The Skin and Lewy Body Disease

Sebum abnormalities

It has recently been discovered that patients with Lewy body disease emit characteristic odors in bodily secretions and excretions before developing many other disease-related symptoms. Reports of “super smellers” correctly identifying PD patients based on odor alone reveal potential diagnostic information [7]. The chemicals associated with body odor are known as volatile organic compounds (VOCs) which derive from the interactions between eccrine, sebaceous, and apocrine glands secretions with resident skin bacteria [8]. This change in odor in PD patients suggests a variation in skin microflora and skin physiology associated with the altered biochemical composition of sebum. Specifically, the unique microbial profile of the skin of PD patients combined with increased sebum secretion may result in changes in the production of these metabolites [9, 10]. This has inspired a few studies to begin investigating alterations in sebum from patients with PD to better understand the characteristic odor.

Sebum lipidomics represents one potential method for the discovery of PD biomarkers through profiling of sebum building blocks. One study used gas chromatography-mass spectrometry (GC-MS) profiling of free fatty acids, fatty alcohols, squalene, cholesterol, vitamin E, triglycerides, and wax esters to identify differences between patients with PD and controls. A general increase of lipid components was found in the PD group when compared to controls, suggesting hyperactive sebum. Similarly, sebum weight and excretion weight values were significantly higher in the PD group, supporting the altered sebum production in these patients [11].

Other studies have utilized metabolomics to identify lipid metabolites and VOCs responsible for the odor in PD. Using liquid chromatography-mass spectrometry (LC-MS) of sebum, one found metabolites in ceramide, triacylglycerol, and fatty acyl classes to be downregulated while glycosphingolipid and fatty acyl metabolites were upregulated in PD samples compared to controls [12]. In a second study, differential expression of triacylglycerides and diglycerides in PD sebum was detected using paper spray ionization coupled with ion mobility mass spectrometry [13]. Another analyzed sebum samples from PD patients and healthy controls using thermal desorption (TD)-GC-MS to identify VOCs associated with the characteristic odor. Perillic aldehyde was observed to be lower in PD samples while eicosane was observed at higher levels in PD samples. Hippuric acid and octadecanal trended toward being significantly increased in PD [9, 10]. Finally, in a validation study using the same technology, VOCs from sebum measured with TD-GC-MS categorized PD and control samples with an accuracy of 84% suggesting that this associated odor profile is unique to PD [10]. In sum, these studies demonstrate promising results in the identification of peripheral and minimally invasive biomarkers for Lewy bodydisease.

Seborrheic dermatitis

Seborrheic dermatitis (SD) is a chronic, inflammatory dermatosis characterized by scaling, erythema, pruritis, induration, and oiliness affecting areas of the skin where sebaceous glands are prominent. The pathophysiology of SD involves a combination of increased sebaceous gland secretion of lipids, colonization with Malassezia yeast, and a susceptible immune system. The prevalence of SD in the general population is estimated to be around 5% [14]. In PD, SD has been reported in up to 60% of patients with the prevalence showing a positive correlation with age and severity of motor symptoms [15]. The etiology of SD in PD remains unclear, but it has been suggested that autonomic dysfunction in PD may lead to increased sebum production, which can facilitate overgrowth of Malassezia species, and in conjunction with reduced facial movements and altered sebum composition, could lead to SD [16]. This is supported by many studies finding that treatment with levodopa reduces sebum secretion and leads to the improvement of SD in these patients [17–20]. SD lesional skin in patients with PD has been shown to have almost double the density of Malassezia yeasts when compared to SD skin from non-PD patients. Combining this with findings of high phosphatase and lipase activity suggests that dense colonization with Malassezia is associated with the pathogenesis of SD in PD [21]. Other potential contributors include gene polymorphisms involved in lipid regulation which confer increased risk of PD and increased diversity of Malassezia species in the skin of PD patients [22, 23].

An important connector between Lewy body disease and SD is inflammation. Neuroinflammation describes the inflammatory response to nervous system injury mediated by activated microglia and astrocytes, cytokines, chemokines, and various second messengers and it is widely recognized to be an important contributor to neurodegenerative diseases [24]. Several postmortem studies have suggested an inflammatory immune response directed by microglia to be involved in the pathogenesis of Lewy body disease. Reports of Lewy body-containing microglia, microglia engulfing Lewy bodies, and increased expression of inflammatory cytokines support this [25–27]. Additionally, altered levels of inflammatory cytokines, including interleukin (IL)-1B, IL-2, IL-4, IL-6, IL-10, and tumor necrosis factor (TNF)-α, have been detected in both blood and cerebrospinal fluid (CSF) from patients with Lewy body disease. Further, the profile of inflammation in the CSF seems to change as Lewy body disease progresses with IL-1B and IL-6 highest during the prodromal phase ofDLB [28].

With respect to SD, Malassezia yeast are capable of inducing keratinocytes to produce proinflammatory cytokines (IL-1a, IL-6, IL-8, IL-12, and TNF-α) [29, 30]. Further, IL-8 production by Malassezia species has been found to be significantly increased in skin biopsy samples taken from SD patients compared to controls. Malassezia induce secretion if IL-8 by keratinocytes, highlighting the pro-inflammatory role of the yeast in SD pathogenesis [31]. The serum of patients with SD has also been studied for inflammatory markers with increased levels of IL-2 and interferon-γ detected when compared to controls [32]. It is possible that the inflammatory profile in Lewy body disease predisposes patients to the development of SD.

Lastly, lipids are essential components of skin structure and function. Ceramides, cholesterol, and free fatty acids comprise the main lipid groups in the stratum corneum of skin with antimicrobial properties and the ability to regulate chemokines and cytokines [33, 34]. Malassezia are lipid-dependent yeasts which thrive on sebum-rich skin. Oleic acid, a Malassezia-produced fatty acid metabolite present in human sebum, has been shown to induce scalp desquamation in susceptible individuals [35]. It is thought that lipases and phospholipases of Malassezia yeasts are essential to its ability to produce free fatty acids which directly contribute to its virulence in SD [36]. When considering this alongside the alterations in sebum present in Lewy body disease, it is possible that the interactions between Malassezia and the lipid composition of sebum is implicated in the pathogenesis of thetwo diseases.

Autoimmune blistering diseases

Bullous pemphigoid (BP) is an autoimmune bullous disorder characterized by bullae and vesicles in the flexural aspects of the limbs and abdomen. The pathogenesis of BP involves autoantibodies directed against hemidesmosomes in the epidermis [37]. Several observational studies have established an association between BP and various neurological disorders, including PD, Alzheimer’s disease, stroke, and multiple sclerosis. A meta-analysis of fourteen studies revealed the association between BP and neurologic disorders, as patients with BP were more than three times more likely to have PD compared to controls. Similarly, patients with BP were more than four times more likely to have dementia of any type [38]. In another meta-analysis of seventeen studies reporting risk between BP and PD only, results showed significantly increased risk of BP in PD patients with a pooled odds ratio of 2.67 compared tocontrols [39].

Pemphigus is another autoimmune blistering disease characterized by painful vesiculobullous lesions involving the skin and mucous membranes. Analogous to that of bullous pemphigoid, the pathogenesis of pemphigus involves autoantibodies reacting against epithelial surface antigens, desmoglein 1 and 3 [40]. In a cross-sectional study assessing prevalence of pemphigus in various neurologic diseases, a two-fold increase in the odds of both dementia and PD were found among patients with pemphigus compared to controls. A genetic predisposition may be involved in this association between pemphigus and PD as both are seen in increased frequency among Ashkenazi Jews [41].

Explanations for these two associations have suggested that cross reaction of autoantibodies against similar antigens in the skin and brain may be involved [42, 43]. It has additionally been postulated that neurodegeneration associated with dementias and PD may lead to generation of autoantibodies [38]. No studies to date have investigated this association with DLB, specifically.

Rosacea

Rosacea is a chronic, inflammatory skin condition characterized by facial flushing, telangiectasias, erythema, papules, pustules, and ocular lesions. Four subtypes of rosacea are recognized: 1) erythematotelangiectatic, 2) inflammatory papulopustular, 3) phymatous, and 4) ocular [44]. A few studies have identified an association between rosacea and PD including a nationwide study of the Danish population with up to 15 years of follow up data. In this study, both an increased risk of new-onset PD and a younger age at PD onset were observed in patients with rosacea. Specifically, PD occurred approximately 2.4 years earlier in patients with rosacea compared to controls. Additionally, a tendency toward increased risk for PD was noted in patients with ocular rosacea [45]. Importantly, however, this study did not perform sensitivity analyses for the remaining subtypes of rosacea.

The association between rosacea and neurologic disorders is not fully understood, but explanations center around matrix metalloproteinases (MMPs). Specifically, skin affected by rosacea has been shown to have increased expression and activation of several MMPs [44]. Increased activity of MMPs in tear fluid from patients with ocular rosacea has also been found [46]. With respect to neurologic diseases, MMPs have been implicated in PD pathogenesis by contributing to dopaminergic neuronal loss and elevated levels of MMP-9 are found in the CSF of patients with certain dementias [47].

An additional pathogenic link between the two diseases is highlighted by neurogenic rosacea. This distinct rosacea subtype is characterized by symptoms of cutaneous burning/stinging, facial flushing, edema, migraine, neuropsychiatric symptoms, and tremor [48]. Due to this, it is postulated that neuronal dysregulation is involved in rosacea pathogenesis through vasomotor instability, release of inflammatory neuropeptides, and neuronal injury. Further, dysautonomia may contribute to the facial flushing observed in both rosacea and PD [48]. No study to date has investigated the relationship between rosacea and DLB.

Melanoma

In general, an inverse association between PD and cancer has been described with the exception of melanoma as several epidemiologic studies have demonstrated an increased risk of melanoma in patients with PD [49]. The largest prospective study of 2,106 patients with PD demonstrated age- and sex-adjusted relative risk of melanoma to be more than seven times that which was expected [50]. A second study found the risk to be bidirectional, with a melanoma diagnosis being associated with 50% increased risk of PD and PD diagnosis associated with a two-fold increase in risk of melanoma diagnosis [51, 52].

Pigment alterations may play a role in this association. Specifically, melanin, a collection of natural pigments, determines skin and hair color and is dysregulated in skin cancers such as melanoma. In melanoma, melanin serves a protective role with decreased melanin density associating with increased incidence of melanoma [49]. Similarly, neuromelanin is a pigment produced by neurons in the substantia nigra and is dysregulated in neurodegenerative diseases like PD and DLB. In both pre- and post-mortem studies, significant loss of neuromelanin-containing neurons has been demonstrated in both PD and DLB cases [53, 54]. While melanin and neuromelanin are distinct pigment types, the synthesis of both depends on 3,4-dihydroxyphenylalanine (DOPA) as an intermediate in rate-limiting steps [49]. Specifically, tyrosine is oxidized to dopamine quinone in melanocytes by tyrosinase with DOPA as an auto-activator. In nigrostriatal dopaminergic neurons, tyrosine is oxidized to DOPA by tyrosine hydroxylase [55]. This unique ability of both dopaminergic neurons and melanocytes to synthesize forms of melanin from common precursors suggests that these biosynthesis pathways may share dysregulated factors which help explain the association between Lewy body disease and melanoma.

Similarly, hair color is one of the most important pigmentation phenotypes and is determined by melanin. Hair color is largely implicated in melanoma as individuals with red hair have three times the risk of melanoma than those with black hair [56]. Interestingly, risk of PD has been found to increase as darkness of hair color decreases. Further, blonde and red hair remain significantly associated with increased risk of PD after adjustments for ethnicity when compared to black hair. Individuals with red hair have twice the risk of PD compared to those with black hair [56]. Genetic analyses further support this link as individuals homozygous for the red hair associated MC1R Arg151Cys allele have an increased risk of PD when compared to individuals with black hair [57].

It is well established that PD is characterized by degeneration of dopaminergic neurons in the substantia nigra (SN). Recently, increased expansion of the echogenic area of the SN has been considered a marker for PD. To better understand the mechanistic link between pigmentation and PD, one study sought to investigate the association between skin pigmentation and SN echogenicity. Interestingly, lighter skin phototype was found to be associated with increased echogenic SN area and increased frequency of abnormal expansion of this echogenic area [58].This again may help to explain the association between PD and melanoma as individuals with lighter skin types are at highest risk for melanoma. The risk of melanoma in patients with DLB has not been studied.

Role of skin biopsy

The diagnosis of DLB and PD remains largely clinical based on recognition of the constellation of classic symptoms. Both diseases are characterized by the accumulation of pathologic p-synuclein in the brain, detection of which would serve as the only definitive diagnosis. Recently, Amprion’s SYNTap, a CSF-based α-synuclein seed amplification assay, received FDA Breakthrough Device Designation for detecting misfolded α-synuclein but has not yet been approved as a biomarker for Lewy body disease [59]. It remains that the identification of p-synuclein in peripheral tissues would serve as an easily accessible source of diagnostic information which could help recognize Lewy body disease in its earlier stages. Recent reviews discuss in detail the studies using p-synuclein in skin biopsies to diagnose synucleinopathies [60, 61]. The identification of p-synuclein in Lewy body disease skin was first demonstrated in postmortem studies with one finding 70% sensitivity and 100% specificity in distinguishing PD patients from controls [3, 62–64]. Since then, several in vivo studies have further demonstrated the utility of p-synuclein identified by immunofluorescence in the skin for diagnosis of PD [65–71]. Others have also determined abnormal α-synuclein deposits, including p-synuclein, in the skin could serve as a diagnostic biomarker for DLB [72–74]. A proximal-distal gradient of p-synuclein deposits has been identified with the highest percentage of positive samples taken from cervical sites followed by distal thigh, and distal leg, respectively [66, 72]. Further, Lewy body synucleinopathies can be differentiated from multiple system atrophy (MSA) based on the cell-type specific location of p-synuclein aggregates in the skin. Specifically, in MSA, p-synuclein is found as Schwann cell cytoplasmic inclusions in skin Remak non-myelinating Schwann cells whereas in PD and DLB, p-synuclein is found in neurons only [75]. Overall, studies utilizing immunofluorescence to detect p-synuclein in skin of patients with Lewy body disease have demonstrated sensitivities ranging from 52–100% and consistent specificities of 100%. Aside from anatomic location of skin biopsy, other factors contributing to differences in sensitivities include sample fixation techniques and section thickness [61].

More recently, additional techniques beyond immunofluorescence have been investigated in an attempt to improve diagnostic utility. One using skin biopsies from both living PD patients and cadavers found that p-synuclein has aggregation seeding activity that was significantly higher than in controls. As was true in prior skin biopsy analyses, the sensitivity and specificity of p-synuclein in PD patients were higher in the posterior scalp and posterior cervical regions than in abdominal skin. This study used real-time quaking-induced conversion (RT-QuIC) and protein misfolding cyclic amplification (PMCA), two assays which were initially developed to detect prion aggregates in CSF [76]. The latter has also demonstrated utility in diagnosing PD using CSF [77]. This skin-based analysis demonstrated comparable diagnostic sensitivity and specificity to that of CSF-based analyses of p-synuclein suggesting that it may serve as a diagnostic biomarker for PD [76]. Additional studies have verified that both immunofluorescence and RT-QuIC of skin biopsies can accurately diagnose Lewy body disease and provide valid alternatives to CSF analysis (Table 1) [78–80].

Utility of skin biopsy for diagnosis may be facilitated by commercially available assays such as CND Life Science’s Syn-One test which uses immunofluorescence to detect phosphorylated α-synuclein in skin punch biopsies. Unpublished data reported sensitivities of 74%, 90%, and 96% were found when considering one, two, or three biopsy sites, respectively. With subsequent validation, the Syn-One test demonstrated 99% accuracy in discriminating Lewy body disease from controls. Although not yet FDA approved, further validation of the test in a multicenter clinical trial will provide additional data to support its clinical use [81, 82].