The transcriptomic analyses of PD patients skin revealed a large set of over 1000 differentially regulated genes. Interdependent biological and functional networks can be distinguished among the differentially regulated genes and the deregulation of these networks demonstrates a state of severe impairment in basal homeostasis, as well as decreased defence mechanisms to cellular stress in PD skin. Our study results correlate with, and further enhance, the existing knowledge of PD associated pathogenesis and additionally provide a possible molecular explanation for the association of PD with melanoma/non-melanoma skin cancers.
“Skin-Brain” crosstalk in PD – disturbed mitochondrial and protein homeostasis in skin
A focal point of our study highlights the relevance of “skin-brain” crosstalk in understanding the pathogenic mechanisms of neurodegenerative diseases, such as PD. Our findings support the assumption that the biomolecular alterations of PD are systemic and are also reflected in other non-neuronal cells of the body, demonstrating the applicability of skin biopsies as an easily accessible ex vivo solid-tissue based model system, for downstream pathogenic studies and for possible diagnostic/biomarker discovery. Notably, our study highlights that PD reflection in skin is dominated by global suppression of central cellular processes (over 80% of genes were downregulated), indicating a state of extensive cellular stress and compensatory molecular changes limiting the physiological functioning to minimum.
The current understanding of PD pathogenesis in the nervous system, is led by impairment of two major biological systems: mitochondrial dysfunction and protein metabolism. One of the most robust findings in our study, is the evident basal defect in cellular bioenergetics, metabolism and the impairment of mitochondrial function. This is initially demonstrated by the suppression of 1/3 of the components of the mitochondrial electron transfer chain in PD skin, including complexes I, III, IV and V. In addition, the cellular redox homeostasis, oxidative stress and antioxidant response are supressed in PD skin. Mitochondrial impairment in PD skin is also demonstrated by downregulation of genes responsible for mitochondrial dynamics and transport. Adding to the impairment, is the suppression of fatty acid biosynthesis and metabolism. PD has been associated with lipid disturbances, including the specific vulnerability of fatty acids to oxidation, leading to an increase in ROS production and enhanced susceptibility to oxidative stress, but also in relation to the direct interactions of α-synuclein with lipids, as its aggregation is known to occur in the presence of lipid membranes and free fatty acids [
24]. Other impaired central biosynthetic and metabolic processes include oxidation of aldehydes, purine and pyrimidine metabolism, steroidogenesis, amino acid, glucose and carbohydrate metabolism, as well as calcium, iron and glycoprotein metabolism. It is noteworthy, that although most of the effector molecules in the cellular metabolism pathways were downregulated in our study, we observed up-regulation of the transcriptional coactivator and central inducer of mitochondrial biogenesis—
PGC-1α, which controls oxidative metabolism through expression of genes in the mitochondrial respiratory chain, and regulates detoxification of reactive oxygen species [
25]. Our finding is in line with a large genome-wide meta-analysis on PD brain samples, which found
PGC-1α to be the main common pathway to be deregulated between 17 different microarray sets, with most of the
PGC-1α controlled genes being suppressed in PD samples [
26]. In addition, our study observed regulation of other transcription factors related to bioenergetics, indicating that the compensatory changes for disturbed bioenergetics and cellular metabolism in PD skin are regulated on the level of central transcriptional control. In addition, mitochondrial DNA regulation has been associated with PD and decreased expression of mitochondrial polymerase –
POLRMT, an enzyme which plays a fundamental role in expression and replication of the human mitochondrial genome and in mitochondrial protein translation, was observed in our study. We also noted decreased expression of several mitochondrial ribosomal proteins, which function as components of the mitochondrial ribosome and regulate the translation of all the essential polypeptides of the oxidative phosphorylation system. Based on the current study results, it can be concluded that a “basal” transcriptomic defect in cellular bioenergetics and metabolism exists in PD skin, this finding correlating with the known understanding of the pathogenic processes of PD in CNS.
The second most robust finding in our study, is the involvement of genes related to the processes of protein metabolism, transport and degradation. Imbalances of protein homeostasis have been associated with PD pathogenesis in the nervous system, however our study demonstrates that protein homeostasis in PD skin is affected already at the level of ribosomal biogenesis and basal eukaryotic translation, as seen by downregulation of a large set of different ribosomal proteins, as well as several eukaryotic translation initiation and elongation factors. The downregulation of ribosomal proteins in response to stress, can be considered compensatory for limitation of energy expenditure and in order to maintain essential cellular functioning, however long-term inhibition can lead to severe cellular damage and death [
27]. Further, a large group of genes playing a role in protein post-translational modifications, protein folding, aggregation and processing in the ER, are affected in PD skin. Another large group includes proteins participating in the cellular trafficking of proteins, vesicles and organelles. Further, the protein degradation machinery of the UPS is evident in PD skin, highlighted by suppression of several proteins involved in ubiquitination and neddylation, as well as downregulation of 11 subunit components of the proteasome, proteasome assembly chaperones and maturation proteins. In addition, several central genes of the autophagic-lysosome cascade are affected, demonstrating impaired autophagic response in PD skin. Also, a large group of cellular and lysosomal phosphatases, peptidases and proteinases are co-ordinately deregulated in PD skin, indicating defective protein degradation machinery and impaired proteolysis. Taken together our findings demonstrate a severe “basal” impairment of protein homeostasis in PD skin, characterized by suppression of protein translation, cellular trafficking, as well as dysfunctional protein quality control by the UPS and impairment of the autophagic response, which all contribute to the vicious cycle of misfolded protein buildup, further contributing to cytotoxicity.
Impaired skin homeostasis, nuclear processes and tumorigenic pathways—the mechanistic link for predisposition to skin cancer in PD patients?
The second focal point of our study enhances the understanding of the mechanistic association between skin cancer and PD, as demonstrated by a basal defect in skin homeostasis, deregulated nuclear processes, as well as dysbalanced cellular signalling, tumorigenic pathways and inflammatory processes—all these alterations possibly contributing to the specific vulnerability of PD skin to mutagenic hazards (such as UV radiation, somatic mutations, genomic instability), which can provide the basis for the mechanistic link to the increased risk of skin cancer in this patient population.
Our study data demonstrates direct alteration of skin physiology in PD patients, characterized by dysregulation of epidermal renewal/keratinocyte differentiation, cornification/desquamation, response to injury and stress, as well as altered structural and molecular composition of dermis. We observed differential regulation of a large group of keratins and several keratin-related proteins, which indicates impaired epithelial differentiation, tissue fragility and structural integrity of PD skin. Our study revealed a coordinated suppression of parallel pathways of the cornification and desquamation processes, highlighted by the suppression of the EDC, which contains 57 genes crucial for the differentiation process located within a tight cluster on chromosome 1q21 (20 genes of EDC supressed) and also ephrin A1, which is a central regulator of epidermal growth, located in close proximity to the EDC on chromosome 1q arm. In addition, we observed the decreased expression of all genes of the stratified epithelium-secreted peptide complex, as well as the cystatin/cathepsin/transglutaminase pathway, which regulates the cross-linking of the CE proteins and influences the desquamation of the stratum corneum. Further we observed suppression of several different junction and desmosome proteins and deregulation of the antimicrobial defence in PD, indicating that the desmosomal adhesions and anchoring junctions are defective in PD, thereby contributing to impairment of structural integrity and barrier function. In addition, the dermal components of skin were affected, characterized by altered levels of several members of the collagen family, as well as deregulation of cytoskeletal remodelling and dynamics, these changes contributing to impairment of tissue elasticity, predisposing to premature aging of skin, and impacting the structural and compositional remodelling. In conclusion, maintenance of skin homeostasis, thru a balanced orchestration of regeneration, cell renewal, differentiation and senescence, is essential to withstand stress and mutagenic hazard and disruption of this equilibrium can lead to skin disease, such as development of cancers.
The second significant category of differentially expressed genes, with possible impact in both PD and skin cancer involves nuclear regulation of cellular processes in PD skin. This is highlighted by suppression of cell cycle proteins, such as several cyclins,
CDKs, activators/inhibitors of
CDKs and others. In skin, cell cycle regulation is essential to maintain the homeostatic balance between proliferation and differentiation, whereby keratinocytes respond to cell cycle insults and DNA damage by deregulation of the cell cycle and induction of terminal differentiation [
28], however chronic dysregulation can lead to enhanced predisposition for development of cancers. Further, our study indicates that several regulators of basal transcription, and a large group of central transcription factors are deregulated in PD skin, which act as direct regulators/corepressors of genes regulating epidermal terminal differentiation (including the EDC), epithelial-mesenchymal transition, programmed cell death, oxidative stress and tumorigenesis. Other deregulated transcription factors play more variable roles regulating cell growth, proliferation, differentiation, longevity, and acting as downstream targets of multiple signalling pathways. Another important aspect of PD and skin cancer crosstalk can be drawn from the observed downregulation of genes related to DNA/mtDNA repair and degradation, which can contribute to buildup of damaged DNA, interfere with normal cellular functioning and also predispose to tumorigenesis. And lastly, in regard to nuclear regulation of cellular processes, a large group of genes in our study was also associated with the epigenetic regulation of gene expression, as seen by deregulation of proteins participating in chromatin remodelling, DNA binding, RNA/DNA processing and transcriptional/post-transcriptional modification, RNA splicing, as well as a set of different micro-RNAs, small- nuclear- and small-nucleolar-RNAs. The deregulation of cell cycle proteins and transcription factors, as well as the suppression of DNA repair processes and modification of epigenetic signalling, might be executed as a compensatory cytoprotective effect against various pro-apoptotic stimuli from the aspect of PD, however the chronic form of such stress can also provide the basis for molecular predisposition to different forms of skin cancers in these patients.
The third important category in our study, in relation to PD and skin cancer crosstalk, is associated to cellular signalling and tumorigenesis. We observed down-regulation of a large set of tumour suppressors and oncogenes in PD skin. Further many central intra- and inter-cellular signalling genes and growth factors were altered in PD skin, including
Ras and other
small-GTPases,
G-protein signalling pathways,
WNT,
NOTCH and several others. Perhaps the most prominently affected signalling pathway in PD skin is related to
WNT signalling, where deregulation of central
WNT effectors, provides one possible mechanistic association between the processes of epidermal renewal/differentiation, tumorigenesis and neurodegeneration.
WNT signalling plays a dominant role in controlling the patterning of skin, influencing the decision of stem cell lineage and in controlling the functioning of differentiated skin cells, and disruption of this signalling pathway has been associated with the development and progression of both melanoma and non-melanoma cancers [
29]. On the other hand
WNT signalling has been strongly associated with PD pathogenesis, functioning in midbrain dopaminergic neuron development, synaptic plasticity and transmission. In addition, suppression of several members of the Ras superfamily indicates dysfunctional growth, differentiation and survival mechanisms in PD skin, including cellular proliferation, cytoskeletal dynamics/morphology, membrane trafficking, cellular adhesion and vesicular transport.
And lastly, our study indicates dysregulation of immune pathways in PD skin. The interplay of pro- and anti-inflammatory signalling, and the discrimination of causal and effector changes in the context of both PD and cancer is complex, however chronic inflammation has been shown to be one of the main factors fostering all stages of neoplasia, but also one of the pathogenic processes in PD progression, thus the basal inflammatory dysfunction associated with PD can thereby contribute to increasing the risk of cancer development in these patients.
In regard to the existing explanations for the basis of PD and melanoma crosstalk, most studies have emphasized the role of the common skin- and neuro-melanin pathways during early development, and possible defects in tyrosine metabolism. Our study did not observe specific changes in regard to melanin pathways or tyrosine metabolism, however as our study design was set up to evaluate the gene expression changes occurring in whole skin of PD, and not specifically the small population of melanocytes, it can be that these changes remained too subtle for detection with our methodology. Follow-up studies utilizing gene expression profiling of cell-type specific samples, could provide further assistance in dissecting the PD related pathways in skin. In addition, it has been suggested that perturbations on the genetic level could contribute to the underlying crosstalk between melanoma and PD, although the majority of our study subjects (11/12 tested) were not carriers of mutations in any common PD associated genes (data not shown), the role of PD-associated genetic variants which can mediate the expression of quantitative trait locus effects in both skin and brain, cannot be excluded. One limitation of our study is the relatively low log2FC levels observed for gene expression, which might pose difficulties in distinguishing the true signal from noise, thus it cannot be excluded that some of the genes with milder expression levels in our pathway analysis might be attributable to noise. This however is a common finding in multiple ex vivo gene expression studies of a chronic disease and as all affected pathways in our study consisted of multiple genes with intertwining functions, the general conclusions can be considered to be valid.