventricular administration of clodronate liposomes and MCAo was carried out four days later. Twenty-four hours post-ischemia, mRNA expression (n = 6 per group) and infiltrating cells were analysed by flow cytometry (n = 9–10 per group) and microscopy (n = 8–10 per group). The brain lesion was studied 24 h post-ischemia by T2w MRI.
Results: Only PVMM expressed CD163+ in control brain and within 24 h post-ischemia. Brain ischemia strongly altered the gene expression profile of PVMM vs. controls (Fig. 1) inducing upregulation of neutrophil chemoattractants, amongst other changes. Clodronate liposomes strongly reduced PVMM without affecting microglia cell number. Rats depleted of PVMM showed reduced mRNA expression of neutrophil chemoattractants and less neutrophils in the ischemic tissue, as independently assessed by flow cytometry and microscopy. The size of the brain lesion is being analyzed.
Conclusion: This study supports the involvement of PVMM in the acute recruitment of neutrophils to the ischemic brain.
Work supported by MINECO (SAF2011-30492). JP and MG have fellowships from AGAUR and the ITN EU program (n°607962), respectively.
BS02-1
Imaging methods
Generalized vessel filter for detecting vessels in medical images
C.-Y. Hsu1, B. Schneller1, N. Narasimhan Sriram1 and A. Linninger1
1University of Illinois at Chicago, Chicago, United States
Abstract
Objectives: Current imaging modalities such as MRI, CT and 3DRA use contrast agents to capture the entire cerebral vasculature, yet information on small vessels and bifurcations are unavailable due to noise, scattering artifacts or weak contrast agents. Our goal is to construct an automated filter that yields increased vessel contrast at all length scales, minimize intensity of non-relevant structures, ensure a fully connected network of vessels and include control parameters for adaptation to varying image sources.
Methods: An automated filtering pipeline is developed containing at its core a multi-scale composite filter (MCF). The MCF functions by computation of the Hessian matrix followed by evaluation of pertinent eigenvalues. The automated package reads a stack of images through a user friendly pop up window, automatically storing the intensities in the image in a 3D matrix. Secondly a parameter file possessing filter controls updated by the user automatically modifies filter to adjust to the image source. This package was tested on images from two sources: magnetic resonance angiography (MRA), 3D rotational angiography (3DRA).
Results: A representative dataset from each imaging modality mentioned in the methods before and after filtration is shown in figure 1. In MRA (figure 1A) the filtered suppresses background noise while amplifying the vessel signal. In case of 3DSA (figure 1B) scattering artifacts are minimized and vessel intensity across the image is equalized. High dice coefficients of >0.77 validate the reliability of the MCF.
Conclusion: MCF enhances vessels at varying length scales while ensuring network connectivity for multiple image modalities. This would allow the fully automatic reconstruction of cerebrovascular networks which can be used for identifying artifacts and construct computational models.
BS02-2
Imaging methods
Influence of relative cerebral blood volume on oxygenation-sensitive T2′ and R2′ mapping in acute ischemic stroke
A. Seiler1, R. Deichmann2, U. Nöth2, W. Pfeilschifter1, J. Berkefeld3, O.C. Singer1 and M. Wagner3
1Goethe University Frankfurt, Neurology, Frankfurt, Germany
2Goethe University Frankfurt, Brain Imaging Center, Frankfurt, Germany
3Goethe University Frankfurt, Neuroradiology, Frankfurt, Germany
Abstract
Background and Purpose: Quantitative T2′ and R2′ (1/T2′) mapping are sensitive to locally increased concentrations of deoxygenated hemoglobin (Hb). Significantly decreased T2′-, respectively increased R2′-values have been observed within perfusion-restricted tissue in acute ischemic stroke and interpreted as the surrogate of an increased oxygen extraction fraction (OEF). R2′ (1/T2′) is assumed to be proportional to the product of OEF and relative cerebral blood volume (rCBV). As local increases of rCBV might substantially influence T2′ and R2′ through accumulation of deoxygenated Hb, we aimed at investigating the relationship between T2′/R2′ and rCBV patients with acute ischemic stroke.
Materials and Methods: Data from 15 patients (5 male, mean age 69 ± 11.6 years) with acute internal carotid artery (ICA) and/or middle cerebral artery (MCA) occlusion were analyzed. T2′, R2′ and rCBV were measured within the infarcted core, oligemic and severely perfusion-delayed areas delineated on time-to-peak (TTP) maps and the relationship between these parameters was examined.
Results: No significant elevations of rCBV were detected within areas with significantly changed T2′ and R2′. A significant negative correlation with rCBV was found for R2′ and rCBV (r = −0.544, p = 0.002), whereas T2′ correlated positively with rCBV (r = 0.546, p = 0.001) in TTP-delayed areas. T2′ and R2′ within hypoperfused tissue remained unchanged at normal or elevated rCBV levels.
[Scatterplot_boxplot_T2′_R2′_rCBV]
Conclusion: Changes of T2′ and R2′ within perfusion-restricted areas in acute ischemic stroke are not caused by local elevations rCBV but best explained by increases of the cerebral OEF. T2′ and R2′ mapping are suitable to detect altered oxygen consumption in acute ischemic stroke and might particularly be useful in patients beyond established time windows and patients with mild or fluctuating symptoms despite persistent vessel occlusion.
BS02-3
Imaging methods
Near-infrared spectroscopy measured vascular reactivity and blood flow autoregulation during intracranial pressure changes
A. Ruesch1, M.A. Smith2, G. Wollstein3, I.A. Sigal2, S. Nelson2 and J.M. Kainerstorfer1
1Carnegie Mellon University, Department of Biomedical Engineering, Pittsburgh, United States
2University of Pittsburgh, Department of Ophthalmology, Pittsburgh, United States
3New York University School of Medicine, NYU Langone Eye Center, New York, United States
Abstract
Objectives: The brain counteracts changes in cerebral perfusion pressure (CPP) and maintains approximately constant cerebral blood flow (CBF) via cerebral autoregulation (CA), which is mediated by pressure reactivity. Within a certain range of CPP, pressure reactivity is implemented through a vasomotor adjustment in vascular resistance. How well CBF is maintained and how well the vasculature reacts and compensates changes in mean arterial blood pressure (MAP) is indicative of the level of autoregulation. While CBF changes and pressure reactivity are related, they are not the same [1] and measurement methods vary. Pressure reactivity is a reliable measure during MAP, but not necessarily during intracranial pressure (ICP) changes. Furthermore, ICP is not measured in many conditions due to invasiveness of the method. Here we combine CBF and pressure reactivity into one CA index to account for such challenges.
Method: Transient, two minute ICP changes (DICP = 10 mmHg) were induced in non-human primates (n = 4) by increasing CSF fluid volume, mimicking Lundberg B-waves. Experiments were repeated at different baseline level of ICP. MAP, ICP, and hemoglobin concentration via near-infrared-spectroscopy were recorded continuously.
Results: We confirmed that induced changes in ICP do not necessarily trigger MAP changes. We found a high correlation between CPP and total hemoglobin (HbT) concentration. In addition, we constructed Lassen's autoregulation curve when using tissue oxygen saturation (StO2) as a surrogate for CBF and HbT as a surrogate for CPP.
Conclusion: Simultaneous, non-invasive measurements of pressure reactivity and CBF autoregulation are achievable via HbT and StO2, respectively. We believe that this combined measurements will lead to determination of the limits of CA during changes in ICP. This is particularly important for conditions where ICP cannot be measured directly.
References:
[1] K.P. Budohoski, et al., “The Relationship Between Cerebral Blood Flow Autoregulation and Cerebrovascular Pressure Reactivity After Traumatic Brain Injury,” Neurosurgery, vol.71, no.3, 2012.
BS02-4
Imaging methods
Non-invasive functional neuroimaging in mice using structured illumination diffuse optical tomography
M. Reisman1, Z. Markow2, A. Bauer3 and J. Culver1,2,3
1Washington University in St Louis, Department of Physics, St. Louis, United States
2Washington University in St Louis, Department of Biomedical Engineering, St. Louis, United States
3Washington University in St Louis, Department of Radiology, St. Louis, United States
Abstract
Objective: Optical intrinsic signal (OIS) imaging has been a powerful tool for capturing functional brain hemodynamics in rodents.1 However, OIS requires scalp retraction and is limited to superficial cortical tissues. Diffuse Optical Tomography (DOT) techniques provide non-invasive imaging, but previous DOT systems are limited by either sparse spatial sampling2 or slow speed.3 Here, we develop a DOT system that combines the high-density spatial sampling of camera-based systems with the rapid imaging of structured illumination4 (SI) to perform non-invasive mapping of functional responses in the mouse cortex.
Methods: The DOT system is comprised of an sCMOS camera and a digital micromirror device (DMD) projector (Fig 1A). SI pattern sequences and measurements were optimized using an analysis of the signal-to-noise ratio of individual measurements relating light intensity to an effective source-detector separation. This was used to identify and remove noisy data from tomographic reconstructions (Fig 1B).
Results: By optimizing the pattern sequence and measurement set included in the reconstructions, we non-invasively observed evoked responses following electrical stimulation of the forepaw. Changes in total hemoglobin showed the expected temporal response (Fig 1C) and spatial response up to a depth of 2 mm (Fig 1D).
Conclusion: SI-DOT overcomes the limitations of previous DOT systems to provide non-invasive three-dimensional imaging of the mouse cortex. The SI-DOT system reported is built on an infrastructure with great flexibility, with analysis techniques designed to enable efficient further optimization to balance imaging speed, resolution, field-of-view, and computation time.
References:
1. White, B. et al. PLoS ONE6, e16322 (2011).2. Culver, J. et al. Opt. Lett.28, 2061 (2003).3. Patwardhan, S. et al. Opt. Express13, 2564–2577 (2005).4. Cuccia, D. et al. Opt. Lett.30, 1354–1356 (2005).
BS02-5
Imaging methods
Stimulus-evoked activations are a subset of resting-state networks: Comparison between fMRI and optical imaging in the rat olfactory bulb
G.J. Thompson1, K. Baker2,3, JV. Verhagen2,3, B.G. Sanganahalli1,4, G.M. Shepherd2 and F. Hyder1,4,5
1Yale University, Radiology and Biomedical Imaging, Magnetic Resonance Research Center (MRRC), New Haven, United States
2Yale University, Neurobiology, New Haven, United States
3The John B. Pierce Laboratory, New Haven, United States
4Yale University, Quantitative Neuroscience with Magnetic Resonance (QNMR) Core Center, New Haven, United States
5Yale University, Biomedical Engineering, New Haven, United States
Abstract
Objectives: Spontaneous fluctuations have been observed in many brain regions with many methods, but, as they are not linked to a task or stimuli, it is difficult to determine their source and function [1].
Methods: We recorded fMRI (at 9.4 T) and intrinsic optical imaging at 565 nm (blood flow) and 617 nm (blood oxygenation) from freely-breathing rats under urethane anesthesia [2]. Data were recorded from the olfactory bulb's glomerular layer under stimulation with three ester odors: isoamyl acetate (IAA), methyl valerate (MV) and ethyl butyrate (EB). We measured odor activation and also 300 s+ baseline scans to measure spontaneous fluctuations, i.e. the “resting state.” Resting state fluctuations' spatial extents were classified into 22 networks for fMRI and 7 networks for optical imaging (3 discarded as noise).
Results: In the glomerular sheet (GS), using fMRI, the resting state networks that correlated with each activation map were different for each odor, though more similar for MV and EB (Figure 1A-B) versus IAA. Results from optical imaging at 617 nm indicated that MV and EB activation maps also correlated strongly with certain networks in a similar pattern (Figure 1C-D). In fMRI, some ICs did not correlate with the studied odors, but also appeared highly organized, suggesting spontaneous fluctuations go beyond the three odors studied herein.
Conclusions: These results indicate spontaneous fluctuations in the olfactory bulb are organized relevantly to odor sensing. These “networks” may combine in various ways depending on the odor presented. Our work provides strong evidence that fMRI-measured spontaneous fluctuations reflect intrinsic organizations that can be recruited based on specific sensory stimuli.
References:
[1] Smith, et al., PNAS U S A, 2009. 106(31): p. 13040–5.[2] Sanganahalli, et al., Neuroimage, 2016. 126: p. 208–18.
BS02-6
Imaging methods
Repeated longitudinal in vivo imaging of cortical microglia under chronic hypoxia in the mice using two-photon microscopy with a closed cranial window technique
I. Kanno1, K. Masamoto1,2, H. Yuki3, T. Sugashi3, M. Unekawa4, Y. Tomita4 and N. Suzuki4
1National Institute of Radiological Sciences, Department of Functional Brain Imaging Research, Chiba, Japan
2University of Electro-Communications, Brain Science Inspired Life Support Research Center, Chofu, Japan
3University of Electro-Communications, Graduate School of Informatics and Engineering, Tokyo, Japan
4Keio University School of Medicine, Department of Neurology, Tokyo, Japan
Abstract
Background: Cerebral hypoxia induces profound vascular remodeling, such as capillary dilation and network formation, in the adult brains. However, little is known about the role of glial cells in the vascular remodeling of the brain.
Objectives: The aim of the present study is to characterize a morphological change in microglia during hypoxia adaptation.
Methods: For visualization of microglia in vivo, CX3CR1-GFP mice (N = 8) in which the cortical microglia expressed green fluorescence protein (GFP) were used, and the GFP-positive microglia and cortical microvessels labeled with sulforhodamine 101 were repeatedly imaged with two-photon microscopy through a closed cranial window developed by Tomita and Seylaz et al. JCBFM 2005. After completion of the surgery for installation of the cranial window over left temporoparietal region, the animals were recovered and housed in a hypoxic room (8–9% oxygen) up to 3 weeks. Before and during hypoxia exposure, the cell size, number density, and number of processes of the microglia were measured with a Matlab software.
Results: Under normoxia conditions, there were no detectable differences in the microglial morphology over 3 weeks from the surgery. No statistically significant differences in the soma area (29.5 ± 2.4 to 28.9 ± 1.7 µm2) and number of processes (8.1 ± 1.4 to 6.7 ± 1.6 number/cell) were observed after 2-week exposure to chronic hypoxia, but a slight decrease in the cell density (8.4 ± 0.8 to 7.3 ± 0.6 × 103 number/mm3). The microglial motility, defined as a displacement of the cell location over a week, was increased after exposure to hypoxia. Some of the cells were found to be accompanied with a capillary sprout and elongation.
Conclusions: Repeated longitudinal in vivo imaging of cortical microglia revealed a close association of microglial migration and microvascular sprout during adaptation to cerebral hypoxia.
BS03-1
Neurovascular coupling
Extensive 3D vascular mapping points to a lack of spatial compartmentalization around neuronal units of the lemniscal pathway
J. Mitiagin1, D. Kain1, H. Jang2 and P. Blinder1,3
1Tel-Aviv University, Neurobiology, George Weiss Faculty of Life Sciences, Tel-Aviv, Israel
2New York University, College of Arts & Science, New York, United States
3Tel-Aviv University, Sagol School of Neuroscience, Tel-Aviv, Israel
Abstract
Recent investigation into the organization of the neurovascular unit around mouse cortical columns pointed to a lack of vascular discretization resulting in neuronal units being embedded in a vasculature continuum; whether this is a canonical organization principle remains an open question.
Objectives:
i) test whether or not there´s a tight spatial vascular compartmentalization around histologically defined neuronal units along the lemniscal pathway.
ii) map the distribution and dynamics of pericytes with respect to cortical columns.
Methods: we combine awake two-photon imaging and large-scale reconstructions of vascular networks surrounding trigeminal “barrelets”, thalamic “barreloids” and cortical “barrels”. We measured different vascular morphological features, branching patterns and soma to nearest vessel distance inside cyto-architectonical identified units and compared these with values obtained by randomly rotating and placing the same units and resampling; this allows us to identify spatial organization patterns around such units. We also measure blood flow dynamics around pericytes (tdTomato expressed under the pdgrfβ-R1 promotor) while considering their relative position with respect to functionally defined neuronal units.
Results: Despite area-related general difference in vascular morphological parameters along the lemniscal pathway, we found no difference between the original and randomly sampled vasculature networks. Pericytes tend to show higher densities inside the boundaries of neuronal units. Initial in vivo results show that microvascular segments associated with pericytes, within the cortical column, have a mixed response to sensory stimulation.
Conclusions: Blood vessels show now spatial patterning that can support tight control of blood flow perfusion, not only around cortical columns as previously shown but along key portions of the lemniscal pathway. Our preliminary morphological evidence points to preferential distribution of contractile elements along the topological vascular tree. Whether they act in a coordinated manner that could support localized changes in blood flow as a response of neuronal activity requires further investigation.
BS03-2
Neurovascular coupling
Depths-dependent flow and pressure characteristics in cortical microvascular networks
F. Schmid1, P. Jenny1 and B. Weber2
1ETH Zurich, Institute of Fluid Dynamics, Zurich, Switzerland
2University of Zurich, Institute of Pharmacology and Toxicology, Zurich, Switzerland
Abstract
A detailed understanding of the flow in cortical microvascular networks (MVNs) is crucial for various fundamental aspects related to brain homeostasis.
We simulate flow in three realistic MVNs from mouse parietal cortex, which have been obtained by two-photon laser scanning microscopy [1]. The numerical model has the unique feature of tracking individual red blood cells (RBCs) [2]. Furthermore, we applied a new approach to assign appropriate boundary conditions over the whole cortical depth of the MVNs [3].
Our study focuses on: (1) The pressure drop in different parts of the network and (2) the trajectories of individual RBCs through the network. Both aspects are analyzed with respect to the cortical depth.
The trajectories of individual RBCs reveal that RBCs tend to exit the capillary bed at the same cortical depth at which entering it. Hence, in the capillary bed the preferential RBC flow direction is horizontal.
Furthermore, we show that close to the cortical surface the pressure drop is largest in the capillary bed, while deeper in the cortex it becomes dominant in the arterioles (Figure 1). We postulate that different regulation mechanisms might be in place depending on cortical depth. Hence, we extended our studies to analyze the effects of various neurovascular coupling scenarios, such as (1) descending arteriole dilation, (2) pre-capillary dilation and (3) capillary dilation.
(A) MVN with five ALs. (B),(C),(D) Mean pressure drop for different vessel types for AL 1, 3 and 5 illustrated over the mean path length. Standard deviations are given by the error bars. PA: pial arterioles, DA + A: descending arterioles and arterioles, C: capillaries, V + AV: venules and ascending venules, PV: pial venules. From [3].
References:
[1] Tsai et al., J.Neurosci., 2009.[2] Schmid et al., AJP-Heart&Circ, 2015.[3] Schmid et al., submitted.
BS03-3
Neurovascular coupling
Two-photon microscopy measurement of cortical functional activation in awake mice
I. Sencan1, T. Esipova2, K. Kilic3, B. Li1, M. Desjardins3, M.A. Yaseen1, H. Wang1, R. Jaswal1, S. Kura1, B. Fu1, D.A. Boas1, A. Devor1,3, S.A. Vinogradov2 and S. Sakadžić1
1Massachusetts General Hospital & Harvard Medical School, Charlestown, United States
2University of Pennsylvania, Philadelphia, United States
3University of California San Diego, San Diego, United States
Abstract
Objectives: The goal of this study was to characterize cortical microvascular pO2 and blood flow changes in response to functional brain activation by whisker and shoulder stimulation in awake mice. In contrast to the previous studies in anesthetized animals, our measurements in awake mice are not affected by the confounding factors of anesthesia on the animal physiology, including the level of cerebral metabolism and the amplitude and speed of neuronal and vascular responses.
Methods: Microscopic imaging was performed through cranial windows in awake, head-restrained C57BLadult mice (5 months old). A custom-designed, multimodal imaging system was utilized to map the activation center and to measure hemodynamic responses in microvascular segments by combining two-photon microscopy of the cortical oxygenation and optical coherence tomography imaging of the cerebral blood flow. In order to perform fast spatio-temporally resolved measurements of pO2, we used a newly-developed oxygen probe PtG-2P, which has significantly higher brightness than the previously used probe PtP-C343. Functional activation was induced by directing the air puffs to a E1 whisker. The stimulation block consisted of a two-seconds-long stimulus at 3 Hz and a 25-second-long inter-stimulation interval. We used identical stimulation parameters to stimulate an animal shoulder.
Results: We mapped the amplitudes and shapes (e.g. initial dip, overshoot, and post stimulus undershoot) of the pO2 changes as a function of the vessel type (e.g., arterioles, capillaries, and venules), diameter, depth, branching order and a distance from the activation center.
Conclusions: Our measurements are the first to map the absolute pO2 transients in the cortical microvasculature in response to a functional activation in awake mice. Our results will help to better understand changes in oxygenation and blood flow on the cortical microvascular scale, will lead to improved understanding of the cerebral physiology, pathophysiology and will improve quantitative interpretation of fMRI signals.
BS03-4
Neurovascular coupling
Assessing the neural and the hemodynamic resting-state functional connectivity in case of neurovascular uncoupling
B. Li1, Q. Huang1, J. Lu1 and P. Li1
1Huazhong University of Science and Technology, Wuhan National Laboratory for Optoelectronics, Wuhan, China
Abstract
Resting-state functional magnetic resonance imaging is widely used to indirectly reflect the functional connectivity of neural activity based on neurovascular mechanisms. However, in case of neurovascular uncoupling, the feasibility of charactering functional connectivity of neural activity by spontaneous hemodynamic fluctuations has not been investigated clearly. The objective of the present study was to compare the hemodynamic resting-state functional connectivity (n = 8) with the neural resting-state functional connectivity (n = 8) under the condition of neurovascular uncoupling induced by cortical spreading depression in mice. The hemodynamic functional connectivity was visualized using optical intrinsic signal imaging and the neural functional connectivity was visualized using voltage-sensitive dyes imaging. Within one hour after cortical spreading depression, although reduced correlations between the bilateral cortex and the increased correlations between the unilateral cortexes were found both in the neural and the hemodynamic resting-state functional connectivity, the neural resting-state functional connectivity was altered to a lesser extent. The spatial correlation analysis revealed that the recovery of the neural resting-state functional connectivity after cortical spreading depression was much faster than that of the hemodynamic resting-state functional connectivity. Moreover, sensory-evoked interhemispheric activity motifs could be represented in the neural resting-state functional connectivity even after cortical spreading depression. These results suggest that under the condition of neurovascular uncoupling, hemodynamic resting-state functional connectivity showed the similar trends as the neural resting-state functional connectivity, but the neural resting-state functional connectivity was altered with a lesser extent and recovered much faster. This study may provide a reference for the experimental research and clinical diagnosis of diseases associated with neurovascular uncoupling by hemodynamic resting-state functional connectivity.
BS03-5
Neurovascular coupling
Optogenetic activation of noradrenergic neurons in the locus coeruleus modulates cortical neurovascular coupling responses
C. Lecrux1, Z. Yao2, E. Mitchell1, A.R. Lim1, A. Adamantidis3, A. Shmuel2 and E. Hamel1
1McGill University, Montréal Neurological Institute, Montréal, Canada
2McGill University, McConnell Brain Imaging Centre, Montréal, Canada
3University of Bern, Deptartment of Preclinical Research (DKF), Bern, Switzerland
Abstract
Neurovascular coupling (NVC) is a fundamental brain function that allows a precise match between activated brain region and increased blood supply. NVC is at the basis of functional brain imaging techniques that rely on hemodynamic signals to assess neuronal activity, and which are used in various brain states both in physiological and pathological conditions. However, NVC responses are poorly characterized under conditions that deviate from baseline physiology. The activity of noradrenergic (NA) neurons originating from the locus coeruleus (LC) is linked to changes in cortical activity underlying brain states, as seen during sleep-wake cycles, and may thus influence NVC responses to other cortical afferent pathways. Here, we investigated the effects of increased LC activity on cortical sensory-evoked NVC responses, using the well-characterized whisker-to-barrel pathway. Dopamine-β hydroxylase (DBH)-Cre mice injected with AAVdj-EF1α-DIO-CHETA-eYFP or AAVdj-EF1α-DIO-eYFP (control) in the LC were used to selectively depolarize NA neurons through optogenetic illumination (blue laser, 473 nm). Cortical maps of hemodynamic responses generated by optical imaging of intrinsic signals over the barrel cortex showed that optogenetic LC NA neurons activation (5–20 mW, 5 Hz, 10 sec) induced a consistent intensity-dependent decrease in the cerebral blood volume (CBV) map (570 nm reflectance), an effect not seen in control animals. Moreover, whisker-evoked increases in CBV were significantly decreased (−35%, p < 0.05, n = 8) by concurrent sub-threshold optogenetic LC stimulation (5 mW, 5 Hz, 10 sec), which alone had no significant effect on baseline CBV. Our results demonstrate a potent role for LC NA neurons as modulators of the NVC response to sensory stimulation in conditions that mimic cortical arousal, which may be important in pathologies where these neurons degenerate, such as Alzheimer's disease. Ongoing studies on the changes in the underlying neuronal activity and cellular network recruited in the barrel cortex should provide insight into the reliability of NVC under such conditions of altered brain state.
BS03-6
Neurovascular coupling
Optogenetic dissection of inhibitory neuron activity contributions to vascular regulation
A.L. Vazquez1, M. Fukuda1 and S.-G. Kim2
1University of Pittsburgh, Radiology, Pittsburgh, United States
2Sungkyunkwan University, Center for Neuroscience Imaging Research, IBS, Suwon, Korea, Republic of
Abstract
Introduction: Recent studies using optogenetic strategies have revealed that inhibitory neurons have a profound impact on blood flow regulation. Inhibitory neurons are very diverse, encompassing numerous sub-types. This work aims to determine the impact sub-populations of inhibitory neuron activity on vascular regulation using cre-driver mouse models to target the specific expression of channelrhodopsin-2 (ChR2).
Methods: Transgenic mice expressing Cre-recombinase in either somatostatin (SST), parvalbumin (PV) or vasointestinal peptide (VIP) neurons were injected with AAV virus (EF1a-DIO-ChR2-EYFP) in somatosensory cortex (forelimb or whisker areas) to drive the expression of ChR2 in these cells. Optogenetic stimuli (1mW@473 nm) were delivered by optic fiber. Measurements of changes in CBF and CBV were acquired by laser Doppler flowmetry (LDF) and optical imaging of intrinsic signal (OIS; 570 nm) under light-isoflurane anesthesia (0.5%) or awake conditions. Sensory activity (forelimb or whisker) was also used as a reference.
Results and Discussion: Two-photon microscopy was used to count the number of cells expressing YFP and their morphology. All mice showed robust expression with more than 40 YFP-positive cells identified for each mouse (Fig1). Photo-stimulation at 5 Hz for 4-sec of PV-CRE and VIP-CRE mouse models evoked negligible changes in CBF under lightly-anesthetisia or awake conditions (Fig2); however, a small increase in CBF might be present for the largest photo-stimulation used (30-ms, blue-lines). Photo-stimulation of SST-CRE mice evoked large changes in CBF that scaled with the photo-stimulus energy. Large rebound activity was recorded after photo-stimulation in these mice, especially under isoflurane, which produced latent and large CBF increases. Mice expressing ChR2 in most inhibitory neurons were also used and showed large increases in CBF. These increases were mostly suppressed with administration of a nitric-oxide-synthase inhibitor (L-NNA). The dependence of these results on the expression profile and type of nitric-oxide-synthase is being investigated.
BS04-1
Therapy & Neuroprotection
A novel preventive therapy for paclitaxel-induced cognitive deficits – preclinical evidence
W. Boehmerle1, P. Huehnchen1, A. Springer2, D. Freyer1 and M. Endres1
1Charité Universitätsmedizin Berlin, Klinik für Neurologie, Berlin, Germany
2FU-Berlin, Chemistry, Berlin, Germany
Abstract
Chemotherapy-induced neurotoxicity occurs frequently and presents an important unmet medical need. Patients often report a cognitive decline in temporal correlation to chemotherapy, particularly for hippocampus-dependent verbal and visuo-spatial abilities. To date, the underlying pathophysiology of post-chemotherapy cognitive impairment is poorly understood. We treated adult C57Bl/6 mice with 12x20 mg/kg paclitaxel, mimicking clinical conditions of dose-dense chemotherapy. In this model, measured paclitaxel concentrations in the hippocampus were 7-fold higher compared to the neocortex and mice developed distinct visuo-spatial learning and memory impairments. Histologic analysis revealed a reduced hippocampal cell proliferation and abnormal differentiation of resident stem and progenitor cells after paclitaxel therapy. In vitro, we observed severe toxicity in neural stem cells after brief exposure to nanomolar concentrations of paclitaxel. In comparison, mature postmitotic hippocampal neurons were less vulnerable. The molecular mechanism underlying paclitaxel induced apoptosis in stem cells involved interaction of neuronal calcium sensor 1 (NCS-1) with the inositol-1,4,5 trisphosphate-receptor, causing an activation of the calcium dependent protease calpain and subsequent caspase-mediated cell death. This apoptotic pathway could be specifically inhibited with lithium, which binds to NCS-1 and protected adult neural stem cells in vitro. In vivo, preemptive treatment of mice with lithium prevented paclitaxel-induced memory deficits and abnormal adult hippocampal neurogenesis.
In summary, we identified a molecular pathomechanism involved in paclitaxel-mediated toxicity against adult neural stem cells, which is independent of paclitaxel's cytostatic mode of action, and established a pharmacological intervention using lithium. These findings provide the framework for a translational clinical approach to prevent paclitaxel-induced neurotoxicity which could be beneficial for all patients receiving paclitaxel chemotherapy.
BS04-2
Therapy & Neuroprotection
Neuroprotective efficacy of poly-arginine-18 (R18) peptides using an in vivo model of perinatal hypoxic ischaemic encephalopathy (HIE)
A. Edwards1,2,3, K. Feindel4, J. Cross1,3,5, R. Anderton1,2, V. Clark1, N. Knuckey1,3,5 and B. Meloni1,3,5
1Western Australian Neuroscience Research Institute, Stroke Research, Nedlands, Australia
2University of Notre Dame, Health Sciences, Fremantle, Australia
3Sir Charles Gairdner Hospital, QEII Medical Centre, Department of Neurosurgery, Nedlands, Australia
4Centre for Microscopy, Characterisation and Analysis, Nedlands, Australia
5Centre for Neuromuscular and Neurological Disorders, The University of Western Australia, Nedlands, Australia
Abstract
Introduction: Excitotoxicity occurs in a number of clinical settings including perinatal hypoxia-ischaemia (HIE). HIE is the leading cause of mortality and morbidity in infants globally. Due to the lack of clinically available pharmacological neuroprotective treatments, there is an urgent need to develop new therapeutic agents to reduce neuronal injury after acute HIE. Currently, neuroprotective poly-arginine peptides provide a novel pharmacotherapuetic approach to mitigate cerebral infarct following HIE, due to their capacity to reduce the deleterious effects of calcium influx1,2.
Aims: 1) Characterise the distribution of cerebral blood flow (CBF) in a modified HIE model using magnetic resonance angiography (MRA) and phase contrast velocity encoding.
2) Investigate the neuroprotective efficacy of poly-arginine peptides (e.g. R18 and R18D) using the modified Rice-Vannucci HIE model.
3) Investigate the role of poly-arginine peptides in mitigating the deleterious cellular calcium influx following in vitro glutamic acid mediated excitotoxicity.
Methods: 1) MRA scans performed using Bruker BioSpec 9.4 T magnet.
2) HIE model: permanent ligation of the external and common carotid artery followed by transient exposure to hypoxia (2.5 h; 92% N2 / 8% O2); peptide delivered intraperitoneally shortly after HIE. In vitro excitotoxic model: cortical neuronal cultures exposed to glutamic acid and calcium indicator Fura-2 for assessment of calcium influx.
Results: 1) MRA detected CBF abnormalities (collateral and anastomotic blood flow); modification of the surgical procedure ameliorated abnormalities in CBF.
2) poly-arginine R18D and R18 peptides significantly by 30–33% (p < 0.05) following HIE.
3) poly-arginine peptides reduce sensorimotor deficits following HIE.
4) poly-arginine peptides reduce in vitro calcium influx by up to 80% following excitotoxicity.
Conclusion: Poly-arginine peptides R18 and R18D are potent neuroprotective agents, which have the capacity to reduce ischaemic infarct lesions and improve functional outcomes following HIE.
References:
1. Meloni et al., 2015. JCBFM. doi: 10.1338/jcbfm.2015.11.2. Meloni et al., 2015. Pharmacol Ther. doi: 10.1016/jpharmthera.2015.06.002.
BS04-3
Therapy & Neuroprotection
Co-administration of nanowired mesenchymal stem cells and cerebrolysin potentiates neuroprotection in Parkinson´s disease following mild traumatic brain injury
A. Sharma1, D.F. Muresanu2,3 and H.S. Sharma1
1Uppsala University Hospital, Surgical Sciences, Anesthesiology & Intensive Care Medicine, Uppsala, Sweden
2University of Medicine and Pharmacy, Clinical Neurosciences, Cluj-Napoca, Romania
3RoNeuro Institute for Neurological Research and Diagnostic, Neurology, Cluj-Napoca, Romania
Abstract
Traumatic brain injury (TBI) enhances chances of Parkinson's disease (PD) in Military personnel. Thus, there is an urgent need to explore the relationship between TBI and PD to find suitable therapeutic measures. Previous studies from our laboratory showed that 1-metyl-4-fenyl-1,2,3,6-tetrahydropyridin (MPTP, 20 mg/kg, i.p.) daily within 2-h intervals for 5 days induced PD like biochemical and behavioural symptoms on the 8th day. In present investigation we examined whether a mild TBI could further exacerbate PD symptoms in MPTP administered animals. Mild TBI was inflicted by dropping a weight of 28.6 g from 20 cm height on the right parietal cortex (impact 0.56 N) of mice under Equithesin anesthesia. In these animals MPTP was administered twice daily for 5 days. Uninjured MPTP treated mice served as controls. Blood-brain barrier (BBB) permeability, brain edema, brain pathology and regional cerebral blood flow (CBF) was examined using standard protocol. Our observations show that PD symptoms in TBI mice were significantly exacerbated on behavioral dysfunction e.g., Rota rod treadmill, gait analysis and walking deficits. In these TBI + PD mice BBB disruption to Evans blue albumin (EBA) and radioiodine was 3- to 5-fold higher, brain edema formation was 2- to 3-fold more and CBF reduction was 40 to 60% greater than PD controls. Neuronal, glial and myelin damages were 180 to 240% higher in TBI + PD group. Nanodelivery of cerebrolysin or mesenchymal stem cells (MSCs) alone reduced the pathophysiological and behavioral symptoms in PD alone but this effect was much less in TBI + PD group. However, a combination of nanowired cerebrolysin (2.5 ml/kg, i.v.) and MSCs (1 million cells) significantly induced neuroprotection in TBI + PD group. These observations are the firs to suggest that nanodelivery of cerebrolysin and MSCs had additive neuroprotective ability in PD not reported earlier.
BS04-4
Therapy & Neuroprotection
ACE2 activity is required to sustain serotonin levels and mediates exercise-induced adult neurogenesis
F. Klempin1,2, M. Bader1,2, R. Santos3 and N. Alenina1
1Max-Delbrück-Center for Molecular Medicine (MDC), Berlin, Germany
3Federal University of Minas Gerais (UFmG), Belo Horizonte, Brazil
Abstract
Objectives: Physical exercise induces the generation of new neurons in the adult mouse hippocampus. Serotonin (5-HT) and angiotensin (Ang) II are important mediators of the pro-mitotic effect of running. While serotonin activity in brain function is becoming understood, less focus has been given on potential peripheral signals that may cause running-induced neurogenesis. Here, we explore the effect of the acute running stimulus on proliferation in conditions of deregulated renin-angiotensin system (RAS).
Methods: Specifically, we took advantage of mice with genetic deletion of the principle regulating enzyme of the RAS, Ang-converting enzyme (ACE) 2, and measured proliferation and differentiation of dentate gyrus precursor cells. ACE2 metabolizes Ang II to Ang-(1–7) and is essential for the intestinal uptake of tryptophan (Trp), the 5-HT precursor.
Results: In ACE2-deficient mice we observed a decrease in brain 5-HT levels, and no increase in the number of BrdU-positive cells following exercise. However, experimentally increased Trp/5-HT levels or blockage of the Ang II AT1 receptors in ACE2-deficient mice could not rescue the effect. Furthermore, mice lacking the Ang-(1–7) receptor, Mas, presented a normal exercise response.
Conclusions: Our results identify ACE2 as a novel factor essential for serotonin metabolism and required for the exercise-dependent modulation of adult neurogenesis, acting independent of 5-HT, AT1, and Mas signaling. These experiments will impact our understanding of the cardiovascular network and its role in translating physical exercise with brain cell genesis.
BS04-5
Therapy & Neuroprotection
A novel strategy of neuroprotection tailored to improve ischemic stroke outcome in aged brains
W. Paschen1 and W. Yang1
1Duke University Medical Center, Anesthesiology, Durham, United States
Abstract
Objectives: Ischemia activates many pathological processes that have been targeted for neuroprotection in ischemic stroke. Such neuroprotection strategies are based on the assumption that blocking pathological processes triggered by ischemia will improve stroke outcome because brain's self-healing capacity will be sufficient to restore impaired cellular functions. However, brain's capacity to activate pro-survival pathways when challenged by ischemic stress dramatically declines with age.1 Importantly, pre-clinical neuroprotection studies are performed primarily in young rodents, and thus, it is conceivable that the success in these animals is limited in elderly stroke patients. To improve ischemic stroke outcome in aged brains therefore, we propose a novel strategy for neuroprotection that boosts a pro-survival pathway activated in the penumbra region in young brains but not in aged brains. To verify the feasibility of this approach, we have focused on O-linked β-N-acetylglucosamine protein modification (O-GlcNAcylation). O-GlcNAcylation is a pro-survival pathway, regulated by the inositol-requiring enzyme-1 branch of the unfolded protein response, and activated in post-ischemic brains of young but not aged mice.1
Method: To investigate the role of O-GlcNAcylation in stroke, 2-months-old (young) and 22-months-old (aged) mice were subjected to permanent middle cerebral artery occlusion (MCAO). O-GlcNAcylation was boosted by thiamet-G, a potent O-GlcNAcylase inhibitor, and evaluated by immunohistochemistry. Stroke outcome was assessed by TTC staining and neurologic deficits evaluation.
Results: O-GlcNAcylation was activated in neurons in the stroke penumbra in young but not in aged mice. O-GlcNAcylation was increased in brains of young and aged mice after thiamet-G treatment, and stroke outcome was significantly improved in aged mice when thiamet-G was given after MCAO.
Conclusions: Boosting a pro-survival pathway to counterbalance the compromised self-healing capacity of aged brains is a novel approach to neuroprotection that may eventually help to remove barriers in translational stroke research.
Reference:
1 J Cereb Blood Flow Metabol 2016;36:393–398.
BS04-6
Therapy & Neuroprotection
Soluble epoxide hydrolase inhibition decreases reperfusion injury after focal cerebral ischemia in rat
R. Tu1,2, J. Armstrong1, K.S.S. Lee3, B. Hammock3 and R. Koehler1
1Johns Hopkins University, Department of Anesthesiology and Critical Care Medicine, Baltimore, United States
2Second Xiangya Hospital, Central South University, Department of Neurology, Changsha, China
3University of California, Department of Entomology, Davis, United States
Abstract
Objectives: Epoxyeicosatrienoic acids (EETs) are produced by cytochrome P450 (CYPs) from arachidonic acid (AA), and its rapid metabolism is mainly through soluble epoxide hydrolase (sEH). EETs exert vasodilatory, anti-inflammatory, anti-apoptotic, and pro-angiogenic effects. Administration of sEH inhibitors before or at the onset of stroke is protective, but the effects of post-treatment at reperfusion when inflammation is augmented has not been as well studied. We tested the hypothesis that 1-trifluoromethoxyphenyl-3-(1-propionylpiperidin-4-yl) urea (TPPU), a potent and specific sEH inhibitor, will protect the brain when administered at reperfusion, possibly by suppressing inflammation and promoting VEGF release.
Methods: Male rats were subjected to 90-min middle cerebral artery occlusion (MCAO) by the intraluminal filament technique, followed by intraperitoneal injection of vehicle or 1 mg/kg TPPU at reperfusion and 24 h later.
Results: Protein expression and activity of sEH increased after reperfusion, and TPPU administration blocked the increase in sEH activity. At 48 h of reperfusion, infarct volume was decreased by approximately 50% (n = 12) and neurological deficits were significantly reduced with TPPU post-treatment. Likewise, performance was improved on the sticky tape test of forelimb sensory function and on the foot fault test (n = 20). TPPU significantly lowered the mRNA expression of IL-1beta by 3.5-fold and TNFalpha by 2.2-fold at 48 h, and elevated TGF-beta by 1.8-fold, whereas IL-10 mRNA was not increased (n = 8). Although trophic factor VEGF-A mRNA was unchanged, the peri-infarct percent of NeuN-positive cells with VEGF colocalization decreased while the percent of GFAP-positive cells with VEGF colocalization increased after reperfusion in the vehicle group. With TPPU treatment, the loss of colocalization with NeuN-positive cells was significantly attenuated and the increase in colocalization with GFAP-positive cells was further augmented (n = 5).
Conclusion: Inhibition of sEH by administration of TPPU at reperfusion can significantly reduce infarction and improve sensorimotor function, possibly by suppressing early pro-inflammatory cytokines, promoting reparative cytokines, and augmenting astrocyte-derived VEGF.
BS05-1
Blood brain barrier
GLP-1 analog raises glucose transport capacity of blood-brain barrier in Alzheimer's disease
M. Gejl1, B. Brock1, L. Egefjord2, J. Rungby3 and A. Gjedde4
1Aarhus University Hospital, Department of Clinical Biochemistry, Aarhus, Denmark
2Aarhus University Hospital, Department of Endocrinology, Aarhus, Denmark
3Bispebjerg University Hospital, Copenhagen, Department of Endocrinology, Copenhagen, Denmark
4University of Copenhagen, Department of Neuroscience and Pharmacology, Copenhagen, Denmark
Abstract
Objectives: Glucose enters the brain tissue from plasma by facilitated diffusion across the two membranes of the endothelium of the blood-brain barrier (BBB), mediated by the glucose transporter 1 (GLUT1). There is evidence in Alzheimer's disease (AD) of reduction of glucose transport across the blood-brain barrier, due to diminished GLUT1 translocation and expression at the BBB. Reduced BBB GLUT1 expression is known to aggravate AD pathology and further impair cognitive function, implying that GLUT1 may be a potential target of therapy directed towards AD neurovascular dysfunction and degeneration.
Hypothesis: The incretin hormone GLP-1 prevents the decline of the cerebral metabolic rate of glucose that signifies cognitive impairment, synaptic dysfunction, and disease evolution in AD, and GLP-1 may directly activate GLUT1 transport in brain capillary endothelium. For this reason, we here claim that the GLP-1 analog liraglutide may prevent the decline of blood-brain glucose transfer in AD.
Methods: In this 26-week test of the hypothesis, we randomized 38 patients with AD to treatment with the GLP-1 analog liraglutide (n = 18) or placebo (n = 20). We determined blood-brain glucose transport capacity (Tmax) with [18F]FDG (FDG) (ClinicalTrials.gov NCT01469351).
Results: In both groups, the Tmax estimates declined in proportion to the duration of AD. The GLP-1 analog treatment very significantly (P < 0.0001) raised the average Tmax estimate in cerebral cortex as a whole compared to the placebo treatment, from 0.72 to 1.1 mmol/hg/min.
Conclusion: The results are consistent with the claim that GLP-1 analog treatment raises GLUT1 activity in the BBB and hence may represent a therapeutic target for neurovascular dysfunction and degeneration in AD.
BS05-2
Blood brain barrier
Nanoagonist-mediated endothelial tight junction opening: a strategy increasing brain drug delivery efficiently and safely
X. Gao1 and C. Li1
1Fudan University, School of Pharmacy, Shanghai, China
Abstract
Objectives: Blood brain barrier (BBB) preciously regulates the movement of molecules and ions between the blood and the brain. However, its extremely low permeability prevents the efficient brain drug uptake. Although heterogeneous BBB breakdown is observed in multiple central nervous system diseases, the concentration of drugs, especially the macro-molecular drugs in the lesions are still far below the therapeutic threshold. Therefore, there is a tremendous need to increase brain drug uptake by tuning BBB permeability efficiently and safely. In this work, we developed a nanoagonist (NA) that temporarily increased BBB permeability by opening inter-endothelial tight junctions (TJs) via signaling adenosine A2A receptor expressed on brain capillary endothelial cells.
Methods: BBB permeability alternations in normal mice with intact BBB were evaluated by in vivo dynamic contrast enhanced magnetic resonance imaging (DCE-MRI). BBB restoration after NA treatment was monitored by in vivo gamma ray imaging. Structural variations of TJs were observed by high resolution transmission electronic microscopy. The spatial interactions between the neurovascular unit components including endothelial cells, basal membrane, pericytes and astrocyte endfeet were studied by immunofluorescence imaging.
Result: DCE-MRI demonstrated remarkable blood-brain barrier permeability enhancements and significantly increased brain uptakes of both small molecular and macromolecular paramagnetic agents after nanoagonist intervention. Gamma ray imaging and transmission electron microscope observed TJ opening followed by spontaneous recovery after nanoagonist administration. Immunofluorescence staining showed the unspoiled basal membrane, pericytes and astrocyte endfeet that tightly wrapped the vascular endothelium. Importantly, side-effects induced by inreversible TJ leakage such as edema and neuronal apoptosis were not observed after nanoagonist intervention.
Conclusion: Nanoagonist-mediated TJ opening provides a promising strategy to enhance brain drug delivery with high efficiency and minimized adverse-effects.
References:
Gao X, et al. Overcoming the blood-brain barrier for delivering drugs into the brain by using adenosine receptor nanoagonist. ACS Nano, 2014, 8(4): 3678–3689.
BS05-3
Blood brain barrier
Paracellular mechanisms may contribute to early blood-brain barrier leakage after cerebral ischemia and reperfusion
X. Jiang1,2, L. Zhang2, Y. Shi2, L. Zhu2, R. Leak3, R. Keep4, M. Bennett5 and J. Chen1,2
1Fudan University, State Key Laboratory of Medical Neurobiology and Institute of Brain Sciences, Shanghai, China
2University of Pittsburgh, Pittsburgh Institute of Brain Disorders & Recovery, Pittsburgh, United States
3Duquesne University, Division of Pharmaceutical Sciences, Mylan School of Pharmacy, Pittsburgh, United States
4University of Michigan, Department of Neurosurgery, Ann Arbor, United States
5Albert Einstein College of Medicine, Dominick P. Purpura Department of Neuroscience, Bronx, United States
Abstract
Objectives: The mechanisms underlying early-onset blood-brain barrier (BBB) hyperpermeability after ischemia/reperfusion (I/R) remain controversial. This study investigated the roles of the transcellular and paracellular pathways in early BBB leakage in both in vitro and in vivo models of I/R injury.
Methods: For in vitro injury, human brain microvessel endothelial cells (HBMECs) cultured to confluence in transwells were subjected to 60 min of oxygen-glucose deprivation (OGD). BBB integrity was assessed by transendothelial electrical resistance (TEER) and permeation of fluorescent dextrans. For in vivo injury, male C57BL/6J mice underwent 60-min transient middle cerebral artery occlusion with time-lapse two-photon microscopic imaging of cortical parenchymal microvessels through a skull window. BBB integrity was evaluated by extravasation of fluorescent tracers dextran-70 kDa and biocytin-tetramethylrhodamine-869Da (TMR-869Da). The role of caveolin-1-mediated transcytosis was investigated using caveolin-1−/− mice and lentivirus-mediated knockdown of caveolin-1 in cultured HBMECs.
Results: Progressive transendothelial leakage to small-sized dextrans (0.95–4.4 kDa) was observed in cultured HBMECs 1–3 h after OGD. Leakage was not blocked by matrix metalloproteinase inhibitors or by caveolin-1 shRNA knockdown (n = 4, p > 0.05), and was not accompanied by cell death. Furthermore, TEER was decreased by 25% after OGD (p ≤ 0.01), consistent with increased paracellular permeability. In vivo two-photon imaging revealed the extravasation of TMR-869Da into cerebral parenchyma as early as 1 h of post-ischemic reperfusion (5/5 wildtype mice) and this leakage was absent in sham-operated mice (n = 3). Despite inhibited endothelial transcytosis in caveolin-1 −/− mice [Reference], I/R-induced extravasation of TMR-869Da was not reduced (n = 4), suggesting paracellular permeability underlies the early passage of small molecules across the injured BBB. Dextran-70 kDa remained within the blood vessels at least during the first 2 h after I/R in both caveolin-1+/+ and caveolin-1−/− mice.
Conclusions: Paracellular rather than transcellular mechanisms contribute to early BBB disruption after brain I/R injury.
[BBB leakage after I/R]
Reference:
Neuron (2014) 82:603–17.
BS05-4
Blood brain barrier
Zinc contributes to acute cerebral ischemia-induced blood-brain barrier disruption
Z. Qi1,2 and K.J. Liu2
1Xuanwu Hospital of Capital Medical University, Cerebrovascular Diseases Research Institute, Beijing, China
2University of New Mexico, Department of Pharmaceutical Sciences, Albuquerque, United States
Abstract
Background: Zinc ions are stored in synaptic vesicles and cerebral ischemia triggers their release from the terminals of neurons. Zinc accumulation in neurons has been shown to play an important role in neuronal death following ischemia.
Objectives: In this study, we investigate whether zinc is involved in ischemia-induced blood-brain barrier (BBB) disruption.
Methods: We investigated the contribution of zinc to ischemia-induced acute BBB disruption and the possible molecular mechanisms using both cellular and animal models of cerebral ischemia.
Results: Zinc greatly increased BBB permeability and exacerbated the loss of tight junction proteins (Occludin and Claudin-5) in the endothelial monolayer under oxygen glucose deprivation conditions. In cerebral ischemic rats, a dramatically elevated level of zinc accumulation in microvessels themselves was observed in isolated microvessels and in situ, showing the direct interaction of zinc on ischemic microvessels. Treatment with a specific zinc chelator N,N,N′,N′-tetrakis(2-pyridylmethyl) ethylenediamine (TPEN), even at 60-min post ischemia onset, could greatly attenuate BBB permeability in the ischemic rats as measured by Evan's Blue extravasation, edema volume and magnetic resonance imaging. Furthermore, zinc accumulation in microvessels activated the superoxide/matrix metalloproteinase-9/-2 pathway, which leads to the loss of tight junction proteins (Occludin and Claudin-5) and death of endothelial cells in microvessels themselves.
Conclusions: Our findings reveal a novel mechanism of cerebral ischemia-induced BBB damage, and implicate zinc as an effective and viable new target for reducing acute BBB damage following ischemic stroke.
Zinc accumulates in isolated microvessles from ischemic areas and chelating zinc reduces BBB permeability in rats with 90-min ischemia/4-h reperfusion. A. Co-staining of selective zinc-specific indicators (NG) with microvessel marker (Tomato Lectin) and nuclear marker (DAPI) B. Evan's Blue extravasation (n = 5). C. Edema volume (n = 5). D. DCE-MRI represents BBB permeability (n = 4).
BS05-5
Blood brain barrier
Identification of two phosphorylation sites essential for Annexin A1 in blood-brain barrier protection after experimental intracerebral hemorrhage in rats
Z. Chen1, Z. Wang1 and G. Chen1
1The First Affiliated Hospital of Soochow University, Suzhou, China
Abstract
Objectives: Annexin A1 (ANXA1) has been reported to exert a blood-brain barrier (BBB) protection [1]. This study was designed to examine the role of ANXA1 in intracerebral hemorrhage (ICH)-induced BBB dysfunction.
Methods: A collagenase ICH model was performed in adult male Sprague Dawley rats.
Results: First, a possible relationship between ANXA1 and ICH pathology was confirmed by a loss of ANXA1 in the cerebrovascular endothelium and serum of ICH rats, and the rescue effects of i.v. administration of human recombinant ANXA1 (rhANXA1) in vivo and ANXA1 overexpression in vitro on the barrier function of brain microvascular endothelial cells (BMVECs) exposed to ICH stimulus. Second, we found that ICH significantly increased the phosphorylation ratio of ANXA1 at the serine/threonine residues. Finally, based on site-specific mutagenesis, we identified two phosphorylation sites (i) ANXA1 phosphorylation at threonine 24 is required for its interaction with actin cytoskeleton, and (ii) Phosphorylation at serine 27 is essential for ANXA1 secretion, both of which were essential for maintaining cytoskeleton integrity and paracellular permeability.
Conclusions: ANXA1 prevents ICH-induced BBB dysfunction in threonine 24 and serine 27 phosphorylation-dependent manners. ANXA1 phosphorylation may be a self-help strategy in BMVECs after ICH; however, that was almost completely abolished by the ICH-induced loss of ANXA1.
References:
[1] Cristante E, McArthur S, Mauro C, et al. Identification of an essential endogenous regulator of blood-brain barrier integrity, and its pathological and therapeutic implications. Proc. Natl. Acad. Sci. USA. 2013; 110: 832–41.
BS05-6
Blood brain barrier
New non-invasive and reproducible method of reversible opening of blood brain barrier for novel pharmacological strategy of treatment of central nervous system diseases
O. Semyachkina-Glushkovskaya1, A. Salmina2, E. Vodovozova3, A. Shirokov4, D. Bragin5, A. Gekaluyk6, M. Ulanova6, V. Fedorova6, E. Saranceva6, D. Zhu7, C. Zhang7, R. Shi7, V. Tuchin6, A. Morgun2, A. Alexeeva3 and J. Kurths7,8,9
1Saratov State University, Biology, Saratov, Russian Federation
2Krasnoyarsk State Medical University, Krasnoyarsk, Russian Federation
3Russian Academy of Sciences, M.M. Shemyakin–Yu.A. Ovchinnikov Institute of Bioorganic Chemistry, Moscow, Russian Federation
4Russian Academy of Sciences, Institute of Bioorganic Chemistry, Saratov, Russian Federation
5University of New Mexico, Albuquerque, Russian Federation
6Saratov State University, Saratov, Russian Federation
7Huazhong University of Science and Technology, Wuhan, China
8Potsdam Institute for Climate Impact Research, Potsdam, Germany
9Humboldt University, Berlin, Germany
Abstract
Here we created a new noninvasive, reproducible, easy performed method for reversible opening of BBB in mouse brain using the combination of audible sound with infrasound in music tracks including following algorithm during 2 h: 60 sec – 110 Db, 370 Hz and 60 sec – pause, then 60 sec 70 Db, 17 Hz and 60 sec pause).
Using confocal, two-photon microscopy and spectrofluorometric analysis we showed that sound exposure on mice during 2 h was associated with the increasing of BBB permeability to Evans blue and dextran as well as for potential drug transporters – fluorescent liposomes (100 nm) that located behind basal membrane. The intensity of opening of BBB developed with time: 30 min after sound effects – in 2,9% mice (7 of 237, weak and medium extravasation); 90 min – in 11,8% mice (28 of 237, weak and medium extravasation); 4 h and 24 h – in 83,1% mice (197 of 237, strong and diffuse extravasation). Other 17% of mice demonstrated individual resistance to sound. Histological analysis of brain showed the weak perivascular edema that developed independently on time of experiment in 100% of mice who demonstrated positive response to sound. We observed in parallel in these mice the increasing of expression of cloudin-5 in the cortical vessels. However, 48 h after experiment we did not find any changes in the BBB permeability and 72 h after sound we did not find any changes in the brain tissues, cerebral vessels and expression of claudin-5.
Thus, the combination of audible sound with infrasound due to his high reproducibility, easy performing, reversibility and high impact to further clinical application can be useful new method for the novel pharmacological strategy of treatment of central nervous diseases.
This work is supported by Grant of Russian Science Foundation No. (16–15-10252).
BS06-1
Plasticity & Recovery
Tissue plasminogen activator promotes axonal sprouting and functional recovery after ischemic stroke
Y. Shi1, S. Ma1, Z. Lu1, H. Pu1, X. Hu1, R. Stetler1, K. Hitchens2, R. Leak3, M. Bennett4 and J. Chen1
1University of Pittsburgh, Pittsburgh Institute of Brain Disorders & Recovery, Pittsburgh, United States
2University of Pittsburgh, Animal Imaging Center, Pittsburgh, United States
3Duquesne University, Division of Pharmaceutical Sciences, Mylan School of Pharmacy, Pittsburgh, United States
4Albert Einstein College of Medicine, Dominick P. Purpura Department of Neuroscience, Bronx, United States
Abstract
Objectives: Emerging evidence suggests that the protective role of tPA extends beyond its well-known thrombolytic activity. We demonstrate that tPA enhances axonal sprouting and long-term recovery after permanent brain ischemia.
Methods: Ischemia was induced in adult male tPA−/− and tPA+/+ mice by distal middle cerebral artery occlusion plus common carotid artery occlusion on the left side. STAIR guidelines were followed. Neurobehavioral assessments were performed up to 35d after ischemia. Gray and white matter (WM) injury was evaluated by immunohistochemistry and ex vivo diffusion tensor imaging (DTI). Axonal sprouting was examined at 35d after ischemia by injecting the contralateral cortex with biotinylated dextran amines (BDA). tPA was delivered intranasally (2 mg/kg) to tPA−/− mice 6 h after ischemia and then daily for 14d to assess its neurorestorative potential. Primary wild-type neurons (DIV3) cultured in microfluidic chambers were treated with tPA (5 ng/mL) for 96 h, and axonal outgrowth was examined with tau immunostaining.
Results: tPA−/− mice exhibited greater sensorimotor (rotarod and cylinder tests) and cognitive (Morris water maze) deficits after ischemia (n = 10/group, p ≤ 0.05 vs. tPA+/+ by ANOVA). tPA−/− mice developed comparable brain infarcts (p = 0.29) but greater WM deterioration, as shown by enhanced demyelination and decreased fractional anisotropy of the external capsule by DTI (n = 4/group, p ≤ 0.05 vs. tPA+/+; Pearson correlation vs. sensorimotor deficits, r > 0.7, p ≤ 0.05). Furthermore, intranasal administration of recombinant tPA to tPA−/− stroke mice ameliorated the sensorimotor and cognitive deficits, improved WM integrity, and increased the sprouting of BDA-labeled axons (n = 4, p ≤ 0.05 vs. vehicle) in the C7 spinal cord (corticospinal tract; Fig.A). Finally, in vitro studies confirmed that tPA enhanced axonal outgrowth of cultured neurons (Fig.B; n = 4).
Conclusions: Endogenous tPA facilitates long-term functional recovery in a non-thrombosis stroke model likely by promoting neuronal remodeling. Intranasal tPA may be a novel neurorestorative therapy for stroke recovery.
[tPA stroke]
BS06-2
Plasticity & Recovery
Experimental neurorehabilitation to study neurovascular remodeling after cerebral ischemia
M. Gabriel-Salazar1, A. Fernández1, A. Morancho1, J. Montaner1 and A. Rosell1
1Vall d'Hebrón Institute of Research (VHIR), Barcelona, Spain
Abstract
Objectives: Neurorehabilitation therapies are the only treatments approved after acute phase of stroke to improve the functional recovery in disabled stroke survivors. These strategies are supposed to participate in brain plasticity by modulating angiogenesis and neurogenesis among other processes. In this context, our aim is to establish a post-stroke neurorehabilitation (NR) model in mice to study the modulation of neurorepair mechanisms.
Methods: After inducing cerebral ischemia by middle cerebral artery occlusion (MCAo) in mice neurorehabilitation therapy was established as: sham (non-rehabilitation), Pasta Matrix (task-oriented approach) and Treadmill (forced physical exercise task). After randomization a total of 36 mice received 2 or 12 days of NR. Infarct size was measured at day 4 post-MCAo by TTC staining in a subgroup of mice together with brain MMPs levels (MMP2, MMP3, MMP8, MMP9 and MMP12) by a MILLIPLEX® Kit and Angiogenin expression. Additionally, after 12 days of NR vessel density was evaluated as lectin+ area after lectin perfusion and neurogenesis as DCX+ cells in the subventricular zone.
Results: At the beginning of rehabilitation (day 4 post-MCAO), infarct volume was similar between Non-Rehabilitation and Rehabilitation groups and brain protein levels of MMP2, MMP3, MMP8 and MMP9 increased in the ischemic cortex compared to corresponding contratalerals. Similarly, Angiogenin expression was higher in the infarcted vs. contralateral cortex, but no differences were found among rehabilitation groups. However, mRNA levels of Angiogenin correlated with infarct size. Finally, 12 days of NR treatment enhanced vessel density (lectin+ area) and migrating neuroblasts (DCX+ cells in the dorsolateral are of the ventricles) being especially higher in mice receiving the physical exercise approach with Treadmill.
Conclusions: Post-stroke Neurorehabilitation can be modeled in experimental cerebral ischemia by modulating cerebral blood vessel density and neurogenesis, thus serving as a strategy to study neurovascular remodeling and brain plasticity in the context of rehabilitation therapies.
BS06-3
Plasticity & Recovery
Elucidating the molecular mechanism behind the long-term cerebral ischemic tolerance mediated by resveratrol preconditioning
N. Khoury1, K. Koronowski1, I. Saul1, K. Dave1, J. Young1 and M. Perez-Pinzon1
1University of Miami Miller School of Medicine, Miami, United States
Abstract
Objectives: Our lab recently showed that resveratrol preconditioning (RPC) promotes tolerance that lasts for 2 weeks in vivo against cerebral ischemic insults. In this study, we aim to identify the transcriptomic and epigenetic modifications altered in the brain by RPC in order to reveal pathways and adaptations that mediate this tolerance.
Methods: We injected 8-week old C57Bl6 male mice with Vehicle/Resveratrol (10 mg/kg); two weeks later we performed an RNA-seq experiment on the mice's cortex. As an in vitro model of cerebral ischemia, we used primary neuronal cultures isolated from rat embryos and then exposed them to an oxygen and glucose deprivation (OGD). To assess cell death levels we used the lactate dehydrogenase (LDH) assay and propidium iodide (PI):DAPI staining. To measure cellular respiration we used the Seahorse Biosciences Technology.
Results: We identified 155 differentially expressed genes (DEG) among which interestingly 126 genes were downregulated by RPC (Fig1A). The DEG clustered into several biological processes including: gene expression, neurotransmitter secretion, ion-transport, and oxidative-phosphorylation. The downregulation in these processes was reminiscent of the phenomenon of metabolic depression, an adaptive mechanism used by hibernating animals to resist severe ischemic conditions. Using our in vitro model, we showed that RPC at 100uM provides tolerance for 6 days against an OGD. At this time-point we then assessed the culture's respiration rate. The glycolytic rate was not altered by RPC, while the maximal respiration rate was significantly reduced (121.55 ± 6.76 to 94.6 ± 7.62% of baseline, p = 0.02)(Fig1B,C).
Conclusions: In conclusion we showed that RPC at the long-term window induces a global downregulation in gene expression and a reduction in the metabolic activity. Ongoing studies will further test this conclusion and determine if this mechanism is behind the RPC-mediated protection.
Funding: AHA pre-doctoral Award #16PRE29170004; R01 NS34773-15
[Effect of RPC on gene expression and metabolism]
BS06-4
Plasticity & Recovery
Delayed inhibition of TRPM2 channels reverses impaired synaptic function after global cerebral ischemia
R. Dietz1, J. Orfila2, G. Deng2, N. Chalmers2, S. Tatum2, R. Traystman3 and P. Herson3
1University of Colorado, Denver, Pediatrics, Aurora, United States
2University of Colorado, Denver, Anesthesiology, Aurora, United States
3University of Colorado, Denver, Anesthesiology & Pharmacology, Aurora, United States
Abstract
Cardiac arrest (CA) is a significant cause of mortality and neurological morbidity. Learning and memory disorders observed following CA are readily explained by the selective vulnerability of hippocampal CA1 neurons. We investigated the role of the non-selective cation channel TRPM2 on ischemia-induced synaptic dysfunction. Adult male and female C57Bl/6 mice (20–25 g) underwent 8 min CA and resuscitation. Hippocampal CA1 function and synaptic plasticity were evaluated using acute brain slices 7 days following CA or sham controls. Synaptic plasticity was measured by long-term potentiation (LTP) of synaptic signals following theta-burst stimulation (TBS). Increase in field excitatory post-synaptic potential slope 60 min after TBS was analyzed to measure LTP. Slices or mice were treated with a novel inhibitor of TRPM2, (tatM2NX), as indicated below. Recordings obtained in slices from male mice 7 days after CA exhibited impairment of LTP compared to the same stimulation that stimulates robust LTP in sham control mice (161 ± 9%, n = 6 in sham; 105 ± 9%, n = 8 in CA). Bath application of tatM2NX (1 µM) reversed the CA-induced impairment of LTP, recovering to 149.8 ± 26% (n = 3; P < 0.05 compared to paired 7-day CA slices). In vivo administration of tatM2NX (20 mg/kg ip on day 6 after CA) reversed CA-induced impairments in LTP, recovering to 171 ± 11% (n = 6; p < 0.05). Similar data was observed in female mice. Memory function was measured using contextual fear conditioning. CA decreased freezing behavior, indicating lack of memory. This was reversed in CA mice given tatM2NX (20 mg/kg, single ip injection 24 hr before testing) on day 7 post-CA compared to vehicle-injected mice. These data indicate that following global cerebral ischemia, synaptic TRPM2 channels are chronically activated, contributing to long-lasting impairments of the remaining hippocampal network. Therefore, inhibition of TRPM2 channels at chronic time points following ischemia may represent a novel strategy to improve functional recovery after cerebral ischemia.
BS06-5
Plasticity & Recovery
Contralesional homotopic activity negatively influences functional recovery after stroke
A. Bauer1, A. Kraft2, G. Baxter1, M. Bruchas3, J.-M. Lee2 and J. Culver1
1Washington University in St. Louis, Radiology, Saint Louis, United States
2Washington University in St. Louis, Neurology, Saint Louis, United States
3Washington University in St. Louis, Anesthesiology, Saint Louis, United States
Abstract
Recent fcMRI studies examining spontaneous brain activity after stoke have revealed disrupted global patterns of functional connectivity (FC). Interestingly, acute interhemispheric homotopic FC has been shown to be predictive of recovery potential. While substantial indirect evidence also suggests that homotopic brain activity may directly impact recovery, results in humans are extremely varied. A better understanding of how activity within networks functionally-connected to lesioned tissue influences brain plasticity might improve therapeutic strategies. We combine cell-type specific optogenetic targeting with optical intrinsic signal (OIS) imaging to assess the effects of homotopic contralesional activity (specifically in excitatory CamKIIa pyramidal neurons) on FC, cortical remapping, and behavior after stroke. Thirty-one mice were housed in enriched cages for the experiment. OIS imaging was performed before, 1, and 4 weeks after photothrombosis of left forepaw somatosensory cortex (S1fp). On day 1 after stroke, 17 mice were subjected to chronic, intermittent optical stimulation of right S1fp for 10 min, 5 days/week for 4 weeks. New cortical representations of left S1fp appeared in non-stimulated mice at week 1, but not in stimulated mice (p = 0.005). Evoked responses were comparable in both groups at week 4 (p = 0.57). Homotopic FC between left and right S1fp regions was equally reduced in both groups (p = 0.012) at week 1. However, in non-stimulated mice, behavioral performance and FC between right S1fp and left perilesional S1 cortex was significantly higher by 4 weeks compared to stimulated mice (p = 0.009). Our results suggest that increased homotopic, contralesional activity in excitatory neurons negatively influences spontaneous recovery following ischemic stroke.
BS06-6
Plasticity & Recovery
Cofilin-actin rod formation in ischemic neurons and axons
R. Swanson1, A. Minnella1 and S.J. Won1
1University of California, San Francisco, United States
Abstract
The disability resulting from ischemic stroke is caused not only by neuronal death, but also by injury to neuronal axons and dendrites. Axons and dendrites have only limited regeneration potential, and often regenerate in aberrant patterns. The factors contributing to ischemic axonal and dendritic injury are not well understood. Cofilin-actin rods are linear aggregates of actin and cofilin that form preferentially in distal neuronal processes. Sustained formation of cofilin-actin rods disrupts local cell function and transport, and in neurodegenerative disorders their formation has been causally linked to neurite degeneration. Both ATP depletion and oxidative stress promote rod formation, suggesting a role for this process in ischemia-induced neurite injury as well. Here we demonstrate that cofilin-actin rods form in neuronal processes after ischemic conditions both in culture and in vivo. Neurons in culture form linear accumulations of rods, as identified by dephosphorylated cofilin, in response to oxygen-glucose deprivation. The rods form in both proximal and distal processes, and persist for many hours after return to normal conditions. Transient middle cerebral artery occlusion similarly produces a robust rod formation. This is most evident shortly after reperfusion but persists for up to 48 hours later. Permanent ischemia induced by photothrombotic occlusion of vessels in cerebral cortex also produced rods, though less extensively than in the perfusion model. However, many rods are in neurites clearly distant from the ischemic region. Studies ongoing aim to establish (a) the relationships between rod formation and neurite degeneration and (b) the effects of blocking rod formation.
BS07-1
Neurological diseases
Cerebrovascular consequences of raised intracranial pressure
M. Czosnyka1, J. Donnelly2, P. Smielewski2, A. Lavinio2 and P.J. Hutchinson2
1Univeristy of Cambridge, Department of Clinical Neurosciences, Cambridge, United Kingdom
2Univeristy of Cambridge, Cambridge, United Kingdom
Abstract
Objective: Intracranial pressure (ICP) is a reference pressure for brain blood flow. It is well documented that raised ICP (>20 mm Hg) after traumatic brain injury (TBI) increases risk of worse outcome, independent on low cerebral perfusion pressure or severity of primary injury.
Material/Methods: We used both direct and indirect methods for CBF and autoregulation monitoring in more than 1000 patients managed in intensive care after TBI. We compared various descriptors of cerebral hemodynamics in patients with elevated ICP (>20 mm Hg) and normal ICP (< 15 mm Hg) averaged for total monitoring time. Transcranial Doppler (TCD) blood flow velocity (BFV), cerebral autoregulation index based on TCD (Mx) and analysis of slow vasogenic waves in arterial pressure and ICP (PRx) were compared between groups. We also analysed diastolic BFV, pulsatility index and direct brain tissue oxygenation (PbtO2) between groups.
Results: Raised ICP is associated with worse cerebral autoregulation (P < 0.001) as assessed with both Mx and PRx indices. Mean arterial blood pressure was greater in intracranial hypertension group (p < 0.005), however this increase did not 100% compensate for a decrease in cerebral perfusion pressure (p < 0.01). Mean BFV was not different but diastolic BFV was significantly lower in intracranial hypertension. TCD pulsatility index wassignificantly greater in raised ICP group. Brain tissue oxygenation did not show significant differences between groups. Mortality in raised ICP group was almost three times greater than in normal ICP group (45% versus 17%)
Conclusion: Raised ICP affects cerebral autoregulation. Without autoregulation working properly, brain is exposed to ischaemic insults, whenever cerebral perfusion pressure falls. Early increase in arterial pressure seems to be a second protective mechanism preventing brain ischemia under depleted autoregulation. It is similar to Cushing response, but starts at relatively low ICP values.
BS07-2
Neurological diseases
A novel mouse surgical model for moyamoya
J. Roberts1, M. Maniskas1, G. Bix1 and J. Fraser1
1University of Kentucky, Lexington, United States
Abstract
Moyamoya disease is characterized by progressive stenosis and occlusion of the internal carotid arteries (ICA) and their major branches. The vascular stenosis is accompanied by the formation of an abnormal vascular network at the base of the brain and two major symptoms include ischemia and hemorrhage. Current treatments are limited to antiplatelet therapy and surgical bypass (direct or indirect). While research has recently discovered a susceptibility gene (RNF213) for idiopathic moyamoya disease, there is still no animal model which mimics the acquired vasculopathy of moyamoya syndrome. We report here for the first time, the use of microcoils placed on the ICA of mice as a model of moyamoya.
Male C57Bl/6 mice (3–8 months old) underwent surgery for the unilateral placement of a microcoil (0.16 mm) onto the proximal ICA. After 14 or 28 days, the blood vessels were examined for changes in size and number of anastomosis using an injection of DiI and brain tissue was examined for ischemia and hemorrhage using cresyl violet and Prussian blue stains.
Following successful placement of the microcoil on the ICA, changes in the diameters of various blood vessels at the circle of Willis were observed. We also report a decrease in the number of anastomosis and the formation of cortical “puffs” consistent with moyamoya. Histological examination shows changes in brain cellular morphology of some mice and indicates the presence of ischemic infarcts.
We report for the first time the development of an animal model with changes which mimic those observed in patients with moyamoya. While further characterization of this model is needed, it may provide a means for the future development of therapeutics for treatment of moyamoya.
BS07-3
Neurological diseases
The effect of diet-induced vitamin D deficiency on acute post-stroke outcome
M. Evans1, H.A. Kim1, T.M. De Silva1, G.R. Drummond1,2, G.R. Zosky3, B.R.S. Broughton1 and C.G. Sobey1
1Monash University, Cardiovascular Disease Program, Biomedicine Discovery Institute, Department of Pharmacology, Melbourne, Australia
2Monash University, Department of Surgery, Southern Clinical School, Melbourne, Australia
3University of Tasmania, School of Medicine, Faculty of Health Science, Hobart, Australia
Abstract
Objectives: Recent findings indicate that patients with low circulating levels of vitamin D experience larger infarct volumes and worse functional outcomes after ischemic stroke compared to those with sufficient levels1,2. However, it is unknown whether a causal relationship exists between low vitamin D levels and poor stroke outcome. This study therefore aimed to assess the effect of vitamin D deficiency on acute outcomes post-stroke.
Methods: Male C57Bl6 mice (6 w-old) were randomly assigned to either a control (n = 17) or vitamin D deficient diet (n = 18) for 4 w prior to stroke. Stroke was induced by middle cerebral artery occlusion for 1 h and was followed by 23 h reperfusion. Neurological deficit scores and hanging grip tests were used to assess functional outcome. Plasma hydroxyvitamin D3 levels were determined by ELISA, infarct volume was assessed by thionin staining and infiltrating neutrophils and T cells were quantified by myeloperoxidase and CD3 immunofluorescence, respectively. In addition, post-stroke infection was assessed from the lung bacterial load.
Results: ELISAs showed that the plasma level of hydroxyvitamin D3 was 85% lower in mice fed the vitamin D-deficient diet compared with the control group (5.2 ± 0.5 versus 29.6 ± 0.8 ng/ml). However, neurological deficit scores and hanging grip times were similar in both groups. Similarly, there were no differences between deficient and control animals in infarct volume (28.9 ± 5.3 versus 27.8 ± 4.6 mm3), numbers of infiltrating neutrophils (56 ± 17 versus 81 ± 15) and T cells (6 ± 1 versus 7 ± 2) or bacterial load within the lungs (1.5 ± 0.4 versus 1.8 ± 0.3 CFU/mg log10).
Conclusions: These data suggest that diet-induced vitamin D deficiency has no impact on acute post-stroke outcomes. However, further studies are required to assess the effect of vitamin D-deficiency on long-term functional outcomes after stroke.
References:
1. Turetsky et al. 2015. J Stroke Cerebrovasc Dis. 24(7): 1555–63.2. Park et al. 2015. Cerebrovasc Dis. 40(1–2): 73–80.
BS07-4
Neurological diseases
Global changes in white matter structure following focal photothrombotic stroke to the sensorimotor cortex
A.E. Meerwaldt1, G.A.F. van Tilborg1, C. van Heijningen1, A. van der Toorn1 and R.M. Dijkhuizen1
1University Medical Center Utrecht, Center for Image Sciences, Biomedical MR Imaging and Spectroscopy Group, Utrecht, Netherlands
Abstract
Objectives: Studies in animal models and patients have shown that loss and recovery of motor function after unilateral stroke are associated with changes in interhemispheric functional connectivity.1 It remains unknown, however, to what extent altered interhemispheric balance relates to structural changes in white matter tracts between the hemispheres. Therefore, in this study we assessed spatiotemporal changes in white matter structure during recovery from unilateral stroke to the sensorimotor cortex.
Methods: Photothrombotic stroke was induced in the sensorimotor cortex of adult male Sprague-Dawley rats. Serial MRI (9.4T) was conducted from 1 week before to 26 weeks after stroke (n = 18). During MRI, rats were anesthesized by mechanical ventilation with 2.0% isoflurane in air/O2. MRI consisted of anatomical imaging and diffusion kurtosis imaging (4-shot 2D-EPI with diffusion-weighting in 30 directions and 4 b-values (0, 590, 1471 and 2485 s/mm2)). Fractional anisotropy (FA) in white matter was statistically analysed voxelwise using FSL Tract-Based Spatial Statistics (TBSS) with family-wise-error-correction. FA values in corpus callosum were tested with factorial repeated measures ANOVA.
Results: TBSS analysis showed significantly lower FA values spread throughout the bilateral white matter from week 1 to week 26 after stroke (Figure). FA in ipsi- and contralateral corpus callosum was significantly lowered at all post-stroke time points (F(2.59, 43.90) = 10.32, p < .001). FA in ipsilateral corpus callosum was significantly lower than in the contralateral hemisphere (F(1.00, 17.00) = 26.62, p < .001).
[Statistical maps of significantly reduced FA.]
Conclusions: Our study shows that focal photothrombotic stroke to the ipsilateral sensorimotor cortex induces widespread changes in white matter structure, not only ipsilaterally, but also extending into the contralateral hemisphere. Globally affected white matter integrity may at least partly explain disturbed interhemispheric functional connectivity after focal unilateral stroke.
References:
1 Dijkhuizen RM et al., Curr Opin Neurol 2014;27:637–43.
BS07-5
Neurological diseases
Detection of radiologic and laboratory features of cerebral Amyloid angiopathy in patients with Alzheimer's disease
P. Fotiadis1, J. Becker2, K. Schwab1, J. Rosand1, A. Viswanathan1, R. Sperling2, K. Johnson2, S. Greenberg1 and E. Gurol1
1Massachusetts General Hospital, Neurology, Boston, United States
2Massachusetts General Hospital, Radiology, Charlestown, United States
Abstract
Background: Cerebral Amyloid Angiopathy (CAA) patients have posterior vascular amyloid deposition and lower Amyloid Beta (Aβ) levels when compared to Alzheimer's Disease (AD). We hypothesized that similar findings would be observed in AD patients with strictly lobar microbleeds (LMB) and/or cortical superficial siderosis (cSS) attributable to CAA [CAA-AD], when compared to AD patients with no hemorrhagic lesion(s) [NH-AD].
Methods: We reviewed MRIs of AD patients with T2*-, FLAIR and 3D T1-weighted MRI as well as Florbetapir PET, CSF Aβ-42 and tau levels, and APOE status within the Alzheimer´s Disease Neuroimaging Initiative (ADNI) database.
Results: The CAA-AD (n = 51) and NH-AD (n = 85) groups were balanced for age, gender, and history of hypertension (all p > 0.2). The APOE4 was more frequently present in the CAA-AD group (78% vs 60%, p = 0.038). Patients with CAA-AD had higher White Matter Hyperintensity volumes (0.73 vs 0.49% intracranial volume [ICV], p = 0.035) and higher occipital-to-global Florbetapir ratio (0.98 vs 0.94, p = 0.02) but similar mean cortical Florbetapir uptake (1.38 vs 1.36, p = 0.57), cortical thickness (2.22 vs 2.20 mm, p = 0.38), and hippocampal volume (0.37 vs 0.38% of ICV, p = 0.24) when compared to NH-AD. In multivariable regression models, higher occipital-to-global Florbetapir ratio (p = 0.009) and presence of APOE4 (p = 0.002) were associated with CAA-AD. In 117 patients with CSF data, CAA-AD (n = 46) had lower Aβ-42 (127 vs 140 pg/ml, p = 0.038) but similar tau levels (131 vs 136 pg/ml, p = 0.68) when compared to NH-AD. Lower Aβ-42 was also associated with CAA-AD (p = 0.024) in multivariable regression models.
Conclusions: Over one-third of AD patients displayed CAA-induced hemorrhagic lesion(s). When compared to NH-AD, they presented higher occipital-to-global Florbetapir uptake, suggesting vascular amyloid binding, and lower CSF Aβ-42 levels potentially related to amyloid sequestering in cortical vessel walls. These results support the possibility that advanced CAA commonly accompanies clinically diagnosed AD, contributing to dementia pathogenesis, and potentially affecting clinical treatment decisions.
On behalf of the Alzheimer´s Disease Neuroimaging Initiative, Los Angeles, United States
BS07-6
Neurological diseases
Graph theoretical heat kernel signatures of structural connectivity are altered in comatose cardiac arrest patients
M.D. Schirmer1, E.S. Rosenthal1, A.W. Chung2, G. Cudemus-Deseda1, B.M. Mills1, M. Villien3, B.L. Edlow1, J.T. Giacino4, J.L. Januzzi1, M.M. Ning1, W.T. Kimberly1, W.A. Copen1, P.W. Schaefer1, N.S. Rost1, D.M. Greer5 and O. Wu3
1MGH and Harvard Medical School, Boston, United States
2Boston Children's Hospital, Boston, United States
3MGH Martinos Center for Biomedical Imaging, Boston, United States
4Spaulding Rehabilitation Hospital and Harvard Medical School, Boston, United States
5Yale School of Medicine, New Haven, United States
Abstract
Objectives: This work aims to detect alterations in heat kernel (HK) signatures of structural connectivity in comatose cardiac arrest patients.
Background: Studying connectivity changes in the diseased human brain using graph theory is becoming a frequently used approach for differentiating healthy controls from patients. To measure alterations in information transport along all possible paths in brain networks of comatose cardiac arrest patients, we propose to characterize their heat diffusion properties using edge-based HK measures.
Methods: Cardiac arrest patients who were comatose after restoration of spontaneous circulation were enrolled in a prospective serial MRI study. Five healthy controls were also scanned. All subjects underwent 3 T MRI diffusion imaging (30 directions; b-values = 1000 and 2000 s/mm2; 10 b = 0 s/mm2 volumes). Network nodes were defined using the AAL template. Structural probabilistic connectivity maps were calculated and only edges with a probability above 10% were considered to reduce noise effects. We investigated edge-based measures, which aim to characterize informative time points in energy transfer within a network (time to maximum energy transfer, tm, and time where the rate of energy transfer stabilizes, tc, i.e. changes are below 2%). Groups were compared using one-way ANOVA and post-hoc Wilcoxon-method.
Results: Ten patients (43 ± 23 years old, 40% male) and five controls (37 ± 19 years old, 40% male) had structural networks estimated. Seven patients woke up by 12 days, three did not. Four were alive at discharge, one was brain dead and five died due to withdrawal of life sustaining treatment. Statistically significant differences, where longer times are associated with worse disorders of consciousness, were found between controls and non-wakers (Ptm = 0.037; Ptc = 0.037) and controls and wakers (Ptm = 0.035).
Conclusion: Using HKs, we identified alterations of structural connectivity in comatose cardiac arrest patients. Longer times may indicate an abnormal information transport, possibly due to global white matter injury.
BS08-1
Mechanisms of cerebral ischemia
Contralesional cortical atrophy and molecular cortical circuit reorganization after cerebral ischemia
F. Hellal1, B. Groschup1, S. Valero-Freitag1, A. Lourbopoulos1, U. Mamrak1, M. Dichgans1 and N. Plesnila1
1Institute for Stroke and Dementia Research, Klinikum der Universität München, Ludwig-Maximilians University (LMU), München, Germany
Abstract
Objectives: Stroke is the leading cause of death and the second leading cause of disability worldwide. The initial injury initiates remote changes further damaging the brain and contributing to persistence and development of clinical symptoms. Unilateral stroke induces interhemispheric misbalance of excitability with persistence of contralateral (healthy) cortex hyperexcitability, which may contribute to lack of functional recovery. However, the cellular and molecular underpinnings of this remote effect are not yet fully understood.
Methods: We sought to characterize the time course (24 hours, 7 days, 14 days, 1 month and 2 months post-stroke) of remote structural and molecular adaptations occurring in response to middle cerebral artery occlusion by confocal imaging and quantitative PCR. We used a murine stroke model in which transcallosal efferent fibers were labeled beforehand by Adeno-associated viruses expressing GFP.
Results: We found a significant thinning of the contralateral cortex associated with decreased spine density in apical dendrites of transcallosal neurons indicative of cortical circuit remodeling. Yet, study of cell death markers, total cell numbers and size of neuronal somas did not show any changes over 2 months. We found dynamic changes in GABAergic and glutamatergic transmission with acute downregulation of GABAA receptor subunits in the ipsilateral cortex. This effect was even more pronounced for receptors mediating phasic inhibition. We also found ipsilateral downregulation of Vglut1 accompanied by an upregulation of Gad65 throughout the progression of stroke. In contrast, the contralateral cortex displayed increased Vglut1 expression and decreased of Gad65 expression indicative of sustained interhemispheric misbalance.
Conclusions: Our data provide new insights in the long-term structural and molecular cortical reorganization after stroke and pinpoint the interhemispheric misbalance as new testable hypotheses how to stimulate recovery of function after long-term stroke.
BS08-2
Mechanisms of cerebral ischemia
Delayed treatment with omega-3 polyunsaturated fatty acids can still promote long-term neurovascular restoration, white matter integrity, and behavioral recovery after ischemic stroke
H. Pu1, X. Jiang1,2, Y. Shi1,3, Z. Wei1, Y. Gao1,2, R.K. Leak4, X. Hu1,2,3 and J. Chen1,2,3
1Pittsburgh Institute of Brain Disorders & Recovery, University of Pittsburgh, Neurology, Pittsburgh, United States
2State Key Laboratory of Medical Neurobiology and Institute of Brain Sciences, Fudan University, Neurobiology, Shanghai, China
3Veterans Affairs Pittsburgh Health Care System, Geriatric Research, Educational and Clinical Center, Pittsburgh, United States
4Mylan School of Pharmacy, Duquesne University, Division of Pharmaceutical Sciences, Pittsburgh, United States
Abstract
Objectives: Prophylactic dietary enrichment with omega-3 polyunsaturated fatty acids (n-3 PUFAs) is well known to improve long-term outcomes of ischemic stroke. The present study sought to investigate the therapeutic potential of delayed post-stroke administration of n-3 PUFAs and the underlying mechanisms.
Methods: STAIR guidelines were followed, including investigator blinding. Adult male C57BL/6 J mice were subjected to transient middle cerebral artery occlusion (tMCAO) for 1 h followed by random assignment to the following groups: 1) vehicle control; 2) delayed docosahexaenoic acid (DHA) injections (10 mg/kg, i.p. 2 h after tMCAO and then daily for 14 d); 3) delayed DHA injections and long-term fish oil dietary supplementation (DHA + EPA [eicosapentaenoic acid], 50 mg/g added to regular chew, beginning 5 d after tMCAO for up to 28 d). Post-stroke neurological deficits, gray and white matter integrity, and neurovascular restorative processes were assessed up to 28 d after tMCAO.
Results: Mice receiving delayed DHA injections showed only marginal amelioration of sensorimotor (rotarod and cylinder tests) and cognitive (water maze) deficits and brain atrophy up to 28d after tMCAO. Importantly, mice receiving combined DHA injections and n-3 PUFA dietary supplementation exhibited greater mitigation of brain atrophy (52.77% vs. vehicle, p ≤ 0.05), improved neurological functions (p ≤ 0.01 vs. vehicle by one or two-way ANOVA, n = 8), and less white matter demyelination (i.e. reduced loss of myelin basic protein). Combined DHA and fish oil treatment promoted post-stroke neurovascular restoration by enhancing neurogenesis and angiogenesis, which significantly correlated with improved cognitive recovery (Pearson r > 0.5, p ≤ 0.01, n = 7). Combined treatment also enhanced oligodendrogenesis and beneficial microglial M2 polarization, contributing to the improved white matter integrity and correlating with sensorimotor recovery (Pearson r > 0.6, p ≤ 0.001, n = 7).
Conclusions: These results demonstrate the efficacy of a delayed fish oil and DHA regimen in promoting long-term neurological recovery after ischemic stroke. Thus, delayed and long-term n-3 PUFA treatment may translate to effective clinical treatments for ischemic stroke.
BS08-3
Mechanisms of cerebral ischemia
DNA hydroxymethylation following focal ischemia
K. Morris-Blanco1 and R. Vemuganti1
1University of Wisconsin-Madison, Neurological Surgery, Madison, United States
Abstract
Objectives: Several epigenetic factors are known to play a role in the pathophysiology of ischemic brain injury. The recently discovered ten-eleven translocation (TET) dioxygenases convert methylated DNA to hydroxymethylated DNA, an epigenetic signature associated with active demethylation and increased gene expression. In the present study, we evaluated the role of TET activity and DNA hydroxymethylation in secondary brain damage following transient focal ischemia.
Methods: Adult C57BL/6 J male mice were subjected to transient focal ischemia by middle cerebral artery occlusion (MCAO). At different reperfusion time points, cortical penumbral tissue was obtained and used to quantify DNA hydroxymethylation by dot blot analysis and mRNA and protein expression of TET isoforms by real-time PCR and western blotting. TET3 isoform was knocked-down in the post-ischemic brain by intracerebral injection of TET3 siRNA. Infarct size was determined using TTC-stained serial brain sections. Profiles of autophagy, apoptosis, and DNA repair genes were analyzed with real-time PCR arrays.
Results: Following transient MCAO, DNA hydroxymethylation was significantly increased between 6 h to 24 h of reperfusion (by 3 fold; p < 0.05; n = 5/group) compared to sham. The mRNA and protein expression of the TET3 isoform, but not TET1 and TET2 isoforms, was also significantly elevated between 6 h to 24 h reperfusion. Knockdown of TET3 decreased post-ischemic DNA hydroxymethylation and increased the infarct size by 37% and mortality by 320% over the negative control (p < 0.05, n = 6/group).TET3 knockdown also resulted in increased expression of autophagy and apoptosis-related genes and decreased the expression of DNA repair genes after MCAO.
Conclusions: TET3 is the major regulator of DNA hydroxymethylation in the penumbra following transient focal ischemia. Furthermore, increased TET3-mediated DNA hydroxymethylation and the resulting regulation of gene expression might be an endogenous neuroprotective mechanism and hence a potential stroke therapeutic target.
BS08-4
Mechanisms of cerebral ischemia
The hypoxisome in neuronal resilience and cell death in stroke
T. Arumugam1 and D.-G. Jo2
1National University of Singapore, Department of Physiology, Yong Loo Lin School of Medicine, Singapore, Singapore
2Sungkyunkwan University, School of Pharmacy, Suwon, Korea, Republic of
Abstract
Ischemic stroke is a major cause of morbidity and mortality, with the outcome largely determined by the amount of hypoxia-related neuronal death in the affected brain regions. Cerebral ischemia and hypoxia activate three prominent signaling pathways (Notch1, NF-kB, P53, Hif-1a and Pin-1) that converge on a conserved DNA-associated nuclear multi-protein complex, the ‘hypoxisome', which controls the expression of genes that can determine the fate of neurons. When neurons experience a moderate level of hypoxia, the hypoxisome up-regulates adaptive stress response genes encoding proteins that promote neuronal survival, but when hypoxia is more severe the hypoxisome upregulates proteins that trigger and execute a neuronal death program. The hypoxisome is a molecular mediator of neuronal hormesis and a target for therapeutic intervention in stroke.
BS08-5
Mechanisms of cerebral ischemia
Inhibition of Mas receptor worsens stroke outcome in a comorbid rat model
E. Reid1, M. Arroja1, W. Holmes1, L. Work1, S. Nicklin1 and C. McCabe1
1University of Glasgow, Glasgow, United Kingdom
Abstract
Background: Hyperactivity of Angiotensin II is implicated in the development of vascular disease. Angiotensin-(1–7) (Ang-(1–7)), acting via the Mas receptor (MasR), counteracts the deleterious actions of Angiotensin II and improves stroke outcome in normotensive rats1. Our aim was to evaluate the therapeutic potential of Ang-(1–7) and examine the effects of MasR blockade post-stroke using a hypertensive rat model.
Methods: Transient middle cerebral artery occlusion (MCAO) was induced in male SHRSP (275–330 g). Rats were randomised to receive intracerebroventricular infusion (commencing on reperfusion and maintained for 7 days) of either: vehicle (aCSF), Ang-(1–7), MasR antagonist (A779) or Ang-(1–7) + A779, n = 10–11 per group. MCAO was confirmed by magnetic resonance angiography (MRA) and lesion volume assessed by diffusion-weighted MRI at 30 min post-MCAO, with reperfusion at 35 min. Day 7 infarct was evaluated by T2MRI and Iba1 labelling of post-mortem brain tissue identified microglia. Vascular effects of Ang-(1–7) were assessed in naïve SHRSP by laser speckle contrast imaging (LSCI) and by myography on isolated basilar arteries. Data presented as Mean ± S.D.
Results: A779 treated rats exhibited reduced tissue salvage following reperfusion, where the reduction in lesion volume from 30 min post-MCAO to day 7 was significantly lower compared to vehicle (37 ± 17% vs 58 ± 16% reduction). Ang-(1–7) treatment had a trend towards increased tissue salvage (71 ± 16% reduction, Figure1A, p < 0.05). Increased numbers of active/ramifying microglia were observed in A779 treated rats in the cortical peri-infarct region and the contralateral cortex (Figure1B). Systemic infusion of cumulative Ang-(1–7) did not induce any changes in cortical CBF, and had no effect on isolated arteries.
Conclusions: MasR activation does not appear to have any direct effect on the cerebrovasculature of SHRSP rats. However, inhibition of MasR worsens stroke outcome, likely due to enhanced pro-inflammatory microglial activation.
Reference:
Mecca AP et al. (2011). Exp Physiol 96:1084–96.
BS08-6
Mechanisms of cerebral ischemia
Neuronal Sirt1 mediates resveratrol preconditioning-induced ischemic tolerance by regulating glycolytic function
K. Koronowski1, I. Saul1, N. El Khoury1, Z. Balmuth-Loris1 and M. Perez-Pinzon1
1University of Miami, Neurology, Miami, United States
Abstract
Objectives: Our previous work demonstrates that resveratrol protects against cerebral ischemia when administered 2 or 14 days prior to injury1. Resveratrol activates Sirt1, an NAD+-dependent deacylase that regulates cellular metabolism. It has been postulated that neuronal Sirt1 directly mediates neuroprotection but it remains to be empirically tested. The objective of this study was to generate inducible, neuronal-specific Sirt1 knockout mice and determine whether neuronal Sirt1 is necessary for resveratrol-induced ischemic tolerance. Additionally, we aimed to identify the metabolic pathways downstream of Sirt1 in the brain.
Methods: Neuronal-specific Sirt1 knockout mice (Sirt1neu−/−) were generated by crossing SLICK-H and Sirt1flox/flox mice. Focal ischemia was evoked by 60 or 45 min middle cerebral artery occlusion (MCAo). Motor function was evaluated using the pole test. Untargeted metabolite profiles were generated by GC-TOF-MS. Transcription factors were screened using a DNA probe-based assay (Signosis).
Results: Tamoxifen treatment resulted in loss of WT Sirt1 protein from brain but not from heart (Figure 1A, n = 3). In WT, RPC reduced infarct volume by 43.7% at 24 hrs following 60 min MCAo, however this effect was lost in Sirt1neu−/− (Figure 1B, one-way ANOVA, n = 5–9). Additionally, RPC improved pole test turn time in WT 7 days following 45 min MCAo (Figure 1C, one-way ANOVA, sham n = 3–5, treatment n = 7–9). Compared to WT, metabolic profiles from Sirt1neu-/- displayed significantly altered glycolysis metabolites (Figure 1D, t-test, p < 0.05, FDR < 0.23, n = 8). Activation of the glycolysis regulator hypoxia inducible factor (HIF) was reduced by 48% in Sirt1neu−/− (Figure 1E, t-test, p = 0.08, n = 3).
Conclusions: We generated and utilized an inducible, neuronal-specific knockout mouse to demonstrate that neuronal Sirt1 is required for RPC-induced ischemic tolerance. Moreover, Sirt1 appears to regulate glycolysis in the brain possibly through HIF.
References:
1. Koronowski et al. Stroke. 2015.
BS09-1
Hemorrhage
Hemispheric and regional asymmetries of early cortical blood flow changes following subarachnoid haemorrhage in mice: Further support for acute vasoconstriction?
T. Hidayatov1,2, S. Hanalioglu1, A.I. Isikay1 and M. Mut Askun1
1Hacettepe University, Faculty of Medicine, Department of Neurosurgery, Ankara, Turkey
2MedEra Hospital, Baku, Azerbaijan
Abstract
Objective: Early brain injury following subarachnoid haemorrhage (SAH) is correlated with overall outcome. Microvascular dysfunction is thought to play a significant role although the exact pathophysiology has not been elucidated completely. In this study, we aimed to investigate spatial and temporal dynamics of early cortical blood flow (CBF) changes after subarachnoid hemorrhage with laser speckle contrast imaging (LSCI).
Methods: Ten Swiss albino mice (25–40 g) were anaesthetised with isoflurane. The cranium overlying both hemispheres was thinned. Subarachnoid hemorrhage was induced with endovascular filament perforation method. Continuous imaging of cortical CBF changes with LSCI was performed before and immediately after SAH induction for 90 minutes.
Results: Immediately after SAH, severe hypoperfusion (62.6 ± 5.1% reduction) was observed in the ipsilateral hemisphere. Ischemia was significantly milder but still remarkable in the contralateral hemisphere (32.8 ± 5.2%; p = 0.004). Within the ipsilateral hemisphere, regional differences were also noted (i.e. ACA territory less affected than MCA territory). A higher incidence of spontaneous spreading depolarizations/spreading ischemia (60% of animals) was detected in the ipsilateral compared to contralateral hemisphere (30% of animals). Hemodynamic responses to spontaneous and induced spreading depolarizations were different between hemispheres. Cerebrovascular reactivity to 5% CO2 was also diminished, albeit not significantly, in the ipsilateral vs contralateral hemisphere (18.7 ± 2.7% vs 31.2 ± 8.8%; p = 0.21).
Conclusion: This study shows that the induction of SAH causes abrupt global cerebral ischemia followed by prolonged oligemia with remarkable asymmetries between hemispheres and different regions. This finding could be better explained by differential spatiotemporal alteration in cerebral vasculature (acute macrovasospasm and/or microvascular disruption) rather than sustained decrease in cerebral perfusion pressure (CPP) per se.
BS09-2
Hemorrhage
Subarachnoid blood acutely induces spreading depolarizations and early cortical infarction in the human and swine brain
J. Hartings1, J. York1, J. Hinzman1, B. Krueger1, M. Winkler2, S. Major2,3,4, V. Horst2, P. Jahnke5, J. Woitzik6, E. Mahoney1, Y. Du7, M. Hagen8 and J. Dreier2,3,4
1University of Cincinnati, Neurosurgery, Cincinnati, United States
2Charité – University Medicine Berlin, Center for Stroke Research Berlin, Berlin, Germany
3Charité – University Medicine Berlin, Neurology, Berlin, Germany
4Charité – University Medicine Berlin, Experimental Neurology, Berlin, Germany
5Charité – University Medicine Berlin, Radiology, Berlin, Germany
6Charité – University Medicine Berlin, Neurosurgery, Berlin, Germany
7University of Cincinnati, Pharmacy, Cincinnati, United States
8University of Cincinnati, Pathology, Cincinnati, United States
Abstract
Early cortical infarction is common in poor-grade patients after aneurysmal subarachnoid hemorrhage, yet there are no animal models of these lesions and mechanisms are unknown. Here we investigated acute sequelae of subarachnoid hemorrhage in the gyrencephalic brain of anesthetized juvenile swine by multi-modal neuromonitoring and post-mortem studies. Subarachnoid infusion of 1–2 ml of fresh arterial blood over a cortical sulcus caused temporal clusters of spreading depolarizations (SDs) in 7/17 (41%) animals and sporadic SDs in 6/17 (35%) in a 6-hr monitoring period. SD clusters were significantly associated with thicker sulcal clots (P < 0.05) and a high likelihood of cortical infarcts (5/7 compared to 2/10 in swine without clusters, P < 0.06). In a second cohort, sulcal infusion of clotted blood produced thicker (median: 6.2 mm) clots, extensive infarction of circumscribed cortex, and SDs in 5/6 animals. The association of SDs with early brain injury was then investigated in 23 patients who underwent early magnetic resonance imaging and frontal electrocorticography after repair of ruptured anterior communicating artery aneurysms. Patients with any frontal lobe brain lesion and those with frontal infarcts only were both significantly more likely to have SDs [10/12 (83%) and 6/7 (86%), respectively) than those without brain lesions (1/11, 9%)(P's < 0.001 and <0.05, respectively). These results demonstrate an association between SDs and early cortical infarction after aneurysmal subarachnoid hemorrhage and establish a clinically relevant model to investigate causal sequences and potential therapeutic interventions.
1Houston Methodist Hospital, Neurosurgery, Houston, United States
Abstract
SAH results in long-term disabilities, the pathogenesis of which remains uncertain. Hindrance of CSF circulation presents possible mechanism interrupting drainage of subarachnoid space (SAS) of hematogenic damaging substances. CSF flow attenuation is partially reversible by administration of tPA suggesting role of fibrin deposition in paravascular route of CSF flow. We explored SAH-induced CSF flow abnormalities after SAH for up to 30 days and role of brain TF in hemorrhage control and CSF flow regulation.
Perforation of circle of Willis was used to trigger SAH in mice. Fluorophores injected into cisterna magna were used to trace CSF flow. Distribution of fluorophores and SAH were scored. Activity of brain TF was blocked by intracerebroventricular administration of specific antibodies. Distribution of TF and fibrin were analyzed using immunohistochemistry.
SAH interrupted CSF flow up to 30 days. Block of CSF flow did not correlate with the size of hemorrhage, and fibrin was observed on the brain surface including areas without blood. Immunohistochemistry demonstrated that TF is associated with astrocytes and glia limitans. Block of TF with antibodies one hour before the SAH increased size of hemorrhage by 50 ± 9%. In sham/naïve animals, block of TF one hour before the fluorophores injection facilitated their spread by 243 ± 21%.
While hemorrhage resolved in 7–10 days after the perforation, SAH-induced block of CSF flow lasted up to 30 days. Increased hemorrhage size after TF block suggests its role in localization of hemorrhage. TF also regulated CSF flow under normal conditions. Brain expressed high levels of TF associated with astrocytes and glia limitans. TF may play critical role in limiting hemorrhage following SAH and slowing of CSF flow. TF may regulate CSF flow in normal conditions. Targeting of the TF system will allow developing of new therapeutic approaches to the treatment of SAH and CSF flow pathologies such as hydrocephalus.
BS09-4
Hemorrhage
Osteopontin attenuates changes in cerebrovascular structure-function relations induced by subarachnoid hemorrhage in rat brain
W. Pearce1, K. Dahlim1, C. Doan1, D. Carreon1, E. Budbazar2, J. Tang2, A. Obenaus3 and J. Zhang2
1Loma Linda University, Center for Perinatal Biology, Loma Linda, United States
2Loma Linda University, Physiology, Loma Linda, United States
3Loma Linda University, Pediatrics, Loma Linda, United States
Abstract
Subarachnoid hemorrhage (SAH) is a common injury involving a complex and incompletely understood etiology. The present study explores the hypothesis that brain injury caused by SAH involves major changes in the structure and function of small cerebral arteries, in addition to its well known effects on neurons, glia and capillary endothelium. To evaluate this hypothesis, studied the effects of endovascular perforation of the bifurcation of the anterior and middle cerebral arteries (MCA) in the adult rat brain. Artery samples were harvested 24 hours post-injury for assessment of structure and function. To test if the 44 kDa extracellular matrix protein Osteopontin (OPN) could serve as a vasoprotectant, some animals were treated with OPN via intranasal administration (5 µg in 50 µl PBS) at 1 hour post-injury. As indicated by MLCK-MLC20 colocalization quantified via confocal microscopy, SAH caused a significant 32% increase in contractile differentiation. This effect of SAH, was attenuated by 66% (11% increase) following treatment with OPN. Simultaneously, SAH decreased the magnitude of the cytosolic calcium response to potassium depolarization by 28 nM (19%), and this effect was reversed and became an increase of 41 nM (49%) after OPN treatment. Additionally, SAH caused a passive dilation of 13 µM compared to untreated Sham control arteries, and after OPN treatment, this dilation was reduced to only 4 µM (69% reduction). Together, these results illustrate that SAH induces vascular injury that is largely prevented by intranasal treatment with OPN. Specifically, SAH promotes passive dilation and increased contractile differentiation, but compromises regulation of cytosolic calcium; these effects are consistent with the development of cerebral vasospasm and/or loss of cerebral autoregulation, both of which are well-known consequences of SAH. The present results strongly suggest that OPN is vasoprotective within the context of SAH, which may contribute significantly to its neuroprotective efficacy.
BS09-5
Hemorrhage
Sanguinate™ (PEGylated-carboxyhemoglobin-bovine) improves cerebral blood flow and oxygen extraction to vulnerable brain regions in patients at risk for delayed cerebral ischemia after subarachnoid hemorrhage
R. Dhar1, H. Misra2 and M. Diringer1
1Washington University in St. Louis School of Medicine, Neurology, Saint Louis, United States
2Prolong Pharmaceuticals, South Plainfield, United States
Abstract
Objectives: Sanguinate is a dual-action oxygen transfer and carbon monoxide(CO)-releasing agent with efficacy in animal models of focal brain ischemia and established safety in healthy volunteers. We performed a dose-escalation study in subarachnoid hemorrhage (SAH) patients at risk for delayed cerebral ischemia (DCI) to evaluate tolerability and explore efficacy of CO in improving cerebral blood flow (CBF) and oxygen transfer in normalizing flow-metabolism balance to vulnerable brain regions.
Methods: 12 subjects were studied over three doses: 160 mg/kg, 240 mg/kg, 320 mg/kg, with safety review prior to proceeding to higher doses. After baseline 15O-PET measurement of global and regional CBF and oxygen extraction fraction (OEF), Sanguinate was infused over two hours; PET was repeated immediately after and again at 24-hours. Vulnerable brain regions were defined as those with baseline OEF ≥ 0.5.
Results: Sanguinate infusion resulted in a significant but transient rise in mean arterial pressure (115 ± 15 to 127 ± 13 mm Hg) that was not dose-dependent. No adverse physiologic or clinical effects were observed with any dose. Global CBF did not rise significantly after Sanguinate (42.8 ± 7 to 46.1 ± 9 ml/100 g/min, p = 0.1). However, in the 28% of regions classified as vulnerable, Sanguinate resulted in a significant rise in CBF (42.2 ± 11 to 51.2 ± 18 ml/100 g/min) associated with a reduction in OEF (0.6 ± 0.1 to 0.5 ± 0.11, both p < 0.001). The increase in regional CBF was only seen with the two higher doses but OEF improved in all tiers. However, response was no longer seen at 24-hours.
Conclusions: We safely administered a novel oxygen transport and vasodilating agent to a cohort of patients with SAH. Sanguinate infusion appeared to improve CBF and flow-metabolism balance in vulnerable brain regions and warrants further study in those at-risk for DCI. Higher or repeat dosing may be required for sustained efficacy.
BS09-6
Hemorrhage
S-ketamine reduces the incidence and modifies the characteristics of spreading depolarization in patients with aneurysmal subarachnoid hemorrhage and in gyrencephalic swine models
E. Santos1, A. Olivares Rivera1, S. Major2, R. Sánchez-Porras1, M. Kentar1, A.W. Unterberg1, O.W. Sakowitz1,3 and J.P. Dreier2
1University of Heidelberg, Neurosurgery, Heidelberg, Germany
2Charité – Universitätsmedizin Berlin, Center for Stroke Research Berlin, Berlin, Germany
3Klinikum Ludwigsburg, Department of Neurosurgery, Ludwigsburg, Germany
Abstract
Electrical and hemodynamic characteristics of spreading depolarizations (SDs) can be affected by the N-methyl-D-aspartate receptor antagonist ketamine, which has been shown to produce neuroprotection in animals with lissencephalic brains. Here, we investigated the effect of S-ketamine on SD characteristics in two porcine models and in patients with aneurysmal subrachanoid hemorrhage (SAH) in whom multimodal neuromonitoring was performed. In both models, SDs were monitored using electrocorticography (ECoG) and large field-of-view movement-compensated intrinsic optical signal (IOS) imaging, which enabled us to perform a long-term analysis of blood volume signals in regions of interest.
In the first animal model (n = 15), SDs were induced with drops of 1 M KCl in both hemispheres at 1 h-intervals during 18 h (Group 1: control, Group 2: ketamine 2 mg/kg/h, Group 3: 4 mg/kg/h, each n = 5). In a second model, the left middle cerebral arteries were transorbitally occluded by surgical ligation (MCAO) and subsequent SDs were observed. Animals were randomly assigned to either receive 5 mg/kg/h ketamine or not (Groups 4 and 5, each n = 5) and were monitored over 30 h. Sixty-seven SAH patients were prospectively monitored, including ECoG (mean 11 days). We retrospectively compared relevant variables of patients who received ketamine (n = 31) vs. no-ketamine. Ketamine reduced the incidence of SDs in both porcine models (reduction to 35.7% in the KCl and to 68.7% in the MCAo model using ketamine 4 mg/kg/h). We also found significant changes in the electrical and hemodynamic characteristics (amplitude, duration, expansion, etc.). Following MCAO, ketamine was less effective in reducing the SD incidence, but it decreased the hypoemic and increased the hyperemic components. In patients, a mean of 2.7 mg/kg/h S-ketamine reduced the SD incidence from 1.87 to 0.68 SDs/day (reduction to 36.6%) (Wilcoxon p < 0.001). Doses above the recommended therapeutic range (>2 mg/kg/h) were more effective. This supports the need of more experimental data before proceeding to a neuroprotective clinical trail.
BS10-1
Cerebral ischemia: clinical
Genetics of white matter hyperintensity burden in patients with ischemic stroke: the MRI-GENIE study
A.-K. Giese1, H. Xu2, K. Ryan2, M.D. Schirmer1, A.V. Dalca3, T. Dave2, J.W. Cole4, P.F. McArdle2, J.P. Broderick5, J. Jimenez-Conde6, C. Jern7, B.M. Kissela5, D.O. Kleindorfer5, R. Lemmens8, A. Lindgren9, J.F. Meschia10, T. Rundek11, R.L. Sacco11, R. Schmidt12, P. Sharma13, A. Slowik14, V. Thijs15, D. Woo5, B.B. Worrall16, O. Wu1, S.J. Kittner4, P. Golland3, J. Rosand1, B.D. Mitchell2 and N.S. Rost1
1Massachusetts General Hospital, Harvard Medical School, Boston, United States
2University of Maryland, School of Medicine, Baltimore, United States
3Computer Science and Artificial Intelligence Lab, MIT, Cambridge, United States
4University of Maryland, School of Medicine and Veterans Affairs, Maryland Health Care System, Baltimore, United States
5University of Cincinnati College of Medicine, Cincinnati, United States
6Universitat Autonoma de Barcelona, Barcelona, Spain
7Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden
8KU Leuven – University of Leuven, Leuven, Belgium
9Lund University, Lund, Sweden
10Mayo Clinic Jacksonville, Jacksonville, United States
11Miller School of Medicine, Miami, United States
12Medical University Graz, Graz, Austria
13Royal Holloway University of London, Egham, United Kingdom
14Jagiellonian University Medical College, Krakow, Poland
15Florey Institute of Neuroscience and Mental Health, Parkville, Australia
16University of Virginia, Charlottesville, United States
Abstract
Introduction: The MRI-Genetics Interface Exploration (MRI-GENIE) study is the first international collaboration that aims to facilitate genetic discoveries in clinical cohorts of patients with acute ischemic stroke (AIS). We have amassed the largest-to-date collection of AIS cases with brain MRI scans and genome-wide genotyping to test the role of genetic susceptibility in MRI-based cerebrovascular traits.
Objective/Hypothesis: To elucidate the genetic architecture of white matter hyperintensity (WMH) burden in AIS patients.
Methods: Using a novel automated algorithm, we extracted WMH volume (WMHv) from clinical MRI scans of 2704 AIS patients (age 63.1 ± 14.7 years, 60.6% male) of European ancestry. Quality control (QC) measures were undertaken per subject and per SNP, excluding subjects with non-European ancestry and poor genotyping, as well as SNPs deviating from Hardy-Weinberg equilibrium and high levels of missingness. Imputation to the Haplotype Reference Consortium (HRC version r1.1) was conducted for 1712 remaining subjects with 2.8 million SNPs on the Michigan Imputation Server. After exclusion of poorly imputed SNPs (R2 < 0.5) and SNPs with minor allele frequency <1%, 7.7 million SNPs remained for further analysis. Genome-wide association testing of natural log-transformed WMHv on the allelic dosage per SNP was adjusted for age, sex and principal components 1–10.
Results: Genome-wide association testing has identified a novel locus on chromosome 2 (T allele at rs72856504) near the LDL Receptor related Protein 1B gene (LRP1B) that was significantly associated with WMHv burden in AIS (β = 0.54, SE = 0.098, p = 3.65*10−8).
Conclusion: We have identified a novel locus (T allele rs72856504) on chromosome 2 near the LRP1B gene, which is specific for WMH in AIS and has not been previosuly described in stroke-free WMH cohorts. A replication effort involving additional independent cohorts of AIS patients with brain MRI and genome-wide genotyping is ongoing.
BS10-2
Cerebral ischemia: clinical
Variable ischemic stress of white matter hyperintensities in stroke patients with large vessel occlusion
E. Nemoto1, S. Zaidi2, K. Uchino3, R. Lin4, H. Kuwabara5, N. Bircher6, D. Sashin7, Y.-F. Chang8, M. Hammer9, V. Reddy9, T. Jovin9, N. Vora10, M. Jumaa11, L. Massaro9, J. Billigen8, F. Borda12, K. Matsubara13 and H. Yonas1
1University of New Mexico, Neurosurgery, Albuquerque, United States
2Promedia Toledo Hospital, Vascular Neurology, Toledo, United States
3Cleveland Clinic, Cerebrovascular Center, Cleveland, United States
4Holy Cross Hospital, Neurology, Pompano Beach, United States
5Johns Hopkins University School of Medicine, Radiology, Baltimore, United States
6University of Pittsburgh School of Medicine, Critical Care Medicine, Pittsburgh, United States
7University of Pittsburgh School of Medicine, Radiology, Pittsburgh, United States
8University of Pittsburgh School of Medicine, Neurosurgery, Pittsburgh, United States
9University of Pittsburgh School of Medicine, Neurology, Pittsburgh, United States
10Riverside Radiology and Interventional Associates, Marion, United States
11Promedica Toledo Hospital, Neurology, Toledo, United States
12New York University, Radiology, New York City, United States
13Research Institute for Brain and Blood Vessels, Akita, Japan
Abstract
Background: In a study evaluating hemodynamic compromise by quantitative positron emission tomography (PET) measurements of oxygen extraction fraction (OEF) and cerebrovascular reserve (CVR) in stroke patients with large vessel occlusion, we report for the first time, that white matter and white matter hyperintensities (WMH) despite an apparent homogeneity vary in ischemic stress. WMH is increasingly recognized as an important factor in cerebrovascular ischemia and dementia ranging from mild cognitive impairment, vascular dementia and Alzheimer's disease and with cardiovascular risk factors (hypertension, diabetes, obesity and dyslipidemia, i.e. metabolic syndrome) that can increase the risk for stroke to 40%.
Methods: We studied seventeen patients who within the past six months, suffered large vessel occlusion (LVO) strokes and with occlusive vascular disease and sixteen controls by 15O2 positron emission tomography (PET) measurements of cerebral blood flow (CBF) with H215O and cerebral metabolic rate for oxygen (CMRO2) with 15O2 gas and derived variables of OEF, OEF reactivity (OEFR) and CVR before and after acetazolamide vasodilatory challenge. Hemispheric WMH volumes were quantitated by manual segmentation of fluid-attenuated inversion recovery (FLAIR) magnetic resonance images (MRI).
Results: Hemispheric WMH volumes were significantly (P < 0.05) higher in stroke patients compared to controls. In stroke patients with occlusive vascular disease, cerebrovascular Reactivity (CVR) was linearly and directly related to OEFR and CBF and CMRO2.
Conclusions: WMH differ in degree of ischemic stress and are prevalent in patients with strokes and occlusive vascular disease exaggerating hemodynamic compromise of WMH.
BS10-3
Cerebral ischemia: clinical
Biological age is a better predictor of 3-months outcome than chronological age in ischemic stroke patients
C. Soriano-Tarraga1, E. Giralt-Steinhauer1, M. Mola-Caminal1, A. Rodríguez-Campello1, R.M. Vivanco-Hidalgo1, E. Cuadrado-Godia1, Á. Ois1, I. Fernández-Cadenas2, J. Jimenez-Conde1 and J. Roquer1
1Institut Hospital del Mar d’Investigacions Mèdiques (IMIM), Neurovascular Group, Barcelona, Spain
2Fundació Docència i Recerca MutuaTerrassa, Hospital Mútua de Terrassa, Stroke Pharmacogenomics and Genetics, Terrassa, Spain
Abstract
Background: Stroke can have great impact on functional status of patients, although there are substantial interindividual differences in recovery capacity. Apart from stroke severity, age is an important outcome predictor after stroke, but aging is not only due to chronological age. There are age-related DNA-methylation changes in multiple CpG sites across the genome that can be used to estimate the biological age (b-Age) that is influenced by lifestyle, environmental factors and genetic variation. We seek to analyze the impact of b-Age in outcome at 3-months after an ischemic stroke.
Methods: We included 511 first-ever acute ischemic stroke patients assessed in Hospital del Mar (Barcelona). Demographic, clinical data and mRs at 3 months were registered. Biological age (b-Age) was estimated with Hannum algorithm, based on DNA methylation in 71 CpGs. IS outcome at 3 months was dichotomized as good (mRs score ≤ 2) or poor (mRs ≥ 3) outcome and they were compared by logistic multivariate analyses.
Results: In logistic multivariate regression of IS outcome at 3 months, b-Age was to significance (p = 0.047) nullifying c-Age (p = 0.396) and stepwise regression kept b-Age (p < 0.001; OR = 1.05 (95% CI 1.03–1.08), initial NIHSS (p < 0.001; OR = 1.27 (95% CI 1.21–1.33)) and rtPA (p < 0.007; OR = 0.37 (95% CI 0.18–0.76)) as explanatory variables, instead of c-Age. This result was validated in an independent IS cohort.
Conclusions: Biological Age, estimated by DNA methylation, is an independent predictor of stroke outcome better than chronological age. “Healthy aging” would ameliorate the outcome after an ischemic stroke.
BS10-4
Cerebral ischemia: clinical
Loci associated with neurological instability after acute ischemic stroke: A multi-ethnic genome-wide association study
L. Ibanez1, L. Heitsch2, C. Carrera3, J. Pera4, A. Slowik4, D. Strbian5, I. Fernandez-Cadenas3, J. Montaner3, C. Cruchaga1 and J.-M. Lee6
1Washington University in St. Louis School of Medicine, Psychiatry, St. Louis, United States
2Washington University in St. Louis School of Medicine, Emergency Medicine, St. Louis, United States
3Vall d'Hebrón Institute of Research (VHIR), Neurology, Barcelona, Spain
5Helsinki University, Neurology, Helsinki, Finland
6Washington University in St. Louis School of Medicine, Neurology, St. Louis, United States
Abstract
Introduction: Following acute ischemic stroke (AIS) neurological deficits can rapidly improve or deteriorate. Mechanisms such as hemorrhagic transformation, reperfusion, or endogenous neuroprotection contribute to this early neurological instability, but little is known about genetic influences.
Methods: AIS patients were prospectively enrolled between 2008 and 2016 at four sites (St. Louis, Barcelona, Helsinki and Krakow). NIHSS scores were obtained within 6 hours and again at 24 hours after stroke onset, to generate an endophenotype quantifying neurological change in the first 24 hours, ΔNIHSS24h = NIHSS6h minus NIHSS24h (negative values represent deterioration; positive values, improvement). Genotyping was generated for common variants, imputing up to 6 million SNPs. The association model used was: ΔNIHSS24h = NIHSS6h + age + gender + glucose + PCA1 – PCA4. Samples were analyzed separately by site/ethnicity; then, the multi-ethnic analysis MANTRA was used to perform a joint analysis correcting for population admixture.
Results: A total of 2,614 patients showed a median NIHSS6h of 8 (IQR 12) and a median ΔNIHSS24h of 2 (IQR 5). The estimated heritability of neurological instability was 14.6%. We identified 2 novel loci that influenced ΔNIHSS24h with genome-wide significance (Figure): rs35116504 (LBF = 5.523) located on chromosome 2 in an intron of FBLN7; and rs6517243 (LBF = 7.568) on chromosome 21 in an intron of RCAN1. FBLN7 encodes fibulin-7, a matrix glycoprotein that is expressed by endothelial cells and regulates angiogenesis. RCAN1 encodes regulator of calcineurin 1 and has been implicated in Alzheimer's disease, Down's Syndrome, and atherosclerosis.
Conclusion: Common genetic variants have a moderate influence on early neurological instability after ischemic stroke. We identified two novel genetic loci that implicate possible vascular mechanisms in early outcomes after AIS. Replication in an independent cohort is ongoing.
[MANTRA]
BS10-5
Cerebral ischemia: clinical
Baseline haemodynamic and optical properties of the newborn brain and the reproducibility of the measurements: a preliminary report from the BabyLux project
M. Giovannella1, B. Andresen2, A. de Carli3, V. Chamizo4, M. Pagliazzi1, M. Rehberger5, D. Contini6, A. Pifferi6,7, L. Spinelli7, R. Donat8, R. Erdmann9, M. Fumagalli3, G. Greisen2, U.M. Weigel4, T. Durduran1,10 and A. Torricelli6
1ICFO – The Institute of Photonic Sciences, Castelldefels, Spain
2Rigshospitalet, Neonatology, Copenhagen, Denmark
3Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Università degli Studi di Milano, Milano, Italy
4Hemophotonics S.L., Castelldefels, Spain
5Fraunhofer Institute for Production Technology IPT, Aachen, Germany
6Politenico di Milano, Dipartimento di Fisica, Milano, Italy
7Istituto di Fotonica e Nanotecnologie, Consiglio Nazionale delle Ricerche, Milano, Italy
8Loop-Competitive Design Network, Sant Cugat del Vallès, Spain
9PicoQuant GmbH, Berlin, Germany
10Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain
Abstract
We present absolute hemodynamic parameters (blood oxygen saturation, StO2; total hemoglobin concentration, tHb; cerebral blood flow index, BFI) and optical properties (absorption coefficient, ma; reduced scattering coefficient, ms') of newborn term babies performed with a prototype hybrid diffuse correlation spectroscopy (DCS) and time resolved near infrared spectroscopy (TR-NIRS) device, developed within the project BabyLux – An optical neuro-monitor of cerebral oxygen metabolism and blood flow for neonatology (http://www.babylux-project.eu).
In particular, we show baseline values for optical and hemodynamic parameters as well as their variability over multiple repositioning measurements. The main goal is to demonstrate a low variability over measurement with this device, aiming at less than 5% for StO2, being variability over repositioning a major drawback in existing cerebral oximeter for neonatology (Hyttel-Sorensen et al. Biomed Opt Express Vol. 2, Issue 11, pp. 3047–3057 (2011) https://doi.org/10.1364/BOE.2.003047).
Twenty healthy term babies are being enrolled in two centers (Rigshospitalet, Copenhagen; Fondazione IRCCS Ca' Granda) during the first day of life. As a part of the BabyLux protocol, a multiple repositioning test has been performed over the fronto-parietal region, placing the probe in the same area for six times and acquiring every time DCS and TR-NIRS signals continuously for 30 s for a total of 180 s acquisition.
Table 1 reports the baseline hemodynamic parameters and the optical properties at 690 nm (one of the three wavelengths used), measured on nine infants on their first day of life. The average coefficient of variability (CV), obtained averaging over the nine babies the CV calculated over the six repositioning, is also reported. Variability in StO2 meets our goal (< 5%) in all the infants recruited.
[Mean, CV for optical and heamodynamics properties]
μ a (1/cm)
CV μ a (%)
μ s′ (1/cm)
CV μ s′ (%)
StO2 (%)
CV StO2 (%)
tHb (μ M)
CV tHb (μ M)
BFI (x10(−8)) (cm2/s)
CV BFI (%)
0.17
9.3
8.3
11.3
69
3
90
9.1
3.4
25.7
BS10-6
Cerebral ischemia: clinical
Continuous monitoring of cerebral hemodynamics during disrupted cerebral auto-regulation in extracorporeal membrane oxygenation therapy
D. Busch1,2, C. Mavroudis1, T. Ko2, J. Lynch1, K. Mensah-Brown1, J. Newland1, T. Boorady1, A. McCarthy1, G. Du Pont-Thibodeau1, J. Connelly1, E. Buckley3, A. Yodh2, T. Kilbaugh1 and D. Licht1
1Children's Hospital of Philadelphia, Philadelphia, United States
2University of Pennsylvania, Philadelphia, United States
3Georgia Institute of Technology, Atlanta, United States
Abstract
Objectives: Extra corporal membrane oxygenation (ECMO) is the ultimate therapy for cardiac and/or respiratory failure, during which the lungs and/or heart are bypassed to permit mechanical blood oxygenation and perfusion. Devastating neurologic injury is a frequent complication of this therapy. ECMO pump-rates directly affect systemic blood pressure and are currently determined by anthropomorphic values and clinical evaluation. Cerebral Auto-Regulation (CAR) is an intrinsic protective mechanism that helps the brain maintain appropriate perfusion and oxygenation despite systemic variations; if disrupted, choice of the ECMO flow rate may be critical to maintain adequate brain perfusion. However, there are currently no tools that allow continuous monitoring of cerebral hemodynamics during ECMO and thus, clinicians are unable to individually optimize flow rates.
Methods: Diffuse optical and correlation spectroscopies (DOS, DCS) provide rapid, quantitative, and non-invasive measurements of blood oxygenation, volume, and flow, as validated against a number of clinical techniques. We have applied these tools to pediatric ECMO patients during manipulation of the ECMO pump rate, and thus blood pressure.
Results: We assessed cerebral hemodynamics during adjustments of ECMO parameters using DOS and DCS in 15 pediatric subjects. We observe both regulated (constant) and passive (blood pressure dependent) cerebral blood flows during titration of ECMO flow. Moreover, we have observed individual patients transitioning between constant and passive flow during the course of therapy and during flow manipulation.
Conclusions: Diffuse optical spectroscopies can provide continuous non-invasive evidence of dysregulated cerebral perfusion during ECMO therapy. Routine clinical non-invasive monitoring of cerebral hemodynamics could permit clinicians to detect and prevent situations of high risk for brain injury. Further, DCS measurements of blood flow could allow clinicians respond to dynamic changes in hemodynamic needs and potentially minimize the risk of either hemorrhagic or ischemic injury, lessening long term burdens of serious neurological injuries among survivors.
BS11-1
Brain immune interaction
Stroke induces exacerbated atheroprogression via alarmin-signaling
S. Roth1, K. Thuß-Silzcak1, A. Geerlof2, D. Vivien3 and A. Liesz1
1University Hospital Munich, Institute for Stroke and Dementia Research, Munich, Germany
2German Research Center for Environmental Health, Helmholtz Center Munich, Munich, Germany
3University of Caen Basse Normandie, Cyceron, Caen, France
Abstract
Objectives: Acute brain lesions induce multiphasic peripheral immune alterations. Subacute immunosuppression is followed by a chronic low-grade inflammation, which can be observed for more than one year after stroke in patients. However, it is unknown whether post-stroke chronic inflammation has an impact on common inflammatory comorbidities of stroke patients, such as atherosclerosis.
Methods: ApoE−/− mice fed with high cholesterol diet were used as an atherosclerosis model. Acute brain ischemia was induced by transient proximal middle cerebral artery occlusion (MCAO). Plaque load in en face whole aortas and aortic valves was assessed histologically. Alarmin-signaling was blocked by either sRAGE (soluble receptor for advanced glycosylation endproducts) or by in vivo knockdown using RAGE siRNA hydrodynamic injection. Aortic immune cell infiltration was analyzed by flow cytometry. In vivo cell proliferation was detected by continuous BrdU administration and cell infiltration was acquired with CCR2-RFP reporter mice.
Results: A significantly increased plaque load in aorta and aortic valve was detected after stroke compared to sham-operated mice, demonstrating exacerbation of atherosclerosis after stroke (Figure). In contrast, neutralization of circulating alarmins by the decoy receptor sRAGE resulted in amelioration of post-stroke atheroprogression, indicating a key role of alarmins in this sterile inflammatory process. Additionally, stroke-dependent mortality was significantly decreased in sRAGE-treated mice. We detected increased cell numbers of Ly6Chigh monocytes as well as lymphocytes in aortas after stroke, which was abrogated after blocking alarmin-signaling with either sRAGE or siRNA against RAGE. Mechanistically, increased post-stroke vascular inflammation was due to increased monocyte infiltration, but not local proliferation of CD11b+ monocytes within the atherosclerotic plaques.
Conclusion: We found exacerbated atherosclerosis after stroke due to peripheral immune activation by brain-released alarmins. These findings are of direct clinical relevance, indicating alarmin-signaling as a potential drug target to prevent post-stroke immunological comorbidities.
[Stefan Roth Figure]
BS11-2
Brain immune interaction
DSC MR Perfusion detects white matter microvascular perfusion abnormalities in X-linked adrenoleukodystrophy
A. Lauer1,2, X. Da3, M. Bo Hansen4, G. Boulouis1, A. Liberato Celso Pedrotti5, J. Kalpathy-Cramer3, P. Caruso5, E. Grabowski6, N. Rost1, K. Mouridsen4, F.S. Eichler1, B. Rosen3 and P.L. Musolino1
1Massachusetts General Hospital, Harvard Medical School, Neurology, Boston, United States
2Goethe University Frankfurt, Neuroradiology, Frankfurt a.M., Germany
3Harvard Medical School, Martinos Center for Biomedical Imaging, Charlestown, United States
4Aarhus University Hospital, Department of Clinical Medicine, Aarhus, Denmark
5Massachusetts General Hospital, Department of Neuroradiology, Boston, United States
6Massachusetts General Hospital, Harvard Medical School, Department of Pediatric Hematology and Oncology, Boston, United States
Abstract
Objectives: X-linked adrenoleukodystrophy (ALD) is a neurodegenerative disorder caused by mutations in the ABCD1 gene that leads in up to 60 % of hemizygotes (HEM) to devastating cerebral inflammatory demyelination (CALD). Mechanisms for conversion to CALD are unknown. Recent work suggests ABCD1 deficiency alters endothelial cell - monocyte interactions.1 We set out to explore if MR-DSC-perfusion can detect microvascular flow abnormalitites in ALD.
Methods: A human monocyte-endothelial adhesion assay was used to assess effects of ABCD1 deficiency in vitro. In vivo, 214 raw DSC-MRI perfusion scans from 43 ALD patients with and without CALD and 18 controls were analyzed to estimate capillary transit-time-distribution (CTH) as previously described.2 Statistical group analyses and mixed effects models were used to explore the effect of CTH upon lesion progression over time adjusting for age, anatomical location and hematopoietic stem-cell-treatment.
Results: In vitro, ABCD1 deficiency causes increased monocyte adhesion predominantly under venular and capillary shear forces. In vivo, HEM exhibited elevated mean CTH in whole WM perfusion compared to controls (3.2 ± 1.4 s vs.1.8 ± 0.5 s; p < 0.01,n=10). Per every CTH second increase the probability of a preselected anatomical WM region of being affected by demyelination increased by approximately 25 %. We found an age dependent effect of ABCD1 deficiency upon most susceptible WM regions for conversion to CALD in HEM as well as in WM at high risk of lesion progression in CALD subjects. Stem-cell-transplantation normalize CTH values in normal appearing WM of CALD subjects (Pre vs. 1 y post treatment:1.4 ± 0.2 s vs.1.1 ± 0.2 s: p=0.01, n=10).
Conclusions: ABCD1 deficiency affects microvascular flow physiology by altering interactions between endothelium and inflammatory cells. DSC-MRI-perfusion confirms microvascular flow abnormalities in patients lacking ABCD1. The temporal and spatial characteristics of these abnormalities strongly suggest a role in the pathophysiology of CALD. CTH may provide a powerful biomarker for identification of patients at high risk of CALD.
References:
1. Musolino et al., 2015.2. Mouridsen et al., 2014.
BS11-3
Brain immune interaction
Adrenergic-mediated loss of splenic innate-like B cells contributes to infection susceptibility after stroke
L. McCulloch1, C.J. Smith2,3 and B.W. McColl1
1The Roslin Institute, University of Edinburgh, Neurobiology, Easter Bush Campus, United Kingdom
2University of Manchester, Stroke and Vascular Research Centre, Manchester Academic Health Science Centre, Manchester, United Kingdom
3UK and Greater Manchester Comprehensive Stroke Centre, Department of Medical Neurosciences, Salford Royal NHS Trust Foundation, Salford, United Kingdom
Abstract
Post-stroke infection is the leading complication suffered by stroke patients leading to an increase in mortality and morbidity after stroke. Infections such as pneumonia are the most common and are typically acquired within the first three days of hospitalisation. Ischemic stroke is known to have suppressive effects on the immune system in both humans and experimental animal models. Reported post-stroke deficiencies in the adaptive immune system do not fully explain the rapid onset of infection described in the clinic. Marginal zone (MZ) B cells in the spleen provide protection against bacterial infection using an antibody-mediated defence mechanism with the speed and low-specificity of the innate immune response. Individuals who lack a spleen due to congenital dysfunction or surgery are susceptible to the similar strains of encapsulated bacteria that typically cause infections in stroke patients. We have shown that experimental stroke in mice rapidly results in a disruption to splenic immune architecture and an extensive loss of lymphocytes. Notably, there is an ablation of MZ B cells resulting in deficiencies in the trapping of blood-borne antigen and reduced levels of protective circulating IgM antibody. Spontaneous bacterial infection, correlating with the extent of MZ B cell loss, occurs in these animals. Circulating IgM levels were similarly suppressed in acute stroke patients and lower levels evident in patients with infection. We have shown these deficits are mediated by adrenergic signalling in experimental stroke, suggesting the involvement of autonomic pathways in brain-immune communication affecting systemic B cell function after stroke. These novel findings suggest that the loss of innate-like functions of B cells after stroke contributes to susceptibility to infection and highlight this pathway as an important target for intervention.
BS11-4
Brain immune interaction
Pre-existing cancer exacerbates cerebral ischemic stroke via Neurophilin-1 mediated recruitment of regulatory T cells into cancer tissue
L. Wang1,2, Y. Zhou1, W. Yu1 and P. Li1
1Renji Hospital, Shanghai Jiaotong University School of Medicine, Dept of Anesthesiology, Shanghai, China
2Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Dept of Anesthesiology, Shanghai, China
Abstract
Objectives: Epidemiological investigations suggest that there's a considerable number of patients with malignant cancer who may develop cerebral ischemic stroke, especially in middle to aged populations [1]. Regulatory T cells (Tregs), have been suggested to play critical roles both in cancer development and stroke pathology. In the present study, we sought to investigate how pre-existing cancer impact the ischemic brain injury and explore the mechanisms underlying the reciprocal interaction between these two diseases.
Methods: B16 or MC38 cells were injected subcutaneously to establish melanoma or colon cancer model in C57BL/6 mice. Distal middle cerebral artery was coagulated (dMCAO) to induce cerebral ischemic stroke 14 days later. Brain infarction and neurological performance were assessed up to 28-day post-stroke. CD4+CD25+and CD4+CD25− cells were transferred into Rag1−/− mice to reconstitute the immune system. Tregs' distribution and neuropilin 1 (NRP-1) expression were analyzed using flow cytometry. NRP-1 on Tregs was blocked in vivo or ex vivo to elucidate the mechanism of pre-existing cancer associated ischemic brain injury aggravation.
Results: The infarct volume of mice with pre-existing cancer developed significantly larger than those of cancer-free mice at 3,7 and 21 days after dMCAO. Brain-infiltration of inflammatory cells and pro-inflammatory cytokine expression in the ischemic brain were both augmented in mice with pre-existing colon cancer. Rag1−/− mice transferred with CD4+CD25− cells together with ex vivo NRP-1 neutralized CD4+CD25+ cells exhibited smaller infarct volume than those transferred together with isotype IgG treated CD4+CD25+ cells. Finally, antibody blocking of NRP-1 in stroke mice with pre-exisiting colon cancer not only hindered Tregs recruitment into cancer tissue, but also attenuated the cerebral ischemic injury without deteriorating cancer burden.
Conclusions: Preexisting cancer exacerbates cerebral ischemic stroke by recruiting Tregs into cancer tissue via the NRP-1 signaling.
References:
1. B. B. Navi et al. Ann Neurol77, 291–300 (2015).
BS11-5
Brain immune interaction
Recombinant Interleukin-33 blocks brain infiltration of pro-inflammatory immune cells and limits ischemic injury
H.A. Kim1, S.R. Zhang1, B.R. Broughton1, C.H. Wong2, T. Magnus3, T.V. Arumugam4, M. Gelderblom3, G.R. Drummond1,5 and C.G. Sobey1,5
1Monash University, Department of Pharmacology, Clayton, Australia
2Monash University, Department of Medicine, Clayton, Australia
3University Medical Center Hamburg-Eppendorf, Department of Neurology, Hamburg, Germany
4National University of Singapore, Department of Physiology, Singapore, Singapore
5Monash University, Department of Surgery, Clayton, Australia
Abstract
Objectives: Stroke triggers a pro-inflammatory Th1-driven response which aggravates brain injury and neurological deficits in the acute phase, and so therapies that promote an anti-inflammatory type response may be useful. We have tested whether acute administration of the Th2-type cytokine, interleukin (IL)-33, elicits anti-inflammatory effects and limits brain injury after ischemic stroke.
Methods: Adult male C57Bl/6 mice (n = 160) were injected with recombinant mouse IL-33 (2 µg) or 1% bovine serum albumin (vehicle) i.p. 24 h before and immediately after 1 h of filament-induced middle cerebral artery occlusion, followed by reperfusion for 23 h. Functional assessments were performed by video tracking of spontaneous activity, the hanging grip test and neurological deficit scoring. Brain sections were stained with thionin for infarct analysis or immunohistochemistry. Flow cytometric analysis of immune cells was performed on cells isolated from brain, spleen and blood. Bacterial content of lung homogenates was assessed following 18 h incubation at 37 °C. In addition to vehicle or IL-33, some mice (n = 36) were injected with ampicillin (100 mg/kg) and gentamicin (4 mg/kg, s.c.) 24 h before, and 1 and 8 h after stroke.
Results: IL-33 increased plasma Th2-type cytokine levels and reduced recruitment of M1-polarised macrophages to the ischemic hemisphere compared with vehicle controls. Furthermore, IL-33 reduced infarct size (by ∼35%) and activation of microglial cells, and blocked infiltration of cytotoxic T cells, particularly natural killer T-like cells, into the post-stroke brain. Of note, mice treated with IL-33 exhibited worse post-stroke survival rates and functional outcomes that were associated with exacerbated bacterial infection in the lungs. However, combination therapy of IL-33 plus antibiotics prevented lung infection and improved functional recovery.
Conclusions: These findings suggest that acute administration of recombinant IL-33 in combination with antibiotic therapy can limit brain injury and prevent post-stroke infection.
BS11-6
Brain immune interaction
Immune response in ischemic stroke regulated by let7i microRNA
G. Jickling1, B. Ander1, N. Schroff1, B. Stamova1, C. Dykstra-Aiello1, D. Liu1 and F. Sharp1
1University of California, Davis, Neurology, Sacramento, United States
Abstract
Objective: Evaluate microRNA let7i in ischemic stroke and its regulation of the immune response.
Methods: A total of 212 patients were studied; 106 with acute ischemic stroke and 106 risk factor matched controls. RNA from circulating leukocytes was isolated from blood collected in PaxGene tubes. Let7i miRNA expression was assessed by TaqMan qRT-PCR. To assess let7i regulation of gene expression in stroke, mRNA from leukocytes was measured by whole genome HT Affymetrix microarray. Given microRNAs act to destabilize and degrade their target mRNA, mRNA that inversely correlated with let7i were identified. To demonstrate let7i post-transcriptional regulation of target genes, a 3'UTR luciferase assay was performed. Target protein expression was assessed by ELISA.
Results: Let7i was decreased in patients with acute ischemic stroke (fold change −1.70, p < 0.00001). A modest inverse correlation between let7i and NIH Stroke Scale at admission (r = −0.32, p = 0.02), infarct volume (r = −0.21, p = 0.04) and plasma MMP9 (r = −0.46, p = 0.01) was identified. The decrease in let7i was associated with increased expression of several of its messenger RNA targets including CD86, CXCL8 and HMGB1. In vitro studies confirm let7i post-transcriptional regulation of target genes CD86, CXCL8 and HMGB1. Functional analysis predicted let7i regulates pathways involved in leukocyte activation, recruitment, and proliferation including canonical pathways CD86 signaling in T helper cells, HMGB1 signaling, and CXCL8 signaling.
Conclusions: Let7i is decreased in circulating leukocytes of patients with acute ischemic stroke. Mechanisms by which let7i regulates inflammatory response post-stroke include targeting CD86, CXCL8 and HMGB1.
BS12-1
Trauma
Nitric oxide (NO) and mitochondria in injured brain: Glutamate node as a possible target in NO-mediated mitochondrial dysfunction after traumatic brain injury (TBI)
M. Üçal1, K. Kraitsy1, A. Weidinger2, J. Paier-Pourani2, S. Patz1, B. Fink3, A. Kozlov2 and U. Schäfer1
1Medical University Graz, Institute of Neurosurgery, Graz, Austria
2Ludwig Boltzmann Institute for Clinical and Experimental Traumatology, Vienna, Austria
3NOXYGEN Science Transfer & Diagnostics GmbH, Elzach, Germany
Abstract
NO has been related to various processes in central nervous system. Our aim is to elucidate the role of NO and its protective or detrimental effects after TBI.
TBI was induced by lateral fluid percussion on adult male rats (mTBI < 2.5 atm < sTBI). DETC-Fe spin trapping and electron paramagnetic resonance were used for quantification of NO after trauma. Hippocampal and cortical nNOS and iNOS were analysed by RT-PCR. G6PDH was used as internal control. Nitrotyrosine levels were assessed by immunofluorescence. Mitochondrial activity was measured in ipsilateral cortical and hippocampal homogenates by high resolution respirometry using glutamate, pyruvate, or succinate.
We show for the first time that average NO levels strongly differ in sub-regions of healthy brain (cerebellum > cortex > hippocampus). Upon a severe TBI, NO levels were almost doubled in all three regions transiently at 4 h post-TBI, but NO-related detrimental effects such as mitochondrial dysfunction, tyrosine nitration and apoptotic/necroptotic cell death were confined to the ipsilateral cortex, despite comparable NO levels in the contralateral regions. Interestingly, NO increase in the ipsilateral cerebral regions was coupled to iNOS expression, while in contralateral regions a change in iNOS or nNOS expression was not observed.
A significant reduction in glutamate-dependent respiration was observed at 4 h in the ipsilateral cortex, indicating a post-TBI NO-mediated inhibition in cortical glutamate utilization. Accordingly, ex vivo exposure of healthy cortex homogenates to aqueous NO (20 µM) confirmed a higher vulnerability of glutamate utilization against NO than that of pyruvate or succinate.
Our data suggest that TBI-induced NO leads to reversible impairment in mitochondrial function through inhibition of glutamate utilization in the injured cortex. Whether this inhibition contributes to post-TBI glutamate excitotoxicity has to be prospectively analysed.
BS12-2
Trauma
Early changes in cerebral blood flow and cerebrovascular reactivity in mild traumatic brain injury
S. Hanalioglu1, T. Hidayetov1, A.I. Isikay1 and M. Mut Askun1
1Hacettepe University, Faculty of Medicine, Department of Neurosurgery, Ankara, Turkey
Abstract
Objectives: Severe traumatic brain injury (TBI) causes reduction of cerebral blood flow (CBF) and loss of cerebrovascular reactivity. However, results of previous research are inconsistent regarding mild TBI. This study aims to investigate early alterations in regional CBF, cerebrovascular reactivity and vascular responses to cortical spreading depolarisations (CSDs) after mild traumatic brain injury in mice.
Material & Methods: Sixteen male Swiss albino mice (20–30 g) were divided into two groups: trauma (n = 10) and sham (n = 6). Closed head trauma was performed under isoflurane anesthesia by using weight drop model (weight: 67 g, height: 5 cm). Cortical blood flow changes were measured using laser speckle contrast imaging. Early CBF changes, vascular reactivity to 5% CO2 inhalation, and CBF responses to induced CSD (with pinprick) were recorded for one hour after the trauma.
Results: Average CBF was 82.2 ± 6.9% in the ipsilateral and 100.9 ± 7.7% in the contralateral hemisphere within 15 minutes after trauma (p = 0.094). However, mean CBF measured in the pericontusional area (75.9 ± 5.1%) was significantly lower than contralateral hemisphere (p = 0.032). At the end of 1-hour recording, mild hypoperfusion (CBF: 74.8 ± 4.6%) was observed in the ipsilateral hemisphere – irrespective of CSD occurrence. There was no significant difference between groups, hemispheres or pre/post-CSD periods in terms of CO2 reactivity. There was no significant difference for vascular changes in response to CSD (net hypoperfusion 52.3 ± 2.6% vs. 56.3 ± 1.9% and duration 44.8 ± 1.8 min vs. 49.8 ± 2.3 min) between two groups.
Conclusion: At the early stages of mild head trauma, ipsilateral hemisphere becomes moderately hypoperfused (20–25%) particularly around the injury epicentre. Nonetheless, in the traumatized hemisphere, vascular reactivity and vascular responses to CSD were not altered significantly. These findings suggest that cerebrovascular responses are largely preserved despite slightly perturbed CBF during the early phase of mild TBI.
BS12-3
Trauma
The lectin complement pathway in human contusions
D. De Blasio1, S. Fumagalli1, L. Longhi2, F. Orsini1, F. Ortolano3, E.R. Zanier1, L. Neglia1, G. Goti4, P. Garred5, E. Picetti6, M. Locatelli7, A. Bernardi4, M. Gobbi1, N. Stocchetti3 and M.-G. De Simoni1
1IRCCS-Istituto di Ricerche Farmacologiche Mario Negri, Milano, Italy
2Azienda Socio Sanitaria Territoriale Papa Giovanni XXIII, Department of Anesthesia and Critical Care Medicine, Neurosurgical Intensive Care Unit, Bergamo, Italy
3Fondazione IRCCS Ca' Granda- Ospedale Maggiore Policlinico, Department of Anesthesia and Critical Care Medicine, Milano, Italy
4Università degli Studi di Milano, Department of Chemistry, Milano, Italy
5University of Copenhagen, Laboratory of Molecular Medicine, Department of Clinical Immunology, Section 7631, Rigshospitalet Faculty of Medical and Health Sciences, Copenhagen, Denmark
6Azienda Ospedaliero-Universitaria di Parma, Division of Anesthesia and Intensive Care, Parma, Italy
7Fondazione IRCCS Ca' Granda-Ospedale Maggiore Policlinico, Department of Neurosurgery, Milano, Italy
Abstract
Objectives: Mannose binding lectin (MBL), an activator of the lectin pathway (LP) of the complement system, has detrimental effect in experimental traumatic brain injury (TBI). Mice genetically depleted for both MBL-A and MBL-C present reduced susceptibility to TBI modeled by controlled cortical impact (1). In addition, in TBI mice the intravenous administration of Polyman9, a multivalent glycomimetic MBL-ligand, is neuroprotective (2). Here, we assessed: 1) the presence of the LP components in human cerebral contusions; 2) their association with clinical events; 3) the ability of Polyman9 to inhibit LP activity in human plasma.
Methods: Expression of LP activators and key enzyme mannose-binding protein-associated serine protease 2 (MASP-2) was assessed by immunohistochemistry in brain tissue samples obtained from 28 TBI patients who underwent therapeutic neurosurgical removal of the contused tissue, within 12 h (early, n = 21) or 1–5 days (late, n = 7) from injury. Control tissues were obtained from 5 non-TBI patients (tumors and non-traumatic cerebral diseases). Using specific ligand coated plates we measured the presence of different LP components in TBI and non-TBI cerebral homogenates.
Results: MBL, ficolin1, ficolin2, ficolin3 and MASP-2 were found inside and outside brain vessels in all the contusions analyzed. No or limited staining was found in control tissues. In TBI patients, MBL was increased by 84% compared to non-TBI patients. Increased levels of MASP-2 were associated with the occurrence of traumatic subarachnoid hemorrhage and/or abnormal pupil reactivity, two prognostic indicators of poor outcome after TBI. Polyman9 dose-dependently inhibited the binding of MBL to its ligands in TBI plasma. This inhibition was associated with reduced C3 cleavage.
Conclusions: Overall these data demonstrate the local presence of LP components in contused human tissue, their clinical relevance and that pharmacological targeting of MBL may be a promising strategy in TBI.
References:
1. Longhi et al. 2014.2. De Blasio et al. 2016.
BS12-4
Trauma
Evaluating the amyloid precursor protein derivative, APP96-110, as a novel therapeutic agent following traumatic brain injury
S.L. Plummer1, E. Thornton1, F. Corrigan1, R. Vink2, R. Cappai3 and C. Van Den Heuvel1
1The University of Adelaide, Translational Neuropathology Laboratory, School of Medicine, Adelaide, Australia
2The University of South Australia, Division of Health Sciences, Adelaide, Australia
3The University of Melbourne, Department of Pathology, Melbourne, Australia
Abstract
Objectives: Following traumatic brain injury (TBI) neurological damage is ongoing through a complex cascade of primary and secondary injury events in the ensuing minutes, days and weeks. Despite its burden on society, the development of successful pre-clinical therapeutics has not translated into efficacious therapies in human clinical trials. However, the delayed nature of secondary injury provides a valuable window of opportunity to help limit the severe consequences of TBI. Recently, the amyloid precursor protein (APP) and its derivative APP96-110 have shown encouraging neuroprotective activity following TBI when administered directly into the brain. However, its efficacy should be assessed following intravenous (IV) administration for clinical success.
Methods: This study assessed the efficacy of IV administration of APP96-110, where a dose response for a single IV dose of 0.005 mg/kg – 0.5 mg/kg APP96-110 at either 30 minutes or 5 hours following moderate-severe diffuse impact-acceleration injury was performed. Male Sprague-Dawley rats were assessed daily for 3 or 7 days on the rotarod to examine motor outcome, with a separate cohort of animals utilised for immunohistochemistry analysis at 3 days post-TBI to assess the degree of axonal injury and neuroinflammation.
Results: Animals treated with either 0.05 mg/kg or 0.5 mg/kg APP96-110 IV after 30 minutes demonstrated significant improvements in motor outcome. This was accompanied by a reduction in axonal injury and neuroinflammation in the corpus callosum at 3 days post-TBI, whereas the 0.005 mg/kg dose had no effect. In contrast, at 5 hours post-TBI treatment with either 0.005 m/kg or 0.5 mg/kg APP96-110 demonstrated significant improvements in motor outcome over 3 days, which was again accompanied by a reduction in axonal injury in the corpus callosum.
Discussion: This demonstrates that APP96-110 remains an effective therapeutic option following IV administration for up to 5 hours post-TBI, and supports its development as a novel therapeutic compound for the treatment of TBI.
BS12-5
Trauma
A new class of non-metal catalytic carbon antioxidants restored cerebral perfusion in traumatic brain injury complicated by systemic hypotension
K. Mendoza1,2,3, L. Cherian-Mathew4, C. Robertson4, L. Nilewski1, W. Sikkema1, J. Tour1 and T.A. Kent2,3
1Rice University, Chemistry, Houston, United States
2Baylor College of Medicine, Neurology, Houston, United States
3Michael E. DeBakey Veterans Affairs Medical Center in Houston, Houston, United States
4Baylor College of Medicine, Neurosurgery, Houston, United States
Abstract
Objectives: Hypotension worsens traumatic brain injury (TBI). Reactive oxygen species (ROS) contribute to loss of autoregulation and poor reperfusion. However, antioxidant therapy has not demonstrated clinical benefit. Most available antioxidants have limited capacity, narrow therapeutic range or require regeneration. We developed a novel carbon nanoparticle (PEG-HCCs) that is a catalytic antioxidant with broad and high capacity ROS-quenching1. PEG-HCCs restored cerebral perfusion and improved functional outcome in mild TBI transformed to severe by hypotension/resuscitation. We hypothesized that the structural features we believe responsible for this remarkable efficacy1 could be mimicked in simpler materials more likely to achieve regulatory approval.
Methods: Graphitic quantum dots (40 nm;PEG-GQDs) were prepared from bituminous coal under fuming nitric acid and compared to graphene-rich pegylated-Perylene diimide (PEG-PDI). 32 Long Evans rats underwent mild cortical compression injury (3 m/s, 2.5 mm deformation). MAP of 40 mmHg was achieved by blood withdrawal and reinfused 80 minutes later simultaneous with intravenous PBS or 4 mg/kg pegylated-GQDs or PDI. Regional cerebral perfusion was measured using Perimed laser speckle.
Results: In-vitro results in cultured neurons and brain endothelial cells indicated equivalent PEG-GQD potency to PEG-HCCs protecting against oxidative injury and both demonstrated catalytic superoxide-quenching activity (not shown). In the mild TBI model, both GQD and PDI (Figure 1a, 1b) were equally effective as PEG-HCCs and completely restored cerebral perfusion even when administered 80 minutes following injury without acute toxicity.
Conclusions: GQDs and PDIs were effective in restoring perfusion. Lesion size and neurological outcome measurements will be presented. Given that perfusion reduction is a major factor in worsening outcome after TBI, these agents deserve consideration as potential clinical therapeutics.
Supported by NINDS R01NS094535 and Acelerox, LLC.
[Cerebral Blood Flow]
Reference:
1. Samuel EL et al, Highly efficient conversion of superoxide to oxygen using hydrophilic carbon clusters. PNAS 112:2343–8, 2015.
BS12-6
Trauma
Transplantation of RADA16-BDNF peptide scaffold with human umbilical cord mesenchymal stem cells forced with CXCR4 and activated astrocytes for repair of traumatic brain injury
and W. Shi1
1Affiliated Hospital of Nantong University, Neurosurgery, Nantong, China
Abstract
Due to the poor self-regeneration of brain tissue, stem cell transplantation therapy is purported to enable the replacement of lost neurons after traumatic brain injury (TBI). The main challenge of brain regeneration is whether the transplanted cells can survive and carry out neuronal functions in the lesion area. The brain is a complex neuronal network consisting of various types of cells that significantly influence on each other, and the survival of the implanted stem cells in brain is critically influenced by the surrounding cells. Although stem cell-based therapy is developing rapidly, most previous studies just focus on apply single type of stem cells as cell source. Here, we found that co-culturing human umbilical cord mesenchymal stem cells (hUC-MSCs) directly with the activated astrocytes benefited to the proliferation and neuron differentiation of hUC-MSCs in vitro. In this study, hUC-MSCs and the activated astrocytes were seeded in RADA16-BDNF peptide scaffold (R-B-SPH scaffold), a specifical self-assembling peptide hydrogel, in which the environment promoted the differentiation of typical neuron-like cells with neurites extending in three-dimensional directions. Moreover, the results showed co-culture of hUC-MSCs and activated astrocytes promoted more BDNF secretion which may benefit to both neural differentiation of ectogenic hUC-MSCs and endogenic neurogenesis. In order to promote migration of the transplanted hUC-MSCs to the host brain, the hUC-MSCs were forced with CXC chemokine receptor 4 (CXCR4). We found that the moderate-sized lesion cavity, but not the large cavity caused by TBI was repaired via the transplantation of hUC-MSCsCXCR4 and activated astrocytes embedded in R-B-SPH scaffolds. The functional neural repair for TBI demonstrated in this study is mainly due to the transplantation system of double cells, hUC-MSCs and activated astrocytes. We believe that this novel cell transplantation system offers a promising treatment option for cell replacement therapy for TBI.
BS13-1
Degeneration and aging
Postsynaptic density 93 alleviates memory deficits in Alzheimer´s disease mouse model by modulation of somatostatin receptor
L. Yu1, Y. Liu1, H. Yang1, X. Zhu1, H. Zhao1, L. Han1 and Y. Xu1
1Drum Tower Hospital, Medical School of Nanjing University, Nanjing, China
Abstract
Objectives: The dysfunction of synapse is closely related to the memory deficits in Alzheimer´s disease (AD). Preliminary studies demonstrated that postsynaptic density 93 (PSD-93) in hippocampus and cortex of 6 month APP/PS1 mice were lower than that of wide-type group, and PSD-93 overexpression could improve cognitive function in APP/PS1 mice. PSD-93 knockout mice showed synaptic plasticity and cognition impairment. However, how PSD-93 affects cognitive function in vivo is unknown.
Methods: To overexpress PSD-93 in vivo, lentivirus-mediated plasmid was injected into the hippocampus of 6-month old PSD-93−/− or APP/PS1 mice. After 28 days, open-flied test (OF), novel objection recognition test (NOR), Morris water maze (MWM), fear condition test (FCT) were undertaken to evaluate the cognitive function. The electrophysiology of treated mice was assessed by long-term depression (LTD), long term potential (LTP) and whole cell record. Spine density was estimated by Golgi staining. Expression of postsynaptic density protein 95 (PSD-95), synaptophysin, SAP-97, SAP-102, somatostatin (SST), somatostatin receptor 1 (SSTR1), SSTR2, SSTR4, CaMKIIα were quantified by western blotting. The distribution of SSTRs in the neuron was determined by immunofluorescence staining.
Results: PSD-93−/− mice showed significant cognitive impairment. Overexpression of PSD-93 improved learning and memory function in PSD-93−/− and APP/PS1 mice. Furthermore, PSD-93-overpression mice exhibited an enhanced level of LTP in CA1 and CA3 regions. Basal synaptic transmission was different in distinct neurons. PSD-93 knockout and overexpression could not affect spine density and levels of PSD-95, SAP-97, SAP-102, SST, synaptophysin. The level of CaMKIIα was decrease followed by PSD-93 knockout. Furthermore, PSD-93 increased the location of SSTRs in the cytomembrane of neuron.
Conclusions: PSD-93 overexpression could attenuate synaptic plasticity and cognitive deficits in PSD-93−/−and APP/PS1 mice, which may be associated with increased cytomembrane location of SSTRs in adrenergic neurons. Therefore, PSD-93 may provide a new target for early diagnosis and treatment for AD.
BS13-2
Degeneration and aging
Stalled blood flow in brain capillaries is responsible for reduced cortical perfusion and impacts cognitive function in mouse models of Alzheimer's disease
O. Bracko1, J.C. Cruz Hernández1, C. Kersbergen1, L. Park2, V. Muse1, M. Haft-Javaherian1, L. Vinarcsik1, Y. Kang1, J. Zhou1, G. Otte1, J.D. Beverly1, E. Slack1, T.P. Santisakultarm1, C. Iadecola2, N. Nishimura1 and C.B. Schaffer1
1Cornell University, Meinig School of Biomedical Engineering, Ithaca, United States
2Weill Cornell Medical College, Feil Family Brain and Mind Institute, New York, United States
Abstract
Objectives: Brain blood flow is decreased by ∼30% in both human patients and animal models of Alzheimer's disease (AD). Although this hypoperfusion likely contributes to the cognitive decline observed in AD, the physiological explanation of this phenomenon remains unclear.
Methods: Two-photon in vivo imaging and blood flow imaging using line scan and ALS-MRI methods in the APP/PS1 mouse model.
Results: In AD mice, about 2% of capillaries have stalled blood flow due to leukocyte adhesion in the lumen, while wild type mice have ∼5 times fewer stalled capillaries. Because a single stalled capillary decreases blood flow in many downstream branches of the vasculature, stalling in only 2% of capillaries results in substantially reduced blood flow. We determined that the majority of the capillary stalls are caused by neutrophils firmly adhered to the interior of the vessel wall. Administration of antibodies against Ly6G led to a ∼60% reduction in the number of stalled capillaries and a ∼30% increase in blood flow in brain penetrating arterioles. Administration of isotype control antibodies produced no changes in the number of stalls or blood flow (Fig1.A and B). A single treatment with anti-Ly6G in AD mice led to significantly improved cognitive performance within a few hours in tests of short-term memory (Fig1.C). Chronic treatment for one month with the same anti-Ly6G antibody improved cognitive performance even further and resulted in a decreased concentration of Aβ40 in the brain of AD mice, suggesting improved clearance of this monomer type (Fig1.D).
Conclusions: These data suggest that brain hypoperfusion likely contributes to cognitive deficits in AD patients and, by impairing vessel-mediated clearance of amyloid-beta, may accelerate disease development. Therapies that interfere with the capillary plugging described here could complement existing ideas for therapies aimed at reducing amyloid.
BS13-3
Sex difference in the neuroprotective effect of microRNA mir363-3p on ischemic stroke
A. Selvamani1 and F. Sohrabi1
1Texas A&M University, Neuroscience and Experimental Therapeutics, Bryan, United States
Abstract
Background: Analysis of circulating microRNA (miR) in young and middle-aged male and female rats revealed that miR363-3p expression was inversely related to infarct volume, such that adult females, who display the smallest infarct volumes, had the highest expression of miR363. The present study is designed to evaluate the role of miR363-3p mimic in middle aged females and males and to determine the mechanism behind the neuroprotective effect of miR-363-3 p.
Methods: Middle aged (12 mo) female and male rats were subject to middle cerebral artery occlusion (MCAo). At 4 h post-stroke, animals received a tail-vein injection of miR-363-3p or scrambled control. Animals were terminated at 5d post-MCAo. At termination, the brain was rapidly removed and processed for TTC (Triphenyl Tetrazolium Chloride) staining to assess infarct volume and behavioral assays were performed pre- and post-stroke. Lectin staining for microvessels was performed on 25um coronal brain sections using Flourescein Lectin, fluorescence images were captured using an inverted scope and microvessel density analyzed using Image J software. Protein lysates from the ischemic tissue was analyzed for caspase-3 expression by Western blot analysis and activity by ELISA. Fluorescein labeled mir363-3p oligo were used to determine cellular uptake of mimic.
Results: Intravenous mir-363-3p mimic reduced MCAo-induced infarct volume and reduced sensory motor impairment in middle aged females but not males. In middle-aged females, mean density of lectin-stained microvessels was significantly greater in the cortex and striatum of the miR-363-3p mimic treated group (p < 0.05). Capsase-3 expression and activity was significantly reduced in mir363-3p-treated middle aged females but not in males. Fluorescein-labeled mir363-3p was localized largely to NeuN positive cells indicating neuronal internalization of the mimic.
Conclusion: Collectively, the data suggests that miR-363-3p improves stroke outcome and preserves microvessel density in middle-aged females but not males and that mir363-3p may exert neuroprotection by suppressing a cell death effector.
BS13-4
Degeneration and aging
Impact of ROCK2 in the cerebral circulation: Implications for large and small vessel disease
M. De Silva1,2,3 and F. Faraci1,2
1University of Iowa, Internal Medicine and Pharmacology, Iowa City, United States
2Iowa City Veterans Affairs Healthcare System, Iowa City, United States
3Monash University, Biomedicine Discovery Institute, Clayton, Australia
Abstract
Abstract
Rho kinase (ROCK) has been implicated in the regulation of vascular function and the pathogenesis of vascular disease. Two isoforms of ROCK (1 and 2) exist and are expressed in vascular cells. Because little is known regarding the effects of specific isoforms, our goal was to examine the functional importance of ROCK2 in several models. First, we examined the impact of ROCK2 during vascular aging. Endothelium-dependent dilation in parenchymal arterioles was significantly impaired in old mice (22 ± 1 mo) compared to adult controls (4–5 mo), but in a pathway-specific manner. Endothelium-dependent hyperpolarization was intact while eNOS-mediated dilation was reduced by ∼50%. Y-27632 (a ROCK1 and ROCK2 inhibitor) or SLX-2119 (ROCK2 inhibitor), restored endothelial function in old mice. Because genetic background is a determinant of vascular disease, we also studied FVB/N mice. Like the C57Bl/6 mice, FVB/N mice exhibited endothelial dysfunction with aging that was reversed by ROCK2 inhibition. Second, myogenic tone in parenchymal arterioles from adult C57Bl/6 mice (baseline diameter = 17 ± 2 µm), was inhibited by SLX-2119 (eg, 1 µmol/L SLX-2119 dilated by 79 ± 4%). Third, we explored the impact of ROCK2 during salt-sensitive hypertension. Mice were treated with deoxycorticosterone (DOCA) and 0.9% salt for 3 weeks. Following DOCA-salt, endothelium-dependent dilation of middle cerebral arteries as well as pial and parenchymal arterioles were reduced compared to sham controls. Endothelium-independent vasodilation was not impaired. DOCA-salt increased mRNA expression of renin angiotensin components (eg, AGT, ACE) in brain and cerebral arteries. Inhibition of angiotensin II (AT1) or mineralocorticoid (MR) receptors or ROCK2 reversed endothelial dysfunction in DOCA-salt treated mice. These findings suggest that ROCK2 has major effects on function of endothelium and vascular muscle in both large arteries and small arterioles during aging and hypertension. The data support the concept that aberrant ROCK2 signaling may be a key contributor to large and small vessel disease.
BS13-5
Degeneration and aging
Using microcoils to mimic cerebrovascular disease leads to early changes in the extracellular matrix and blood-brain barrier
J. Roberts1, M. Maniskas1 and G. Bix1
1University of Kentucky, Lexington, United States
Abstract
Bilateral carotid artery stenosis (BCAS), an experimental model of vascular dementia, leads to white matter lesions and cognitive impairment. With time, pathology worsens and spreads to the hippocampus and cortex. While some variability in the temporal and spatial distribution of brain injury may result from inter-mouse strain differences in cerebrovascular anatomy, coil size employed and surgical technique, it is generally accepted that hippocampal and cortical pathology is not significantly present prior to 30 days. However, as changes in the extracellular matrix (ECM) integrity and blood-brain barrier (BBB) permeability are known to precede more overt brain pathology in a variety of diseases, we hypothesized that BBB changes could occur earlier after BCAS in the hippocampus, striatum and cortex and be a precursor of longer term pathology in these regions.
Three month old male C57Bl/6 mice underwent BCAS with 0.18 mm microcoils or sham surgery and changes in the ECM and BBB were analyzed by collagen IV and perlecan (vascular basement membrane components), claudin-5 and occludin (tight junction proteins), Evan's blue (permeability marker), GFAP and CD11b (glial cell markers), and Ki-67 (cell proliferation marker) immunohistochemistry, protein and RNA expression levels after 3, 7, 14, or 21 days.
Significant changes in properties of cerebrovascular integrity were detected within 7 days compared to sham animals, in both the striatum and hippocampus. Increased astrocyte and microglia activation was also observed. While few changes were observed in the cortex, surprisingly some of the animals did experience cortical ischemic infarcts within 14 days.
This study demonstrates for the first time that changes in the BBB occur shortly after BCAS in multiple regions throughout the brain and suggests that such changes might underlie the gradual development of BCAS non-white matter pathology.
BS13-6
Degeneration and aging
Brain tissue oxygenation changes with age in awake mice
M. Moeini1,2,3, X. Lu1,2, T. Lam3, S. Bélanger1,2, A. Kakkar3 and F. Lesage1,2
1Ecole Polytechnique de Montreal, Biomedical Engineering Institute, Montreal, Canada
2Montreal Heart Institute, Montreal, Canada
3McGill University, Department of Chemistry, Montreal, Canada
Abstract
Objectives: Previously established correlation between vascular disorders and cognitive impairments [1] suggests that brain tissue oxygenation changes associated with age-dependent compromised cerebrovascular function may contribute to cognition decline. This work aimed to study age-related changes in resting brain tissue oxygenation in awake mice, removing age-related anesthesia confounds.
Methods: Brain tissue pO2 was measured in Young (Y, 9 months, n = 7), Middle-aged (M, 15 months, n = 7) and Old (O, 27 months, n = 7) C57Bl/6 mice using two-photon phosphorescence lifetime microscopy [2] and O2-sensitive phosphorescent dye PtP-C343 [3]. A custom titanium bar was implanted on the skull and a thinned-skull window [4] was created over the left barrel cortex. Awake imaging was done on a treadmill wheel which allowed free movement of limbs while the head was restrained. Animals were trained on the wheel for 4 sessions to habituate to the head restrain and minimize stress. pO2 measurements up to 250 µm deep were used to obtain the average pO2, pO2 spatial heterogeneity (quantified as standard deviation/mean) and tissue pO2 gradients from arterioles.
Results: With increased age, we found decreased tissue pO2 (Y: 38.2 ± 0.1 mmHg, M: 34.5 ± 0.1 mmHg, O: 29.2 ± 0.1 mmHg), increased pO2 spatial heterogeneity (Y: 27.8%, M: 32.3%, O: 36.3%), and an increase in the fraction of sampled points with hypoxic pO2 (< 5 mmHg) (Y: 0.01%, M: 0.28%, O: 0.87%). The analysis of pO2 gradients is still in progress to study arterial oxygen supply.
Conclusions: We measured for the first time changes in brain tissue oxygenation with age and showed a decrease and potential for hypoxia. These results are of significant importance as they may contribute in neurodegenerative diseases and cognition decline in elderly people.
References:
[1] Yaffe et al., Circulation, 129:1560, 2014.[2] Sakadžić et al., Nat. Meth., 7:755, 2010.[3] Finikova et al., ChemPhysChem, 9:1673, 2008.[4] Shih et al., J. Vis. Exp., 61:e3742, 2012.
BS14-1
Brain energy metabolism
Non-invasive, multi-modal monitoring of bispectral index, cerebral oxygen metabolism and cerebral blood flow under general anesthesia
M. Kacprzak1, S. Tagliabue1, C. Lindner1, I. Chochron da Prat2, A. Sanchez-Guerrero3, O. Martinez Silva2, J.L. Hollmann1, U.M. Weigel4, M. de Nadal2, J. Sahuquillo3 and T. Durduran1,5
1ICFO – Institut de Ciències Fotòniques, Castelldefels, Spain
2Vall d’Hebron University Hospital, Department of Anesthesiology, Barcelona, Spain
3Vall d’Hebron University Hospital, Department of Neurotraumatology and Neurosurgery Research Unit, Barcelona, Spain
4HemoPhotonics S.L., Castelldefels, Spain
5Instituciò Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain
Abstract
The bispectral index, cerebral blood flow and metabolic rate of oxygen were monitorized simultaneously by hybrid, noninvasive optical platform during propofol-induced anesthesia
Introduction: The depth of propofol-induced anesthesia is often monitorized by an electroencephalographic measure of brain activity called BIS (bispectral index), which outputs information about cerebral metabolic rate of oxygen (CMRO2) (Rigouzzo2008). Hybrid diffuse optics can non-invasively measure microvascular hemodynamics via the oxygen saturation and a cerebral blood flow (CBF) index and CMRO2 (Durduran2014). Propofol reduces CMRO2 and leads to a coupled decrease in relative CBF, when neurovascular coupling is preserved. We sought to compare two methods during controlled anesthesia in healthy brain.
Methods: A combination of near-infrared time-resolved spectroscopy (TRS) and diffuse correlation spectroscopy (DCS) were measured bilaterally on the frontal lobes during extracraneal surgical procedures under general anesthesia with target-controlled infusion of propofol. The BIS was on the left brain hemisphere. A Pearson correlation test was performed on the data.
Results: A representative time evolution of BIS, rCBF and rCMRO2 from the left brain hemisphere in one subject is shown in figure1.
Figure1. Representative time evolution of (a) the BIS, (b) rCBF and (c) rCMRO2.
Nineteen patients were included in the study. The entire dataset correlates significantly (p < 0.001) between rBIS and rCBF both for raw (R = 0.33) and de-noised (R = 0.67) analysis. rBIS and rCMRO2 correlate significantly (p < 0.001) in a similar manner. Individual subjects show varying levels of correlation that is anecdotally confirmed to be related to the type of surgery and the mean-BIS value.
Conclusion: The first comparison of non-invasively, optically derived CMRO2 and CBF by hybrid diffuse optics with the established BIS readings demonstrates the potential of diffuse optical techniques for monitoring the depth of anesthesia and validates CMRO2 as a metabolic index.
BS14-2
Brain energy metabolism
Neurometabolic changes during cortical spreading depolarization: high-resolution measurements with a lactate-glucose dual microbiosensor array
C.F. Lourenço1, A. Ledo1, G. Gerhardt2, J. Laranjinha1,3 and R.M. Barbosa1,3
1Center for Neuroscience and Cell Biology, University of Coimbra, Coimbra, Portugal
2Center for Microelectrode Technology, University of Kentucky, Lexington, United States
3Faculty of Pharmacy, University of Coimbra, Coimbra, Portugal
Abstract
Spreading depolarization (SD) is a slow propagating wave of strong depolarization of neural cells, implicated in several neuropathological conditions, such as traumatic brain injury (TBI). The breakdown of brain homeostasis is accompanied by significant hemodynamic and metabolic alterations, which impact on neuronal function. In this regard SD does not occur as an epiphenomenon in TBI, but can elicit further neuronal injury, often worsening the outcome. The magnitude and profile of the metabolic disturbance impacts both neuronal viability and clinical outcome. In parallel, the measurement of neuronal activity has been proven to be of utmost relevance to detect neurological dysfunctions in the clinical setting following TBI. In particular, the spectral profile of the electrical events have been suggested to provide relevant information about SD, including its prediction.
In this work we developed a novel dual-biosensor based on platinum multisite microelectrode arrays to measure lactate and glucose fluctuations with high spatial and temporal resolution and explored its ability to provide simultaneous neurometabolic and electrophysiological information during episodes of experimentally-induced spreading in anesthetized rats. By using amperometric recordings, with a single array, we successfully measured changes in lactate and glucose extracellular concentrations simultaneously with local field potential-related currents and in correlation with cerebral blood flow changes. We observed a significant drop in glucose concentration matched to a rise in lactate and ongoing with pronounced changes in the spectral profile of LFP-related currents. The high temporal resolution of the microbiosensor revealed multiphasic and complex fluctuations particularly in the extracellular lactate concentration, after the SD. This multimodal approach, encompassing metabolic, electric and hemodynamic responses can be valuable tool for a better understanding of metabolic changes in physiological and pathological conditions and their impact in neuronal function.
BS14-3
Brain energy metabolism
Glial-neuronal interactions following Tiagabin administration in wild type and GAD65 knockout mice
A.B. Walls1, L.H. Nilsen2, E. Eyjolfsson2, A. Schousboe1, U. Sonnewald2 and H.S. Waagepetersen1
1University of Copenhagen, ILF, Copenhagen, Denmark
2NTNU, Department of Neuroscience, Trondheim, Norway
Abstract
Objectives: The objective was to determine the role of GAD65 – the small isoform of the enzyme glutamate decarboxylase – for GABA synthesis. The selective GAT1 inhibitor, Tiagabin, was employed to inhibit reuptake of GABA from the synaptic cleft following neurotransmission and we hypothesize that this will augment the need for GABA synthesis.
Methods: GAD65 knockout and wild type mice were injected i.p. with Tiagabin (1 mg/kg) and one hour later, the mice were injected i.p. with a combination of [1-13C]glucose and [1,2-13C]acetate. After another 15 minutes the brains were subjected to microwave fixation and the cerebral cortices was isolated. Subsequently, tissue extracts were analyzed by 3H and 13C NMR spectroscopy.
Results and Conclusions: Glial glutamine is an important precursor for synthesis of neuronal glutamate and GABA. Biosynthesis of GABA from glial glutamine appears to be increased in the presence of Tiagabin, as the amount of [1,2-13C]GABA is maintained while the amount of [4,5-13C]glutamine is attenuated during Tiagabin treatment. This further implies that the two parts of the GABA/glutamate-glutamine cycle, i.e. 1) transfer of glutamine from astrocytes to neurons and 2) release of GABA from neurons and subsequent uptake into astrocytes, operate independent of each other. In addition, it appears that the increased extracellular GABA concentration resulting from Tiagabin treatment leads to a general neuronal hypometabolism of glucose in the TCA cycle observed by a decrease in the amounts of [4-13C]glutamate, [2-13C]aspartate and [3-13C]aspartate. In wild type mice the amount of [2-13C]GABA was reduced to a similar extent as its precursor [4-13C]glutamate by Tiagabin treatment indicating an unaltered precursor-product correlation. In GAD65 knockout mice, the amount of [2-13C]GABA was not affected by Tiagabin although the amount of its precursor [4-13C]glutamate was reduced, proposing that the effect of Tiagabin on GABA synthesis can be associated with GAD65.
BS14-4
Brain energy metabolism
Coupling of the glutamate-glutamine cycle rate with both glial and neuronal oxidative metabolism in the visual cortex of the Tupaia belangeri
S. Sonnay1, J. Poirot2, N. Just3, A.-C. Clerc1, R. Gruetter4, G. Rainer2 and J.M.N. Duarte1
1Ecole Polytechnique Federale de Lausanne (EPFL), Laboratory for Functional and Metabolic Imaging (LIFMET), Lausanne, Switzerland
2University of Fribourg, Department of Medicine, Visual Cognition Laboratory, Fribourg, Switzerland
3University Hospital Münster, Münster, Germany
4Ecole Polytechnique Federale de Lausanne (EPFL), Laboratory for Functional and Metabolic Imaging (LIFMET), Department of Radiology, Lausanne, Switzerland
Abstract
Objective: Brain energy metabolism results from cellular cooperation between neurons and astrocytes. Yet, the actual metabolic contribution of glial in supporting glutamatergic neurotransmission remains to be established [1]. We took advantage of the columnar characteristics of the Tupaia belangeri primary visual cortex (V1) to measure metabolic changes induced by continuous stimulation of V1.
Methods: 9 tree shrews under isoflurane anaesthesia (0.5–0.7%) were randomly allocated to groups (stimulation, n = 5; rest, n = 4). Magnetic resonance spectroscopy (MRS) and imaging (MRI) were performed at 14.1 T under continuous stimulation or at rest in a VOI of 35 µL or 105 µL for 1H and 13C MRS [2] during infusion of [1,6-13C]glucose, respectively. The VOI was localized in V1, as assessed by blood oxygenation level-dependent (BOLD) functional MRI. Visual stimulation was performed by randomly delivering lines in 4 orientations and 2 directions at 5–7 Hz with spatial frequency 0.04–0.05 cycle/degree. Data were analyzed with LCModel [3] and fitted to a two-compartment model of energy metabolism [4][5].
Results: Cortical brain activity resulted in a decrease in both brain glucose concentration (−17%; −0.34 µmol/g; P < 0.001) and phosphocreatine/creatine ratio (−8%; −0.07; P < 0.05) after 15 minutes of stimulation. Analysis of 13C MRS data at the individual level showed close relationships between the neurotransmission rate (VNT) and cerebral metabolic rate of oxidative glucose consumption (CMRglc(ox), Pearson-r = 0.82, P = 0.006), glial (VTCAg, Pearson-r = 0.81, P = 0.008) and neuronal (VTCAn, Pearson-r = 0.64, P = 0.066) oxidative metabolism. At the group level, 20% increase in VNT (+0.038 ± 0.042 µmol/g/min; P = 0.077) resulted in 24% (+0.063 ± 0.057 µmol/g/min; P = 0.007) and 12% (+0.061 ± 0.032 µmol/g/min; P < 0.001) increase in VTCAg and VTCAn, respectively, resulting in a net 14% increase in CMRglc(ox) (+0.058 ± 0.032 µmol/g/min; P < 0.001).
Conclusion: These results support the active partnership between neurons and astrocytes as both respond to glutamatergic neurotransmission by increasing in the same proportions their oxidative metabolism.
References:
[1] Lanz et al. (2013) FrontEndocrinol, 4:156.[2] Henry et al. (2003) MRM, 50:684.[3] Henry et al. (2003) NMRBiomed, 16:400.[4] Duarte et al. (2011) FrontNeuroenergetics, 3:3.[5] Sonnay et al. (2016) JCBFM, 36(5):928–40.
BS14-5
Brain energy metabolism
Exposure to recurrent hypoglycemia alters brain metabolism in diabetic rats
V. Shukla1, N. Dewan1, A.K. Rehni1, K.B. Koronowski1, H. Stradecki1, T.J. Garrett2, M.A. Perez-Pinzon1 and K.R. Dave1
1University of Miami, Miller School of Medicine, Neurology, Miami, United States
2University of Florida, College of Medicine, Pathology, Immunology and Laboratory Medicine, Gainseville, United States
Abstract
Objectives: Recurrent hypoglycemia (RH) is common in diabetic patients receiving glucose lowering therapies1. Exposure to RH leads to cognitive impairments2. Despite the significant effect of RH on hippocampal function, underlying mechanisms are not known. Our goal was to determine the effect of RH on hippocampal metabolism. The effects of RH on hippocampal metabolome and key glycolysis enzymes were evaluated.
Methods: Rats were made diabetic using streptozotocin. Hyperglycemia was corrected by insulin-pellet implantation. These animals were considered as “insulin-treated-diabetic” (ITD) rats. ITD rats were randomly divided in ITD + RH (diabetics on insulin therapy experiencing RH), and ITD + RH + Glucose (co-infusing insulin and dextrose to control for the effect of exogenous insulin used in RH group) groups. Mild/moderate RH was induced once a day (3 hrs duration) for 5 consecutive days. Hippocampal metabolome was evaluated using Thermo Q-Exactive Oribtrap mass spectrometer with Dionex UHPLC. Activities and substrate kinetics of key glycolytic enzymes (hexokinase3, phosphofructokinase4, and pyruvate kinase5) were evaluated.
Results: Several statistically significant differences in metabolites levels belonging to major metabolic pathways (e.g glycolysis, Krebs cycle, glycogen synthesis and amino acid metabolism) were noticed in ITD + RH group (n = 10) compared to ITD + RH + Glucose group (n = 10). Detailed analysis will be presented at the conference. Hexokinase (22%, p < 0.05) and phosphofructokinase (34%, p < 0.001) activities were also significantly elevated in ITD + RH group (n = 7) compared to ITD + RH + Glucose group (n = 6). Enzyme kinetics results analysis is in progress.
Conclusions: This is the first study evaluating the effect of RH on hippocampal metabolism and demonstrates profound effect of RH on hippocampal metabolism. Understanding how RH impacts hippocampal metabolism may help attenuate adverse effects of RH on hippocampal function.
Acknowledgement: NIH(NS073779),Southeast Center for Integrated Metabolomics
The brain tumor microenvironment: a little sweet, but a little cool
D. Coman1, Y. Huang1, P. Herman1, G. Kaneko1, J.U. Rao1, M. Parent1, S. Maritim2, J.J. Walsh2 and F. Hyder1,2
1Yale University, School of Medicine, Radiology & Biomedical Imaging, New Haven, United States
2Yale University, School of Medicine, Biomedical Engineering, New Haven, United States
Abstract
Objectives: Aerobic glycolysis and endothelial dysfunction are brain tumor hallmarks. Aerobic glycolysis is defined as glucose metabolism (CMRglc) in excess of oxidative phosphorylation (CMRO2), whereas endothelial dysfunction leads to abnormal blood flow (CBF). To understand differences between tumor and normal tissues, we measured CMRO2, CMRglc, and CBF in rat brains bearing gliomas using several MR methods. Because CMRO2, CMRglc, and CBF showed differences between tumor and normal tissues, a steady-state bioheat transfer model was used to calculate intratumoral-peritumoral temperature gradients. To confirm predicted temperature gradients, we used Biosensor Imaging of Redundant Deviation in Shifts (BIRDS) for high-resolution temperature (T) and extracellular pH (pHe) imaging of 9L, RG2 and U251 brain tumors.
Methods: CMRO2 was assessed by 13C turnover from glucose to glutamate. CMRglc was determined by dissociating 19F signals from fluorodeoxyglucose and fluorodeoxyglucose-6-phosphate. CBF was measured by arterial spin labelling (ASL) with independent labeling of tumor and normal tissues. The bioheat transfer model utilized measured CMRO2, CMRglc and CBF to calculate temperature distributions. T and pHe imaging was obtained with BIRDS, which utilizes shifts of nonexchangeable protons from macrocyclic chelates (e.g.,1,4,7,10-tetraazacyclododecane-1,4,7,10-tetrakis-methylenephosphonate or DOTP8−) complexed with thulium ion.
Results: The measured CMRO2 was 4–5 times lower, CMRglc was 3–4 times higher while CBF showed ∼40% reduction in tumor vs. normal tissue. The bioheat transfer model suggested a cooler tumor (∼1°C). The pHe and T maps confirmed that tumor's core was more acidic (>0.6 pHe units) and hypothermic (∼1°C) compared to surrounding tissue (Figure).
Conclusions: The results suggest a more acidic and cooler tumor microenvironment which reflects elevated aerobic glycolysis. Accurate pHe and thermal imaging of tumor in relation to its surroundings might help optimize chemotherapy and hyperthermia treatments and monitor response to therapies targeting tumor metabolism.
Supported by NIH(R01EB-011968, R01CA-140102, P30NS-052519).
BS15-1
Neuroinflammation in cerebral ischemia 2
M1 microglia aggravate white matter injury in a rat model of chronic cerebral ischemia
L.-Y. Zhang1, Z.-Y. Wei2, Y. Zhu3, J.-J. Pan4, L.-P. Wang1, M. Mamtilahun4, C. Qin4, Y.-Y. Ma1, L. Jiang4, Y.-Y. Song1, J.-Y. Liu4, L.-L. Luo4, Y.-T. Wang4, Z.-J. Zhang4 and G.-Y. Yang4
1Ruijin Hospital affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China
2Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
3Shanghai Jiao Tong University, Shanghai, China
4Med-X Research Institute, Shanghai Jiao Tong University, Shanghai, China
Abstract
Objectives: Chronic cerebral hypoperfusion and white matter injury are key factors for vascular dementia[1]. Microglia are activated under chronic cerebral hypoperfusion and may participate in the white matter disintegration[1]. However, the molecular mechanism of microglia activation involved in chronic ischemic white matter injury is unclear. Here, we explored the effect of activated microglia on the white matter injury in a rat model of chronic brain ischemia.
Methods: Fifty adult male Sprague-Dawley rats underwent bilateral common carotid artery occlusion for 1, 2, and 4 weeks. Cerebral blood flow was determined by magnetic resonance imaging. Neurobehavioral assessment, immunostaining and CLARITY imaging were performed to explore how microglia activation contributed to white matter injury.
Results: The cerebral blood flow decreased to 45% of the baseline in chronic ischemia using magnetic resonance imaging. Spatial learning and memory deficits began at 2 weeks and sustained to 4 weeks in chronic ischemic rats compared to the control (p < 0.05). In chronic ischemic rats, the myelin density and the expression of myelin basic protein decreased, while activated microglia and the proportion of M1 microglia increased (p < 0.05). CLARITY imaging showed that M1 microglia adhering to myelin increased in chronic ischemic rats (p < 0.05). Furthermore, we found that chemokines including CCL2, CCL3, CCL5, adhesion molecule CD11b, phagocytosis-related molecule CD68 were up-regulated in chronic ischemic rats (p < 0.05).
Conclusions: Our results suggest that chronic cerebral hypoperfusion can lead to white matter injury and cognitive impairment. M1 microglia mediate white matter injury by adhering to and phagocytosing myelin under chronic ischemia.
References:
1. Farkas E et al. Brain Res Rev. 2007;54(1):162–80.
BS15-2
Neuroinflammation in cerebral ischemia 2
Astrocyte-specific gene transfer of insulin-like Growth Factor (IGF)-1 in middle aged female rats improves stroke outcomes
F. Sohrabji1 and A. Okoreeh1
1Texas A&M University Health Science Center, Neuroscience and Experimental Therapeutics, Bryan, United States
Abstract
Introduction: Middle-aged female rats sustain larger strokes than younger female rats, and this is associated with an age-related loss of circulating and parenchymal brain levels of insulin like growth factor-1 (IGF-1). IGF-1 synthesis is decreased in astrocytes, a cell type that is critical post stroke recovery. Here we tested the hypothesis that astrocyte-specific IGF-1 gene transfer to middle-aged females will improve stroke outcomes.
Methods: Middle-aged (10–12 mo old, acyclic) female rats were injected into the striatum and cortex with adeno-associated virus serotype 5 (rAAV5) packaged with the coding sequence of the hIGF-1 gene downstream of an astrocyte-specific promoter (GFAP). The rAAV-control consisted of an identical shuttle vector construct without the hIGF-1 gene. Six-eight weeks later, animals underwent 90 minute transient middle cerebral artery occlusion (MCAo) via intraluminal suture. At 1d post stroke, flow cytometry was used to determine the type and extent of peripheral immune cells trafficked into the brain. In parallel studies, animals were tested for performance on sensory motor tasks at 2d and 5 days after MCAo.
Results: rAAV-mediated IGF-1 expression was confirmed in astrocytes with RT-PCR. Flow cytometry analysis of immune cells in the brain at 24 h post stroke found that proportion of Treg cells was greater in animals with rAAV-IGF-1 as compared to rAAV-controls. Additionally, while there was no difference in the proportion of M2 microglia, rAAV-IGF-1 enhanced M2 infiltrating macrophages. At 2d and 5d post stroke, stroke induced sensory motor impairment was reduced in animals with rAAV-IGF-1 as compared to rAAV-controls.
Conclusion: Targeted enhancement of IGF-1 in astrocytes of middle-aged females improved stroke-induced behavioral impairment concomitant with recruitment of anti-inflammatory cell types to the ischemic brain.
BS15-3
Neuroinflammation in cerebral ischemia 2
Single penetrating artery occlusion leads to widespread white matter reorganization across the entire brain in a CX3 chemokine receptor 1 dependent manner
A. Lubart1,2, A. Benbenishty1,2 and P. Blinder1,2
1Tel Aviv University, Sagol School for Neuroscience, Tel Aviv, Israel
2Tel Aviv University, Neurobiology, Tel Aviv, Israel
Abstract
Background: Micro-stroke, occlusion of a small blood vessel, triggers physiological and structural changes in neuronal circuits adjacent to the infarct area. These changes affect stroke recovery and might be manipulated to lead to efficient neural repair. Microglia cells play an important role in brain homeostasis and they are the first to respond to brain damage by proliferation and migration towards the injury. CX3 chemokine receptor 1 (CX3CR1) is important in modulating the communication between neurons and resident microglia/infiltrating macrophages in the brain. The aim of this study was to investigate how microstroke affects neuronal reorganization throughout the entire brain, in wild type and CX3CR1 deficient mice.
Methods: Wild-type (WT), heterozygous (CX3CR1 GFP/+) and CX3CR1-deficient (CX3CR1 GFP/GFP) mice were subjected to photothrombotic occlusions of a single artery, combined with 2-photon imaging through cranial window.The ischemic brain damage and structural changes were monitored by magnetic resonance imaging. Neuronal regeneration and microglia/macrophage activation were analyzed by immunostaining.
Results: Small blood vessel occlusion produced widespread alterations in mean diffusivity (MD) and fractional anisotropy (FA) across the entire brain, both in the ipsi- and contra-lesional hemispheres, at 5 weeks post occlusion. The cortical occlusion led to increased expression of doublecortin (DCX), indicating post-stroke neurogenesis in the ipsilesional hemisphere. CX3CR1-deficient mice produces smaller lesions than in CX3CR1 GFP/+ mice. Micro-strokes produced widespread microglial/macrophage accumulation in the subcortical white matter of WT and CX3CR1 GFP/+ mice but not in CX3CR1 GFP/GFP mice.
Conclusion: Occlusion of a single penetrating artery is sufficient to produce widespread structural alterations in the entire brain. These changes are associated with extensive remodeling in the ipsilateral and the contralateral hemispheres. CX3CR1 expressed by microglia/macrophages might play a role in these alterations due to its effect on infarct size and activation of microglia/macrophages in the white matter following occlusion.
BS15-4
Neuroinflammation in cerebral ischemia 2
Progression of inflammation and brain metabolism after cerebral ischemia – inflammatory cells consume glucose
H. Backes1, M. Walberer2, M. Schroeter3 and R. Graf1
1Max Planck Institute for Metabolism Research, Cologne, Germany
Objectives: Cerebral ischemia leads to the development of neuroinflammation. In this long-term and follow-up study of individual rats we analyzed the dynamics of transient and chronic inflammation and investigated its relation to regional glucose metabolism.
Methods: In male Wistar rats, cerebral ischemia was induced by occlusion of the left middle cerebral artery (MCA) using SiO2 macrospheres, a rat model for embolic stroke. Blood flow (CBF) was measured using [15]O2-H2O PET before and 5, 30, and 60 minutes after induction of ischemia followed by a [18F]FDG PET scan at 75 minutes. During the following 6 weeks inflammation and metabolism were repeatedly measured using [11C]PK11195 and [18F]FDG at day 2, 7, 14, 21, 42.
Results: After permanent occlusion of the MCA, the ischemic territory was established within 30 minutes. We did not observe CBF changes between 30 and 60 minutes. Inflammation developed predominantly in the border zone of the ischemic region with low intensity at day 2, peak intensity at day 7 and steady reduction thereafter (primary inflammation). Chronic inflammation in the ipsilateral thalamus also peaked at day 7 but persisted at a constantly high level up to day 42 (secondary inflammation). Repeated measurements of cerebral metabolism showed that the region of primary inflammation was infarcted and that metabolism transiently returned during the presence of inflammation (Figure). We hypothesized and confirmed by histology that the transient return of metabolism in the primary inflammation region was caused by glucose consumption of inflammatory cells.
[Inflammation and metabolism after stroke]
Conclusions: Glucose consumption of inflammatory cells can considerably contribute to regional brain metabolism. Since neurodegenerative diseases are often accompanied by neuroinflammation and reduced metabolism, the impact of glucose consumption of inflammatory cells on the analysis of hypometabolism needs to be further investigated.
BS15-5
Neuroinflammation in cerebral ischemia 2
Transgenic expression of human Interleukin-37 reduces post-stroke functional deficit and lung infection in mice
S.R. Zhang1, C.G. Sobey1, G.R. Drummond1, C.A. Nold-Petry2, M.F. Nold2 and H.A. Kim1
1Monash University, Pharmacology, Clayton, Australia
2Hudson Institute of Medical Research, The Ritchie Centre, Clayton, Australia
Abstract
Objectives: Post-stroke Th1-driven inflammation may exacerbate brain injury and neurological complications as well as impact systemic immune system effects, and thus may be a target for therapy. Here we have tested in mice whether transgenic expression of the anti-inflammatory human cytokine interleukin (IL)-37, of which no mouse homolog has yet been identified, might modulate brain injury and functional deficits after ischemic stroke.
Methods: Age-matched adult male mice expressing the human IL-37 gene (IL-37tg; n = 21) and wild-type C57BL/6 controls (WT; n = 24) were anesthetised with ketamine (80 mg/kg) and xylazine (10 mg/kg) i.p. and underwent filament-induced middle cerebral artery occlusion for 1 h followed by reperfusion for 23 h. Functional assessment was performed by video tracking of spontaneous activity (ANYmaze software), and brains were then frozen and sections (30 µm) were stained with thionin for infarct analysis. Bacterial content of homogenates of lungs was assessed following 18 h incubation at 37 °C. 3,3'-Diaminobenzidine staining was performed on frozen brain sections (10 µm) for quantification of myeloperoxidase+ neutrophils.
Results: IL-37tg mice had lower mortality (9.5% vs 16.7%) and exhibited markedly less functional impairment after stroke than WT controls, travelling 9-times farther over 5 min (1.21 ± 0.41 m vs 0.13 ± 0.03 m, respectively, n = 18–20; P < 0.01). IL-37tg mice also tended to have 30% smaller infarcts (25.2 ± 2.8 mm3 vs 35.5 ± 5.0 mm3; P = 0.10) with 20% fewer neutrophils (41 ± 18 vs 53 ± 20 cells/section; n = 4–5) than WT controls. These differences were also associated with 80% fewer aerobic bacteria in lungs of IL-37tg than WT mice (8,339 ± 3,209 CFU/mg vs 39,603 ± 9,613 CFU/mg; P < 0.01).
Conclusions: These data suggest that endogenous expression of IL-37 can limit both functional deficits and secondary lung infections in mice after stroke. The ability of IL-37 to suppress post-stroke morbidity may represent a novel therapeutic direction for potential use in stroke patients.
BS15-6
Neuroinflammation in cerebral ischemia 2
Beta4 integrin plays a vasculo-protective role during neuroinflammation
J. Welser1, R. Kant1 and R. Milner1
1The Scripps Research Institute, La Jolla, United States
Abstract
Objectives: Beta4 integrin is an endothelial cell adhesion molecule that is strongly upregulated on cerebral blood vessels during neuroinflammation 1. In light of the key protective role for beta4 integrin in mediating barrier integrity in the epidermis 2, the goal of this study was to determine whether this integrin plays a similar role in the central nervous system as an inducible protective mechanism that stabilizes the blood-brain barrier (BBB) under conditions of stress.
Methods: We employed the mouse model of multiple sclerosis, experimental autoimmune encephalomyelitis (EAE), which displays BBB breakdown, neuroinflammation and demyelination. Using this model we defined the time-course of beta4 expression with EAE progression and evaluated the role of beta4 integrin in vasculo-protection. Functional relevance was examined in transgenic mice in which endothelial expression of beta4 integrin was ablated using the Cre-Lox approach (beta4-EC-KO mice 3). The impact of the absence of beta4 was evaluated by measures of clinical disease severity and histopathology.
Results: Endothelial expression of beta4 integrin was strongly upregulated with EAE progression and maximal levels attained at the peak of EAE disease. Absence of endothelial beta4 integrin resulted in worse clinical disease in the EAE model and histological evaluation revealed enhanced neuroinflammation in beta4-EC-KO mice, as seen by increased numbers of CD45, MHC II, and Mac-1 positive cells within the cerebral parenchyma.
Conclusions: Taken together, our data support the hypothesis that endothelial beta4 integrin is an inducible protective mechanism that enhances BBB integrity and reduces leukocyte infiltration during neuroinflammatory disease.
References:
1. Milner et al, Mol. Cell. Neurosci. (2006), 33: 429–440.2. van der Neut et al, Nature Genetics (1996), 13: 366–369.3. Welser-Alves et al, Arterioscl. Thromb. Vasc. Biol. (2013), 33: 943–953.
BS16-1
Imaging applications
Quantitative measurement of blood velocity changes in single brain cortical microvessels during sensory stimulation using functional ultrasound imaging
C. Brunner1, G. Montaldo1, E. Mace2 and A. Urban1
1NERF, a Joint Research Initiative between IMEC, VIB and KU Leuven, Leuven, Belgium
2Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland
Abstract
When neurons are activated, they require more nutrients and more oxygen and consequently there is an increase of cerebral blood flow in the activated region. This phenomenon is called functional hyperemia or neurovascular coupling and is used to indirectly track neuronal activity in various brain imaging methods such as functional MRI. Functional ultrasound imaging (fUSi) is a recent imaging modality offering a high spatiotemporal resolution (100 µm pixel size, 400 ms) and a large field of depth (few cms). The current methods to detect hemodynamics with ultrasound are based on 2 classical Doppler modes: the color and the power Doppler. However, these modes are not optimal to measure the increase of cerebral blood volume (CBV), red blood cells velocity (RBC-V) and cerebral blood flow (CBF) related to neurovascular coupling. We developed a novel method called differential functional ultrasound imaging (d-fUSi) with a maximal signal-to-noise ratio based on a differential measure of the Doppler spectrum obtained in each pixel. D-fUSi has an unprecedented sensitivity allowing efficient detection of brain activity in response to ultrashort sensory stimulations without temporal averaging. Here, we demonstrated that fUSi allows efficient visualization of brain vasculature and precise quantification of both RBC-V and CBF in cortical microvessels of anesthetized rats. Automatic segmentation identified around 600 cortical vessels per cm2 including 55% of penetrating arterioles (PA) and 45% of ascending venules (AV). Within all cortical vessels recorded less than 7% were activated during stimulation of the forepaw. Furthermore, we quantified hemodynamic variations in activated vessels and detected increases of up to 60% of the RBC-V and 100% of the CBF in some individual PA. By monitoring properly multiple key parameters of brain hemodynamic, fUSi provides quantitative view of the neurovascular coupling that is fully complementary to the microscopic view provided by current optical imaging systems.
BS16-2
Imaging applications
Quantitative measurements of spinal cord blood flow of an animal model of relapsing-remitting MS
M. Tachrount1, A. Davies2, F. Rosianu1, R. Desai2, D. Thomas1, K. Smith2 and X. Golay1
1UCL Institute of Neurology, Dept. of Brain Repair and Rehabilitation, London, United Kingdom
2UCL Institute of Neurology, Dept. of Neuroinflammation, London, United Kingdom
Abstract
Objectives: The presence of hypoxia has been reported in Experimental-Autoimmune-Encephalomyelitis (EAE) animal model[1]. However, the mechanism and the role of hypoxia are still unclear. To the best of our knowledge, no quantitative measurements of the blood flow have been reported so far. The purpose of this work was to measure the spinal cord blood flow (SCBF) of EAE rats using Arterial Spin Labelling (ASL) MRI at different phases of the disease progression.
Methods: EAE was induced by injecting dark Agouti rats rMOG in IFA. Control animals received an emulsion of IFA and saline[1]. Thirteen animals (6 controls and 7 EAE) were scanned under anaesthesia using a 9.4T Agilent scanner before immunization and at different stages of the disease: onset, remission, and relapse. Perfusion maps with an in-plane spatial resolution of 190x190um2 were obtained using an optimised pre-saturation-FAIR-Q2TIPS ASL preparation combined with a reduced FOV readout sequence [3].
Results: The averaged SCBF before immunisation (n = 13) in GM and WM was 346.7 ± 26.1 ml / 100 g/min (CVGM = 7.5%) and 151.7 ± 23.1 ml/100 g/min (CVWM = 15.2%), respectively (Fig1). There were no significant differences between baseline and control measurements at different time points. Compared to the controls, EAE animals showed significant decrease of SCBFGM during the onset (−30.6%, p < 0.0001) and the relapse (−47.0%, p < 0.0001) phases. There were no significant differences between EAE and controls within WM except the last time point (−42.8%, p < 0.0001).
Conclusions: This study demonstrated for the first time on EAE animal model that neurological deficits are strongly correlated with impaired blood flow (r = −0.851). It confirms that the perfusion plays a key role in the progression of the disease. Further studies are needed to investigate the hypoperfusion's causes and to assess the energy metabolism during the progression of the disease.
References:
[1] Davies et al. Ann. Neurol. 2013; 74(6):815–25.[2] Tachrount et al. Proc. ISMRM2015 (#6758).
BS16-3
Imaging applications
Laminar cortical blood flow in mice is quantified by dynamic contrast optical coherence tomography
C. Merkle1 and V. Srinivasan1
1University of California Davis, Biomedical Engineering, Davis, United States
Abstract
Objectives: Dynamic Contrast Optical Coherence Tomography (DyC-OCT), a new technique based on cross-sectional imaging of the kinetics of an intravascular tracer, has previously been shown to enable transit-time measurements at the microscopic scale. The objectives of this study are to adapt the DyC-OCT methodology to quantify laminar blood flow and volume, and to perform simultaneous measurements of transit-time distributions, flow, and volume across cortical laminae in the mouse brain.
Methods: DyC-OCT imaged time-resolved changes in the dynamic scattering signal during the passage of an exogenous tracer through the imaged field-of-view. A highly scattering lipid emulsion was used as the tracer. A 1325 nm spectral/Fourier domain OCT microscope repeatedly imaged the mouse somatosensory cortex and tracked the passage of a bolus injection, which caused a transient increase in scattering signal within the vasculature. Indicator-dilution theory was then applied to the dynamic scattering signal to extract quantitative information about hemodynamics.
Results: DyC-OCT revealed clear laminar differences in transit time, flow rate, volume, and hematocrit. The middle cortical layers of the somatosensory cortex were previously shown to have the earliest microvascular transit times and lowest microvascular transit time heterogeneity. In this study, layer 4 was additionally shown to have the highest quantified flow rate (mL/100 g/min) and volume (mL/100 g), based on an estimated 36% arteriolar hematocrit and 25% microvascular hematocrit. Though absolute values of flow changed depending on the choice of inputs and assumed hematocrits, observed trends across layers (Figure) were always consistent.
[Figure 1 – Laminar Blood Flow]
Conclusions: DyC-OCT was developed to quantify flow, volume, and transit time distributions in brain microvascular networks. DyC-OCT will be a useful tool to investigate laminar differences in hemodynamics that accompany the distinct metabolic needs of individual neocortical layers, and to better inform interpretation of laminar functional Magnetic Resonance Imaging signals.
BS16-4
Imaging applications
Identifying sources of initial deficit and post-stroke recovery in mice using early MR-Imaging and correlation analysis
F. Knab1, S. Koch2, U. Grittner3, R. Bernard1, A. Schneider3, S. Mueller2, T.D. Farr4, P. Böhm-Sturm2, U. Dirnagl1 and C. Harms1
1Charité – University Medicine Berlin, Department for Experimental Neurology, Berlin, Germany
2Charité – University Medicine Berlin, Charité Core Facility 7T Experimental MRIs, Berlin, Germany
3Charité – University Medicine, Berlin, Germany
4University of Nottingham, School of Life Sciences, Nottingham, United Kingdom
Abstract
Introduction: Post-stroke recovery remains a poorly understood process. We correlated behavioural analysis with various lesion parameters to better understand the source of the initial deficit and post-stroke recovery.
Methods: 60 C57/Bl6 mice were trained in the staircase test, and underwent either 45 min of middle cerebral artery occlusion (MCAO n = 17), sham surgery (n = 17) or photothrombosis (PT n = 9). MR-Images were acquired 24 h after the surgery and co-registered to the Allen-Brain-Atlas. We correlated regional damage with the degree of initial deficit (performance on days 2–6) and residual deficit (performance on days 11–21). We used the coefficient of determination R2 to obtain the proportion of performance variance that can be explained by lesion volume and location.
Results: The largest difference in performance was found between the MCAO and sham group (mean difference: 41.2%, (30.7–51.7%), p < 0.001), and the smallest between PT and sham (mean difference: 19.9% (6.8–33.0%), p = 0.0049). In the MCAO group, the strongest correlation was observed between initial deficit and total stroke volume (R2 = 0.530), followed by various other regions: the retrohippocampal region (R2 = 0,510), cortical subplate (R2 = 0,470) and midbrain (R2 = 0,399). In the PT group, initial deficits correlated most with damage to the primary somatosensory cortex (R2 = 0.615) and striatum (R2 = 0.583). Most of the variance in post-recovery performance in MCAO can be explained by damage in the midbrain (R2 = 0,542) and the hippocampal region (R2 = 0,375). In the PT it can be explained by damage in the hippocampal region (R2 = 0,585) and in the thalamus (R2 = 0,307). Surprisingly, primary motor cortex damage correlated poorly with deficits in both groups.
Conclusions: Anatomy based correlation analysis can be used for further investigating initial deficit and post-stroke recovery and should be repeated with histological images. We may identify interesting targets for neuroprotection. The role of the primary motor cortex in functional deficits and recovery might be of less importance than expected.
BS16-5
Imaging applications
Dual-calibrated fMRI measurement of resting capillary and venous blood volumes
M. Germuska1, A. Merola1 and R. Wise1
1Cardiff University, CUBRIC, Cardiff, United Kingdom
Abstract
Objectives: We present a novel analysis methodology to estimate capillary (CBVcap) and venous (CBVv) blood volumes from dual-calibrated fMRI data.
Methods: Dual-calibrated fMRI allows for the estimation of OEF, CBF, venous-weighted blood volume and CMRO2. In this work we incorporate flow-diffusion modelling of oxygen extraction1 to derive voxel-wise estimates of CBVcap. The kinetics of oxygen extraction are based on extended Bohr-Kety-Crone-Renkin equations and Michaelis-Menten kinetics. We also incorporate the assumption of an idealised vascular network, such that capillary transit-time heterogeneity is linearly related to the mean transit time (MTT). By cascading the kinetic models, OEF is found to a have a monotonic relationship with MTT. Thus, allowing for the calculation of CBVcap from the resting OEF and CBF estimates.
Data (in 10 healthy volunteers) were acquired on a 3T MRI system using a pulsed ASL acquisition scheme (TE1 = 2.7 ms TE2 = 29 ms, TR = 2.2 s, TI1 = 700 ms, TI2 = 1500 ms). Respiratory challenges consisted of 3 periods of hypercapnia (5% CO2) and 2 periods of hyperoxia (50% O2) interleaved with room air (total acquisition time 18 minutes.)
Results: The mean grey matter CBVcap was 1.61 ± 0.29 ml/100g, with a mean CBVv of 1.55 ± 0.16 ml/100g. Maps of CBVv showed significant contributions from large veins, which were absent from maps of CBVcap (see figure). Across subjects grey matter values of CBVcap were correlated with CBF and CMRO2 (R2 0.78 and 0.96, p < 0.001). CBVv was not significantly correlated with any measured parameter.
[In-vivo maps of CBVv and CBVcap]
Conclusions: We present a novel method for non-invasive mapping of CBVcap. The strong correlation between CBVcap and CMRO2 suggests an important role for CBVcap in assessing cerebral metabolic function and neurovascular coupling.
References:
[1] Angleys et al. 2015. JCBFM 35, 806–817.
BS16-6
Imaging applications
MRI detection of iron oxide-labeled B cells in perilesional tissue after stroke in mice
A. van der Toorn1, P. Yanev1,2, S.B. Ortega2, G.A.F. van Tilborg1, R.M. Dijkhuizen1 and A.M. Stowe2
1University Medical Center Utrecht, Center for Image Sciences, Biomedical MR Imaging and Spectroscopy Group, Utrecht, Netherlands
2UT Southwestern Medical Center, Dept. of Neurology & Neurotherapeutics, Dallas, United States
Abstract
Objectives: B cells may contribute to lesion size reduction and functional recovery improvement after stroke1, but their mechanism of action is understood incompletely. To enable characterization of the migration pattern of B cells after stroke, we evaluated the potential of MRI to detect the distribution of iron-labeled B cells after injection in post-stroke mouse brain.
Methods: Highly purified (>95% CD19+) splenic B cells from adult male C57BL/6 mice were labeled with 20 µg/mL of an iron oxide-based superparamagnetic and fluorescent contrast agent (Molday ION™ Rhodamine B (MIRB)) overnight at 37 °C and 5% CO2. Recipient mice received perilesional intraparenchymal (∼2.0X105/ 2 µl PBS) or intraventricular (∼8.0X106/4 µl PBS) injection of MIRB-labeled B cells or vehicle in the ipsilesional hemisphere, 2 or 4 days after photothrombotic stroke. Mice were sacrificed 2 days post-injection followed by ultrahigh-resolution MRI (9.4 T; BSSFP, 3D gradient echo; 75 mm isotropic voxels) of extracted brains.
Results: Clear contrast enhancement (hypointensity) was observed at the injection site and in the lesion borderzone after intraparenchymal injection of labeled B cells.
[Figure 1. Sagittal post-mortem MRI of mouse brains]
Mice that were injected intraventricularly showed contrast enhancement throughout the CSF, but mostly near the lesion. The large iron-induced susceptibility effect ('blooming'), however, prevented observation of possible migration outside the ventricles.
Conclusions: Our MRI data clearly revealed presence of MIRB-labeled B cells with superior anatomical correlation and spatial resolution. Contrast enhancement patterns suggested migration of B cells towards the lesion site, with predominant perilesional accumulation where these cells may exert their beneficial effects. Combined serial in vivo MRI of labeled B cells and lesion progression may contribute to elucidation of interactions between adaptive immune responses and CNS injury after stroke.
References:
1. Selveraj U.M. et al. (2016) Neurotherapeutics 13 (4), 729–747.
BPS01-1
BrainPET: Novel radiotracers
A small bispecific antibody-based construct based on bapineuzumab as a PET tracer for amyloid beta pathology in brain
X.T. Fang1, G. Hultqvist1, S.R. Meier1, D. Sehlin1 and S. Syvänen1
1Uppsala University, Public Health and Caring Sciences, Uppsala, Sweden
Abstract
Alzheimer's disease is characterized by the accumulation and aggregation of amyloid beta (Aβ) in the brain. Current in vivo methods of imaging Aβ are lacking the sensitivity required to track disease progression. The goal was to create an antibody-based radioligand suitable for immunoPET targeting Aβ in the brain.
We designed a small recombinant bispecific antibody construct targeting the Aβ N-terminus (the murine version of bapineuzumab) and the transferrin receptor (TfR) for active transport across the blood-brain barrier (BBB). Following expression in mammalian cells, we investigated functional binding to both Aβ and TfR with affinity ELISA. We radiolabeled the bispecific construct with iodine-124 or iodine-125 and i.v. injected it into APP-transgenic mice (tg-ArcSwe) and wildtype (wt) mice. Mice were perfused three days post-injection, brains and other organs were harvested to measure brain uptake and biodistribution.
Our construct bound with high affinity to Aβ1-40 (KD: 0.20 nM) both pre- and post-radiolabeling and with moderate affinity to TfR (KD: 12.19 nM). Two hours post-injection, wt mice had a high concentration of antibody present in brain (mean%injected dose 1.41% which is comparable to small molecular radioligands), indicating successful BBB passage. Three days post-injection, 18 m.o. tg-ArcSwe had a 40-fold higher brain retention compared with age-matched wt mice. There was a small concentration of antibody present in kidney (%injected dose/gtissue 0.20%) and spleen (0.19%). We then labeled our construct with iodine-124, performed PET, and successfully imaged amyloid pathology 24 h post-injection.
In summary, we have successfully created a small bispecific antibody capable of crossing the BBB to bind to its target in the brain and it functions as a PET radioligand. Lastly, our construct design may be used for other brain targets also.
BPS01-2
BrainPET: Novel radiotracers
Novel PET radioligands show that COX-1 is constitutively expressed and that COX-2 is induced by inflammation in rhesus macaque
and S. Shrestha1
1National Institute of Mental Health & Neurosciences, Bethesda, United States
Abstract
Background: The cyclooxygenase (COX) system comprises two isoforms, COX-1 and COX-2, which are key enzymes in neuroinflammation. Distinction of COX-1 as a ‘constitutive isoform' and COX-2 as an ‘inducible isoform' has been postulated although the literature shows mixed findings. We recently developed two PET radioligands: 11C-PS13 for COX-1 and 11C-MC1 for COX-2. The purpose of this study was to demonstrate in vivo expression of COX-1 and COX-2 with the new radioligands in rhesus monkeys.
Methods: The study was divided into two parts: whole-body PET/CT in normal monkeys and brain PET in monkeys with neuroinflammation. To induce transient neuroinflammation, lipopolysaccharide (LPS) was injected into the right putamen of monkeys. After intravenous injection of either radioligand into rhesus monkeys, dynamic whole-body or brain PET scans were obtained. To measure the specific uptake, we performed both baseline and blockade scans. As blockers, we used non-radioactive PS13, non-radioactive MC1, and aspirin, a preferential inhibitor of COX-1.
Results: In normal condition, 11C-PS13 showed specific uptake in spleen, gastrointestinal tract, and brain while 11C-MC1 showed only in brown adipose tissue. After putaminal injection of LPS, 11C-PS13 uptake was not increased in brain while 11C-MC1 uptake was significantly increased.
Conclusions: Our results suggest that COX-1 is constitutively expressed in major organs while COX-2 is only induced by inflammation. Different pattern of expression between two COX isoforms could be successfully demonstrated by 11C-PS13 and 11C-MC1. Thus, these two radioligands show promise as biomarkers for measuring inflammation in various disorders, as well as target engagement of therapeutic drugs.
BPS01-3
BrainPET: Novel radiotracers
Tracer kinetic analysis of [11C]Cetrozole as a PET tracer for aromatase in the human brain
M. Jonasson1,2, E. Comasco1, P. Nordeman1, H. Wilking2, H. De Grauw1, K. Takahashi3, G. Antoni2, I. Sundström Poromaa1 and M. Lubberink1,2
1Uppsala University, Uppsala, Sweden
2Uppsala University Hospital, Uppsala, Sweden
3RIKEN Center for Life Science Technologies, Kobe, Japan
Abstract
Objectives: Aromatase is an enzyme that converts testosterone and androstenedione to estradiol and estrone in the brain. The aim of this study is to evaluate tracer kinetic models for quantitative analysis of the novel aromatase PET ligand [11C]Cetrozole.
Methods: Data from seven 90 min dynamic [11C]Cetrozole PET scans was used and arterial blood was sampled for measurement of blood radioactivity and metabolite analysis. VOI-based analysis was performed using single-tissue (1TCM) and two-tissue (2TCM) reversible plasma-input compartment models and reference tissue models, SRTM and reference Logan, with cerebellum as reference region. Five VOIs based on a probabilistic template on co-registered T1-MRI images were used for the evaluations; thalamus, amygdala, hypothalamus, putamen and raphae. Correlation and agreement of DVR-1 and BPND values of plasma-input and reference methods were assessed by linear regression.
Results: The 2TCM performed best according to Akaike criteria in 31/35 cases, however, it was not possible to determine the individual parameters with any robustness. 1TCM distribution values agreed well with 2TCM distribution values, and was used for validation of the reference tissue models. Highest 1TCM DVR-1 values were found in thalamus (DVR-1 = 0.41–0.63). Correlation and agreement between 1TCM DVR-1 and SRTM BPND values of all five VOIs was high (R2 = 0.97, slope = 0.97), as well as between 1TCM and reference Logan DVR-1 (R2 = 0.94, slope = 1.05), as shown in Figure 1.
Conclusion: The reference tissue models SRTM BPND and reference Logan DVR-1 showed high correlation to plasma-input 1TCM DVR-1 values. However, more subjects are needed for the evaluation and also to validate the use of cerebellum as a reference tissue.
Figure 1: One tissue compartment model (1TCM) DVR-1 versus SRTM BPND and reference Logan DVR-1
BPS01-4
BrainPET: Novel radiotracers
First-in-human evaluation of a novel agonist radiotracer for PET imaging of the kappa opioid receptor
Y. Huang1, M. Naganawa1, N. Nabulsi1, S. Li1, S.-F. Lin1, D. Labaree1, J. Ropchan1, H. Gao1, A. Shirali1, R. Carson1 and D. Matuskey1
1Yale University, PET Center, Department of Radiology and Biomedical Imaging, New Haven, United States
Abstract
Objectives: The kappa opioid receptor (KOR) is implicated in a variety of stress-related disorders and thus an important target for disease mechanism research and drug development. The first KOR agonist tracer, 11C-GR103545, is characterized by slow kinetics in humans (1). We report here the first-in-human evaluation of a novel KOR agonist PET radiotracer, 11C-EKAP.
Methods: Six healthy human subjects (3 M and 3 F) underwent test-retest scans with 11C-EKAP on the HRRT scanner. An additional 12 subjects (6 M and 6 F) underwent a pair of baseline and blocking (after 150 mg of oral naltrexone) scans. Metabolite-corrected arterial input functions were measured for analysis of time-activity curves (TACs) in brain regions. Regional distribution volume (VT) was estimated using one- and two-tissue compartment models (1TC and 2TC) and the multilinear analysis-1 (MA1) method.
Results: Metabolism of 11C-EKAP was fairly quick, with ∼32% of parent tracer in plasma at 60 min post-injection. Plasma free fraction was ∼25%. In the brain, 11C-EKAP displayed fast and reversible kinetics, with peak uptake at 10–20 min post-injection in most regions. Regional TACs were fitted better with the 2TC than 1TC model, and fitting with MA1 was generally good (Fig. 1). Mean VT values estimated by MA1 (t* = 20) were 14.2, 10.3, 9.9, 8.0, 6.4, 5.4 mL/cm3, respectively, for the cingulate cortex, putamen, frontal cortex, caudate, cerebellum and thalamus (n = 6). Minimum scan time was 80 min for stable VT estimates. Mean test-retest variability of VT was ∼10% across all brain regions. Pre-treatment with 150 mg naltrexone decreased VT values across brain region, with non-displaceable distribution volume (VND) estimated at ∼3.5 mL/cm3.
Conclusions: The novel KOR agonist 11C-EKAP exhibits excellent kinetic and imaging properties and is a promising radiotracer for PET imaging of kappa opioid receptor in humans.
[Time activity curves and model fitting]
References:
1. Naganawa M. et al. Neuroimage 2014; 99:69.
BPS01-5
BrainPET: Novel radiotracers
Serotonin release measured in the human brain: A PET study with [11C]Cimbi-36 and d-amphetamine challenge
D. Erritzoe1, A. Colasanti2, G. Searle3, Y. Lewis3, J. Passchier3, S. Azeem3, J. Beaver3, D. Nutt1, G. Knudsen4, R. Gunn1,3 and E. Rabiner2,3
1Imperial College, London, United Kingdom
2King's College, London, United Kingdom
3Imanova Ltd, London, United Kingdom
4University of Copenhagen, Copenhagen, Denmark
Abstract
Introduction: Estimation of 5-HT release in the human brain using PET has proved to be challenging. A citalopram challenge using the 5-HT2 agonist ligand [11C]Cimbi-36 was not successful, possibly due to marginal increases in 5-HT in the cortical projection areas following single dose SSRI administration. D-amphetamine is a safe experimental challenge in humans that releases dopamine and noradrenaline, and has been shown to increase extracellular 5-HT levels in pre-clinical experiments. As [11C]Cimbi-36 has >1000-fold selectivity for the 5-HT2A over dopaminergic and noradrenergic targets, we tested d-amphetamine as a paradigm to evaluate synaptic 5-HT changes with PET.
Methods: Six healthy male volunteers received [11C]Cimbi-36 PET scans before and after a single oral dose of d-amphetamine (0.5 mg/kg). Arterial blood samples were collected and a metabolite corrected arterial plasma input function was determined. Dynamic PET data were acquired over 90 minutes, corrected for attenuation, scatter and subject motion, and the total volume of distribution (VT) in the frontal cortex and the cerebellum derived from kinetic analysis using MA1. The frontal cortex binding potential (BPNDfrontal) was calculated as VTfrontal/ VTcerebellum-1. 5-HT release was quantified as ΔBPND = 1- BPNDfrontal post-dose/ BPNDfrontal baseline. Statistical inference was tested by means of a paired Students t-test evaluating a reduction in post-amphetamine [11C]Cimbi-36 BPND.
Results: Following d-amphetamine administration, [11C]Cimbi-36 BPNDfrontal was reduced by an average of 17% (p = 0.03).
Conclusions: [11C]CimbiCimbi-36 may be sensitive to synaptic 5-HT release in the human brain and combined with a d-amphetamine challenge can enable the evaluation of the human brain 5-HT system in neuropsychiatric disorders.
BPS01-6
BrainPET: Novel radiotracers
Quantitative assessment of synaptic density using the SV2A ligand 11C-UCBA in humans
M. Lubberink1,2, L. Appel1,3, K. Lindskog2, T. Danfors1,3, M. Sprycha3, J. Daging4,5, J. Eriksson3,6, E.-M. Larsson1,3, E. Kumlien4,5 and G. Antoni3,6
1Uppsala University, Surgical Sciences, Uppsala, Sweden
2Uppsala University Hospital, Medical Physics, Uppsala, Sweden
3Uppsala University Hospital, Medical Imaging Centre, Uppsala, Sweden
4Uppsala University, Neurosciences, Uppsala, Sweden
5Uppsala University Hospital, Neurology, Uppsala, Sweden
6Uppsala University, Medicinal Chemistry, Uppsala, Sweden
Abstract
Objectives: Quantitative imaging of the synaptic vesicle glycoprotein-2A (SV2A) with PET can be used as a measure of synaptic density, changes of which occur in many neurodegenerative diseases. 11C-UCBA has previously been validated as an SV2A tracer in preclinical studies. The aim of this work was to evaluate tracer kinetics and simplified methods for quantification of 11C-UCBA in humans.
Methods: Six subjects (4 epilepsy patients, 2 controls) underwent a 90-min PET/MR scan starting with injection of 5 MBq/kg 11C-UCBA. Arterial blood was withdrawn for measurement of blood concentrations and metabolite analysis. Images were reconstructed using zero-echo-time-MR-based attenuation correction. A VOI template was defined on a T1-MRI image and transferred to the dynamic PET images, including a centrum semiovale VOI as potential reference tissue. Data were analysed using single-tissue (1T-2k), two-tissue irreversible (2T-3k) and reversible (2T-4k) models, as well as SRTM and plasma- and reference-Logan methods. SUVR at various time points was compared to modeling outcomes.
Results: Plasma and brain kinetics of 11C-UCBA were slow, with peak activity in brain after 80 min and parent fraction 40% at 90 min. Data were best described using the 2T-4k model, but this could not provide robust VT or BPND values. Mean plasma-Logan-VT was 18.6 ± 6.7 in grey matter VOIs and 2.9 ± 0.2 in centrum semiovale. Plasma-Logan-DVR using centrum semiovale as reference tissue correlated well with 2T-4k-DVR (R2 0.93), and reference-Logan-DVR showed good correlation with plasma-Logan-DVR (R2 0.72). SUVR70–90 correlated well with reference-Logan-DVR (R2 0.98; Figure 1).
Conclusions: Slow kinetics of 11C-UCBA caused poor robustness of reversible compartment models, resulting in Logan analysis being preferred for quantitative analysis. Further studies are necessary to assess the ability of different analysis methods to detect changes in SV2A. SUVR70–90 may be a simplified marker of synaptic density using 11C-UCBA.
BPS02-1
BrainPET: Neurotransmitter systems
In vivo dopamine imaging using [11C]raclopride positron emission tomography in a chemogenetic mouse model
A.L. Cremer1, R. Lippert2 and H. Backes1
1Max Planck Institute for Metabolism Research, Multimodal Imaging of Brain Metabolism, Cologne, Germany
2Max Planck Institute for Metabolism Research, Neuronal Control of Metabolism, Cologne, Germany
Abstract
Objectives: In vivo measurement of dopamine (DA) release using [11C]raclopride PET in the mouse brain. [11C]raclopride is sensitive to changes in DA concentrations due to its relatively low binding affinity to D2/D3 receptors (Fig. 1A). Based on this principle we analyzed dynamics in [11C]raclopride PET as potential surrogate marker for dopamine release rates.
Methods: A group of 7 transgenic mice with chemogenetically activatable dopaminergic neurons (DAT-Cre tg/wt; DREADD fl/wt (Designer Receptors Exclusively Activated by Designer Drugs)) underwent two [11C]raclopride PET scans of 60 minutes using a bolus-plus-constant-infusion injection protocol. Clozapine-N-oxide (CNO) activates dopaminergic neurons and induces dopamine release in these DREADD mice. During one scan the mice received CNO injected intra-peritoneally 10 minutes after the start of the scan; during the second scan the mice received isotonic saline (NaCl) as control. Images were reconstructed in 60 time frames of 1 minute each using the MAP reconstruction algorithm.
Results: Activation of dopaminergic neurons resulted in an increase of [11C]raclopride signal alterations. Voxelwise paired testing revealed clusters of statistically significant differences in [11C]raclopride dynamics after CNO and NaCl injection (Fig. 1B).
Conclusions: Chemogenetical activation of dopaminergic neurons in mice induced changes in [11C]raclopride dynamics. We conclude that temporal dynamics of [11C]raclopride can be used as a surrogate marker for alterations in dopamine release.
BPS02-2
BrainPET: Neurotransmitter systems
Open-Field-PET: Reproducibility of detecting transient changes in D2/D3 receptor occupancy during drug competition studies on freely moving animals
G. Angelis1, G. Hart2, J. Eisenhuth1, A. Parmar3, K. Popovic1, K. Clemens2, G. Pascali3, G. Perkins3, A. Kyme1, W. Ryder1, R. Fulton4 and S. Meikle1
1The University of Sydney, Brain and Mind Centre, Faculty of Health Sciences, Sydney, Australia
2University of New South Wales, School of Psychology, Sydney, Australia
3Australian Nuclear Sciences and Technology Organization, Life Sciences Division, Kirrawee, Australia
4The University of Sydney, School of Physics, Sydney, Australia
Abstract
We have developed and previously presented a novel methodology for imaging awake rats1, which enables simultaneous measurements of changes in brain function and animal behaviour in response to pharmacological or environmental stimuli. The objective here was to assess the reproducibility of detecting transient changes in receptor occupancy induced by competitive binding at D2/D3 receptor cites.
Four adult Sprague-Dawley rats were injected with 48.1 ± 18.4 MBq (0.84 ± 0.16 nmol) of [11C]raclopride via a chronic indwelling jugular catheter and imaged for 60 min on the microPET Focus 220. During imaging the animals were free to move within a small enclosure (20x12 cm2), while their motion was tracked by a pair of optical cameras located on either side of the gantry2. Each rat acted as its own control. On day 1 the animals underwent a baseline scan, where a saline solution was injected via the catheter 20 min after radiotracer injection. On day 2 the animals received 2 mg/kg of unlabelled raclopride, again 20 min after radiotracer injection. Dynamic PET data were corrected for motion and regional time-activity curves from striatum and cerebellum were analysed using the lp-ntPET method3, which models transient changes in receptor occupancy by the radioligand.
For those animals administered with unlabelled raclopride we were able to reproducibly measure the displacement of [11C]raclopride from D2/D3 receptors, both in terms of activation onset (td = 19.5 ± 1.2 min) and magnitude (k2a = 1.85 ± 0.2 times the baseline). Additionally, no statistical significant displacement was detected for those animals that received a vehicle injection.
Our methodology allows for reproducible measurements of exogenous drug-induced radioligand displacement from D2/D3 receptors. Work is in progress to extend the evaluation of this methodology to measure endogenous dopamine release induced by amphetamine or an external stimulus and to correlate this temporally with behavioural outcomes.
BPS02-3
BrainPET: Neurotransmitter systems
Simultaneous PET/MRI depicts changes in serotonin transporters, glucose metabolism and multimodal brain connectivity patterns after pharmacological stimulation
M. Amend1, T. Watabe2, A. Thielcke1, R. Stumm1, J. Hatazawa2, B. Pichler1 and H. Wehrl1
1Eberhard Karls University Tuebingen, Medical Faculty and University Hospital, Department of Preclinical Imaging and Radiopharmacy, Werner Siemens Imaging Center, Tuebingen, Germany
2Osaka City University Graduate School of Medicine, Department of Nuclear Medicine and Tracer Kinetics, Osaka, Japan
Abstract
Objectives: The effects of pharmacological stimulations using 3,4-Methylenedioxymethamphetamine on serotonin transporter, glucose metabolism, hemodynamic parameters and multimodal brain connectivity were investigated by simultaneous PET/MRI, allowing exact spatial and temporal correlation of the modalities. Additionally, we applied new infusion protocols of PET tracers to detect changes between baseline and stimulation in a single session.
Methods: PET/MRI was performed on a preclinical 7 T scanner equipped with a PET insert. Male Lewis rats (n = 12) were anesthetized (1.5% isoflurane) and body temperature was stabilized at 37 °C. A BOLD-fMRI sequence (EPI, TE: 18 ms, TR: 2000 ms) was acquired over 100 min. Simultaneously, PET acquisitions were performed with bolus and constant infusion of [11C]DASB (bolus 64 ± 8 MBq / 0.48 ml + 15 µl/min) and a constant infusion of [18F]FDG (total: 117 ± 4.6 MBq / 0.8 ml) two days later. MDMA (3.2 mg/kg), respectively PBS for the control group, were i.v. injected after 40 min. Data were analyzed using SPM 12, GLM and PMOD. Changes in BOLD-fMRI and PET connectivity pattern were investigated by principal component analysis and correlation maps.
Results: TACs of [18F]FDG were increased after MDMA stimulation (Fig. 1A). Normalized [18F]FDG SUVs were significantly (p < 0.05) increased in Caudate Putamen and Orbitofrontal Cortex (Fig. 1B). BPND of [11C]DASB revealed a decrease after MDMA challenge (Fig. 1C). Largest signifikant decreases of BPND (p < 0.01) were observed in Thalamus, Midbrain, and VTA (Fig. 1D). The PBS injected control group revealed no increased [18F]FDG SUVs, nor decreased [11C]DASB BPND. Functional (BOLD-fMRI), molecular ([11C]DASB-PET) and metabolic ([18F]FDG-PET) correlation matrices of brain regions showed changes after MDMA administration (Fig. 1E) and detailed networks were identified.
Conclusions: Simultaneous PET/MR using constant infusion of PET tracers is promising to encode the effects of pharmacological challenges on the brain on a functional, molecular and metabolic scale delivering detailed views of networks and time scales impacted.
BPS02-4
BrainPET: Neurotransmitter systems
Low 5-HT1B receptor binding in the migraine brain: A PET study
M. Deen1,2,3, H.D. Hansen2, A. Hougaard1, S. da Cunha-Bang2,3, M. Noergaard2, C. Svarer2, S.H. Keller4, C. Thomsen5, M. Ashina1,3 and G.M. Knudsen2,3
1Rigshospitalet, Danish Headache Center, Dept. of Neurology, Glostrup, Denmark
2Rigshospitalet, Neurobiology Research Unit, Dept. of Neurology, Copenhagen, Denmark
3University of Copenhagen, Faculty of Health and Medical Sciences, Copenhagen, Denmark
4Rigshospitalet, Department of Clinical Physiology, Nuclear Medicine and PET, Copenhagen, Denmark
5Rigshospitalet, Department of Radiology, Copenhagen, Denmark
Abstract
Background: The pathophysiology of migraine may involve dysfunction of serotonergic signaling. In particular, the 5-HT1B receptor is considered a key player due to the efficacy of 5-HT1B receptor agonists for treatment of migraine attacks. To date, no study has investigated in vivo cerebral 5-HT1B receptor binding in the migraine brain.
Aim: To examine the cerebral 5-HT1B receptor binding in migraine patients without aura and compare it to controls.
Methods: 16 controls and 18 migraine patients outside of migraine attacks were examined with positron emission tomography (PET) and [11C]AZ10419369 for quantification of cerebral 5-HT1B receptor binding. Patients who reported migraine attacks <48 hours after the scan were excluded, leaving 14 migraine patients for the final analysis. The 5-HT1B receptor binding was assessed in predefined regions of interest involved in pain modulation.
Results: Patients had lower 5-HT1B receptor binding across regions of interest compared to controls (p = 0.04). In a whole-brain voxel-based analysis, we found that duration since last migraine attack correlated positively with 5-HT1B receptor binding in the dorsal raphe and in the midbrain.
Conclusion: Migraine patients have low 5-HT1B receptor binding in pain modulating regions, reflecting decreased receptor density. Further investigations are needed to determine whether these changes are migraine-specific or related to repeated activation of pain modulating areas. In the dorsal raphe 5-HT1B receptors may be temporarily downregulated during a migraine attack.
BPS02-5
BrainPET: Neurotransmitter systems
Individual waking EEG power is negatively related to adenosine receptor density measured with PET
D. Elmenhorst1, E. Hennecke2, E.-M. Elmenhorst2, T. Kroll1, D. Aeschbach2 and A. Bauer1
1Forschungszentrum Jülich, Institute for Neurosciences and Medicine INM-2, Jülich, Germany
2DLR, Aerospace Medicine, Cologne, Germany
Abstract
The spectral components of quantitative human wake EEG is a highly genetically determined phenotype which has been demonstrated in twin and test-retest studies. The alpha range (8–12 Hz) dominates the wake EEG. Its occurrence was found to be related to cerebral energy metabolism by an inverse relation in e.g. cortical fMRI BOLD and thalamic FDG PET studies. Adenosine is a neuromodulator directly linked to the energy metabolism by ATP breakdown. Individual differences in alpha rhythm occurrence might be based on adenosine receptors availability.
The objective of the current study was to investigate whether adenosine A1 receptor density is related to the alpha component in the wake EEG.
Methods: 14 healthy male volunteers (mean age 27.7 ± 5.4 years) participated in a dynamic 18F CPFPX bolus/infusions-PET study with blood sampling and metabolite correction. Regional distribution volumes (VT) were determined by calculating the tissue to plasma ratio during the steady state phase. Wake EEG was acquired repeatedly over two days and the alpha power determined by fast Fourier transformation (µV2) averaged over the sessions and in the respective range. Adenosine receptor availability in atlas based brain regions in proximity to the EEG derivations and known to be relevant in sleep wake regulation were correlated with the alpha density (Pearson's product moment correlational analyses).
Results: Significant negative correlations were found between the alpha power in the occipital recording (O2 electrode) and striatal (r = −0.59; p = 0.025) and thalamic (r = −0.60; p = 0.02) adenosine receptor availability. Similar associations were found for the central recordings (C4) and the anterior cingulate cortex (r = −0.64; p = 0.013) occipital cortex (r = −0.59; p = 0.03) and the thalamus (r = −0.6; p = 0.038).
Conclusion: Our findings suggest an individual functional relationship between A1 adenosine receptor availability and EEG alpha power density.
BPS02-6
BrainPET: Neurotransmitter systems
Endogenous mu-opioid receptor system mediates anticipatory reward processing in humans: A combined PET-fMRI study
L. Nummenmaa1, T. Saanijoki1, J. Tuulari1, L. Tuominen2, J. Hirvonen1, P. Nuutila1 and K. Kalliokoski1
1University of Turku, Turku PET Centre, Turku, Finland
2MGH Martinos Center for Biomedical Imaging, Boston, United States
Abstract
Objectives: Palatable foods carry a strong motivational power. Consequently, mere sight of a delicious cake might trigger a strong urge to eat even when the individual would be physiologically fully satiated. The endogenous opioid system and particularly the mu-opioid receptor (MOR) system is involved in hedonic and motivational processing of food, yet its contribution to anticipatory reward processing in remains unresolved. Here we tested directly whether individual differences in MOR predict anticipatory reward responses to food cues in humans.
Methods: We scanned 33 healthy non-obese individuals with positron emission tomography (PET) and [11C]-carfentanil, a ligand specific to mu-opioid receptors (MORs). Subsequently, the subjects underwent a functional magnetic resonance imaging (fMRI) experiment where they viewed pictures of appetizing and bland foods to trigger anticipatory reward responses. Availability of MORs in key components of the reward circuit (amygdala, thalamus, ventral and dorsal striatum, orbitofrontal and cingulate cortices) was computed, and subsequently used to predict fMRI reward responses to palatable versus non-palatable foods.
Results: Viewing the palatable versus non-palatable foods resulted in consistent activation in the brain's homeostatic and reward circuits, including amygdala, ventral striatum and hypothalamus. MOR availability in the reward circuit was negatively associated with the magnitude of the anticipatory reward responses measured with fMRI, most notably in orbitofrontal and cingulate cortices. Responses to non-hedonic objects (cars) were unaffected by MOR availability.
Conclusions: Individual variation in MOR expression predicts anticipatory reward responses when seeing palatable foods. MOR availability may be an important factor explaining why some people engage in food consumption when encountering food cues, and it might constitute a risk factor for weight gain and obesity.
BPS03-1
BrainPET: Quantification
[18F] ASEM PET imaging of the alpha 7 nicotinic cholinergic receptor: test retest and sex differences
D.F. Wong1, H. Kuwabara1, J. Roberts1, J. Brasic1, C. Mishra1, K. Kitzmiller1, M. McDonald1, L. Gapasin1, A. Nandi1, G. Wand2, A. Gjedde3 and A. Horti1
1Johns Hopkins University School of Medicine, Radiology, Baltimore, United States
2Johns Hopkins University School of Medicine, Medicine, Baltimore, United States
3University of Copenhagen, Neuroscience & Pharmacology, Copenhagen, Denmark
Abstract
Background/Objectives: Nicotinic cholinergic receptors nAchR have been studied principally for the more prevalent alpha4/beta2 subtype. We have been studying an alpha 7 selective subtype preclinically and in humans (Wong et al., 2014). As part of our validation, we are carrying out human occupancy studies, as well as extended test retest studies in men and women. In this presentation we concentrated on the age range under 50 to examine test retest and variability between men and women
Methods: Eight healthy non-smokers without neuropsychiatric disorders (age: 37.6 ± 12.3 years; 2 females) successfully completed test and retest scans with [18F] ASEM. In addition, a total of fifteen male (age: 40.6 ± 12.9; range: 18–57) and eight female (age: 25.6 ± 5.3; range: 20–34) healthy non-smokers were studied to understand gender differences of α7-nAChRs. Values of distribution volume (VT) were obtained in 22 MRI-derived regions (with left and right regions merged) using the plasma reference graphical analysis (Logan et al., 1990). Then, regional distribution ratios (pons-reference) were obtained. Estimates of test-retest variability (TRV) of VT and DVR were calculated for individual regions (TRV = mean across subjects of |test-retest|/mean of test and retest, in percentage).
Results: Regional estimates of TRV ranged from 5.1 to 7.3% for VT and from 1.7 to 3.8% for DVR. A two-way analysis of variance (ANOVA) showed strong gender (F = 76.37; p < 10−16; d.f. = 1) and region (F = 15.27; p < 10−38; d.f. 21) effects. Subsequent t-test suggested gender differences in the amygdala, cerebellum, lingual gyrus, parietal cortex, putamen, and ventral striatum (p < 0.05). However, after matching for ages for females (younger), we observed no gender differences.
Conclusion: This novel alpha 7 radiopharmaceutical shows promising reasonable test retest variability in an extended human sample with no definitive sex differences.
BPS03-2
BrainPET: Quantification
An automated algorithm to quantify brain amyloid load
A. Whittington1, D.J. Sharp1 and R.N. Gunn1,2,3
1Imperial College London, Brain Sciences, Medicine, London, United Kingdom
2Imanova Ltd, London, United Kingdom
3University of Oxford, Institute of Biomedical Engineering, Oxford, United Kingdom
Abstract
Objectives: The cortical standardised uptake value ratio (SUVr) derived from PET amyloid tracers has been used to quantify the β-amyloid burden in the brain using a range of different cortical target regions. In previous work, we have shown that the amyloid load (AβL) in AD, defined as the percentage of the maximal Aβ concentration, is constant across the whole cerebral cortex in an individual. We introduce an automated method of quantifying AβL, using the whole PET image volume, as a useful and robust biomarker for studies in neurodegenerative disease.
Methods: The method uses two template images in MNI152 space for the tracer non-specific binding (NS) and the Aβ carrying capacity (K). An individual PET amyloid scan is spatially normalised with DARTEL1 into MNI152 space, converted to an SUVr image by normalisation to the grey matter cerebellum reference region, and an image-wise regression performed to calculate AβL;
SUVr = AβLK + NS
In this work, we used previously derived [18F]-AV45 NS and K template images2 and applied the automated algorithm to 769 [18F]-AV45 PET scans from the ADNI database (HC = 211, EMCI = 223, LMCI = 203 and AD = 132) to estimate AβL.
Results: AβL was calculated automatically in less than one minute for each scan. The median AβL for AD was 45.5% compared to 29.2% for HC and there was an effect size of 0.62 (hedges g) between the two groups. The median AβL of both EMCI and LMCI was 34%.
[Amyloid Load calculation and results]
Conclusions: We have introduced a rapid and robust algorithm to calculate AβL that makes full use of the imaging data. AβL has the potential to be a valuable biomarker for neurodegenerative disease, either for stratification into clinical trials or the assessment of therapies.
References:
1. Ashburner, J.Neuroimage38,95–113 (2007).2. Whittington, A., Sharp, D. & Gunn, R.J. Nucl. Med.57,346 (2016).
BPS03-3
BrainPET: Quantification
Validation of an image-derived input function method for 15O-water PET/MR brain scans
M. Lubberink1,2, M.M. Khalighi3, L. Appel1,4, M. Engström5, G. Antoni4,6 and G. Zaharchuk7
1Uppsala University, Surgical Sciences, Uppsala, Sweden
2Uppsala University Hospital, Medical Physics, Uppsala, Sweden
3GE Healthcare, Applied Science Laboratory, Menlo Park, United States
4Uppsala University Hospital, Medical Imaging Centre, Uppsala, Sweden
5GE Healthcare, Applied Science Laboratory, Uppsala, Sweden
6Uppsala University, Medicinal Chemistry, Uppsala, Sweden
7Stanford University, Radiology, Stanford, United States
Abstract
Objectives: Accurate measurement of the arterial input function (AIF) is essential in quantitative analysis of cerebral blood flow (CBF) using 15O-water PET. The aim of the present work was to validate a newly proposed PET/MR method for estimation of image-derived input functions (IDIF)1 by comparison to arterial blood sampling.
Methods: Four subjects underwent 10 min dynamic PET/MR scans starting with the injection of 5 MBq/kg 15O-water. Arterial blood was sampled continuously and a sagittal vascular (inhance 3D velocity) MR image was acquired to establish the total volume of the cervical arteries. First-pass PET images were used to estimate spill-over and spill-in masks. IDIFs were estimated by dividing the total activity in the PET-based cervical artery mask by the MR-based arterial volume. Blood sampler data were corrected for delay and dispersion. Area under the curve (AUC) of IDIF and sampler curves were compared. A VOI mask was placed over a T1-MRI image and transferred to the dynamic PET images and CBF values were computed for each VOI based on IDIF and sampler data.
Results: AUC of IDIFs were on average 9% lower than sampler curves in the first pass-peak, and 11% lower for the whole scan. A high within-subject correlation (R2 > 0.98 for all subjects) between IDIF- and sampler-based CBF values was found, but slope between IDIF- and sampler-based CBF values varied between 0.72 and 1.03 across subjects.
Conclusions: The proposed IDIF method shows promising results, with a good representation of the arterial blood curves and a high within-subject correlation between IDIF- and sampler-based CBF values. However, in this small sample IDIFs tended to result in underestimation of CBF values and need further optimisation. Use of IDIFs would allow for truly non-invasive quantitative 15O-water CBF measurements using PET-MR and have possible applications for other tracers.
References:
1 Kahlighi et al., ISMRM 2016, p4321.
BPS03-4
BrainPET: Quantification
Comparison of Oxygen-15-water and arterial spin labeling measures of cerebral blood flow in healthy subjects and patients with cerebrovascular disease using TOF-enabled simultaneous PET/MRI
A.P. Fan1, P. Gulaka1, M.M. Khalighi2, J. Guo1, D. Holley1, H. Gandhi1, B. Shen1, P. Singh1, J. Park1, F.T. Chen1 and G. Zaharchuk1
1Stanford University, Radiology, Stanford, United States
2GE Healthcare, Menlo Park, United States
Abstract
Objectives: We compared 15O-water PET and arterial spin labeling CBF measurements in healthy subjects and cerebrovascular disease patients on a hybrid PET/MRI scanner, and explored different MR attenuation correction (MRAC) methods.
Methods: PET/MRI (Signa, GE Healthcare) was performed in 44 subjects (21 healthy subjects, 16 Moyamoya disease, and 7 carotid stenosis). The protocol included neck MRA for image-derived input function1 and 2-point Dixon and zero-TE (ZTE) scans for MRAC2. CBF was imaged twice at baseline to assess reproducibility, and once after 1 g acetazolamide. For each CBF imaging block, 15O-water injections (550–925 MBq) were performed simultaneously with 3 pseudo-continuous ASL (pcASL) sequences: single 2000 ms delay, sequential multidelay (5 delays, 700–3000 ms), and Hadamard-encoded multidelay (7 sub-boluses). Contrast perfusion-weighted imaging (PWI) was performed.
Results: In healthy subjects, all ASL CBF maps were qualitatively similar to PET, and showed CBF increases after acetazolamide. Relative reduced CBF was noted on single-delay ASL within basal ganglia and watershed regions, which was mitigated with multidelay ASL. In patients, single-delay ASL adequately measured CBF in about half of cases, while CBF errors due to arterial delay occurred in the rest. Improvements were seen with multidelay ASL. In a few patients with long delays, all ASL sequences underestimated CBF in affected regions relative to PET. Reproducibility of the modalities was similar, though ASL white-matter variability was higher. ZTE-based MRAC increased 15O-water CBF by 5–10% compared to atlas-based MRAC.
[CBF maps by PET/MRI in Moyamoya patient]
Discussion: Simultaneous PET/MRI is an ideal approach to assess CBF as it reduces errors related to physiological changes. Multidelay ASL improved CBF measurements, but longer scantime necessitates reduced spatial resolution. In patients with severely delayed flow, longer label/delay or velocity-selective ASL may be required3.
References:
1 Khalighi et al., Proc ISMRM 2016.2 Weisinger et al., MRM 2016.3 Wong et al., MRM 2006.
BPS03-5
BrainPET: Quantification
Modelling TSPO brain PET data including endothelial binding: considerations on tracer affinities
G. Rizzo1, M. Veronese2, C. Wimberley3, S. Lavisse3, M. Bottlaender3, P. Bloomfield4, O. Howes4,5, F.E. Turkheimer2 and A. Bertoldo1
1University of Padova, Department of Information Engineering, Padova, Italy
2King’s College London, Department of Neuroimaging, IoPPN, London, United Kingdom
3Paris-Sud University, CEA-DSV-I2BM, Paris, France
4Imperial College London, Institute of Clinical Sciences, London, United Kingdom
5King's College London, Department of Psychosis Studies, IoPPN, London, United Kingdom
Abstract
Objectives: The 18 kDa translocator protein (TSPO) is a marker for microglia activation and the main target for positron emission tomography (PET) ligands for neuroinflammation studies. Quantitative PET imaging of TSPO is a challenge [1]. Previous works showed that accounting for TSPO endothelial binding improves quantification of PET data [2–4]. This work aims to explore the relationship between the TSPO vascular component and tissue binding of tracers with different affinity including [11C]PK11195, [11C]PBR28 and [18F]DPA714.
Methods: First, we quantified [11C]PK11195 data from six healthy subjects with a one tissue compartmental model including TSPO endothelial binding (1TCM-1K), selected through model comparison. Then, we compared [11C]PK11195 Kb estimates (that describe the vascular binding) with previously published Kb estimates for [18F]DPA714 [4] and [11C]PBR28 [5]. We also compared, across tracers, the grey matter and white matter kinetic differences to evaluate tissue similarity and amenability to cluster analysis to determine a pseudo reference-region.
Results: The endothelial binding models improved [11C]PK11195 data description data (Fig.1A) and in the majority (66%) of the regions the 1TCM-1K was selected as the optimal model.
Across tracers, the vascular component estimates Kb were proportional to tracer affinity (Fig.1B), whereas the similarity of grey and white matter kinetics was related to affinity and was minimal for [18F]DPA714 (Fig.1C).
Conclusions: The presence of endothelial binding is common to all TSPO tracers and proportional to their affinity. Medium affinity tracers such as [18F]DPA714 maximize relative tissue tracer concentration hence grey and white matter tissue contrast and, besides improving signal-to-noise ratio, improve the potential for supervised reference tissue analysis.
References:
[1] Turkheimer et al. Biochem Soc Trans 2015.[2] Tomasi et al. JNM 2008.[3] Rizzo et al. JCBFM 2014.[4] Wimberley et al. NRM 2016.[5] Bloomfield et al. J Am Psych 2016.
BPS03-6
BrainPET: Quantification
Endothelial modelling of [11C]PBR28 brain PET data: a validation using XBD173 blocking and tissue analysis
M. Veronese1, T. Reis Marques2, P.S. Bloomfield3, G. Rizzo4, N. Singh1, D. Jones5, E. Agushi6, D. Mosses6, A. Bertoldo4, O. Howes2,3, F. Roncaroli6 and F. Turkheimer1
1King's College London, Neuroimaging, London, United Kingdom
2King's College London, Department of Psychosis Studies, London, United Kingdom
3Imperial College London, Institute of Clinical Sciences, London, United Kingdom
4Padova University, Department of Information Engineering, Padova, Italy
5Salford Royal Foundation Trust, Department of Cellular Pathology, Salford, United Kingdom
6University of Manchester, Division of Neuroscience and Experimental Psychology, Manchester, United Kingdom
Abstract
Introduction: The 18 kDa translocator protein (TSPO) is a marker of microglia activation in the central nervous system and represents the main target of radiotracers for the in vivo quantification of neuroinflammation with positron emission tomography (PET). TSPO PET is methodologically challenging given the heterogeneous distribution of TSPO in blood and brain [1]. Our previous studies with the TSPO tracers [11C]PBR28 and [11C]PK11195 demonstrated that a model accounting for TSPO binding to the endothelium improves the quantification of PET data [2,3]. Here we performed a validation of the kinetic model with the additional endothelial compartment through a displacement study.
Methods: Seven subjects with schizophrenia, all high-affinity binders, underwent two [11C]PBR28 PET scans before and after oral administration of 90 mg of the TSPO ligand XBD173. In addition we also studied TSPO expression in vessels using 3D reconstructions of histological post-mortem data of frontal lobe and cerebellum.
Results: Comparison with standard 2-tissue compartmental model confirmed that accounting for the endothelial modelling provides a more efficient data description and a better signal compartmentalization. With the vascular model, the TSPO blockage produced the expected changes in the concentration of the specific tissue compartment and vascular TSPO, whereas non-displaceable compartment remained unchanged (Figure). The results of the endothelial modelling were confirmed by the immunohistochemistry tissue analysis, demonstrating that TSPO positive vessels account for 30% of the vascular volume in cortical and white matter.
Conclusions: Accounting for endothelial TSPO gives a more precise and accurate quantification of [11C]PBR28 brain PET data that is consistent with TSPO neurobiology. The results are generalizable for any TSPO PET ligand.
References:
[1] Turkheimer et al. Biochem. Soc. Trans. 2015.[2] Tomasi et al. JNM 2008.[3] Rizzo et al. JCBFM 2014.
BPS04-1
BrainPET: Dementia
Antibody-based PET radioligands for imaging of amyloid-beta protofibrils
S. Syvänen1, X.T. Fang1, G. Hultqvist1, J. Fälting2, G. Antoni1,3, L. Lannfelt1 and D. Sehlin1
1Uppsala University, Uppsala, Sweden
2BioArctic AB, Stockholm, Sweden
3Uppsala University Hospital PET Centre, Uppsala, Sweden
Abstract
Objectives: The aim of the study was to develop antibody-based radioligands for PET imaging of soluble amyloid-β (Aβ) protofibrils, which are suggested to cause neurodegeneration in Alzheimer´s disease (AD) [1]. Antibodies are large molecules with low brain distribution. Thus, they must be engineered for more efficient passage across the blood-brain barrier (BBB) for use as PET radioligands. We have previously shown this concept to be feasible [2].
Methods: Protofibril selective antibody mAb158 [3] was fused with either TfR antibody 8D3 [4] or fragments of 8D3. Three different bispecific antibodies were generated. Binding to TfR enabled receptor mediated transcytosis across the BBB while the mAb158-moiety bound to Aβ in the brain parenchyma. The bispecific antibodies were then labeled with 124I and used for PET imaging in APP-transgenic and wild-type mice.
[TfR mediated transcytosis]
Results: All three bispecific antibodies bound both Aβ protofibrils and TfR, and displayed up to 80-fold better BBB transport compared to mAb158. Monovalent binding to TfR resulted in more efficient transport across the BBB compared to bivalent TfR binding. There was a clear difference between PET images obtained in AD and wild-type mice at three days post injection. The PET signal correlated closely with levels of Aβ protofibrils measured in brain homogenate.
Conclusions: Protofibrils were visualized in vivo with PET. To our knowledge this is the first time antibody-based ligands have been successfully used for imaging of a target inside the brain. In a longer perspective, the use of bispecific antibodies as PET ligands may enable imaging of proteins involved in diseases of the brain for which radioligands are lacking today.
References:
1. Esparza, et al. Ann Neurol (2013).2. Sehlin, et al. Nature Communications (2016).3. Englund, et al. J Neurochem (2007).4. Kissel, et al. Histochem Cell Biol (1998).
BPS04-2
BrainPET: Dementia
Investigation of the selectivity of Tau PET radioligand THK5351 in vitro and in vivo
Q. Guo1, B. Otterstaetter2, S. Benninghoff2, D. Reuter1, J. Wang1, M. Skaddan1, A. Tovcimak1, C. Schroeder1, G. Krishnan1, K. Wilcox1, R. Comley1, P. Makidon1, M. Voorbach1, A. Haupt2, M. Mugnaini2 and L. Martarello1
1AbbVie, North Chicago, United States
2AbbVie Deutschland GmbH & Co. KG, Neuroscience Research, Ludwigshafen, Germany
Abstract
Objectives: Tau protein has been investigated as a target for imaging and treatment of neurodegeneration diseases. Multiple tau PET radioligands have been developed in recent years; however the question about their selectivity in vivo remains. Here, we aim to evaluate of the off-target binding of THK5351 in human Alzheimer's disease (AD) tissue in vitro and in young and aged cynomolgus macaques (cynos) in vivo.
Methods: Homogenates from the entorhinal cortex from an AD donor (Braak stages V) were prepared and incubated with [3H]THK5351. Competitive binding assays were performed in the presence of various concentrations of THK5351 as well as clogyline and selegiline, in order to evaluate the total specific binding as well as binding to monoamine oxidase A and B (MAO-A and MAO-B) respectively.
Dynamic [18F]THK5351 PET scans were performed in 10 young (8–10yrs) and 4 geriatric (24–26yrs) cynos. 3 old cynos had a second scan post i.v. administration of 1 mg/kg selegiline. The total volume of distribution (VT) was estimated using kinetic modeling with metabolite corrected plasma input function.
Results: The in vitro binding assay demonstrated high levels of [3H]THK5351 binding to AD homogenates. 100 nM selegiline blocked 70% of the total binding. A 20% increase in VT was found in globus pallidus (p = 0.02) and striatum (p = 0.06) in aged cynos compared to the young. Following selegiline pretreatment, there was 20–60% reduction in VT across brain regions in aged cynos (Figure 1), suggesting MAO-B binding.
Conclusion: This dataset provides in vitro and in vivo evidence of THK5351 binding to MAO-B, the level of which is reported to increase with age. This non-tau binding should be considered as a confounding effect in longitudinal clinical trials using THK5351.
[Figure 1. VT parametric maps of [18F]THK5351]
BPS04-3
BrainPET: Dementia
Optimization of specific binding quantification using dynamic and static PET acquisition and comparison of 18F-THK-5317 and 18F-THK-5351
T. Betthauser1, P. Lao1, D. Murali1, T. Barnhart1, S. Furumoto2, N. Okamura2, C. Stone1, S. Johnson1 and B. Christian1
1University of Wisconsin – Madison, Madison, United States
2Tohoku University, Sendai, Japan
Abstract
Objectives: This work sought to determine optimal quantification strategies for estimating specific binding with dynamic and static imaging of PET tau radioligands 18F-THK-5351 and 18F-THK-5317.
Methods: Twenty-eight participants (mean = 71 ± 7 yrs) underwent 90-minute dynamic 18F-THK-5351 or 18F-THK-5317 PET scans and T1-w MRI. Ten participants were scanned with both tracers (321 ± 46 days between PET scans). Participants included individuals at-risk for Alzheimer's disease (AD) (18 APOE-ɛ4 carriers, 15 parental history) and included cognitively stable (n = 9), cognitively declining (n = 10), amnestic MCI (n = 3) and probable AD (n = 6). Bland-Altman and regression analyses were carried out to compare SUVR and DVR estimates determined using Logan graphical analysis, MRTM2, and basis function SRTM. The effects of t* (and k2 for Logan) were investigated for MRTM2 and Logan. The effects of scan duration and scanning window were determined for DVR and SUVR, respectively. Regressions analyses were performed to compare within-subject DVR and SUVR estimates for the ten participants that received both THK scans.
Results: MRTM2 and Logan DVR estimates were stable for t* = 30 min and t* = 40 min, respectively. Logan estimates were sensitive to combinations of t* and k2. Between-method DVR estimates were generally in agreement for 90-minute dynamic scans for both tracers, with the exception of SRTM, which failed to converge in some ROIs for THK-5317. Scan durations of 60-minutes using MRTM2 and 70-minutes using Logan produced DVR estimates consistent with 90-minute derived estimates. SUVR became temporally stable 50–70 minutes post-injection with SUVR from 30–50 minutes post-injection indicating the highest correlation with DVR. General agreement was observed between THK-5351 and THK-5317 DVR and SUVR estimates for individuals scanned with both tracers with THK-5351 having higher values.
Conclusions: THK-5351 has more favorable imaging characteristics for tau imaging compared to THK-5317. Shortened 60-minute dynamic scans are suitable for DVR estimation using MRTM2 for both THK-5351 and THK-5317 in this subject population.
BPS04-4
BrainPET: Dementia
Selegiline reduces brain [18F]THK5351 uptake
K.P. Ng1,2, T.A. Pascoal1, S. Mathotaarachchi1, M.S. Kang1, M. Shin1, J. Therriault1, S. Levasseur1, K. Horowitz1, P. Gravel3, G. Massarweh3, J.-P. Soucy3, S. Gauthier1 and P. Rosa-Neto1
1McGill University Research Centre for Studies in Aging, Montreal, Canada
2National Neuroscience Institute, Singapore, Singapore
3McGill McConnell Brain Imaging Centre, Montreal, Canada
Abstract
Objectives: [18F]THK5351 is a quinolone-derivative imaging agent with high affinity to tau paired helical filaments (PHF) and good kinetic profile.1 However, [18F]THK5351 also binds to brain regions with negligible concentrations of PHF (so called off-target binding). Given the previously reported affinity between quinolones and monoamine oxidase B (MAO-B), we conducted a study testing the effect of MAO-B inhibitor selegiline on [18F]THK5351 standardized uptake value (SUV).
Methods: 5 mild cognitive impairment (MCI; 70.1 ± 7.2 y.o; MMSE[29–30]), 2 Alzheimer's disease (AD; 59.5 ± 5.6 y.o; MMSE[15–16]) and 1 progressive supranuclear palsy (PSP; 68 y.o; MMSE = 29) individuals had a baseline [18F]THK5351 scan (6.6 ± 0.3 mCi). All participants had a second [18F]THK5351 scan (6.7 ± 0.4 mCi) conducted 1 hour following an oral dose of 10 mg selegiline at a mean interval of 7.8 ± 6.3 days from the baseline scan. Images were obtained using the ECAT HRRT (High Resolution Research Tomograph). The primary outcome measure was SUV, obtained using tissue radioactivity concentration 50 to 70 minutes following [18F]THK5351 injection normalized by body weight and injected tracer dose.
Results: At baseline, the mean SUVs were highest in the basal ganglia (0.64 ± 0.11) and thalamus (0.62 ± 0.14), brain regions rich in MAO-B. In the follow-up scan, the SUVs in the brain regions were reduced in the range of 36.7% – 51.8%, with the greatest decline noted in the thalamus (51.8%) and basal ganglia (51.4%) (Figure 1), presumably due to the blocking of MAO-B binding sites by selegiline.
Conclusion: Quantification of [18F]THK5351 binding is affected by concomitant use of selegiline, particularly those of off-target binding. These results merit careful studies focusing on the underlying basis of this phenomenon, particularly interactions between [18F]THK5351 and MAO-B binding sites.
Reference:
1 Harada et al. “18F-THK5351: a novel PET radiotracer for imaging neurofibrillary pathology in Alzheimer disease." Journal of Nuclear Medicine 57.2 (2016): 208–214.
BPS04-5
BrainPET: Dementia
PET tau and amyloid levels in default mode network synergistically determine clinical status in predementia stages of Alzheimer's disease
T.A. Pascoal1, S. Mathotaarachchi1, M. Shin1, M.S. Kang1, K.P. Ng1, J.-P. Soucy2, S. Gauthier1 and P. Rosa-Neto1
1McGill University, Montreal, Canada
2Montreal Neurological Institute, Montreal, Canada
Abstract
Objective: To test the hypothesis that the synergy between, rather than the independent or additive effects of, amyloid and tau pathologies determines clinical status in the predementia stages of Alzheimer's disease (AD).
Methods: We assessed 80 ADNI individuals (41 cognitively normal individuals and 39 mild cognitive impairment) with magnetic resonance, [18F]florbetapir amyloid and [18F]AV1451 tau positron emission tomography (PET) as well as clinical status assessed with Clinical Dementia Rating Sum-of-Boxes. Using a novel PET analytical framework that allows complex multimodal voxel-wise operations (1), a voxel-wise interaction model was build to test the main and interactive effects of amyloid and tau PET on clinical status, accounting for age, gender, education, APOE ɛ4, and grey matter density at every voxel. We further carried out an analysis of variance to compare the synergistic with the mediation and additive models.
Results: We found that the synergism between amyloid and tau pathologies in the precuneus, posterior cingulate, inferior parietal, lateral temporal and medial prefrontal cortices were associated with worse clinical status in the predementia stages of AD (P < 0.001). In addition, analysis of variance strongly supported that the model with the interaction term best described clinical status, as compared to reduced models testing (A)amyloid; (B)tau; and (C)amyloid plus tau with P < 0.0001 in all three cases.
Conclusions: These results highlight the concept that the synergy between amyloid-β and tau determines clinical status in the predementia stages of AD (2). Importantly, the regions where such a synergy determined worse clinical status were confined to the functional hubs of the brain's default model network, which are postulated fundamental to the large-scale brain organization.
References:
(1) Mathotaarachchi et al. (2016). VoxelStats: A MATLAB Package for Multi-Modal Voxel-Wise Brain Image Analysis. Front Neuroinform.(2) Pascoal et al. (2016). Amyloid-beta and hyperphosphorylated tau synergy drives metabolic decline in preclinical Alzheimer´s disease. Mol Psychiatry.
BPS04-6
BrainPET: Dementia
Amyloid-β plaque accumulation and glucose hypometabolism in non-demented adults with Down syndrome demonstrate pattern of association observed in Alzheimer's disease
P. Lao1, T. Betthauser1, J. Price2, W. Klunk2, P. Bulova2, S. Hartley1, R. Hardison2, R. Tumuluru2, D. Murali1, C. Mathis2, A. Cohen2, T. Barnhart1, D. Tudorascu2, D. Devenny3, S. Johnson1, B. Handen2 and B. Christian1
1University of Wisconsin-Madison, Madison, United States
2University of Pittsburgh, Pittsburgh, United States
3New York Institute for Basic Research in Developmental Disabilities, New York City, United States
Abstract
Objective: The aim of this work is to investigate the association between amyloid-β plaque accumulation and neuronal dysfunction underlying the memory deficits characteristic of Alzheimer's disease (AD) in adults with Down syndrome (DS), a population predisposed to the overproduction of amyloid-β.
Methods: Twenty-three non-demented adults (38.6 ± 7.0 yrs; 12 M,11 F) with DS underwent [11C]PiB and [18F]FDG PET scans (124.7 ± 102.6 days apart). Standard uptake value ratios (SUVRs) were calculated for [11C]PiB and [18F]FDG using 50–70 min and 40–60 min post-injection, respectively. Both tracers used cerebellar gray matter as reference region. Parametric SUVR images were spatially normalized using tracer-specific templates and analyzed in MNI space. Biological parametric mapping (BPM; MATLAB2011a, SPM5) was used to investigate voxelwise associations between PiB and FDG SUVR images, adjusted and unadjusted for chronological age and sex using a linear regression model. A cluster size threshold of 1000 voxels was used at α = 0.001 significance level, uncorrected.
Results: Statistically significant negative associations between PiB SUVR and FDG SUVR were found in the precuneus and temporal cortex (r = 0.62–0.84) in unadjusted as well as in the adjusted analysis. There were no significant clusters of positive association between PiB and FDG SUVRs.
Conclusions: An elevated amyloid-β plaque burden was associated with glucose hypometabolism, suggesting that neuronal dysfunction is spatially concordant with amyloid-β plaque accumulation in these non-demented adults with DS. Furthermore, hypometabolism in the precuneus and temporal cortex is often observed in AD pathogenesis, and is spatially distinct from the hypometabolism in the anterior cingulate and frontal cortex observed with aging (Loessner 1994; Herholz 2002).
Research Support: R01AG031110, P30 HD03352, U01AG051406
BPS05-1
BrainPET: Neurology
Chemogenetic modulation of the pedunculopontine nucleus coupled with 11C-PHNO uptake reveals striatal dopamine release accompanies profound motor recovery in a rodent model of Parkinson's disease
P. Sharma1, L. Wells2, C. Coello2, S. Lanzarone2, S.-P. Tang2, I. Rabiner2, R. Gunn2, G. Gillies1, D. Dexter1 and I. Pienaar3
1Imperial College London, London, United Kingdom
2Imanova Ltd, London, United Kingdom
3Northumbria University, Newcastle upon Tyne, United Kingdom
Abstract
Introduction: The pedunculopontine nucleus (PPN) has prominent extrapyramidal motor associations (particularly gait and postural control). It is also notable for its dense population of cholinergic neurons, which are known to be depleted in Parkinson's disease (PD). Deep brain stimulation (DBS) of the PPN has been variably efficacious in restoring postural and gait stability, inviting intensive study for its refinement. We have previously shown that selective PPN cholinergic stimulation leads to dramatic motor recovery in a rodent model of PD. Here we report for the first time, a chemogenetic approach coupled with 11C-PHNO uptake in a rodent PD model to determine the effect of selective PPN-cholinergic stimulation on striatal dopaminergics, whose potentiation may underlie motor recovery in PPN DBS.
Methods: 18 rodents were inoculated with an adenoviral vector carrying the sequence for a human G-protein receptor ('hM3Dq'), into the left PPN; and either lactacystin toxin injection (n = 10) or saline sham injection (n = 8) into the left substantia nigra. Baseline and 5-week post-surgery behaviours were conducted; the latter with pre- and post- hM3Dq stimulation with Clozapine-N-Oxide (CNO). At 5 weeks, rodents underwent gamma counts of extracted striatal tissues, with or without PPN stimulation.
Results: Significant motor recovery was seen on behavioural assessment with PPN stimulation. Gamma counts of extracted striatal tissue expressed as standardised uptake values (SUVs) normalised to cerebellum show a pre-stimulation mean difference of 54.1% (SEM ± 6.7%) between lesioned and unlesioned sides. Upon cholinergic PPN stimulation, the mean difference between lesioned and unlesioned sides fell to 30.2% (SEM ± 4.5%; p = 0.03), implying endogenous dopamine release occurs in the ipsilateral striatum as a result of stimulation.
Conclusions: The highly novel combination of chemogenetics with radiotracer uptake is a powerful new tool, which here links PPN cholinergic activity with striatal dopamine release, associated with profound motor recovery in a rodent PD model.
BPS05-2
BrainPET: Neurology
Altered adenosine 2A and dopamine D2 receptor availability in the 6-hydroxydopamine-treated rats with and without levodopa-induced dyskinesia
X. Zhou1, J. Doorduin1, P. Elsinga1, R. Dierckx1, E. de Vries1 and C. Casteels1,2
1University Medical Center Groningen, Groningen, Netherlands
2University of Leuven, Leuven, Belgium
Abstract
Several lines of evidence imply alterations in adenosine signaling in Parkinson's disease (PD). Here, we investigated cerebral changes in adenosine 2A receptor (A2AR) availability in 6-hydroxydopamine (6-OHDA)-lesioned rats with and without levodopa-induced dyskinesia (LID) using PET with [11C]preladenant. In parallel dopamine type 2 receptor (D2R) imaging with [11C]raclopride PET and behavioral tests for motor and cognitive function were performed.
Methods: Parametric A2AR and D2R binding potential (BPND) images were reconstructed using reference tissue methods. All images were analyzed using volume-of-interest (VOI) and voxel-based approaches.
Results: On the behavioral level, 6-OHDA-lesioned rats showed asymmetry in forepaw use and deficits in spatial memory and explorative behavior as compared to the sham-operated animals. 15-Days of levodopa treatment induced dyskinesia but did not alleviate motor deficits in PD rats. Intranigral 6-OHDA injection significantly increased D2R binding in the lesioned striatum (BPND: 2.69 ± 0.40 6-OHDA vs. 2.31 ± 0.18 sham,+16.6%; p = 0.03), whereas levodopa treatment did not affect the D2R binding in the ipsilateral striatum of the PD rats. In addition, intranigral 6-OHDA injection decrease the affect-non-affected BPND ratio of [11C]preladenant in striatum (ratio: 0.94 ± 0.03 6-OHDA vs. 1.00 ± 0.02 sham; −6.1%; p = 0.01). Levodopa treatment significantly increased A2AR binding in the affected striatum (BPND: 6.02 ± 0.91 L-DOPA vs. 4.90 ± 0.76 saline; +23.4%; p = 0.02). In PD rats with LID, positive correlations were found between D2R and A2AR BPND values in the ipsilateral striatum (r = 0.88, ppeak = 8.56.10−4 uncorr), and between AIM score and the D2R BPND in the contralateral striatum (r = 0.98; ppeak = 9.55.10−5 uncorr).
Conclusion: PET imaging of striatal A2AR indicates that A2AR availability changes during the course of PD development and treatment. The observed correlations of striatal D2R availability with A2AR availability and with AIM score may provide new knowledge on striatal physiology and new possibilities to further unravel the functions of these targets in the pathophysiology of PD.
BPS05-3
BrainPET: Neurology
Effects of MPTP on serotonergic neuronal systems and mitochondrial complex I activity in living brain: A PET study in conscious rhesus monkeys
and H. Tsukada1
1Hamamatsu Photoincs KK, Central Research Laboratory, Hamamatsu, Japan
Abstract
Objectives: The objective of the present PET study was to assess the effects of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) on serotonergic neuronal system in comparison with dopamine and MC-I activity in the living brain of rhesus monkey (Macaca mulatta) under conscious condition.
Methods: Parkinson's disease (PD) model monkeys were prepared by repeated MPTP administration. 11C-DASB for serotonin transporter (SERT), 18F-MPPF for serotonin 1A receptor (5-HT1AR), 11C-PE2I for dopamine transporter (DAT), 6-11C-methyl-m-tyrosin (6MemTyr) for dopamine synthesis, 11C-raclopride for dopamine D2 receptor (D2R), or 18F-BCPP-EF for MC-1 was intravenously injected into normal and MPTP monkeys. Serotonin as well as dopamine parameters were calculated using time activity curves in the cerebellum as input functions. The total distribution volume (VT) of 18F-BCPP-EF was determined using Logan plot graphical analysis using metabolite-corrected plasma input function.
Results: No significant differences in the postsynaptic parameter of D2R binding were seen in either striatum or substantia nigra pars compacta (SNc). In contrast, significant reductions of the presynaptic dopamine parameters were detected by PET in both striatum and SNc of MPTP monkeys compared with normal ones. The binding of SERT, but not 5-HT1AR, revealed to be lower in the extra-striatal regions including the frontal cortex, mid brain, and limbic system. MPTP-induced diffuse reductions of MC-I activity was observed throughout the brain.
Conclusions: These results obtained by multiparametric PET measurements demonstrated that the chronic MPTP treatments induced the reductions of not only dopaminergic system in nigrostriatal pathway but also of SERT in mid-brain and cortical regions of PD model monkey. These results suggested that the neurotoxicity of MPTP was not exclusive in nigrostriatal pathway as predicted from the MC-1 damage in the extra-striatal regions of brain.
BPS05-4
BrainPET: Neurology
PET imaging of [11C]PBR28 in patients with Parkinson´s disease does not indicate increased binding to TSPO
K. Varnäs1, Z. Cselényi2, A. Jucaite2, C. Halldin1, P. Svenningsson1, L. Farde1,2 and A. Varrone1
1Karolinska Institutet, Clinical Neuroscience, Stockholm, Sweden
2AstraZeneca, Translational Science Centre, PET CoE at Karolinska Institutet, Stockholm, Sweden
Abstract
Previous imaging studies using the 18 kDa translocator protein (TSPO) PET radioligand [11C]PK11195 as marker of activated microglia have shown elevated TSPO binding in the brain of patients with Parkinson's disease (PD; 1). The initial findings have, however, not been supported by a recent investigation using the second generation TSPO radioligand [18F]-FEPPA (2). In the present study, we used another second generation radioligand, [11C]PBR28, for quantitative assessment of brain TSPO in 16 control subjects and 16 PD patients. All subjects underwent PET measurements using [11C]PBR28 and the high-resolution research tomograph (HRRT) system. In all PD patients, a PET measurement with the dopamine transporter radioligand [18F]FE-PE2I was also performed to confirm the presence of nigrostriatal degeneration. The [11C]PBR28 total distribution volume (VT) was estimated using the wavelet-aided parametric imaging approach (3), based on a multi-linear version of Logan graphical analysis. Regions of interest included nigrostriatal regions, limbic cortices and the brainstem. Based on genotype analysis of the TSPO rs6971 polymorphism, sixteen subjects (8 control subjects and 8 PD patients) were identified as high-affinity binders and the remaining 16 subjects were identified as mixed-affinity binders. For all regions evaluated, there was a main effect of genotype on [11C]PBR28 VT, but no statistically significant effect group or group x genotype interactions. VT values for [11C]PBR28 did not correlate with [18F]FEPE2I binding to the dopamine transporter in the putamen or the substantia nigra. The findings do not support the hypothesis of elevated brain TSPO binding in PD.
References:
1. Gerhard, A. Clin Transl Imaging (2016) 4:183–190.2. Koshimori, Y. et al. PLoS One. (2015) doi: 10.1371/journal.pone.0138721.3. Cselényi, Z et al. Neuroimage (2006) 32: 1690–708.
BPS05-5
BrainPET: Neurology
Inter-ictal [15O]H2O/PET differentiates between drug-resistant and seizure-free mesial temporal lobe epilepsy patients
J. Ryan1, M. Feldmann1,2, M. Walker1,2, V. Charalambous1, J. Anton-Rodriguez1, S. Wang1, M. Bauer1,3, R. Hinz1, M. Koepp2 and M.-C. Asselin1
1The University of Manchester, Manchester, United Kingdom
2Institute of Neurology, London, United Kingdom
3Medical University of Vienna, Vienna, Austria
Abstract
Hallmarks of mesial temporal lobe epilepsy (TLE) include inter-ictally reduced glucose metabolism, measured semi-quantitatively using [18F]FDG/PET, and ictally increased regional cerebral blood flow (rCBF), measured by subtraction [99mTc]HMPAO/SPECT. The sensitivity of PET and SPECT to detect focal abnormalities highly depends on seizure frequency and proximity to last seizure, which are difficult to predict. State (seizure)-independent diagnostic tools would facilitate assessment.
Objective: To assess whether quantitative rCBF measurements from a single [15O]H2O/PET scan can differentiate between drug-resistant (DR) and seizure-free/drug-sensitive(DS) TLE patients.
Methods: Fifteen healthy volunteers (HV), eight DS and 13 DR TLE patients with unilateral hippocampal sclerosis underwent a 6-min inter-ictal [15O]H2O scan on the HRRT brain PET camera with arterial blood sampling. Segmented atlas-based regions of interest (ROIs: frontal lobe FL, temporal lobe TL, combined hippocampus and amygdala Hi/Am, whole brain gray matter GM) were defined, the one-tissue compartment model was used to estimate rCBF and the asymmetry index (AI) calculated.
Results: rCBF differed significantly depending on group (p = 0.0151) and region (p < 0.0001). Compared to HV, DR patients demonstrated higher rCBF across all regions except for the ipsilateral Hi/Am where rCBF was marginally higher in HV compared to both patient groups (fig. 1a). We found asymmetrical CBF (fig. 1b) with lower CBF ipsilaterally in TL regions of DR patients (−6.81 ± 6.8%, p = 0.017) and even greater CBF asymmetry in the Hi/Am (−15.1 ± 14%, p = 0.017). There was no difference in hippocampal volumes between DR and DS patients, and both patient groups were taking anti-epileptic drugs, excluding the possibility of the underlying pathology or treatment causing the rCBF differences between patients.
Conclusions: Active epilepsy results in inter-ictal global increases of CBF, which is attenuated in the epileptic focus. Asymmetries of mesial-temporal CBF detected in the majority of unilateral mesial TLE patients raises the possibility of quantitative [15O]H2O PET being a localising diagnostic marker even in the inter-ictal state.
[rCBF in mTLE]
BPS05-6
BrainPET: Neurology
In vivo tau and amyloid pathology in Corticobasal Degeneration (CBD)
H. Wilson1, F. Niccolini1, S. Hirschbichler2, G. Pagano1, R. Erro2, T. Yousaf1, A. Whittington3, R. Gunn3,4, E. Rabiner4,5, K. Bhatia2 and M. Politis1
1King's College London, Department of Basic and Clinical Neuroscience, Neurodegeneration Imaging Group, London, United Kingdom
2UCL Institute of Neurology, Sobell Department of Motor Neuroscience, London, United Kingdom
3Imperial College London, Division of Brain Sciences, Department of Medicine, London, United Kingdom
4Imanova Ltd, Centre for Imaging Sciences, Hammersmith Hospital, London, United Kingdom
5King's College London, Institute of Psychiatry, Psychology and Neuroscience, Department of Neuroimaging, London, United Kingdom
Abstract
To assess in vivo the anatomical distribution of Tau aggregates with [18F]T807 PET and β-amyloid retention with [18F]Florbetapir PET molecular imaging, in patients with Corticobasal Degeneration (CBD).
Six CBD patients (3 males; mean age ± SD: 71.3 ± 7.4 years; mean disease duration ± SD: 5.2 ± 2.9 years) and 6 healthy controls (3 males; mean age ± SD: 72.5 ± 6.5 years) underwent a [18F]T807 PET scan (mean injected dose ± SD: 164.3 ± 5.1 MBq) and [18F]Florbetapir PET scan (mean injected dose ± SD: 158.1 ± 1.6 MBq) and a 3-T MRI scan. Image processing and kinetic modelling was carried out using MIAKAT™. Standardised uptake value ratio (SUVR) were generated for [18F]T807 SUVR80–100 relative to cerebellum crus grey matter and [18F]Florbetapir SUVR50–60 relative to cerebellum grey matter.
CBD patients showed 10–30% increases in cortical tau deposition compared to the group of healthy controls. In CBD patients, [18F]T807 SUVR80–100 were significantly increased in the medial frontal cortex (+10%; P < 0.05), posterior medial frontal cortex (+13%; P < 0.05), supplementary motor area (+ 21%; P< 0.05), precentral gyrus (+17%; P < 0.05), parietal lobe (+21%; P < 0.05), parietal lobule (+30%; P < 0.01), postcentral gyrus (+17%; P P < 0.05), precuneous (+22%; P < 0.05), cuneus (+22.2%; P < 0.05), occipital fusiform gyrus (+26%; P < 0.01) and lingual gyrus (+15%; P < 0.05). Asymmetrical tau deposition with greater [18F]T807 retention contralateral to the most affected side was observed in CBD patients. CBD patients had [18F]Florbetapir uptake within normal levels.
Cortical aggregates of tau pathology are present in CBD patients with low amyloid burden. The presence of tau depositions could have implications for cognitive impairment. Collection and analysis of arterial metabolite data for [18F]T807 is ongoing.
BPS06-1
BrainPET: Psychiatry
Evaluation of the effect of glucocorticoids on TSPO expression in nonhuman primate brain using positron emission tomography
A. Colasanti1,2, Q. Guo3, P.B. Jacobson3, D. Reuter3, A. Tovcimak3, J.-Q. Wang3, C. Schroeder3, M. Voorbach3, P. Makidon3, J.D. Beaver3, R.N. Gunn4, E.A. Rabiner4 and R. Rajagovindan3
1Brighton and Sussex Medical School, University of Sussex, Falmer, Brighton, United Kingdom
2Institute of Psychiatry, Psychology and Neuroscience, King's College London, London, United Kingdom
3AbbVie, North Chicago, United States
4Imanova Ltd, London, United Kingdom
Abstract
Background: Both acute and chronic treatments with glucocorticoids have been demonstrated in rodents ex-vivo to prime the responsiveness of hippocampal microglial cells to a subsequent immune challenge with LPS. Positron emission tomography (PET) with 18 kDa translocator protein (TSPO) radioligands such as [18F]PBR111 enables assessment of microglia and thus inflammation in vivo. Here we investigate the effects of glucocorticoids on microglia in vivo using [18F]PBR111 in healthy non-human primates.
Methods: Five female cynomolgus macaques had a 120 min dynamic [18F]PBR111 PET scans with arterial blood sampling before and after sub-chronic treatment with prednisolone (a loading dose of 30 mg/kg; p.o. on day 1 and once daily 10 mg/kg p.o. for the ensuing 6 days). Logan graphical method was used to estimate the total distribution volume (VT). The hippocampus was chosen as a priori region of interest based on previously reported findings. The whole brain was also studied to assess whether the effects of glucocorticoids on microglia were localized or global. The distribution volume ratio (DVR) using white matter as a pseudo reference region was also investigated.
Results: [18F]PBR111 VT in the whole brain and hippocampus were increased after treatment with prednisolone in all five subjects (Z = 2.02, p < 0.05). The median increases were 8.3% in the hippocampus (range: 6%-71%;) and 10.7% in the whole brain (range: 2%-74%). There was a trend toward DVR increase in the hippocampus (Z = 1.75, p = 0.08) but not in the whole brain.
Conclusion: Sub-chronic treatment with high dose of glucocorticoids in nonhuman-primates induced an increase in [18F]PBR111 brain binding primarily in the hippocampus, which may reflect the priming effect of glucocorticoids on microglia. Our experimental paradigm may be a useful approach to quantify microglial priming in the living brain.
BPS06-2
BrainPET: Psychiatry
Spatially coupled abnormalities in striatal dopamine density and functional connectivity in depression using simultaneous PET/MRI
T. Hjørnevik1,2,3, J.P. Hamilton4, M.D. Sacchet5, B. Knutson6, D. Holley2, B. Shen2, M. Khalighi7, F.T. Chin2, G.H. Glover2, G. Zaharchuk2 and I.H. Gotlib6
1Oslo University Hospital, The Intervention Centre, Oslo, Norway
2Stanford University, Department of Radiology, Stanford, United States
3The Norwegian Medical Cyclotron Centre, Oslo, Norway
4Linköping University, Center for Social and Affective Neuroscience, Linköping, Sweden
5Stanford University, Department of Psychiatry and Behavioral Sciences, Stanford, United States
6Stanford University, Department of Psychology, Stanford, United States
7GE Healthcare, Applied Sciences Laboratory, Menlo Park, United States
Abstract
Objectives: Using simultaneous PET/MRI, we investigated striatal D2-receptor availability and its relation with neural functional connectivity in Major Depressive Disorder (MDD).
Methods: 16 adult patients diagnosed with MDD and 14 healthy controls (HCs) were imaged with [11C]Raclopride PET and resting-state fMRI (rs-fMRI) using a simultaneous PET/MR scanner (Signa, GE Healthcare). D2-receptor binding potential (BPND) was measured from a 15-minute dynamic acquisition, using the voxel-based simplified reference tissue model (SRTM) with cerebellum as reference tissue (PMOD 3.7). rs-fMRI was acquired simultaneously during two separate six-minute epochs and corrected for artifacts from respiratory, cardiac, and regional variations in coil sensitivity, and both PET and rs-fMRI were warped to standard MNI space. We conducted voxel-wise, group-level analysis of BPND (SPM12). Then, using a seed region based on differential ventral striatal BPND between HC and MDD, we constructed functional connectivity maps from rs-fMRI.
Results: We found increased (t-test; cluster >50; p < 0.05) BPND in MDD in the bilateral putamen, right caudate, and bilateral frontal cortices (blue; Fig1A). Occipital and temporal cortices showed reduced D2-receptor density (red). Although [11C]Raclopride is a low-affinity PET tracer, the high sensitivity (23.3 cps/kBq1) of the GE PET/MRI scanner may allow detection of extra-striatal D2 binding. Moreover, using rs-fMRI, we found greater functional connectivity between left and right ventral striatal regions in MDD than in HC (Fig1B).
[Altered BPnd and functional connectivity in MDD]
Conclusions: Using a high-sensitivity, simultaneous PET/MRI scanner, we show increased D2-receptor availability bilaterally within both striatal and extra-striatal regions in MDD. Given the lower D2-receptor density outside the striatum, findings in these regions should be viewed with caution. We also found increased functional connectivity between bilateral ventral striatal regions in MDD, suggesting that cross-hemispheric communication within the striatum is related to increased dopamine receptor availability.
References:
1 Grant et al., Med Phys 2016.
BPS06-3
BrainPET: Psychiatry
Association between cortical thickness and MAO-A binding in subjects suffering from seasonal affective disorder
R. Seiger1, G.M. James1, M. Spies1, C. Vraka2, L. Nics2, M. Hienert1, C. Philippe2, P. Baldinger-Melich1, W. Wadsak2,3, M. Mitterhauser2,4, S. Kasper1, D. Winkler1 and R. Lanzenberger1
1Medical University of Vienna, Department of Psychiatry and Psychotherapy, Vienna, Austria
2Medical University of Vienna, Department of Biomedical Imaging and Image-guided Therapy, Division of Nuclear Medicine, Vienna, Austria
3Center for Biomarker Research in Medicine (CBmed), Graz, Austria
4Ludwig Boltzmann Institute for Applied Diagnostics, Vienna, Austria
Abstract
Objectives: Seasonal affective disorder (SAD) is a subtype of affective disorders in which patients exhibit depressive symptoms in the winter months. Alterations in serotonergic neurotransmission have been shown in SAD [1]. Studies indicate that the monoamineoxidase-A (MAO-A) may exert an influence on gray matter volume in distinct brain regions [2]. However, whether there is a direct association between cortical thickness and MAO-A binding in cortical brain areas in SAD patients remains to be determined.
Methods: 13 SAD patients (mean age ± SD = 33.08 ± 10.34) and 13 age matched healthy controls (36.69 ± 11.66) underwent MRI (T1-weighted, 0.88x0.47x0.47) and PET scanning (3.13x3.13x4.25). Voxelwise kinetic modeling was carried out in PMOD 3.5 using the Logan plot approach. Co-registration with the respective MR scan and surface projection of PET data was carried out with Freesurfer 5.3. ROI-based extraction of MAO-A binding potential and cortical thickness values were delineated using the Desikan-Killiany atlas.
Results: In the SAD cohort, moderate to strong positive correlations between cortical thickness and MAO-A binding in the superior frontal cortex (Spearman's rho = 0.53; P = 0.07, two-tailed) and the caudal anterior cingulate cortex (r = 0.76; P = 0.002) have been observed. Additionally, weak correlations were evident in the control group in certain brain regions but did not reach statistical significance.
Conclusions: Our data suggest a distinct structure-function relationship between cortical thickness and MAO-A binding in specific brain regions of subjects suffering from SAD disorder as well as in healthy controls. Lower correlations in the control group indicate weaker associations between cell density and MAO-A protein concentrations.
References:
[1] Spindelegger C, Stein P, Wadsak W, et al. Light-dependent alteration of serotonin-1A receptor binding in cortical and subcortical limbic regions in the human brain. World J Biol Psychiatry. 2012;13(6):413–22.[2] Cerasa A, Gioia MC, Labate A, et al. MAO A VNTR polymorphism and variation in human morphology: a VBM study. Neuroreport. 2008;19(11):1107–10.
BPS06-4
BrainPET: Psychiatry
Inverse relationship between central serotonin and noradrenaline transporter availability in humans with high BMI range – a potential biological mechanism in obesity
S. Hesse1,2, M. Rullmann1,2,3, A. Bresch1, M. Heinicke1, C. Schincke4, F. Vettermann5, M. Hankir2, J. Luthardt1, G.-A. Becker1, M. Patt1, P.M. Meyer1, M. Fasshauer2,4, M. Blueher1, W. Fenske1,2, F. Then Bergh4, A. Hilbert2,6, Y.-S. Ding7 and O. Sabri1,2
1University of Leipzig, Department of Nuclear Medicine, Leipzig, Germany
2Integrated Research and Treatment Centre (IFB) Adiposity Diseases, Leipzig, Germany
3Max Planck Institute for Cognitive and Human Brain Sciences, Leipzig, Germany
4University of Leipzig, Department of Neurology, Leipzig, Germany
5Ludwig-Maximilians-Universität Munich, Department of Nuclear Medicine, Munich, Germany
6University of Leipzig, Department of Medical Psychology and Medical Sociology, Leipzig, Germany
7New York University School of Medicine, Departments of Radiology and Psychiatry, New York, Germany
Abstract
Objective: The brain neurotransmitters serotonin (5-HT) and noradrenaline (NA) are both implicated in the regulation of appetite and energy balance. Disturbances of these systems, i.e. in key areas of feeding control such as the hypothalamus, lead to eating disorders and obesity. The mechanism by which an impaired 5-HT/NA signalling contributes is still unclear.
Methods: A PET study using either central 5-HT transporter (5-HTT)- or NA transporter (NAT)-selective [11C]DASB/[11C]MRB in 65 lean-to-highly obese individuals (BMI range: 19.1–54.1 kg/m2). 5-HTT/NAT binding potential BPND was obtained as PET outcome measures and correlated with BMI.
Results: 5-HTT BPND and NAT BPND tend towards an inverse relationship with increasing BMI (e.g., in the midbrain: Fig. 1A; p = 0.13 and p = 0.06, respectively). Such an opponent association is supported by correlative data with neurobehavioral scores (Y-FAS, FEV II), pharmacological stress test results, and follow-up BPND after 6-months diet. Although the participants were free of major depressive disorder (as assessed by structured clinical interview), correlative analyses showed a relationship between midbrain NAT BPND and Beck Depression Inventory (BDI) as an index for sub-threshold depression (r = −0.408; p = 0.07) as well as a correlation between BMI and BDI in both groups (5-HTT: p < 0.01; NAT: p = 0.19; Fig. 1B).
Conclusion: High 5-HTT and low NAT are associated with high BMI. This indicates a potential mechanism in the pathogenesis of human obesity reflecting high NA tone and low 5-HT levels. Together with certain eating behaviours and other biomarkers (pharmacogenotypes) the hypothesized association may lead to more individualised approaches if confirmed. It also questioned the presumed mechanism of action of the drug sibutramine, which is a combined NAT and 5-HTT inhibitor. Sub-threshold depression is an important modulator but may not explain the findings alone.
BPS06-5
BrainPET: Psychiatry
Prenatal stressors alter neuroadaptation of the 5-HT1A serotonergic system following fixed-dose ethanol exposure in non-human primates
P. Lao1, T. Betthauser1, D. Tudorascu2, A. Hillmer3, T. Barnhart1, J. Larson1, C. Moore1, M. Schneider1 and B. Christian1
1University of Wisconsin-Madison, Madison, United States
2University of Pittsburgh, Pittsburgh, United States
3Yale University, New Haven, United States
Abstract
Objectives: The aim of this work was to assess the neuroadaptive response of the serotonergic system to fixed-dose ethanol exposure.
Methods: Twenty-three rhesus monkeys (Macaca mulatta;19.1 ± 0.8 yrs;12M,11F) were reared under identical conditions to examine the individual and combined effects of prenatal alcohol exposure (PN-Alc) and prenatal stress (PNS) on neurodevelopment (7controls, 6PN-Alc, 5PNS, 5PN-Alc + PNS). Subjects underwent an alcohol-naïve, pre-drinking [18F]mefway PET scan, then self-administered a fixed dose of ethanol for 9 months, and underwent a post-drinking [18F]mefway scan. Subjects then underwent an ad libidum phase of ethanol self-administration. Distribution volume ratios (DVRs) were estimated with MRTM2 using cerebellar gray matter as reference region and data 40–90 minutes post-injection. Nine regions of interest (ROIs) from the 112RM-SL atlas (McLaren 2009) were chosen for their involvement in salience and reward circuitry. The Wilcoxon signed rank test was used to assess differences in pre- and post-drinking DVR. The Kruskal-Wallis test was used to assess differences in DVR between prenatal treatment groups. Spearman's correlation coefficients were calculated to investigate associations between DVR and ad libidum ethanol consumption.
Results: Controls demonstrated significant increases between pre- and post-drinking DVR in all 9 investigated ROIs (p < 0.05), while the PN-Alc, PNS, or PN-Alc + PNS did not show any significant differences. Furthermore, in the control group, ad libidum ethanol consumption was positively associated with pre-drinking DVR in the anterior cingulate (ρ = 0.79 [0.10, 0.97]) and lateral prefrontal cortex (ρ = 0.79 [0.10, 0.97]). Pre-drinking DVR did not differ between prenatal treatment groups.
Conclusions: While prenatal alcohol exposure and/or stress did not affect the neurodevelopment of the 5-HT1A serotonergic system, it did lead to a differential neuroadaptive response to moderate alcohol self-administration, with widespread increases in [18F]mefway binding that was blunted in subjects with PN-Alc or PNS.
Research Support: NIH AA017706, P30 HD03352
BPS06-6
BrainPET: Psychiatry
Comparing m-opioid receptor availability and opioid/β-endorphin release between individuals with gambling disorder, alcohol dependence and healthy volunteers using [11C]carfentanil PET and dexamphetamine challenge
S. Turton1, I. Mick1, J. Myers1, A. Colasanti2, H. Bowden-Jones3, L. Clark4, E.A. Rabiner5,6, R.N. Gunn5,7, D.J. Nutt1 and A. Lingford-Hughes1
1Imperial College London, Neuropsychopharmacology Unit, Centre for Psychiatry, London, United Kingdom
2Institute of Psychiatry, Psychology and Neuroscience, King's College London, Centre for Affective Disorders, London, United Kingdom
3Imperial College London, National Problem Gambling Clinic, CNWL NHS Foundation Trust, London, United Kingdom
4University of British Columbia, Centre for Gambling Research, Department of Psychology, Vancouver, Canada
5Imanova Ltd, London, United Kingdom
6King's College London, Centre for Neuroimaging Sciences, London, United Kingdom
7Imperial College London, Centre for Restorative Neuroscience, Division of Brain Sciences, London, United Kingdom
Abstract
Objectives: There is evidence of endogenous opioid system dysregulation in both gambling disorder (GD) and alcohol dependence (AD). [11C]Carfentanil is a selective mu-opioid (MOR) agonist PET radioligand sensitive to endogenous opioid/β-endorphin release following oral dexamphetamine.
We examined the hypothesis that both AD and GD are associated with blunted opioid/β-endorphin release and differences in baseline MOR availability compared with healthy volunteers (HV).
Methods: 13 male AD (abstinence mean 282 days, range 59–720 days), 14 male GD (abstinence mean 47 days, range 3–128 days) and 20 male HV were recruited. Both HV and GD consumed <21 UK units/168 g of alcohol per week. Participants underwent two [11C]carfentanil PET scans, one before and one 3 hours following a 0.5 mg/kg oral dose of dexamphetamine. Dynamic PET data were acquired on a Siemens Biograph 6 for 90 minutes. [11C]Carfentanil regional binding potential (BPND) values (determined from a simplified reference tissue model, with the occipital cortex as the reference) were quantified using MIAKAT™ (www.miakat.org). An omnibus mixed-model ANOVA assessed BPND, ΔBPND, (=Post-amphetamineBPND – Pre-amphetamineBPND)/Pre-amphetamineBPND), ROI (n = 9, see Figure 1) and Group (AD, GD, HV).
ROIs were chosen a priori as areas of importance in addiction, and with significant amphetamine-induced reductions of [11C]carfentanil BPND in previous studies.
Results: There is a significant main effect of Group on [11C]carfentanil ΔBPND between HV and AD (p = 0.001), HV and GD (p = 0.015) but not between GD and AD (p = 0.193), though there is a non-significant trend of smaller ΔBPND values in AD (Figure 1). There is no significant effect of Group on pre-amphetamine [11C]carfentanil BPND (p = 0.055).
Conclusions: Both abstinent AD and GD have blunted opioid/β-endorphin release compared to HV, but no differences in MOR availability. There is no significant difference in the degree of blunted opioid/β-endorphin release between AD and GD.
