Granulocyte- Macrophage Colony-Stimulating Factor Reverses Immunosuppression Acutely Following a Traumatic Brain Injury and Hemorrhage Polytrauma in a Juvenile Male Rat Model

Abstract

Traumatic brain injury (TBI) is a common cause of morbidity and mortality in children. We have previously shown that TBI with a concurrent extracranial injury reliably leads to post-injury suppression of the innate and adaptive immune systems. In patients with post-injury immune suppression, if immune function could be preserved, this might represent a therapeutic opportunity. As such, we examined, in an animal injury model, whether systemic administration of granulocyte macrophage colony-stimulating factor (GM-CSF) could reverse post-injury immune suppression and whether treatment was associated with neuroinflammation or functional deficit. Prepubescent male rats were injured using a controlled cortical impact model and then subjected to removal of 25% blood volume (TBI/H). Sham animals underwent surgery without injury induction, and the treatment groups were sham and injured animals treated with either saline vehicle or 50 μg/kg GM-CSF. GM-CSF was administered following injury and then daily until sacrifice at post-injury day (PID) 7. Immune function was measured by assessing tumor necrosis factor-α (TNF-α) levels in whole blood and spleen following ex vivo stimulation with pokeweed mitogen (PWM). Brain samples were assessed by multiplex enzyme-linked immunosorbent assay (ELISA) for cytokine levels and by immunohistochemistry for microglia and astrocyte proliferation. Neuronal cell count was examined using cresyl violet staining. Motor coordination was evaluated using the Rotarod performance test. Treatment with GM-CSF was associated with a significantly increased response to PWM in both whole blood and spleen. GM-CSF in injured animals did not lead to increases in levels of pro-inflammatory cytokines in brain samples but was associated with significant increases in counted astrocytes. Finally, while injured animals treated with saline showed a significant impairment on behavioral testing, injured animals treated with GM-CSF performed similarly to uninjured animals. GM-CSF treatment in animals with combined injury led to increased systemic immune cell response in whole blood and spleen in the acute phase following injury. Improved immune response was not associated with elevated pro-inflammatory cytokine levels in the brain or functional impairment.

Introduction

Traumatic brain injury (TBI) represents a major cause of morbidity and mortality in children worldwide.¹ An important sequela of severe trauma is post-injury nosocomial infection, which itself places patients at higher risk for morbidity and mortality.^2,3 TBI is an independent risk factor for the development of post-injury infection, particularly ventilator-associated pneumonia.³ Furthermore, the combination of an acute TBI and an extra-cranial injury has been associated with an even higher rate of post-injury infection in children following severe trauma.⁴

We, and others, have shown that systemic immune function is frequently impaired in the aftermath of severe acute traumatic injury in adults and children.^5,6 This impairment is known to affect both the innate and adaptive arms of the immune system, and a higher degree of immune impairment has been repeatedly associated with increased risk of post-injury infection.^7,8 In order to investigate these relationships further, we developed a combined injury model in juvenile rats that includes an experimentally induced frontal brain contusion plus a controlled systemic hemorrhage.⁹ Using this model, we are able to reliably induce reductions in both innate and adaptive immune function following injury.^9,10

Trauma-induced immune suppression may be reversible through the systemic administration of immunostimulatory drugs. Recombinant human granulocyte macrophage colony-stimulating factor (GM-CSF), a drug known to increase myeloid cell numbers and function, has been used to improve innate immune function in this setting in vitro and in vivo.^11,12 It is unclear if these improvements in innate immune function are associated with secondary improvement in adaptive immune function. The effects of systemic administration of GM-CSF on neuroinflammation and behavioral testing outcomes are similarly unclear.

We therefore designed this study, using our combined injury, juvenile rat model, to test the hypothesis that post-injury systemic treatment with GM-CSF will reverse adaptive immune suppression without inducing neuroinflammation or worsening motor coordination. Following the combined injury, we administered daily doses of intraperitoneal (IP) GM-CSF and examined lymphocyte function in whole blood and splenocytes. We also examined animals for evidence of neuroinflammation and neurodegeneration using histology, immunohistochemistry, and ELISA, and we examined motor coordination in these animals using Rotarod testing.

Methods

Animal care

Protocols and techniques for all experiments involving live animals were in accordance with National Institutes of Health Guidelines and were approved by the Institutional Animal Care and Use Committee at the Research Institute at Nationwide Children’s Hospital. Juvenile Sprague Dawley male rats (post-natal day [PND] 28 at the time of surgery; Charles River, Wilmington, MA) were used for all experiments and housed in separate cages with ad libitum access to water and chow. Rats at this age are equivalent to a preadolescent human.¹³ Treatment groups consisted of at least 4 animals per group, and groups included: sham injury treated with saline vehicle (Sham Vehicle), TBI plus extra-cranial hemorrhage (TBI/H) treated with saline vehicle (TBI/H Vehicle), sham injury treated with GM-CSF (Sham GM-CSF), and TBI/H treated with GM-CSF (TBI/H GM-CSF). Immediately post-injury and daily for 6 days, animals scheduled to receive recombinant rat GM-CSF were administered 50 μg/kg GM-CSF (AF-400-23, PeproTech, Cranbury, NJ), which was suspended as a concentration of 10 μg/mL in sterile saline and given via IP injection with a 27-gauge needle. The GM-CSF dose chosen has been used in prior TBI rodent studies,¹⁴ and daily dosing of GM-CSF was selected as it is similar to the manner in which human recombinant GM-CSF is used clinically.¹⁵ Vehicle-treated animals received equal volume of sterile saline alone. The overall experimental design is also outlined graphically (Fig. 1).

FIG. 1.

Overview of the experimental design. Animals underwent either sham-surgery or injury induction at post-natal day of life 28 (D0). The first dose of either vehicle or GM-CSF was given immediately following surgery/injury and daily for 6 days. Blood was obtained at pre-injury day 1 and on post-injury days 1,3, and 7. Rotarod performance test was done on post-injury day 2. Animals were euthanized at post-injury day 7, and spleen and brain were harvested. GM-CSF, granulocyte-macrophage colony-stimulating factor.

Controlled cortical impact

Rats were anesthetized with a mixture of 5% isoflurane and oxygen and were placed in a stereotactic device (Kopf, Tujunga, CA) with the head fixed in a neutral position. Isoflurane was maintained at 2% for the duration of the surgery. For all surgical procedures, standard sterile technique was used. A midline incision was performed, and the scalp was retracted to expose the cranium. A circular bilateral craniectomy was made at midline (2 mm anterior to bregma), exposing the bilateral frontal cortex. For TBI, injury was induced using an Impact One Instrument (Leica, Deer Park, IL). The impactor tip was 3 mm in diameter, and injury was delivered at a velocity of 4 m/s with a dwell time of 500 ms and a depth of 3 mm. For sham injury, animals were anesthetized in a similar fashion and duration, and incision and craniectomy were performed without injury induction.

Fixed-volume hemorrhage

Following induction of TBI, an experimentally induced systemic hemorrhage was performed during the same anesthetic event. After cranial incision closure, animals were removed from the stereotactic device and positioned supine. In a sterile fashion, an incision was made over the right inguinal area, and the femoral artery and vein complex were localized. The femoral vein was cannulated with a 27-gauge needle for aspiration of 25% EBV. Estimated blood volume (EBV) for each rat was calculated based on body weight (70 mL/kg).¹⁶ For sham surgeries, animals were anesthetized using isoflurane for a similar duration, and inguinal incision and dissection were performed without removal of blood.

Euthanasia and post-injury blood and tissue collection

Euthanasia was performed at PID 7. Animals were anesthetized using a mixture of isoflurane and oxygen. Cardiac puncture was performed in a sterile fashion to collect blood in EDTA (ethylenediaminetetraacetic acid) coated microtainers (02–669-33, BD Biosciences, Franklin Lakes, NJ) for stimulation. Spleens were also harvested and placed in chilled phosphate-buffered saline (PBS). Portions of the brain were collected either in chilled PBS for later digestion or in 4% paraformaldehyde (PFA, ThermoFisher, Waltham, MA) in PBS.

Adaptive immune function testing

Whole blood was removed at pre-injury day 1 and on PID 1, 3, and 7 (n ≥ 5 per experimental group). For blood removal at the pre-injury and PID 1 and 3 time points, saphenous vein puncture under sterile conditions was used. For the PID 7 time point, blood from the cardiac puncture was used. All collected blood was placed in EDTA-coated microtainers. For splenic samples (n ≥ 5 per experimental group), tissue was homogenized by passing tissue through a 50 μm filter with 3 mL of PBS containing 2% fetal bovine serum (ThermoFisher).

For ex vivo stimulation experiments, 75 μL aliquots of blood or 100 μL of spleen homogenate were incubated for 18 h at 37°C with 10 μg/mL of the lymphocyte stimulant pokeweed mitogen (PWM) (Sigma Aldrich, St. Louis, MO). Samples were then centrifuged at 1000 x g for 5 min, and supernatants were collected and stored at −70°C. TNF-α in the PWM-stimulated supernatants was measured by ELISA (R&D Systems, Minneapolis, MN).

Brain tissue digest and Bio-Plex analysis

In a subset of harvested brains (n ≥ 6 per experimental group), the frontal lobe (at and near the craniectomy site) was identified, separated, and placed in chilled PBS. For injured animals, frontal lobe tissue was further divided into lesion and peri-lesional areas (Supplementary Fig. S1), which were defined as within 250 μm of the center of the impact site (lesion) or within 250–500 μm from the impact site (perilesional). Regions were weighed, and for every gram of tissue, sections were incubated in 10 mL of N-PER reagent with 1:100 dilution of Halt Protease Inhibitor Cocktail (ThermoFisher, Waltham, MA). Samples were homogenized with a Dounce homogenizer on ice with 10–20 strokes per section. Samples were then incubated for 10 min on ice. After incubation, samples were transferred to microcentrifuge tubes and spun at 10,000 x g for 10 min. Supernatants were then collected and frozen at −70°C, as previously described,¹⁷ to allow for batch analysis. Cytokine levels from the brain homogenates included interleukin (IL)-1α, IL-6, GM-CSF, interferon gamma (IFN-γ), TNF-α, and IL-10. Samples were batch-analyzed using a rat-specific Cytokine Assay Bio-Plex plate on a Bio-Plex 200 System (Bio-Rad, Hercules CA) using High PMT, RP1 settings.

Histology/immunohistochemistry

Brain tissue for these experiments (n ≥ 9 per experimental group) were fixed in PFA for 24 h, switched to 70% ethanol, and embedded in paraffin. Paraffin-embedded tissue was then sectioned at 10 μm thickness and mounted onto slides. The region of interest was at the site of the craniectomy for uninjured animals. For injured animals, regions of interest were the lesion (within 250 μm of the center of the impact site) and the peri-lesional area (within 250–500 μm from the impact site). Slides underwent heat-induced epitope retrieval utilizing 0.1M pH 6.0 sodium citrate buffer in a Digital Pressure Cooker (Nesco, Two Rivers, WI) for 20 min. After blocking with secondary-specific neutral buffer, slides were incubated overnight at 4°C with antibodies against the microglial marker Ionized calcium binding adaptor molecule 1 (IBA1) (Fujifilm Wako, Tokyo, JP) at a 1:500 dilution or the astrocyte marker Glial fibrillary acidic protein (GFAP) (Abcam, Cambridge, UK) at a 1:500 dilution. Slides were washed 3 times for 5 min in PBS, then incubated for 2 h at RT with a fluorescent conjugated secondary antibody Anti-Rabbit 594 (ThermoFisher, Waltham, MA) at a 1:500 dilution. Slides were washed 3 times for 5 min in PBS, then autofluorescence was quenched (Autofluorescence Quenching Kit, Vector Labs, Newark, CA). Slides were mounted with DAPI Fluoromount-G (SouthernBiotech, Birmingham, AL). Slides were viewed on a Leica DMI4000 B inverted fluorescence microscope using LAS X software (Leica, Wetzlar, Germany).

Neuronal cell count was assessed using cresyl violet staining. Tissue was fixed and sliced as noted above. Slides were deparaffinized by washing three times in xylene for 2 min, then rehydrated 3 min at a time in 95% ethanol, 70% ethanol, and distilled water. Slides were incubated inside a staining chamber suspended in a 60°C water bath for 10 min, in a solution of 0.1% cresyl violet (Cresyl Violet Acetate J64318, ThermoFisher) with 0.3% acetic acid in distilled water. Subsequently, slides were dehydrated in 70% ethanol for 3 min, 95% ethanol for 1 min, followed by a 1-sec immersion in 100% ethanol. Finally, slides were kept in xylene for 90 min before coverslips were mounted onto the slides using Permount Mounting Medium (Electron Microscopy Sciences, Hatfield, PA).

Cell counting for IBA1 and GFAP was performed using ImageJ Software (Version 1.53, National Institutes of Health, Bethesda, MD). ImageJ was used to automatically calculate the area of sections, and cells were quantified by the localization of DAPI-stained nuclei and immunopositive signal. For neuronal cell count, the count was performed by identifying individual cresyl violet-stained cells and nuclei using standard morphological findings.¹⁸ Cell counts were determined from 6 microscopic fields, and results are presented as cells per square millimeter, as previously described.¹⁹

Rotarod performance test

The Rotarod performance test in juvenile rats was performed as previously described²⁰ and was used to assess motor function after injury with an accelerating Economex Rotarod (Columbus Instruments, Columbus, OH). Rats (n ≥ 6 per experimental group) were acclimated to the Rotarod 3;days prior to injury by placing each on a stationary rod for 5 min. Training commenced, consisting of 3 trials, starting at 1 rpm and accelerating to 30 RPM over 150 s. Time-to-fall over 150 s was recorded, and animals were given a 5-min break between each trial. Experimental testing occurred on PID 2 and consisted of 3 trials with the Rotarod gradually increasing from 1 to 30;rpm over 150 s with a 5-min break between each trial. Time to fall was recorded, and the trial was stopped after 150 s if a fall did not occur.

Statistical analysis

Results were analyzed using Prism Version 8.2.0 (GraphPad Software, San Diego, CA). Data are expressed as mean ± standard error of the mean (SEM) of separate experiments and compared by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Overall comparisons are reported with the overall F-value of the ANOVA and the corresponding p-value with post-hoc comparison performed if the overall p-value was statistically significant. The difference between two treatments was considered to be significant at p ≤ 0.05.

Results

Adaptive immune function in whole blood and spleen

PWM-induced TNF-α production from whole blood (Fig. 2) showed no significant differences in the four treatment groups at the pre-injury time point. At PID 1 (F = 3.74, p = 0.021), both injured and sham animals treated with GM-CSF showed higher PWM-induced TNF-α production compared to TBI/H Vehicle animals (p = 0.035 and 0.045, respectively).

FIG. 2.

Measurement of peripheral blood responsiveness to lymphocyte stimulant. Peripheral blood samples from animals were incubated with PWM and then tested for their responsiveness by examining levels of TNF-α. There was a main effect of GM-CSF to increase measured TNF-α levels in response to PWM stimulation at post-injury day 1. The symbol * indicates p < 0.05, as compared to TBI/H Vehicle (n ≥ 5 per experimental group). GM-CSF, granulocyte-macrophage colony-stimulating factor; PWM, pokeweed mitogen, TNF-α, tumor necrosis factor alpha; TBI/H, traumatic brain injury plus hemorrhage.

Lymphocyte responsiveness in the spleen was assessed in a similar fashion on PID 7 (F = 4.60, p = 0.011, Fig. 3). PWM-induced TNF-α production was significantly higher in splenocytes from the TBI/H GM-CSF animals compared to the TBI/H Vehicle group (p = 0.023).

FIG. 3.

Measurement of splenic responsiveness to lymphocyte stimulant. At post-injury day 7, splenic samples were obtained, homogenized, and incubated with PWM to stimulate lymphocyte activity. Levels of TNF-α were examined by ELISA. Animals treated with TBI/H GM-CSF were noted to have significantly increased levels of TNF-α, as compared with TBI/H Vehicle animals (* indicates p < 0.05, respectively); n ≥ 5 per experimental group. ELISA, enzyme-linked immunosorbent assay; GM-CSF, granulocyte-macrophage colony-stimulating factor; PWM, pokeweed mitogen, TNF-α, tumor necrosis factor alpha; TBI/H, traumatic brain injury plus hemorrhage.

Cytokine levels in brain homogenates

Cytokine levels in brain homogenates on PID 7 were largely similar between treatment groups (Fig. 4). When comparing injured animals to sham animals (F = 8.61, p = 0.0143), IL-1α levels were higher in the peri-lesional region in the TBI/H Vehicle group (p = 0.003), as compared with Sham GM-CSF. Compared with sham animals, levels of IL-10 at the lesion (F = 5.084, p = 0.001) of TBI/H Vehicle animals were significantly lower than Sham treated with (p = 0.032) or without GM-CSF (p = 0.037). There was no statistically significant increase in pro-inflammatory cytokine levels in the lesion or perileional brain samples of animals that received GM-CSF.

FIG. 4.

Measurement of brain cytokine levels. At post-injury day 7, brain tissue was harvested and frontal lobe tissue at the craniectomy was dissected. For injured animals, tissue at the craniectomy site was divided into lesion and peri-lesional areas, and tissue was assessed using multiplex ELISA to evaluate cytokine levels. Significant differences from Sham Vehicle or Sham GM-CSF were denoted (* and † indicates p < 0.05) and were seen only in IL-1α and IL-10; n ≥ 6 per experimental group. ELISA, enzyme-linked immunosorbent assay; GM-CSF, granulocyte-macrophage colony-stimulating factor; IFN-γ, interferon gamma; IL, interleukin; PWM, pokeweed mitogen, TNF-α, tumor necrosis factor alpha; TBI/H, traumatic brain injury plus hemorrhage.

Histology

Using brain tissue harvested at PID 7, microglia were identified and counted using IBA1 antibody for immunohistochemistry (F = 9.89, p < 0.0001, Fig. 5A, D). For the Sham animals, there was no difference in the number of IBA1-positive cells when comparing Sham Vehicle and Sham GM-CSF animals. At the lesion site, both TBI/H Vehicle and TBI/H GM-CSF groups had higher counts of IBA1-positive cells compared to Sham Vehicle (p < 0.0001). At the peri-lesional area, TBI/H Vehicle and TBI/H GM-CSF groups had non-significantly higher counts of IBA1-positive cells compared to Sham Vehicle (p = 0.1854 and p = 0.051, respectively). When comparing IBA-1 counts between TBI/H Vehicle and TBI/H GM-CSF at the lesion and peri-lesional areas, no significant difference was noted.

FIG. 5.

Brain tissue immunohistochemistry. Sliced brain tissue samples from animals at post-injury day 7 were examined for neuroinflammation using IBA1 and GFAP antibody, and nuclei were identified using DAPI. Neuronal counts were performed using cresyl violet stain. Cell counts are shown in A–C with representative micrographs in D–F (400× magnification, scale bar is 63 μm). For injured animals, the lesion and peri-lesional areas were examined separately. (A, D) IBA1 cells counts were significantly higher in the lesion in TBI/H Vehicle and GM-CSF treated animals, as compared with Sham Vehicle. (B, E) Astrocytes (GFAP+ cell counts) were significantly higher in TBI/H GM-CSF peri-lesional tissue, as compared with Sham Vehicle. (C, F) Neuronal cell counts were significantly lower in the TBI/H Vehicle group at the lesion, as compared with Sham Vehicle. For cell counts, * indicates p < 0.05 and ** indicates p < 0.0001, as compared with Sham Vehicle (n ≥ 9 per experimental group). GFAP, Glial fibrillary acidic protein; GM-CSF, granulocyte-macrophage colony-stimulating factor; IBA1, Ionized calcium binding adaptor molecule 1; TBI/H, traumatic brain injury plus hemorrhage.

Astrocytes were identified and counted using GFAP antibody (F = 5.554, p = 0.0004, Fig. 5B, E). No significant differences were noted between the two Sham treatments. No significant differences were noted between Sham treatments and the lesion. At the peri-lesional area, the TBI/H GMCF group did show significant increases in GFAP+ cells as compared to either Sham group (p < 0.0001), but not when compared to TBI/H Vehicle (p = 0.06).

Neuronal counts were made using cresyl violet staining to identify Nissl substance (F = 5.854, p = 0.0003, Fig. 5C, F). No significant difference was noted between Sham Vehicle and Sham GM-CSF. At the lesion, a significant decrease in counted cells in TBI/H Vehicle group was noted when comparing this group with Sham Vehicle (p = 0.001), and a non-significant decrease was noted when comparing the Sham Vehicle group with the TBI/H GM-CSF group (p = 0.37). Compared to Sham Vehicle, significantly decreased neuronal counts were not noted in the peri-lesional area in either injury group.

Rotarod performance test

In PID 2 Rotarod testing (F = 6.125, p = 0.004), animals from the TBI/H-Vehicle group demonstrated a fall from the Rotarod device at a significantly earlier timepoint, as compared to the Sham-Vehicle group (p = 0.002, Fig. 6). No significant difference in latency to fall was noted between the Sham-Vehicle group and the TBI/H GM-CSF group (p = 0.2).

FIG. 6.

Rotarod performance test. The Rotarod performance test was used to evaluate motor coordination in the four treatment groups. The mean latency to fall is shown. There was a significant difference noted between Sham Vehicle and TBI/H Vehicle (*p < 0.05, as compared with Sham Vehicle); n ≥ 6 per experimental group. TBI/H, traumatic brain injury plus hemorrhage.

Discussion

As shown previously, TBI/H was associated with an early reduction in post-injury immune function in our juvenile animal model. Post-injury treatment with GM-CSF prevented the development of adaptive immune suppression on PID 1, as measured by the cytokine response to an ex vivo lymphocyte stimulant (PWM) in whole blood. The effect of GM-CSF on augmenting adaptive immune function was still present by PID 7 in splenocytes. There was not an associated increase in pro-inflammatory cytokine levels in the brain, in either the lesion or peri-lesional areas in GM-CSF-treated animals. Both injured animal groups (vehicle and GM-CSF treated) had significant increases in numbers of microglia, but TBI/H GM-CSF animals did demonstrate significantly more GFAP+ cells than TBI/H Vehicle. Functional assessment with the Rotarod performance test showed that post-injury administration of GM-CSF was not associated with a worsening of motor coordination. In fact, GM-CSF-treated animals had similar performance to uninjured Sham animals. Taken collectively, these data suggest that post-injury treatment with GM-CSF is associated with an increase in systemic adaptive immune function without deleterious neuroinflammatory or motor coordination performance sequelae.

In patients with severe trauma, especially injuries that include TBI, the systemic inflammatory response that is induced by the injury is typically accompanied by a nearly-concurrent compensatory anti-inflammatory response that can render a patient functionally immunocompromised.⁶ When severe, this has been termed immunoparalysis, and it can affect both circulating innate and adaptive immune cells as well as those in lymphoid organs such as the spleen.^5,21 Reversal of these systemic immunosuppressive effects through the use of immunostimulatory approaches has the potential to reduce the risk of post-injury infection and to improve outcomes in critically injured patients. A theoretical concern, however, is that enhancement of systemic immune function could lead to increased inflammation at the site of injury, potentially limiting recovery.²²

Systemic immune suppression following injury results in an initial surge of inflammation, but this is coupled with an anti-inflammatory response characterized by reduced function of circulating leukocytes (e.g., diminished release of TNF-α when stimulated).²¹ From our prior research, we have shown that the TBI plus hemorrhage induces a decrease in monocytes (measured as a percentage of total leukocytes) and a corresponding reduction in both innate and adaptive immune function,^9,10 consistent with prior clinical findings of patients with post-traumatic immunoparalysis.⁷ In patients undergoing chemotherapy, GM-CSF has been shown to expand both the total leukocyte count and the absolute neutrophil count.²³ In in vitro studies, GM-CSF has been shown to stimulate myeloid cell proliferation, improve cell survival, and restore cellular responsiveness via multiple pathways including NF-κB²⁴ and PI-3/AKT.²⁵ However, GM-CSF has also been shown expand regulatory T cells (e.g., CD4⁺CD25⁺Foxp3⁺ cells),²⁶ and these dual effects may explain why we did not see more evidence of neuroinflammation with GM-CSF treatment.²⁷ Our future research will utilize flow cytometry to determine whether the efficacy of GM-CSF treatment in this injury model can be explained by its role in increasing leukocyte counts, cellular function, or both. We will also focus on how exogenous treatment with GM-CSF may affect expansion of T cell populations.

While GM-CSF is well known as a hematopoietic cytokine capable of inducing granulocyte differentiation,²⁸ it also serves as a communication bridge between the innate and adaptive immune response through the release of GM-CSF by T-cells.²⁹ Levels of GM-CSF are noted to be elevated at sites of tissue inflammation,³⁰ and experimental evidence suggests GM-CSF may have a role in the pathogenesis of certain auto-immune diseases.^31,32 As such, there may be concerns regarding the use of GM-CSF in a disease process where neuroinflammation might exacerbate injury. In our data, we did note that injured animals had an increased IBA1 cell count at the lesion, but this was not significantly increased in injured animals treated with GM-CSF. Proliferation of microglia has been noted in hippocampal slice cultures treated with GM-CSF, but this finding was not associated with increased inflammatory response.³¹ In fact, recent data suggest that GM-CSF-mediated activation of microglia was suggested with reduced viral titers and increased survival in a murine model of West Nile virus encephalitis.³³ In a rat model of acute spinal cord injury, treatment with GM-CSF was associated with cortical reactivation and recovery of tactile abilities, along with a decrease in pro-inflammatory cytokine expression and reduced glial scarring at the lesion site.³⁴ Kelso et al. showed that treatment with GM-CSF was associated with decreased lesion volumes and increased cortical tissue sparing following controlled cortical impact in mice.¹⁴ Similarly, our data suggest that pro-inflammatory cytokine levels were not increased in the brain tissue of GM-CSF treated animals, though there was significant increase in counts of GFAP positive cells in the peri-lesional area in TBI/H GM-CSF animals compared with sham-injured animals. Astrocytes have been shown to secrete GM-CSF.³⁵ Additionally, astrocytes have been shown to possess GM-CSF receptors, and treatment with GM-CSF may lead to astrocyte proliferation.^35,36 One role for GM-CSF is likely communication between astrocytes and microglia, with in vitro interactions between these cell populations leading to A2 and M2 polarization, and this phenomenon, in part, might explain the lack of neuroinflammation seen with GM-CSF treatment.³⁷ Finally, astrocyte proliferation and scar formation at 5 to 14 days following central nervous system injury has been shown to segregate surrounding tissue from inflammatory cells at the injury site.³⁸

Functional testing of the four treatment groups revealed no significant differences in performance between GM-CSF treated (either sham or injured animals) and the Sham Vehicle animals, as assessed by the Rotarod performance test. These findings agree with prior results which showed that GM-CSF knockout mice showed behavioral testing impairments after lateral fluid percussion injury.³⁹

There are several limitations to this research. Our current data show that the responsiveness of circulating and splenic lymphocytes can be augmented through systemic, post-injury treatment with GM-CSF; however, we did not measure other aspects of adaptive immune function (e.g., cytotoxic killing, antibody production). More detailed characterization of how GM-CSF treatment influences immune function across innate and adaptive immune cells is ongoing in our laboratory. Also, it is not clear that the improvement in adaptive immune function that we observed is relevant to the prevention of post-operative infection. Future work will involve induction of experimentally-induced pneumonia to determine if GM-CSF treatment can prevent or lessen post-injury infection. Additionally, we will also measure lesion volume to determine if systemic immunomodulation has an effect on brain injury. We only used male rats in the current set of experiments. There is some evidence that sex plays a role in modulating the inflammatory response,^40,41 and both sexes will need to be used in this future work. While we did not observe increases in neuroinflammation or functional impairment in the setting of post-injury GM-CSF therapy in our model, it is important to note that our behavioral evaluation was limited to motor coordination function testing. Additional experiments evaluating a broader array of neurocognitive function (e.g., memory, problem-solving), are needed. Lastly, our preclinical injury model is performed in prepubescent rats to make it more relevant for pediatric practice; however, it is unclear if our results are generalizable to older animals.

Conclusion

Our experiments have shown that systemic, post-injury administration of GM-CSF is associated with augmentation of systemic adaptive immune function in our model of experimental TBI plus hemorrhage in juvenile male rats. Treatment with GM-CSF in this model was not associated with the induction of neuroinflammation and resulted in preserved motor coordination. These findings are deserving of further study using GM-CSF and other immunomodulators and are supportive of the concept of post-injury immune function monitoring and modulation.

Transparency, Rigor, and Reproducibility Summary

The study design and analytic plan were not preregistered. Sample size was planned to be 6 rats per group, based on our prior observations of a 25% reduction in TNF-α production in response to PWM with 80% power and alpha = 0.05. Forty-two were used for the outlined experiments, none were excluded for technical reasons, 2 died, and all were randomly assigned to treatment groups to maintain a balanced preinjury weight. All 40 rats received the predetermined therapy, and complete data were obtained. Investigators who administered the therapeutic and performed outcome assessments were blinded to group. The first therapeutic administration occurred immediately after injury. A prior dose–response relationship was performed (Supplementary Fig. S2), and the selected dose was based on previously published results¹⁴ and the preliminary dose–response data. Surgeries, laboratory, and behavioral testing were done using highly protocolized techniques with a single operator.

Footnotes

Authors’ Contributions

E.S. Conceptualization, Methodology, Validation, Formal analysis, Investigation, Writing—Original, Writing—Review & Editing. T.W. Methodology, Validation, Formal analysis, Investigation, Writing—Original, Writing—Review & Editing. M.H. Conceptualization, Methodology, Validation, Formal analysis, Investigation, Writing—Review & Editing. All authors had full access to all data in this study and take complete responsibility for the integrity of the data and accuracy of the data analysis.

Author Disclosure Statement

Partner Therapeutics (manufacturer of recombinant human GM-CSF) provides M.H. with complimentary study drug for a clinical trial for which he is the principal investigator. Partner Therapeutics has no involvement in the project that is the subject of this article.

Funding Information

National Institute of Neurological Disorders and Stroke (NINDS) of The National Institutes of Health (K08NS119878) and The Abigail Wexner Research Institute at Nationwide Children’s Hospital

Supplementary Material

Supplemental Figure S1

Supplemental Figure S2

References

Schrieff-Elson

, Steenkamp

, Hendricks

, et al. Local and global challenges in pediatric traumatic brain injury outcome and rehabilitation assessment. Childs Nerv Syst, 2017; 33(10):1775–1784

Glance

, Stone

, Mukamel

, et al. Increases in mortality, length of stay, and cost associated with hospital-acquired infections in trauma patients. Arch Surg, 2011; 146(7):794–801

, Liu

, Xiao

, et al. Incidence, risk factors, and outcomes of ventilator-associated pneumonia in traumatic brain injury: A meta-analysis. Neurocrit Care, 2020; 32(1):272–285

Sribnick

, Hensley

, Moore-Clingenpeel

, et al. Nosocomial infection following severe traumatic injury in children. Pediatr Crit Care Med, 2020; 21(5):443–450

Sharma

, Shultz

, Robinson

, et al. Infections after a traumatic brain injury: The complex interplay between the immune and neurological systems. Brain Behav Immun, 2019; 79:63–74

Sribnick

, Popovich

, Hall

. Central nervous system injury-induced immune suppression. Neurosurg Focus, 2022; 52(2):E10

Muszynski

, Nofziger

, Greathouse

, et al. Innate immune function predicts the development of nosocomial infection in critically injured children. Shock, 2014; 42(4):313–321

Bietar

, Lehmann

, Stadnyk

. Effects of CNS injury-induced immunosuppression on pulmonary immunity. Life (Basel), 2021; 11(6)

Sribnick

, Weber

, Hall

. Innate immune suppression after traumatic brain injury and hemorrhage in a juvenile rat model of polytrauma. J Neuroimmunol, 2019; 337:577073

10.

Sribnick

, Warner

, Hall

. Traumatic brain injury and hemorrhage in a juvenile rat model of polytrauma leads to immunosuppression and splenic alterations. J Neuroimmunol, 2021; 361:577723

11.

Wolach

, van der Laan

, Maianski

, et al. Growth factors G-CSF and GM-CSF differentially preserve chemotaxis of neutrophils aging in vitro . Exp Hematol, 2007; 35(4):541–550

12.

Rosenbloom

, Linden

, Dorrance

, et al. Effect of granulocyte-monocyte colony-stimulating factor therapy on leukocyte function and clearance of serious infection in nonneutropenic patients. Chest, 2005; 127(6):2139–2150

13.

Sengupta

. The laboratory rat: Relating its age with human's. Int J Prev Med, 2013; 4(6):624–630

14.

Kelso

, Elliott

, Haverland

, et al. Granulocyte-macrophage colony stimulating factor exerts protective and immunomodulatory effects in cortical trauma. J Neuroimmunol, 2015; 278:162–173

15.

Paul

, Ram

, Kugler

, et al. Subcutaneous versus intravenous granulocyte colony stimulating factor for the treatment of neutropenia in hospitalized hemato-oncological patients: Randomized controlled trial. Am J Hematol, 2014; 89(3):243–248

16.

Diehl

, Hull

, Morton

, et al. European Federation of Pharmaceutical Industries Association and European Centre for the Validation of Alternative Methods. A good practice guide to the administration of substances and removal of blood, including routes and volumes. J Appl Toxicol, 2001; 21(1):15–23

17.

Sjolin

, Kultima

, Larsson

, et al. Distribution of five clinically important neuroglial proteins in the human brain. Mol Brain, 2022; 15(1):52

18.

Garcia-Cabezas

, John

, Barbas

, et al. Distinction of neurons, glia and endothelial cells in the cerebral cortex: An algorithm based on cytological features. Front Neuroanat, 2016; 10:107

19.

Pruss

, Kopp

, Brommer

, et al. Non-resolving aspects of acute inflammation after spinal cord injury (SCI): Indices and resolution plateau. Brain Pathol, 2011; 21(6):652–660

20.

Geddes

, Sribnick

, Sayeed

, et al. Progesterone treatment shows benefit in a pediatric model of moderate to severe bilateral brain injury. PLoS One, 2014; 9(1):e87252

21.

Hall

, Greathouse

, Thakkar

, et al. Immunoparalysis in pediatric critical care. Pediatr Clin North Am, 2017; 64(5):1089–1102

22.

Witcher

, Bray

, Chunchai

, et al. Traumatic brain injury causes chronic cortical inflammation and neuronal dysfunction mediated by microglia. J Neurosci, 2021; 41(7):1597–1616

23.

, Hua

. Evaluation of effectiveness of granulocyte-macrophage colony-stimulating factor therapy to cancer patients after chemotherapy: A meta-analysis. Oncotarget, 2018; 9(46):28226–28239

24.

Ebner

, Bandion

, Binder

, et al. GMCSF activates NF-kappaB via direct interaction of the GMCSF receptor with IkappaB kinase beta. Blood, 2003; 102(1):192–199

25.

Comalada

, Xaus

, Sanchez

, et al. Macrophage colony-stimulating factor-, granulocyte-macrophage colony-stimulating factor-, or IL-3-dependent survival of macrophages, but not proliferation, requires the expression of p21(Waf1) through the phosphatidylinositol 3-kinase/Akt pathway. Eur J Immunol, 2004; 34(8):2257–2267

26.

Kared

, Leforban

, Montandon

, et al. Role of GM-CSF in tolerance induction by mobilized hematopoietic progenitors. Blood, 2008; 112(6):2575–2578

27.

Bhattacharya

, Thiruppathi

, Elshabrawy

, et al. GM-CSF: An immune modulatory cytokine that can suppress autoimmunity. Cytokine, 2015; 75(2):261–271

28.

Borriello

, Galdiero

, Varricchi

, et al. Innate Immune Modulation by GM-CSF and IL-3 in Health and Disease. Int J Mol Sci, 2019; 20(4)

29.

Ingelfinger

, De Feo

, Becher

. GM-CSF: Master regulator of the T cell-phagocyte interface during inflammation. Semin Immunol, 2021; 54:101518

30.

Becher

, Tugues

, Greter

. GM-CSF: From growth factor to central mediator of tissue inflammation. Immunity, 2016; 45(5):963–973

31.

Lotfi

, Thome

, Rezaei

, et al. Roles of GM-CSF in the pathogenesis of autoimmune diseases: An update. Front Immunol, 2019; 10:1265

32.

Cook

, Pobjoy

, Steidl

, et al. Granulocyte-macrophage colony-stimulating factor is a key mediator in experimental osteoarthritis pain and disease development. Arthritis Res Ther, 2012; 14(5):R199

33.

Stonedahl

, Leser

, Clarke

, et al. Treatment with granulocyte-macrophage colony-stimulating factor reduces viral titers in the brains of west Nile virus-infected mice and improves survival. J Virol, 2023; 97(3):e0180522

34.

Thomaty

, Pezard

, Xerri

, et al. Acute granulocyte macrophage-colony stimulating factor treatment modulates neuroinflammatory processes and promotes tactile recovery after spinal cord injury. Neuroscience, 2017; 349:144–164

35.

Choi

, Lee

, Lim

, et al. Human astrocytes: Secretome profiles of cytokines and chemokines. PLoS One, 2014; 9(4):e92325

36.

Guillemin

, Boussin

, Le Grand

, et al. Granulocyte macrophage colony stimulating factor stimulates in vitro proliferation of astrocytes derived from simian mature brains. Glia, 1996; 16(1):71–80

37.

Kim

, Son

. Astrocytes stimulate microglial proliferation and M2 polarization in vitro through crosstalk between astrocytes and microglia. Int J Mol Sci, 2021; 22(16)

38.

Wanner

, Anderson

, Song

, et al. Glial scar borders are formed by newly proliferated, elongated astrocytes that interact to corral inflammatory and fibrotic cells via STAT3-dependent mechanisms after spinal cord injury. J Neurosci, 2013; 33(31):12870–12886

39.

Shultz

, Tan

, Wright

, et al. Granulocyte-macrophage colony-stimulating factor is neuroprotective in experimental traumatic brain injury. J Neurotrauma, 2014; 31(10):976–983

40.

Casimir

, Heldenbergh

, Hanssens

, et al. Gender differences and inflammation: An in vitro model of blood cells stimulation in prepubescent children. J Inflamm (Lond), 2010; 7:28

41.

Lamason

, Zhao

, Rawat

, et al. Sexual dimorphism in immune response genes as a function of puberty. BMC Immunol, 2006; 7:2

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

6.08 MB

3.85 MB