Introduction
Worldwide, individuals experiencing homelessness face significant inequalities in health that contribute to shorter life expectancy and increased morbidity. Reference Stafford and Wood1 Regular, preventative medical care is often inaccessible, leading homeless individuals to engage with their healthcare system only in acute states, or in emergency. 2 Homelessness is an unforgiving situation, with homeless individuals often needing to prioritize access to shelter or food over healthcare. Reference Stafford and Wood1 Furthermore, while the circumstances of homelessness often directly lead to poor health, social and lifestyle factors associated with an individual experiencing homelessness in the first place Reference Shelton, Taylor, Bonner and van den Bree3 can become entwined with risk factors for other diseases. Pertinent to our study, there is an increasing appreciation for the potential role of a broad range of these lifestyle factors in the development of neurodegenerative diseases, including Parkinson’s disease (PD). Thus, while caffeine intake, smoking, moderate alcohol consumption, and exercise may reduce the risk of PD, Reference Hernán, Takkouche, Caamaño-Isorna and Gestal-Otero4–Reference Xu, Fu and Le6 greater alcohol consumption has been shown to increase the risk of PD. Reference Liu, Guo and Park5
Despite one study suggesting that development of PD does not directly correlate with socioeconomic status, Reference Kivimäki, Batty and Pentti7 others have found that older adults in the early stages of neurodegenerative disease trapped in vulnerable circumstances, such as low socioeconomic status or poor living environment, find themselves at a greater risk of homelessness. Reference Piña-Escudero, López, Sriram, Longoria Ibarrola, Miller and Lanata8 This constellation of inaccessible preventative medical care, convergence of lifestyle factors related to PD, and possible increased risk of homelessness in older adults in the early stages of a neurodegenerative disease emphasizes the need to study PD in the homeless population.
Alpha-synuclein (aSyn) aggregation is a neuropathological hallmark of PD Reference Spillantini, Schmidt, Lee, Trojanowski, Jakes and Goedert9 and multiple system atrophy. Reference Spillantini, Crowther, Jakes, Cairns, Lantos and Goedert10 aSyn is a neuronal protein abundant in brain and also found in peripheral tissues such as the gastrointestinal tract and nerve fibers innervating the heart and the skin. Reference Li, Zhou, Zhang, Chen and Yuan11 To better understand the relationship between homelessness and aSyn pathology, here we studied the prevalence and distribution of pathological aSyn throughout the central nervous system and peripheral organs in 44 consecutive homeless individuals available for autopsy.
Materials and Methods
Cohort
Forty-four consecutive individuals available for autopsy, who met the study inclusion criteria of being homeless at the time of death, were recruited between November 2017 and February 2018 at the Department of Pathology, Forensic and Insurance Medicine, Semmelweis University Budapest, Hungary, shown in supplemental table 1.
For our study, we selected cases that fulfill the criteria of the first group (“rooflessness: without a shelter of any kind, sleeping rough”) of the European Typology of Homelessness and Housing Exclusion (ETHOS) definition on homelessness developed by the European Federation of National Organisations Working with the Homeless (FEANTSA) in 2017. 12 Unfortunately, detailed information on how long the people included in our study cohort had been living without shelter is lacking. In Hungary, in the case of the death of homeless persons where the circumstances of the death are unclear and/or violent death is suspected, a forensic full body autopsy is mandatory. Detailed neuropathological examinations focusing on neurodegenerative pathologies were performed only in the context of this study. The study protocol was approved by the Semmelweis University Regional and Institutional Committee of Science and Research Ethics (Nr. 257/2022).
Tissue Collection
Tissues were collected from 18 regions of the brain and spinal cord, as well as the right and left olfactory bulb, the cauda equina, the extramedullary portion of the vagus nerve, and 27 sites of peripheral organs, shown in supplemental table 2.
All 44 cases were stained and assessed for 5G4 immunoreactivity in the brain regions required for Braak staging of Lewy-related pathology as well as in the olfactory bulb, submandibular gland, parotid gland, esophagus, and stomach, regions 1–23 shown in supplemental table 2.
All cases with positive 5G4 immunoreactivity in brain were then stained for 5G4 in all remaining available tissue regions, regions 24–49 shown in supplemental table 2.
Immunohistochemistry
A two-step approach was used. First, immunohistochemistry was performed for 5G4, an antibody which selectively recognizes disease-associated forms of aSyn Reference Kovacs, Wagner and Dumont13 using methods described in detail elsewhere. Reference Martinez-Valbuena, Visanji and Kim14 Briefly, 4 μm thick formalin-fixed paraffin-embedded tissue sections were immunostained for 5G4 (1:4,000; Roboscreen, Leipzig, Germany) using a Dako automated immunostainer, EnVision detection, and Peroxidase/DAB according to the manufacturer’s instructions (Agilent Technologies, Santa Clara, USA). In cases with immunoreactivity for 5G4, we confirmed the results using an antibody detecting phosphorylated-aSyn (ps129; 1:10,000; Fujifilm Wako Pure Chemical Corporation, Richmond, VA, USA) (supplemental Figure 1). Sections were counterstained for hematoxylin. In the second step, in cases with positive staining for 5G4 in the medulla oblongata we additionally immunostained the frontal and temporal cortices, anterior cingulate cortex, hippocampus, amygdala, posterior part of the basal ganglia, midbrain, pons, and cerebellum. In the three aSyn-positive cases, we screened for additional neurodegenerative diseases, including chronic traumatic encephalopathy (CTE) following criteria provided by consensus studies for CTE, Reference Bieniek, Cairns and Crary15,Reference McKee, Cairns and Dickson16 Alzheimer’s disease-related neuropathologic change, Reference Hyman, Phelps and Beach17 Lewy body disease, Reference Attems, Toledo and Walker18 including Braak stages, Reference Braak, Del Tredici, Rüb, de Vos, Jansen Steur and Braak19 corticobasal degeneration, Reference Dickson, Bergeron and Chin20 progressive supranuclear palsy, Reference Roemer, Grinberg and Crary21 argyrophilic grain disease, Reference Saito, Ruberu and Sawabe22 aging-related tau astrogliopathy (ARTAG), Reference Kovacs, Ferrer and Grinberg23 primary-age-related tauopathy, Reference Crary, Trojanowski and Schneider24 limbic-predominant age-related TDP-43 encephalopathy, Reference Nelson, Dickson and Trojanowski25 and frontotemporal lobar degeneration (FTLD) subtypes. Reference Mackenzie, Neumann and Bigio26,Reference Mackenzie, Neumann and Bigio27 Further immunostaining was performed using the following antibodies anti-tau (pS202/pT205; 1:1000; Thermo Scientific; regions: frontal and temporal cortices, hippocampus, posterior part of the basal ganglia, amygdala, midbrain), anti-phospho-TDP-43 (pS409/410; 1:2000; Cosmo Bio, Tokyo, Japan; regions: frontal cortex, hippocampus, amygdala), and anti-Aβ (6F/3D; 1:50; Dako, Santa Clara, CA; regions: frontal, temporal cortices, hippocampus, anterior type of the basal ganglia, midbrain, cerebellum).
Results
Cohort Demographics
The total number of autopsies, comprising cases with natural and non-natural causes, performed in the Department of Pathology, Forensic and Insurance Medicine, Semmelweis University, Budapest, Hungary, in 2017 was 2563 (including 98 homeless individuals) and, in 2018, it was 2603 (including 107 homeless individuals). We performed the consecutive case collection in the months of November and December in 2017 (in this period 21 homeless individuals) and January and February in 2018 (in this period 23 homeless individuals). Neurological examination was not performed in our cohort.
Forty-four cases (6 female and 38 male) that met inclusion criteria underwent autopsy, shown in supplemental Table 1. The median age was 58 years (range 32–67), and median postmortem delay was 4 days (range 0–14).
The immediate cause of death was defined as the most serious organ lesion found at autopsy and taking into account the results of additional investigations (i.e., histopathology, alcohol levels, and toxicology). The direct cause of death was accepted as due to hypothermia if a body temperature value was available and/or if autopsy signs of hypothermia were detectable.
The most common cause of death was hypothermia (21 cases), followed by chronic obstructive pulmonary disease (5 cases), bronchopneumonia with abscess (3 cases), ischemic cardiomyopathy (3 cases), pulmonary embolism (3 cases), duodenal ulcers with perforation, peritonitis (2 cases), lobar pneumonia (2 cases), and 1 case each from drug overdose, multiorgan disease, ethylene glycol poisoning, liver failure, and leg ulcer, and sepsis.
Staging of Lewy-Related Pathology in Positive Cases
Screening for aSyn pathology did not reveal multiple system atrophy-related pathology in the medulla oblongata in any case. Lewy-related pathology was discovered in the brains of 3/44 cases, all 3 being male, age range 56–60 years shown in Table 1 and Figure 1. The cause of death for all three subjects was hypothermia, with Wischnewski spots present in two of three individuals. One case was Braak stage 2 (case Nr. 10; 60 years old), with Lewy bodies present in the medulla oblongata, tegmentum, and pons (locus coeruleus). Two cases were Braak stage 4 (case Nr. 11: 56 years old and case Nr. 15: 59 years old), with Lewy bodies additionally present in the brainstem and amygdala. Only one case (Nr. 15; Braak stage 4) showed evidence of Lewy pathology in the spinal cord and cauda equina, shown in Figure 1. The extramedullary portion of the vagus nerve was available only on one side in each case. aSyn immunoreactive deposits were seen only in case Nr. 15 (Braak stage 4). The olfactory bulb was available in two cases (case Nr. 10 and 15). In both cases, a mild degree of aSyn-positive deposits was identified, shown in Figure 2.
Representative microphotographs of alpha-synuclein pathology in the brain and spinal cord. a–o, Immunohistochemistry for alpha-synuclein (5G4). a–c, Case 10; d–f, case 11; g–o, case 15. a, d, g, Dorsal vagal nucleus in the medulla oblongata; b, e, h, locus coeruleus of the pons; c, i, substantia nigra in the midbrain; f, j, amygdala; k, CA2 in the hippocampus; l, anterior cingulate cortex; m, cervical spinal cord; n, thoracic spinal cord; o, lumbar spinal cord (inset: cauda equina). b, e, Arrows indicate neuronal cytoplasmic inclusion and arrowheads indicate Lewy neurites. Scale bar = 50 μm for a–f and j–o, 100 μm for g–i, and 10 μm for the inset of o.
Representative microphotographs of alpha-synuclein (aSyn) pathology in the olfactory bulb and peripheral organs. a–l, Immunohistochemistry for alpha-synuclein (5G4). a, g, j, case 10; h, k, case 11; b–f, i, l, case 15. a, b, Olfactory bulb; c, adrenal medulla; d, skin; e, g–i, submandibular gland; f, Auerbach plexus of the esophagus; j–l, esophagus. a, In case 10, very few aSyn-positive deposits are identifiable only in the high-magnification view. b, In case 15, scattered aSyn-positive deposits are observed. Note that the severity of aSyn deposition in both cases is relatively mild considering the Braak Lewy-related pathology staging in these cases (2 and 4, respectively). c–f, In the peripheral organs, no alpha-synuclein immunoreactive deposits are observed. g–l, In contrast, all three cases showed intracytoplasmic alpha-synuclein immunoreactivity in the macrophages in some peripheral organs. Scale bar = 50 μm (a–f), 10 μm (g–l).
Neuropathological findings in all Lewy pathology positive cases. ARTAG= ageing-related tau astrogliopathy; CAA= cerebral amyloid angiopathy
Neurodegenerative Co-pathologies in aSyn-Positive Cases
Two Lewy pathology positive cases were Braak neurofibrillary tangle stage I and one case showed only fine granular cytoplasmic AT8 immunoreactivity of neurons in the transentorhinal area (Braak stage 1b), shown in Table 1. None of the three Lewy pathology positive cases had any evidence of tau pathology including ARTAG. Furthermore, all were devoid of TDP-43 pathology or cerebral amyloid angiopathy, shown in Table 1.
Prevalence and Distribution of Lewy-Related Pathology in Peripheral Tissues
Twenty-seven sites of peripheral organs were examined for Lewy pathology. All three cases had a complete lack of unequivocal disease-associated synuclein deposits outside the central nervous system, shown in Figure 2.
Peripheral 5G4 Positive Mast Cells/macrophages
All 44 cases were assessed for 5G4 immunoreactivity in the submandibular gland, parotid gland, esophagus, and stomach. The three cases with positive Lewy-related pathology, as well as an additional five cases which did not show cerebral Lewy-related pathology, had evidence of 5G4 positive tissue macrophages, shown in Table 2 and Figure 2.
Cases with the presence of 5G4 positive tissue macrophages in the submandibular and parotid glands, esophagus, and stomach
Discussion/Conclusion
To our knowledge, this is the first study of the presence and distribution of aSyn pathology in the homeless. Lewy-related pathology was present in the brains of 3/44 cases (7%). All positive cases were male, one Braak stage 2 for Lewy-related pathology (60 years old) and two stage 4 (56 and 59 years old). One of the Braak stage 4 cases had Lewy-related pathology in the spinal cord, cauda equina, and extramedullary portion of the vagus nerve. Examination of 27 sites of peripheral organs found all three cases with brain Lewy-related pathology were devoid of peripheral organ neuritic aSyn pathology. Interestingly, all three Lewy-related pathology positive cases, as well as an additional five cases, which did not show cerebral Lewy-related pathology, had evidence of 5G4 positive tissue macrophages. Importantly, clinically detected movement disorders Reference Kim, Procyshyn and Jones28 and cognitive disorders Reference Mullady, Castellanos and Lopez29 suggestive of LBD have been reported in the homeless. A neuropathology-based study in a cohort of medical examiner subjects found only 2% of cases with LBD pathology. Reference Uryu, Haddix, Robinson, Nakashima-Yasuda, Lee and Trojanowski30 However, unlike our work, this study did not evaluate exclusively homeless individuals, and the neuropathological screening for neurodegenerative pathology focused on the hippocampus, while we focused on the lower brainstem to detect aSyn pathology.
Due to the nature of our study recruitment, and to the fact that homeless individuals tend to seek healthcare only in acute states or emergency, 12 clinical evaluation of the neurological state was not performed in our cohort. Since there is a paucity of data on similar cohorts, we have compared our findings with studies reporting the frequency of incidental Lewy body disease (iLBD) and other autopsy-based studies without clinical information. The term iLBD comprises the presence of Lewy pathology in individuals lacking a clinical diagnosis of a Lewy body disorder and is fraught with complications arising from differences in sampling strategy and case selection Reference Parkkinen, Soininen, Laakso and Alafuzoff31 as well as technical factors. Thus, in our study, it is more precise to use the term aSyn pathology detected at autopsy without clinical information. aSyn pathology is reported in as low as 3.8% and as high as 17% of neurologically normal people aged over 60. Reference Bloch, Probst, Bissig, Adams and Tolnay32,Reference Gibb and Lees33 Existing literature suggests the prevalence of iLBD increases with age. One study noted an increase from 4.5% to 19.4% between individuals in their 40s and 90s Reference Parkkinen, Soininen and Alafuzoff34 and a second from 3.8% to 12.8% between individuals in their 60s and 90s. Reference Gibb and Lees33 In our homeless cohort, aged 32–67, we found evidence of aSyn pathology in three cases (7%). As described above, extrapolating between studies is complex; however, observing LBD pathology in three individuals aged 56, 59, and 60 may be more prevalent than anticipated based on prior studies, being almost twice as high as the reported prevalence of 3.8% in individuals in their 60s. Reference Gibb and Lees33 Furthermore, all positive cases in our cohort showed aSyn pathology beyond the medulla oblongata, with one case being Braak stage 2 and two being Braak stage 4. One of the cases (Nr. 11) showed an atypical distribution pattern where the locus coeruleus showed the highest amount of pathology, while the substantia nigra showed less than the amygdala. Although this is compatible with Braak stage 4, the lesser involvement of the substantia nigra is unusual. Furthermore, we noted that classical Lewy bodies were sparsely detectable on hematoxylin and eosin staining while immunostaining for disease-associated aSyn revealed more neuritic and fine granular cytoplasmic deposits. This might suggest a different type of seeding of pathological aSyn, as recent studies demonstrate that differences in aSyn seeding are associated with heterogeneity in the progression of LBD. Reference Martinez-Valbuena, Swinkin and Santamaria35
Considering potential reasons for these observations in individuals ranging in age from 56 to 60, we speculate that as homeless people frequently suffer poor nutrition, their immune function and intestinal flora may differ from people with well-balanced nutrition. This might facilitate the intestinal route of PD pathogenesis proposed by Braak’s dual-hit hypothesis Reference Hawkes, Del Tredici and Braak36 leading to a faster progressing disease. However, if this were the case, one might expect aSyn pathology to be observed in the enteric nervous system, which was not the case. Indeed, it is noteworthy that all three cases in our homeless population were devoid of peripheral organ synuclein pathology in the 27 regions examined. Previous studies have found that iLBD cases also exhibit peripheral aSyn pathology. Reference Bloch, Probst, Bissig, Adams and Tolnay32,Reference Beach, Adler and Sue37 Borghammer recently proposed that the pathogenesis of PD be classified into two subtypes, with one initiating in the central nervous system and spreading to the peripheral organs (CNS-first), and another where aSyn pathology originates in the peripheral nervous system before invading the CNS (PNS-first). Reference Borghammer, Horsager and Andersen38–Reference Borghammer and Van Den Berge40 Although clearly our findings need to be explored in larger independent cohorts, our data might suggest that homeless individuals are more likely to develop CNS-first disease.
Determining the prevalence of LBD-type pathology in the homeless clearly warrants further studies in larger independent cohorts. Should such studies find an increased prevalence of LBD in the homeless, particularly at a younger age, a possible explanation could be overlap in environmental and lifestyle factors associated with both homelessness and PD. In our study, all three LBD cases had emphysema, moderate to severe vascular disease, and steatosis. The case with the most severe aSyn pathology (Nr. 15; Braak stage 4) also had a diagnosis of alcohol dependence and evidence of pyelonephritis, with overconsumption of alcohol being a known catalyst for PD progression and mortality. Reference Paul, Chuang and Shih41 Conversely, one must also consider the possibility that early changes in a person’s socioemotional functioning, personality, and behavior caused by the development of Lewy body disease may leave individuals more vulnerable to the evolution of events that lead to homelessness. Reference Piña-Escudero, López, Sriram, Longoria Ibarrola, Miller and Lanata8
Finally, we noted peripheral tissue macrophages and mast cells positive for abnormal aSyn deposition in the submandibular and parotid glands, esophagus, and stomach of 8/44 individuals. We interpret this finding as nonspecific labeling of macrophage enzymes; however, a single study has raised the possibility that these might have pathological relevance. Reference Gray, Munoz, Gray, Schlossmacher and Woulfe42 In PD patients, duodenal biopsy samples frequently show immunoreactive pathology using 5G4. Reference Skorvanek, Gelpi and Mechirova43,Reference Emmi, Sandre and Russo44 We cannot exclude the possibility that the detectability of aSyn aggregates in postmortem samples of peripheral organs is different than in biopsy samples. However, (i) the 5G4 antibody reliably detects aSyn pathology even following long formalin fixation (>10 years) Reference Kovacs, Wagner and Dumont13 and (ii) we did observe labeling of mast cells/macrophages as reported in the biopsy samples. Reference Skorvanek, Gelpi and Mechirova43,Reference Emmi, Sandre and Russo44
To our knowledge, this is the first study of the presence and distribution of aSyn pathology in the homeless. As the number of homeless individuals continues to increase worldwide, we should consider the impact of an increasing number of individuals with undiagnosed neurodegenerative impairments on this already vulnerable population who face significant inequities in health. It is important to explore differences in the care of homeless individuals in different countries to better understand the association of lifestyle factors with neurodegenerative diseases. A summary of the care of the homeless in Hungary has been recently reported. Reference Nagy-Borsy, Vági, Skerlecz, Szeitl, Kiss and Rákosy45 In this study, the authors concluded that “seeking health care is usually not a priority for them as they struggle with other life-threatening conditions such as lack of food and lack of shelter.” Reference Nagy-Borsy, Vági, Skerlecz, Szeitl, Kiss and Rákosy45 Thus, screening for neurodegenerative diseases Reference Kovacs46 in forensic practice might help us understand the spectrum of these conditions in a population not otherwise systematically examined. Finally, future studies comparing population-based autopsy cohorts with and without homelessness status could shed light on the influence of the socioeconomic status on neurodegenerative pathologies.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/cjn.2023.291.
Statement of authorship
Krisztina Danics and Naomi P. Visanji are contributed equally.
KD, NPV, and GGK contributed to the study conception and design. Material preparation, data collection, and analysis were performed by, KD, NPV, SI, GS-K, and GGK. The first draft of the manuscript was written by NPV and SM, and all authors commented on the version of the manuscript. All authors read and approved the final manuscript.
Funding
This study was funded by the Edmond J. Safra Foundation and the Blidner Family Foundation. These sources had no role in the preparation of data or the manuscript.
Competing interests
Krisztina Danics, Shojiro Ichimata, Sarika Mathur, and Gabriella Sára-Klausz have nothing to declare. Naomi Visanji received funding from the Blidner Family Foundation. Dr Kovacs has served as an advisor for Biogen, received royalty for 5G4 synuclein antibody and publishing royalties from Wiley, Cambridge University Press, and Elsevier, received grants from Edmond J Safra Philanthropic Foundation, Rossy Foundation, National Institutes of Health, Michael J. Fox Foundation, Parkinson Canada, The MSA Coalition, Ontario Research Fund and Canada Foundation for Innovation.
Statement of ethics
This study protocol was reviewed and approved by the Regional and Institutional Research Ethics Committee of Semmelweis University, approval number Nr.257/2022 who granted an exemption from requiring written informed consent.
Open access
