Keratins (KRT), the major components of the epithelial cytoskeleton, are responsible for maintaining the structural stability and integrity of keratinocytes. So far, more than 54 mammalian keratins have been identified, contributing to ~30–80% of total protein and forming the ~10 nm intermediate filaments (IFs) in keratinocytes ^[1][2][3][7,8,9]. These highly diverse keratins are subdivided into two classes based on their pH: The acidic type I keratins (KRT9-KRT40) and the neutral–basic type II keratins (KRT1-KRT8) ^[4][5][6][10,11,12]. Keratins form a heterodimer between one type I keratin and one type II keratin, while self-assembling into antiparallel, staggered tetramers, forming an intermediate filament through longitudinal and lateral interaction ^[3][9]. Genes encoding type I and II keratins are clustered into two distinct chromosomal regions including chromosome17q12–q21 for type I keratins (except K18) and chromosome12q11–q13 for all type II keratins and KRT18 ^[7][8][9][13,14,15]. Despite their distinct locations on the genome, specific pairs of type I and type II keratins exhibit a highly specific and consistent expression pattern within a specific epidermal cell layer ^[5][10][11,16]. For example, in the interfollicular epidermis, KRT14-KRT5 is the major type I-type II keratin pair expressed in proliferative basal keratinocytes, whereas differentiated keratinocytes in the suprabasal layers downregulate KRT14-KRT5 and express KRT10-KRT1 as the major keratin pair. During wounding, stressed keratinocytes rapidly induce de novo transcription of KRT16/17-KRT6, whose expression is normally restricted to epidermal cells of glabrous skin, the oral mucosa, and several appendages ^[1][7].

Epidermal keratinocytes are constantly exposed to physical stress, and therefore a primary role of the keratin intermediate filament is to act as a flexible scaffold and provides resilience, enabling epidermal cells to resist mechanical stress ^[11][17]. Similar to other IFs, keratins have a head-rod-tail structure, with a central α-helical rod domain that is essential for intermediate filament formation ^[12][13][18,19]. Keratin IFs extend from the desmosomes to the nuclear membrane, providing tissue resilience to resist environmental stresses ^[14][20]. In addition to providing structural support to epithelial cells, growing evidence has shown that keratins also regulate cell proliferation, migration, adhesion and inflammatory features of keratinocytes ^[1][10][15][7,16,21]. Mutation or abnormal expression of keratin proteins is associated with a variety of skin diseases, such as skin blistering diseases (Epidermolysis bullosa simplex and Epidermolytic hyperkeratosis), psoriasis and skin tumors (squamous cell carcinoma and basal cell carcinoma). 

Studies from Coulombe’s and other groups using null mouse models has established the role of KRT6 or KRT16 in maintaining cell adhesion and optimal cell migration. Mice with germline deletion of Krt6a and Krt6b, the two isoforms of Krt6, appear normal at birth, but die during the first week after birth due to massive oral epithelial blistering, which results in death due to poor nutrition ^[16][75]. Mice overexpressing human KRT16 develop skin lesions concomitant with alterations in keratin filament organization and in cell adhesion at one week after birth ^[17][18][76,77]. In an ex vivo skin explant culture model, Krt6α/β null keratinocytes exhibited an enhanced epithelialization potential due to increased migration ^[19][78]. Loss of KRT6α/β was accompanied by a decrease in KRT16 protein expression without a change in K16 mRNA levels, suggesting that K6 may have been necessary for maintaining the stability of its keratin binding partner KRT16 ^[19][78]. However, when Krt6α/β null skin was grafted onto immunocompromised mice, Krt6 null KCs exhibited epithelial fragility after incisional wounding, during which swelling and lysis of the null keratinocytes was observed. In addition, this phenotype became exacerbated when the grafted skin was first subjected to chemical irritation followed by mechanical wounding ^[19][78]. In a different study, overexpressing human KRT16 in mice led to a delay in wound closure in vivo, and the transgenic K16 skin explants also exhibited a significant reduction in keratinocytes outgrowth ^[20][79]. KRT6 interacts directly with Src kinase, dampening its kinase activity and the migratory potential of keratinocytes during wounding ^[21][80]. A recent study from Coulombe’s group showed that KRT6 interacts with myosin IIA to limit migration potential of keratinocytes, and KRT6 is also required to maintain the expression of desmoplakin, which mediates the attachment of IFs to desmosomes, to maintain keratinocytes’ cell-cell and cell-matrix adhesion ^[15][21]. Together, these results suggest that the KRT6-KRT16 keratin pair may be required for maintaining the resilience necessary to withstand the rigors of a wound site at the cost of a delay in epithelialization, and thus highlights the role of these keratins in collective cell migration.

Studies have shown that KRT17, but not KRT6 or KRT16, plays a role in driving keratinocyte hyperproliferation during wounding or psoriasis. Induction of KRT6, KRT16 and/or KRT17 during wound closure is known to occur at the expense of differentiation marker KRT10 and KRT1. Culombe’s group showed that the capacity of KCs to enact this switch was already acquired in mouse embryos at E11.5 ^[22][81], a time that was well ahead of the onset of differentiation (~E13.5) and epidermal barrier formation (~E16.5). In addition, only KRT17 null embryos, but not KRT6 null embryos, exhibited significant delay in wound closure compared to WT controls ^[22][81]. The lack of a wound closure phenotype in KRT6 null embryos may be due to the concomitant upregulation of the expression of related type II KRT5, which may have functional redundancy to KRT6 ^[22][81].

KRT17 null keratinocytes are smaller in size than wild-type cells in vivo and in culture, and protein translation is depressed in KRT17 null cells, correlating with decreased mTOR/AKT signaling, which is essential for cell growth and protein synthesis ^[23][82]. It has also been shown that the KRT17 intermediate filament network interacts with STAT3 to facilitate STAT3 phosphorylation and nuclear transportation and induces cyclin D1 expression, leading to keratinocyte hyperproliferation ^[24][83]. The absence of KRT17 delays the onset of epidermal hyperplasia, which is preceded by reduced inflammation and a shift of cytokine profiles from a Th1-Th17 dominant profile to a Th2 dominant profile in the mouse model of basal cell carcinoma ^[25][84]. This study suggests that KRT17 promotes keratinocyte proliferation by polarizing the immune response towards Th1 and Th17 dominated profile, which is known to boost keratinocyte hyperproliferation.

KRT17 is ectopically expressed in numerous pathological skin conditions associated with robust inflammation, including wounds, various skin tumors, virus-induced warts and psoriasis ^[26][25][52,84], suggesting that KRT17 may also play an important role in immune regulation. Studies have demonstrated that KRT17 modulates the expression of an array of inflammatory cytokines that shapes the development of the adaptive T cell responses.

It has been reported that KRT17 interacts with the RNA-binding protein hnRNP K, a member of the heterogeneous nuclear ribo-nucleoprotein family of DNA/RNA-binding protein involved in gene expression ^[27][85]. This interaction is required for cytoplasmic localization of hnRNP K and its binding to an array of mRNAs encoding for CXCR3 ligands, such as CXCL9, CXCL10 and CXCL11, and the subsequent expression of these proinflammation mRNAs ^[27][85], suggesting a role for KRT17 in RNA export and/or processing. Using a high resolution microscopic technique, recent studies have also identified KRT17 inside the nucleus of tumor epithelial cells ^[28][29][86,87]. Nuclear KRT17 is found to be associated with promoter regions of proinflammatory cytokines, such as CXCL5/10/11, CCL2/19 and IFNG, and KRT17 is also associated with the transcription regulator AIRE and NFκB, indicating the role of KRT17 in chromatin binding and transcription.

Several peptides derived from the KRT17 protein have been identified to serve as an autoantigen to stimulate T cells in psoriasis, and these KRT peptides are now recognized as the major target antigens recognized by autoreactive T cells isolated from psoriasis patients ^[26][30][31][52,88,89]. KRT17 peptides share similar epitopes with virulent factors derived from group A Streptococci, a bacterium that often triggers or exacerbates psoriasis ^[26][30][31][52,88,89]. These antigenic peptides can be first captured by immature dendritic cells (DCs) and subsequently trigger DC maturation and migration to lymphoid organs, where DCs present antigens to and activate both CD4+ and CD8+ T cells ^[26][30][52,88]. Upon contact with these autoreactive T cells in the skin, KRT17 peptides stimulate T cell proliferation and IFNγ production in psoriasis. However, it is still not clear how KRT17 peptides are processed and released from keratinocytes. Future studies will be needed to identify the protease and related secretory mechanisms that are responsible for the release of these autoantigenic KRT17 peptides from psoriatic keratinocytes.

As mentioned earlier, keratins are among the first wave of alarmins produced by keratinocytes during wound healing, and these keratins play a vital role in wound repair by providing a framework for cell anchoring, resilience against mechanical trauma, regulating keratinocyte proliferation/differentiation and inflammatory immune responses. Microarray analysis of nonhealing chronic wounds identified KRT16, KRT17, KRT14 and KRT6B among the top 15 downregulated genes in chronic wounds compared to normal skin ^[32][90], suggesting that dysregulation of these keratins contributes to defective healing. Therefore, keratin supplementation holds great promise in promoting wound healing, and keratin scaffolds have now received extensive investigations as a new biomaterial for skin regeneration. Keratin dressings have been shown to accelerate wound closure and epithelialization possibly by enhancing the activation of keratinocytes in mice ^[33][34][91,92]. More recently, scientists found that the addition of human adipose stem cells to keratin scaffolds could further accelerate skin wound healing, re-epithelialization and tissue remodeling ^[35][93]. Together, the keratin scaffold represents a new natural biomaterial with great therapeutic potential for wound healing. Future studies are needed to determine the exact molecular mechanisms involved in the wound healing process.

Conventional treatment options for psoriasis, such as topical corticosteroids, topical vitamin A or D analogues, systemic methotrexate (antimetabolite drug) or cyclosporine therapies and UVB phototherapy, are associated with a broadband immunosuppressive effect and/or organ toxicities, which can cause severe health problems with long term use ^[36][94]. New biological therapies have gained popularity due to increased safety and effectiveness compared to conventional therapies. These FDA-approved biological drugs include monoclonal antibodies or inhibitors against TNFα (etanercept, adalimumab and infliximab), IL12 and IL23 (Ustekinumab), IL17A (Secukinumab and Ixekizumab), IL17 receptor A (Brodalumab) and IL23 (Guselkumab, Tildrakizumab, and Risankizumab) ^[37][47]. In general, the majority of these psoriasis therapies have been reported to reduce KRT17 expression in keratinocytes ^[26][52], suggesting that KRT17 may be critical to psoriasis pathogenesis, and therefore a potential target for therapy. Indeed, knock downing of KRT17 by KRT17-specific antisense oligonucleotides or siRNA inhibits proliferation in keratinocytes in vitro and it is effective in reducing epidermal hyperplasia and inflammatory cell infiltration on transplanted psoriatic tissue in a mouse model ^[38][71]. Together, KRT17 holds great potential as a new target for the treatment of psoriasis in the future.

Pachyonychia congenita (thick nails, plantar keratoderma), a rare keratinizing, autosomal dominant inherited skin disorder, is caused by mutations in KRT6, 16 and 17 genes ^[1][7]. In clinical trials, KRT6a N171K mutant siRNA, which specifically and potently targets the mutant mRNA without affecting wildtype KRT6a mRNA, has been shown to have clinical efficacy in treating pachyonychia congenita ^[39][95]. The development of keratin mutant allele-specific interventions holds great potential not only for treating pachyonychia congenita, but also for treating other inflammatory skin disorders caused by autosomal mutations in keratin genes.