Abstract
Orthostatic hypotension (OH) is a common non-motor symptom in Parkinson’s disease (PD) and is linked with increased mortality risk among the elderly. Although the locus coeruleus (LC) is the major source of noradrenaline (NA) modulation in the brain, its role in the pathogenesis of OH in PD remains largely elusive. Here we examined 44 well characterized postmortem brains of PD patients and available clinical data to explore the relationship between OH and LC pathology in PD. Our results failed to indicate that the LC pathology as well as the substantia nigra pathology were robustly associated with the presence of OH in PD patients, suggesting targeting LC norepinephrinergic system alone may not be sufficient to treat OH in PD.
INTRODUCTION
Non-motor features of Parkinson’s disease (PD) are increasingly recognized and include constipation, autonomic dysfunction, cognitive impairment, and psychiatric symptoms [1]. Orthostatic hypotension (OH) is a common non-motor symptom in PD. The OH affects 30% –58% of PD patients [2], with a 36% increased mortality risk among the elderly [3]. The OH has been linked to peripheral noradrenergic cardiovascular denervation and also in a certain degree to central noradrenergic involvement [4]. As the major source of noradrenaline (NA) modulation in the brain, locus coeruleus (LC) undergoes pathological alterations in the early stage of PD. It is known that LC noradrenergic neurons degenerate among PD patients by up to 70% of its numbers and that NA concentrations decrease to a comparable extent [5]. Meanwhile, intraneuronal α-synuclein burden in the noradrenergic neurons of the LC precedes and may be of even greater magnitude than that in the dopaminergic neurons at substantia nigra (SN) [6, 7]. Recently, one neuroimaging study has demonstrated that PD patients with defined OH exhibited decreased LC norepinephrinergic neuromelanin signal and reduced noradrenergic function [8]. Therefore, targeting LC norepinephrinergic system is proposed as a treatment strategy for OH in PD. However, the role of LC pathology in the pathogenesis of OH in PD remains largely unclear.
In the current study we aimed to explore the relationship between OH and LC pathology in autopsy brains of PD patients. Establishing this relationship in humans would add to our understanding of the mechanism underlying OH symptom in α-synucleinopathies and explore the utility of LC noradrenergic augmentation as a therapeutic mechanism.
MATERIALS AND METHODS
Patient selection and clinical information
We reviewed autopsy reports and available clinical information on consecutive PD patients who had autopsies performed during 2014–2019 in the Johns Hopkins Parkinson’s Disease Research Center. The presence or absence of OH was established retrospectively based on the clinical diagnosis. We only included cases with medical records that contained regular reports of clinical developments. This study was approved by the institutional review board. All material has been collected from donors for or from whom a written informed consent for a brain autopsy and the use of the material and clinical information for research purposes had been obtained. Information on demographics, past medical history, and microscopic description of the brains were collected. Forty-four PD patients who (1) met clinical criteria for PD and (2) had neuropathology-confirmed Lewy body disease (LBD) (i.e., brainstem-predominant, limbic (transitional), or neocortical (diffuse) variants) [9, 10] were enrolled. Autopsy data were analyzed and correlated with clinical features.
Neuropathological assessment
Autopsies were conducted by the neuropathologists at Johns Hopkins Hospital. Brains were examined externally, fixed for two weeks in 10% buffered formaldehyde. Tissue blocks for microscopic examination were processed, embedded in paraffin, and cut at 10μm thickness. All tissue sections were stained with hematoxylin and eosin (H&E). Selected sections are stained with silver (Hirano method) and immunostained for α-synuclein, amyloid-beta, and phospho-tau. The neuropathological assessment and diagnostic formulation were performed according to the third report of the DLB Consortium [10] and the 2012 NIA-AA criteria [9].
Statistical analysis
Demographic characteristics (age, brain weight, duration of disease) and pathological scores were demonstrated as mean with standard deviations. Gender distribution and other binary or nominal variables between patient groups were compared with the Pearson chi-square or Fisher exact test when appropriate. Continuous, normally distributed data were compared between groups with an independent t-test. We performed logistic binary regression to evaluate the strength of the association between different pathological scores and OH. Data were analyzed with IBM SPSS Statistics 19 software with the significance level set at p < 0.05.
RESULTS
Forty-four patients with clinical and neuropathologically confirmed PD/LBD were divided into two groups, PD with OH and PD without OH. We compared demographic and clinical data between the two groups. As shown in Table 1, there were no statistic differences in age at death (p = 0.614), age at onset (p = 0.925), duration of disease (p = 0.660), sex (p = 1.000), postmortem interval (p = 0.361) and brain weight (p = 0.386) between the two groups. Meanwhile, the binary regression model using PD with or without OH as outcome, coupled with age at death, age at onset, duration of disease as covariates showed that none of them had statistic association with OH (all p > 0.05). These suggested that age at death, age at onset and duration of disease had no influences on OH in PD patients.
Demographic and clinical data of PD patients
To further explore the probable involvement of LC in the pathogenesis of OH, we compared the pathological variables in the LC between the two groups. As shown in Table 2, neither the severity of neuronal loss (p = 0.545) nor the Lewy body scores (p = 0.617) in the LC demonstrated any statistic differences between the two groups. Additionally, there was also no significant difference in the tauopathology (p = 0.710) in the LC between the two groups. To explore the underlying central pathological mechanism, we also analyzed the pathological variables in the SN between the two groups. Similar to the results in LC, no significant differences were shown in the severity of neuronal loss (p = 0.281), the Lewy body scores (p = 0.966) and tauopathology (p = 0.918) between the two groups. The binary regression model using PD with or without OH as outcome, coupled with neuronal loss scores in the SN, neuronal loss scores in the LC, Lewy body scores in the SN, Lewy body scores in the LC, tauopathology in the SN and tauopathology in the LC as covariates showed that neither the LC pathology nor SN pathology was significantly associated with OH (all p > 0.05).
Comparison of pathological variables
B, brainstem-predominant; L, limbic (transitional); N, neocortical (diffuse) variant.
DISCUSSION
Here we explored the neuropathologic link between LC pathology and presence of OH in PD patients. Contrary to our expectations, our results failed to indicate that the LC pathology was associated with the presence of OH in PD patients. Meanwhile, the SN pathology also showed similar results. OH is one of the most frequent and troublesome autonomic symptoms in PD. Previously, LC atrophy in PD patients has been frequently associated with non-motors symptoms. Though central noradrenergic system contributes to OH, the exact sites of the central lesion that result in cardiovagal failure is not fully understood. Lewy pathology has been found in the dorsal motor nucleus of the vagus, the medullary reticular formation, the raphe nuclei, and the LC [11]. These brainstem nuclei are associated with the modulation of autonomic nervous system output and are affected early in PD. Cortical centers responsible for autonomic control are also involved by the pathological process in PD. Our results demonstrate that the SN or LC lesions alone is probably not sufficient to cause OH in PD. However, it is worth noting that the patient sample collected at Johns Hopkins Hospital, a large tertiary and quaternary care center, might be biased towards late/advanced LBD stage. This has reflected in the predominant neocortical (diffuse) variant of LB pathology in our cohort and the relatively low OH rate in our patients compared with previously reported estimates [12]. In addition, changes at biochemical and molecular levels may still reveal an association between OH and LC/SN pathology. Such possibilities should be further examined in future studies. Nevertheless, our study is supported by observations in other neurodegenerative diseases that affect different parts of the brainstem to varying degrees, as OH is not common in corticobasal degeneration and progressive supranuclear palsy but is a core symptom of multiple system atrophy (MSA). Indeed, it is worth noting that two autopsy case reports of Lewy body diseases with significant autonomic failure including OH exhibited extensive damage to the autonomic nervous system including the brainstem, spinal cord, and sympathetic ganglia [13, 14]. This widespread central and peripheral noradrenergic dysfunction implies that the central noradrenergic degeneration might be the main contributor to the process of OH in patients with MSA, whereas the peripheral noradrenergic degeneration may play a more significant role in the process of OH in PD.
According to Braak’s hypothesis [15], α-synuclein burden in the LC precedes in the SN. LC neuronal loss seems to surpass that of the SN, even in the prodromal phases [7]. Notably, two latest studies suggest that pathological α-synuclein may lead to neuronal dysfunction sufficiently to cause OH before overt neuronal loss in MSA [16] and pure autonomic failure [17]. Meanwhile, PD patients with OH exhibit decreased LC neuromelanin signal on magnetic resonance studies [8], suggesting that the LC might be involved in the pathogenesis of OH at an early stage of PD. The postmortem pathology of brain in PD patients represents the end stage of disease. Therefore, the lack of correlation at the pathological level at the end stage could not rule out the LC is still involved in the pathogenesis of OH at an earlier stage of PD.
LC noradrenergic denervation is a complexed process and include both loss of noradrenergic neurons and reduced catecholamine enzymes (tyrosine hydroxylase and dopamine beta-hydroxylase) activities [18]. It remains unknown whether the aggressive management of the noradrenergic system can benefit PD patients with OH. Studies examining whether drugs targeting the LC noradrenergic system may treat OH in PD are still exploratory at this stage. Our data indicates that targeting LC norepinephrinergic system alone may not be sufficient to treat OH in PD. Future studies should focus on comprehensive study the central and peripheral noradrenergic dysfunction in PD.
CONFLICT OF INTEREST
The authors have no conflict of interest to report.