Overview of the molecular determinants contributing to the expression of Psoriasis and Psoriatic Arthritis phenotypes

Abstract Psoriasis and psoriatic arthritis are multifactorial chronic disorders whose etiopathogenesis essentially derives from the alteration of several signalling pathways and the co‐occurrence of genetic, epigenetic and non‐genetic susceptibility factors that altogether affect the functional and structural property of the skin. Although shared and differential susceptibility genes and molecular pathways are known to contribute to the onset of pathological phenotypes, further research is needed to dissect the molecular causes of psoriatic disease and its progression towards Psoriatic Arthritis. This review will therefore be addressed to explore differences and similarities in the etiopathogenesis and progression of both disorders, with a particular focus on genes involved in the maintenance of the skin structure and integrity (keratins and collagens), modulation of patterns of recognition (through Toll‐like receptors and dectin‐1) and immuno‐inflammatory response (by NLRP3‐dependent inflammasome) to microbial pathogens. In addition, special emphasis will be given to the contribution of epigenetic elements (methylation pattern, non‐coding RNAs, chromatin modifiers and 3D genome organization) to the etiopathogenesis and progression of psoriasis and psoriatic arthritis. The evidence discussed in this review highlights how the knowledge of patients' clinical and (epi)genomic make‐up could be helpful for improving the available therapeutic strategies for psoriasis and psoriatic arthritis treatment.


| BACKG ROU N D
The effectiveness of skin as protective organ is provided by a peculiar structure specifically designed to counteract a variety of dangerous events, including mechanical stress, transepidermal water loss, penetration of microorganisms or physico-chemical irritants.
Alteration in the structural properties of skin or the malfunctioning of the differentiation and maturation process among the skin layers can result in the onset of cancers or other skin-related disorders.
This kind of pathologies is very common in the general population, affecting from 30% to 70% of individuals of any age. 1 In this context, psoriasis (Ps), psoriatic arthritis (PsA) and atopic eczema (AE) are only few examples of a wide spectrum of disease phenotypes involving the skin at different levels. Interestingly, all of them show a multifactorial aetiology characterized by the contribution of genetic and non-genetic factors that interact together to determine a higher or lower susceptibility to develop a specific disease phenotype. [2][3][4][5] In particular, this review will be focused on Ps and PsA, which are both chronic disorders characterized by the dysregulation of inflammatory and immune-mediated responses. Ps affects more than 100 million individuals worldwide. The prevalence rate is highly variable depending on the geographic area, ranging from 0.09% to 11.43%. 6,7 On this subject, the ultraviolet (UV) exposure deserves particular attention since the disease prevalence has been supposed to be correlated with the variable exposure to UV light and latitude. In fact, Ps has been found to be more frequent at higher latitude and in locations less exposed to UV light (such as in Arctic Kasach'ye displaying a prevalence of 11,8%) compared to the prevalence observed in areas closer to the equator (0% in pacific Samoa Islands). 8 One of the possible mechanisms through which UV exposure might exert a protective role is the ability of influence the production and concentration of vitamin D. As a matter of fact, the serum levels of vitamin D have been reported to be low in psoriatic patients compared to healthy individuals in several studies. 9 Of note, variants within genes related to vitamin D metabolism (such as VDR, GC,CYP2R1 and DHCR7) have been found to interact with latitude and sun exposure, altogether contributing to the concentration of serum vitamin D levels in a cohort of Australian origin. 10 Moreover, the higher UV exposure plays a protective role also for AE, for which the local environment plays a role in the positive/negative selection of pathogenic genetic variants. 4 Taken together, these data suggest that UV radiations are able to interact with intrinsic factors, involving directly skin integrity and influencing Ps cutaneous manifestations development.
In fact, UVB radiation influences antigen-presenting cells (APCs), such as Langerhans cells and Langerhans cell-like dendritic cells which, in turn, modulate the keratinocytes functions through cytokines and epidermal cell-derived thymocyte-activating factor (ETAF) production. 11 In particular, the inhibition of Langerhans cells endocytic activity by UVB has been supposed to impair cell migration and maturation, suppressing the antigen-specific immunosensitivity and, ultimately, decreasing the immuno-inflammatory responses in the skin. 11,12 Moreover, similar mechanisms of skin photoprotection through alteration of Langerhans cells functions have been reported for UltraViolet-Solar Simulated (UV-SSR) and UVA radiations. 12 Generally, patients affected with Ps present papulosquamous lesions (ie psoriatic plaques) which are characterized by variable morphology, distribution, severity and progression. Psoriatic plaques are usually itching, stinging and painful and are located on the scalp, elbows, knees, navel and lumbar area. Moreover, Ps can involve the oral mucosa, the soles of the feet and palms of the hands and nails. The main molecular signatures of Ps are the over-production of cytokines that activate Th 17 and Th 1 cells which, in turn, trigger the amplification of inflammation and hyperproliferation of keratinocytes. 13 Especially, the IL-23/Th 17 / IL-17 axis seems to play a crucial role in the initiation of the inflammatory events in Ps, whereas Th 1 /IFN-γ and pro-inflammatory cytokines appear to be typical of the chronic phase. 14 Moreover, the dysfunction of regulatory T cells (Tregs) is an essential feature F I G U R E 1 Subset of known common and differential associated genes between Psoriasis and Psoriatic Arthritis of Ps because of their impaired function in suppressing activation of responder T cells (Tresp) in affected patients. 15 Approximately 20%-30% of psoriatic patients develop PsA that affect the distal joints, the entheses and the axial skeleton over the skin. In fact, PsA patients present a dysregulation of bone morphogenesis and metabolism. 16 The estimated prevalence of PsA ranges between 0,3% and 1%. 17,18 Although PsA and Ps show phenotypic differences, they share several susceptibility factors, especially genes mainly involved in immuno-inflammatory responses and epidermal differentiation ( Figure 1). In particular, the human leucocyte antigen (HLA)-Cw*06:02 allele has been described as the first susceptibility factor for both Ps and PsA. 19 Of note, HLA-Cw*06:02 was found to mediate autoimmunity against melanocytes, through the ability of its protein product to present ADAMTS-like protein 5 (ADAMTSL5). In fact, this complex is recognized by epidermal CD8 + T cells, which directly target melanocytes and produce inflammatory cytokines (such as TNF-α and IL-17) that, in turn, are able to alter melanocytes functions and proliferation, leading to dysregulation of skin homeostasis. 20 Therefore, the alteration of melanocytes functions because of this autoimmune response could be regarded as a process by which HLA-Cw*06:02 predispose to Ps, highlighting this disease as an autoimmune condition whose pathogenic mechanisms also depend from the individual genetic make-up. 20,21 Several studies have identified susceptibility HLA-related supported by the fact that they are able to alter the antigen recognition and binding properties, triggering thereby the activation of signalling cascades and the corresponding immune response. 19 Moreover, the effect of the HLA profile on the response to treatments has been evaluated in order to identify a target therapy suitable for each patient. 23 Over HLA genes, other genetic contributors have been specifically associated with PsA but not with Ps, such as killer-cell immunoglobulin-like receptor (KIR2D 19q13.42), interleukin 4, (IL4, 5q31.1), kinesin family member 3A, (KIF3A, 5q31.1), and beta-1,3-N-acetylglucosyaminyltransferase 2, (B3GNT2, 2p15). [24][25][26] Given the existence of common and differential susceptibility genes and molecular pathways between Ps and PsA, this review will be addressed to further explore differences and similarities in the etiopathogenesis and progression of both disorders, with a particular focus on genes involved in the maintenance of the skin structure and integrity (keratins and collagens), modulation of patterns of recognition (through Toll-like receptors and dectin-1) and immuno-inflammatory response (by NLRP3-dependent inflammasome) to microbial pathogens. In addition, special emphasis will be given to the contribution of epigenetic elements (methylation patterns, non-coding RNAs, chromatin modifiers and 3D genome organization) to the etiopathogenesis and progression of Ps and PsA. Altogether, the investigation of genetic and epigenetic factors associated with Ps and PsA may provide additional knowledge concerning the disease mechanisms and pave the way for the development of novel diagnostic protocols and more effective treatments.

| MAINTENAN CE OF S K IN S TRUC TURE AND INTEG RIT Y: THE ROLE OF KER ATIN G ENE S
Keratins take part in cell proliferation and inflammatory response, by mediating the interactions either among cells or cells and extracellular matrix (ECM). 27  of KRTs in the etiopathogenesis of Ps is consistent with the rapid turnover, and the hyperproliferation of keratinocytes typically observed in psoriatic lesions. On this subject, the deletion of KRT1 has been showed to result in the dysregulation of the skin barrier caused by the release of S100A8 and S100A9 heterodimers which, in turn, trigger the Toll-like receptor 4 (TLR4)-dependent activation of NLR family, pyrin domain-containing 3 (NLRP3) inflammasome and, ultimately, the initiation of the inflammatory cascade. 31 Furthermore, KRT17 has been supposed to act as an antigen, triggering an autoreactive T cell immunity response, the release of the interferon γ (IFNγ) and thereby the inflammation of epidermis. 32 Moreover, it seems that KRT17 expression can be directly induced by IL-17 through STAT1-and STAT3-dependent mechanisms. 33 KRT16 may activate the inflammatory response by regulating the production of glutathione (GSH) through the nuclear factor erythroid-derived 2-related factor 2 (NRF2) and, as a consequence, modulate the response to oxidative stress in keratinocytes. 34 Over the role of keratins in the skin structure maintenance, highly penetrant mutations in KRTs can make tissues more fragile and unorganized, causing specific genetic disorders (such as epidermolytic ichthyosis, type I and II pachyonychia congenita). 27,35 However, the potential impact of KRTs polymorphisms on the susceptibility to Ps and PsA is still unclear and deserves to be further investigated.

| MAINTENAN CE OF S K IN S TRUC TURE AND INTEG RIT Y: THE ROLE OF COLL AG EN G ENE S
Collagens are the most abundant fibrous proteins of the ECM and provide stability, resilience and strength to human tissues. 36,37 The structural properties of collagens are given by its triple-helix conformation able to form fibrils or reticuli, which altogether allow the interaction with other ECM components and cell surface proteins. Therefore, disruption of the structure and function of collagens because of highly penetrant mutations can affect every tissue or organ and cause several pathological phenotypes. In particular, mutations within COL17A1 cause junctional epidermolysis bullosa, non-Herlitztype, 38 whereas mutations in COL7A1 are responsible for dominant or recessive dystrophic epidermolysis bullosa (Table 1). 39 Furthermore, COL1A1, COL3A1 and COL5A1 (Table 1) are physiologically expressed in the skin and are involved in the proper maintenance of blood vessels. Their dysfunction, indeed, may account for vessels fragility and joint hypermobility. 40 Other genes coding for collagen VI chains, namely COL6A5 and COL6A6 (Table 1), are characterized by a more limited localization in adult tissues. 41 COL6A5 is mainly expressed in the skin, although it has been detected in adult testis, lung and intestine. 42 This gene has been investigated for genetic association with AE, although results are highly controversial. [43][44][45][46] In particular, a recent study on a cohort of 428 Mediterranean individuals (namely, Italian, Egyptian and Greek subjects) affected by AE did not find any association between COL6A5 and AE in these populations. 47 However, the expression of COL6A5 in the papillary dermis 42 suggested a potential association with Ps and PsA, given the fact that the papillary dermis can be affected in both disorders. Supporting this hypothesis, we found a significant association of rs12488457 (A/C) variant located in COL6A5 with Ps and PsA in a recent study on the Italian population. 48 .
The investigation on COL8A1 and COL8A2 (Table 1) unveiled interesting insights concerning the role of ECM remodelling and angiogenesis in Ps and PsA. 49 Previous studies found that a polymorphism (rs13081855, G/T) of COL8A1 is associated with an increased susceptibility to exudative Age-related macular degeneration (AMD), and it is likely to interact with the pro-angiogenic factor VEGFA. 50,51 Concerning skin disorders, COL8A1 has been found to be strongly associated with AE in the previously mentioned study on the Mediterranean populations. 47  encode important components of cartilage and are known to play an important role in joints maintenance and bone formation. 36,37 In this context, mutations in these genes have been associated with different forms of chondrodysplasia 36 and increased levels of collagen X have been found in the serum of PsA patients. 52 On this subject, we found that the COL10A1-rs3812111 (T/A) variant was exclusively associated with PsA, suggesting the possible contribution of bone metabolism-related factors in the differential etiopathogenesis of PsA compared to Ps. 48 Given these data and considering the role of collagen genes in the maintenance of bone homeostasis, their investigation may provide interesting insights into different etiopathogenetic pathways underlying Ps and PsA. 53

| G ENE S AFFEC TING RECOG NITI ON PAT TERNS AND IMMUNO -INFL AMMATORY RE S P ON S E TO MI CROB IAL PATHOG EN S
The knowledge of the human microbiome represents a new standpoint in the study of complex disorders. In particular, human microbiome is characterized by bacteria, archaea, viruses and eukaryotic microbes that live in the human body. This microbial community is characterized by a variable composition, whose modifications may lead to different signatures among healthy and affected subjects.
Human microbiome participates in the metabolism and interacts with the immune system and the skin barrier. In particular, skin microbial community can be highly heterogenous, although Staphylococcus spp., Corynebacterium and Propionibacterium represent most of bacterial populations. 54 However, the microbial load may also be influenced by age, gender and lifestyle, suggesting that individuals may be stratified according to their microbiome profile. 55,56 The interactions between microbiome and skin are crucial for the maintenance of the barrier function, the defence from pathogens and tissue repair. On this subject, the proper maintenance of the skin homeostasis is contributed to by the production of anti-inflammatory molecules and anti-microbial proteins (AMPs) 54  might also be related to an excessive response to microbial pathogens. 59 It is a matter of fact that microbial infections strongly increase the susceptibility to cutaneous disorders, especially Ps and PsA. 17,60 Interestingly, the presence of specific microorganisms on the skin of psoriatic patients may exacerbate the activation of adaptive immunity and promote the evolution towards the arthropathic form. 53 In fact, a higher prevalence of C Albicans infections has been found in psoriatic patients, 61,62 The recognition mechanisms in response to fungine infections include different receptors, such as dectin-1 and TLRs. Dectin-1, encoded by CLEC7A (Table 1) is a lectin expressed on immune cells that are able to identify the saccharide wall antigens located on the surface of C Albicans. 62 In vitro models showed an up-regulation of dectin-1 together with an alteration of Ps-associated pathways. 19,63,64 Interestingly, the stop codon polymorphism c.714T > G (rs16910526, p.Tyr238Ter) within CLEC7A has been found to increase the risk of developing chronic mucocutaneous candidiasis. As a matter of fact, defects in CLEC7A expression and activity have been found to correlate with dysregulation of inflammatory cytokines, including reduction of IL-17, which is able to induce the production of anti-microbial peptides. Moreover, the down-regulation of IL17 production may prevent the treatment with IL17 inhibitors. 62 TLRs (TLR2, TLR4, TLR6, TLR3 and TLR9) ( Table 1) instead are able to recognize fungine polysaccharides and fungine nucleic acids, respectively. 65,66 The genetic association between TLR2/TLR4 and Ps remains controversial and needs to be further investigated. 67 and caspase-1 (coded by CASP1) ( Table 1). 73 Genetic studies re- Several studies suggested that nail inflammation could be a specific marker of PsA because of close proximity between the nail, an enthesis (namely the distal interphalangeal joint, DIP) and the collateral ligaments around nail. 78 However, further studies are still needed to clarify the role of microbiome and immune-related genes (CLEC7A, TLRs, NLRP3, PYCARD and CASP1, Table 1) in the etiopathogenesis of Ps and its evolution towards PsA. Concerning the contribution of DNA methylation to Ps or PsA, deregulation of the global methylation pattern has been found in psoriatic lesions. [83][84][85] In particular, the promoting region of Epidermal Differentiation Complex (EDC locus, 1q21.3) genes has been found hypomethylated. This is in line with the overexpression of epidermal differentiation genes in psoriatic patients, namely S100 calcium-binding protein (S100A3, S100A5, S100A7, S100A12, 1q21.3) and

| EPI G ENE TI C S
late cornified envelope protein 3A (LCE3A, 1q21.3). 83 86 In this context, epigenetic modifications have been showed to influence the shaping and the rearrangements of 3D genome architecture in health and disease. Therefore, it would be interesting to understand whether specific epigenetic modifications may induce conformational changes at chromatin level which, in turn, may confer a higher/lower susceptibility to Ps or PsA. Of note, EDC locus has been shown to present a peculiar 3D organization in keratinocytes compared to thymocytes, where this locus is transcriptionally silent. 87 In addition, the deregulation of chromatin remodelling genes involved in cell proliferation and differentiation has been described in subjects with Ps. On this subject, the SATB1 gene (Table 1) acts as 'genome organizer' in epidermal differentiation promoting the transcription of EDC genes. 88 CTCF gene (Table 1) (Table 1), were significantly correlated with the disease activity and the excessive inflammatory response. According to these results, lnc-RP11-701H24.7 and lnc-RNU12 were proposed as potential biomarkers for PsA. 94 Over lncRNAs, an altered expression of miRNAs (such as miR-203a, miR-146a, miR-146b, miR-125b, miR-21, Table 1)  Such a treatment approach proved to be much more cost-effective than designing drugs ab initio. Conversely, the knowledge of differential processes discriminating Ps from PsA is critical to better understand the specific hallmarks of disease and improve the diagnosis and the selection of the optimal therapy. In conclusion, the genetic, epigenetic and molecular factors discussed in this review may provide interesting insights concerning the biological processes underlying the etiopathogenesis of Ps and PsA, highlighting possible therapeutical targets and candidate biomarkers to be used for diagnostic and prognostic purposes. On this subject, the genes (Table 1)  F I G U R E 2 Overview of the networks underlying the etiopathogenesis of Psoriasis and Psoriatic Arthritis discussed in this manuscript. In particular, the interactions among genetic (genes coding for tissue structural components, microbial recognition and inflammatory mediators), epigenetic (epigenetic changes and epigenetic modifiers) and non-genetic (environment and microbiome) factors contribute to the onset and progression of both the diseases and may also account for peculiar blended phenotypes (such as the severe form characterized by nail, scalp and intergluteal skin involvement) features 99,100 improve the patient's quality of life and prevent as much as possible harmful complications.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data and materials were included in the manuscript.