Intrinsic Networks Regulating Tissue Repair: Comparative Studies of Oral and Skin Wound Healing

Abstract

Wound repair is a systematic biological program characterized by four overlapping phases: hemostasis, inflammation, proliferation, and remodeling. Notwithstanding differences between species and distinct anatomical sites, the fundamental phases in the wound healing process are conserved among mammalian species. Oral wound healing is defined as an ideal wound healing model because it resolves rapidly and without scar formation. Understanding the regulation and contribution of the different molecular events that drive rapid wound healing in oral mucosa compared with those of the skin will help us define how these lesions heal more efficiently and may provide new therapeutic strategies that can be translated to the clinical settings for patients with chronic nonhealing wounds. Although all epithelial tissues have remarkable ability for tissue repair, the efficiency of such repair differs between epithelia (oral mucosa vs. cutaneous). This prompts the long-standing, fundamental biological and clinically relevant questions as to why and how does the oral mucosa achieve its enhanced wound healing capacity. In this review, we focus on (1) distinct innate wound healing capabilities of the oral mucosa, (2) lessons learned from comparative transcriptomic studies of oral mucosa versus skin, and (3) translation of findings to therapeutics for enhanced wound healing.

DEVELOPMENTAL PERSPECTIVE: INTRINSIC WOUND HEALING CAPABILITIES OF DIFFERENT EPITHELIAL SURFACES

The skin and the oral mucosa present common characteristics derived from their shared lineage from ectoderm: both are stratified squamous epithelia overlying a dermal layer of mesodermal origin (Fig. 1). Both squamous epithelia undergo a stratification process in which the keratinocytes in the basal layer differentiate toward the apical surface maturing through the spinous and granular layers. A distinction between these stratified squamous epithelia is that whereas the skin is keratinized, depending on the region of the oral cavity, the epithelium of the oral mucosa might be nonkeratinized (buccal mucosa) or orthokeratinized/parakeratinized (hard palate and gingiva). This is key during the proliferative phase of wound repair as oral keratinocytes have been shown in human and other mammals to be more proliferative and migratory than cutaneous keratinocytes (Schrementi et al. 2008; Turabelidze et al. 2014; Iglesias-Bartolome et al. 2018). Furthermore, the adnexal structures of the hair follicles and sebaceous glands, comprised of cutaneous keratinocytes, are absent in the oral mucosa. Although hair follicle–derived keratinocytes have been shown to contribute to cutaneous healing, their contribution is minimal to the bulk of interfollicular keratinocytes (Langton et al. 2008). The oral mucosa has its own adnexal structure (major and minor salivary glands) that generate saliva, a possible key constituent of the oral repair program. The skin is under constant fluctuating conditions in air temperature and humidity, whereas the oral cavity houses a controlled humid and warm environment. Increasing the moisture of cutaneous wounds has been shown to improve healing (Svensjö et al. 2000). Saliva itself has been shown to contain a plethora of pro-healing peptides and growth factors and is critical to successful oral wound healing (Sciubba et al. 1978; Bodner et al. 1991; Larjava et al. 2011; Brand et al. 2014). Although there are body site–specific variations in the surface dampness of the skin due to, in large part, sebum and sweat (e.g., axillary regions vs. exposed skin of the forearm), the skin is largely a dry tissue (Gibbs and Ponec 2000). Altogether, superficial wounds (i.e., those affecting the epithelial layer alone or with some superficial mesenchyme) in both the skin and oral mucosa are effectively ameliorated by the intrinsic capabilities of their epithelial cells. Deeper wounds, those that scar in the skin, involve a more complex interplay of various cell populations.

Figure 1.

Hematoxylin and eosin (H&E) staining of human skin (left) and buccal oral mucosa (right) biopsies (scale bar equivalent for each biopsy). Top panels detail epithelial layers of both tissues (scale bars, 100 µm).

The thicker oral epithelium has an underlying connective tissue composed of the lamina propria and the submucosa (that may contain fat cells, salivary glands, and muscle fibers). In contrast, the connective tissue underlying the skin epidermis is composed of the dermis, which is characterized by the presence of hair follicles and sebaceous glands. The lamina propria of the oral mucosa is more vascularized than the dermis (Szpaderska et al. 2005; Glim et al. 2014, 2015). Oral wounds tend to have fewer new blood vessels than skin wounds, and excess vessels in the oral wounds are more efficiently “pruned” after wound resolution compared to the skin, where the new vessels may become a permanent fixture of the healed tissue (Mak et al. 2009; Wietecha et al. 2013). The extracellular matrices (ECMs) of the two tissues are highly similar but differ dramatically in response to wounding. The lamina propria and dermis are comprised mostly of collagens I and III with other matrix proteins contributing to site-specific features (e.g., higher amounts of elastin in labile buccal mucosa vs. rigid hard palate) (Winning and Townsend 2000; Frantz et al. 2010; Hsieh et al. 2010; Glim et al. 2014). Interestingly, the lamina propria ECM has higher levels of hyaluronan, tenascin-C, and chondroitin sulfate than the adult dermis (Meran et al. 2007; Wong et al. 2009; Glim et al. 2013). During the wound repair cascade, the balance of fibrosis and matrix resolution significantly differs between the two tissue types. The wound bed ECM following fibroblast migration and during early remodeling is comprised of fibronectin and an immature network of collagen I/III (Hinz 2007; Xue and Jackson 2015). Cutaneous fibroblasts are more reactive to a wound stimulus and, as a result, produce more fibronectin and collagen I/III (Shannon et al. 2006; Hara-Saito et al. 2020). These matrix proteins are more durable in the healed skin than the oral mucosa and are major contributing components of scars (Corr and Hart 2013; Marshall et al. 2018). Importantly, the resolution of oral wounds is associated with return of the ECM to a homeostatic baseline, with the organization of collagen I/III and ECM components generally returning to the prewounded state (Mak et al. 2009). This could be because of the divergent resolution of the inflammatory response and the intrinsic features of the tissue-resident fibroblasts. Oral fibroblasts secrete a panel of matrix metalloproteinases (MMPs) and inhibitors of metalloproteinases (TIMPs) that are more favorable to maturation and reorganization of immature ECM and clearance of excess matrix constituents (Stephens et al. 2001; Rohani and Parks 2015). Both oral and skin wounds were found to be full of activated myofibroblasts (α-smooth muscle actin positive), although oral wounds were found to have even higher populations of these fibroblasts thought to contribute to fibrosis (Lygoe et al. 2007). These findings suggest tissue-specific molecular features of the fibroblasts can dictate healing outcomes, and recent studies have shown that there is a far larger diversity of fibroblast populations than previously recognized (Rinkevich et al. 2015; Mah et al. 2017; Guerrero-Juarez et al. 2019; Worthen et al. 2020; Foster et al. 2021).

Resident immune cells in the skin and oral mucosa play a role maintaining tissue homeostasis and for dictating key checkpoints to wound repair. Generally, the key distinctions differentiating the cutaneous and oral inflammatory programs are (1) the responsiveness of immunocyte populations to the site of injury, (2) the phenotype of responding immune cells, and (3) how fast does the inflammatory reaction peak and resolve. Most studies point toward either an earlier influx or no difference in the rate of neutrophil and macrophage migration into nascent oral wounds versus cutaneous wounds (Szpaderska et al. 2003; Glim et al. 2015). Similarly, later-arriving T cells travel more readily to oral wounds (Iglesias-Bartolome et al. 2018; Park et al. 2018). More proinflammatory cytokines and chemokines are released into the wound milieu of cutaneous sites compared to the oral mucosa, and the differentiation of responding cell types toward an inflammatory state, such as the switch of proinflammatory M1 macrophages to the anti-inflammatory M2 phenotype, is a feature of skin wounds (Chen et al. 2010; Glim et al. 2015). Ultimately, oral mucosal injuries are the sites of rapid, but self-limiting deployment of various immunocyte populations that release a robust complement of pro-healing and antimicrobial factors into the wound bed. The delayed immune cell resolution in the skin also contributes significantly to the failure of dermal fibroblasts to properly regenerate damaged ECM as, for example, the higher population of postwounding dermal macrophages secretes pro-fibrotic factors (e.g., TGF-β) that can push the activity of resident fibroblasts to a pro-fibrotic state (Walton et al. 2017).

Independently of these distinctions, the oral mucosa surface has the ability to heal faster and without noticeable scarring. At the same time, oral cavity wound healing occurs in an environment that sustains ongoing trauma due to mastication and with constant exposure to bacteria (Dutzan et al. 2017). The unfavorable ECM, haphazard vascularization, extended inflammatory response, and incapability of epidermal keratinocytes to regenerate lost adnexa lead to a healed, but scarred, skin. The oral mucosa rarely scars in the same sense as the skin, usually as a consequence of extensive surgical procedures, various diseases, or severe infection (Escudero-Castaño et al. 2008; Fierz et al. 2013; Politis et al. 2016; Evans 2017). Indeed, completely lost salivary glands are often not regenerated following large oral injuries, but partially damaged ones have a significant capacity to regenerate, notwithstanding the compensation of surrounding, healthy tissue to adequately provide saliva (Carpenter and Cotroneo 2010; Emmerson et al. 2018; Rocchi et al. 2021). This raises the interesting possibility that the networks present in the immuno-privileged oral mucosa are necessary for its enhanced tissue repair capacity and begets the need for a better understanding of these networks (Novak et al. 2008; Park et al. 2018).

GLEANING INSIGHTS INTO INNATE ORAL WOUND HEALING CAPACITY WITH COMPARATIVE TRANSCRIPTOMIC STUDIES

Many studies since the beginning of the twenty-first century have sought to elucidate the molecular differences of the oral mucosa and the skin (Fig. 2). With the advent of technologies permitting large-scale transcriptomic, proteomic, and refined microscopic studies, there is growing insight into how the basic biology of the two tissue types—oral mucosa and skin—respond differently to physical insult. Traditional standard-of-care wound dressing and newer skin reconstructs or substitutes have been employed extensively for the treatment of large wounds and burns but rely on the natural healing progress of the surrounding skin to ultimately reconstitute the lost tissue (Stone Ii et al. 2018). Few “targeted” therapeutics such as small molecules or exogenous growth factors like platelet-derived growth factor (PDGF) have been Food and Drug Administration (FDA)-approved for the treatment of cutaneous wounds and have limited clinical efficacy (Steed 2006; Yamakawa and Hayashida 2019). Understanding the basic molecular biology of the entire oral mucosa and skin as organs with dynamic intra- and intercellular interactions is key to future clinical progress using targeted therapeutics. Ultimately, a goal of these stand-alone or comparative studies is to unlock the potential mechanisms that would facilitate rapid, efficient, and scarless wound healing in the skin.

Figure 2.

Progression of the healing cascade in cutaneous and oral wounds. Healthy skin and oral mucosa maintain barrier function and satisfy local physiologic functions (top panels). Progression of healing and tissue remodeling after wound closure differentiate the two tissues (middle panels). (1) The oral mucosa keratinocytes migrate and proliferate more rapidly into the wound bed. (2) The overall inflammatory response to wounding is far more transient in oral wounds with less immune infiltrate and rapid resolution of the active inflammatory state. (3) Pro-fibrotic fibroblasts are largely absent in the healing oral mucosa and different lineages contribute to earlier and more rapid extracellular matrix reorganization (ECM) than in the skin. (4) Although the oral mucosa is more vascularized in the healthy state (top panels), oral wounds have less neovascularization/angiogenesis into the wound bed than cutaneous wounds. Healed wounds in the skin contrast significantly with regenerated wounds in the oral mucosa (bottom panels). Adnexal structures in the skin (hair follicles, sebaceous glands, sweat glands, nerves, etc.) are lost in healed wounds that leave a deep scar. The fibrotic ECM and inability of adult cutaneous cells to transdifferentiate into the necessary adnexal precursors precludes the reformation of these structures in extensive wounds and may inhibit full tissue recovery in more superficial wounds. In contrast, oral wounds regenerate to their physiologic baseline (top panel, right) without the formation of fibrotic ECM and with the return of site-dependent adnexal structures.

Molecular Factors Discriminating Acute Oral versus Cutaneous Wound Healing

Clinically, rapid wound healing has long been observed in the oral mucosa. Interest in leveraging the regenerative repair mechanism of oral tissues for the advancement of therapeutics has spurred research efforts. Since the late 1970s, these studies have demonstrated that rapid and scarless wound repair is a highly conserved feature of oral mucosal tissues (desJardins-Park et al. 2019; Pereira and Sequeira 2021). The advent of newer sequencing technologies has permitted massive parallel sampling of both cutaneous and oral sites, paving the way for the meticulous characterization of molecular players implicated in the wound healing cascade (Cole et al. 2001; Cooper et al. 2005; Brem et al. 2007; Charles et al. 2008; Liang et al. 2016; Ramirez et al. 2018; Stuart and Satija 2019). Initial transcriptomic studies of wound healing capitalized on complementary DNA (cDNA) microarray technology to assess transcriptional networks mediating the wound healing process in skin (Cole et al. 2001; Li et al. 2001; Spies et al. 2002; Cooper et al. 2005). Many of these early studies determined gene expression profiles of human acute skin wounds at various time points postwounding, yielding substantial insights into each stage of the wound healing cascade. During the initial inflammatory response, for example, genes such as RHO, HP1, NMBR, LCN2, HBB, and Lilrb4a were up-regulated (Cole et al. 2001; Li et al. 2001). These also identified numerous genes whose role in the wound healing response had previously been unknown, such as CCNC, CCNE1, and TTK (Li et al. 2001; Spies et al. 2002). As the field of comparative transcriptomics has expanded, groups have taken advantage of this technology to assess injury-induced molecular changes in the skin and to compare these changes to those seen in the oral mucosa. As a result of the thorough characterization of the cellular distinctions between oral and cutaneous tissues in both healthy and wounded conditions, different groups have identified keratinocytes as a primary driver of the improved wound healing process in the oral mucosa (Chen et al. 2010; Turabelidze et al. 2014; Iglesias-Bartolome et al. 2018). Consequently, we will focus on the contribution of keratinocytes to wound repair and how baseline differences between oral and cutaneous keratinocytes critically prime oral tissue to regenerate following injury.

Evidence to suggest that differences in wound repair are driven by intrinsic differences in the cells of the oral mucosa and skin epithelium first came from a study done by Chen et al. (2010). This study comprehensively assessed the genetic signatures of both oral (lateral tongue) and cutaneous wounds at multiple stages of the wound healing process (Chen et al. 2010). Although oral mucosal and skin wounds exhibit similar patterns of gene expression changes over the course of wound healing, fewer genes are differentially expressed in the oral mucosa in response to wounding, and these gene expression changes are resolved more quickly in oral mucosal wounds. At all points following injury that were investigated, the groups of differentially expressed genes between the skin and oral mucosa were distinct. At early time points, some of the up-regulated genes unique to the skin included proinflammatory molecules (interferons and chemokines), pattern recognition receptors, and matrix metallopeptidases. By contrast, oral mucosa uniquely up-regulated genes that negatively regulate proliferation during the early stages of wound healing. At later stages, the skin had increased expression of procollagens, serine-type endopeptidase activity, and metalloendopeptidase activity. Importantly, they applied their transcriptomic findings to show that in vitro, oral keratinocytes express less IL-6 and TNF-α in response to IL-1β stimulation than cutaneous keratinocytes. This finding established their hypothesis that the large differences in the genetic response to wounding was due in part to baseline differences in the keratinocyte populations at each site. Turabelidze et al. sought to ascertain the differences between keratinocyte populations by performing a transcriptomic comparison of oral and skin epithelia (Turabelidze et al. 2014). Genes involved in cellular growth and proliferation, cell cycle, and tissue development were enriched in oral keratinocytes compared to skin keratinocytes. Based on these findings, they isolated oral and skin keratinocytes and performed functional in vitro assays. Oral keratinocytes were found to have greater proliferative and migratory capacities than skin keratinocytes. When stimulated with various wound-associated factors, including IL-6 or hypoxia, oral keratinocytes responded less markedly, producing lower levels of VEGF, IL-15, and AKT-3. In a follow-up study, microRNA (miRNA) sequencing of unwounded and wounded skin and oral mucosa tissues revealed baseline differences between the two sites, as well as fewer changes in miRNA expression during oral mucosa wound healing compared with the skin (Simões et al. 2019). MiR-10a/b, for example, is minimally expressed in the oral mucosa compared to skin and is further down-regulated in the skin upon wounding.

Iglesias-Bartolome et al. (2018) has more recently shown that healthy acute healing is largely driven by keratinocyte activity (Fig. 3). This study implemented a similar biopsy-induced wounding and rebiopsying of the wound site as Chen et al. but the clinical protocol in Iglesias-Bartolome et al. obtained paired sequential biopsies during the healing process, allowing for the longitudinal comparison of oral and cutaneous wound healing at a molecular level in the same healthy human volunteers over time. Importantly, the total tissue transcriptome was compared via RNA sequencing (RNA-seq) at healthy, early, and late wounding time points between the skin and oral (buccal) mucosa. In addition to corroborating previous evidence that cutaneous wounds have higher levels of inflammation and keratinocyte differentiation than oral wounds, Iglesias-Bartolome et al. revealed that healthy oral mucosa exists in a primed state that is only achieved in the skin upon wounding. Keratinocyte activation upon skin wound healing has been marked with the expression of certain keratins, including KRT6 and 17, and has been shown to describe the concurrent processes of keratinocyte dedifferentiation, migration, and proliferation into the wound bed (Freedberg et al. 2001; Morasso and Tomic-Canic 2005). Upon further comparison with publicly available RNA-seq data sets derived from human and murine skin and oral keratinocytes in vitro (Szpaderska et al. 2003; Chen et al. 2010; Lizio et al. 2015), an oral signature gene set was hypothesized to be critical for the role of the enhanced wound healing capabilities of the oral keratinocytes.

Figure 3.

Transcriptomic analysis reveals unique oral signature transcription factors that permit ideal wound healing in oral mucosa. A representative heatmap of human skin and buccal mucosa gene expression profiles (left panel). Expression of oral transcription factors PITX1 (red) and SOX2 (green) drives activity of distinct transcriptional networks (middle panel). Expression of these transcription factors is mostly absent at baseline in skin. Downstream targets of PITX1- and SOX2-regulated transcriptional networks contribute to rapid and scarless wound healing by equipping oral keratinocytes with distinct functional attributes in both unwounded and wounded tissue (right panel).

Notable in the oral signature gene set were the transcription factors SOX2, PITX1, PITX2, and PAX9. These factors were highly expressed at baseline in the oral mucosa and absent from the skin with only nominal expression of some of these factors during cutaneous wound healing. Extensive work has shown that SOX2 is critical for the establishment and maintenance of stem cell populations across tissue types and can dedifferentiate various cell types to pluripotent stem cells when introduced with the other Yamanaka factors OCT4, KLF4, and MYC (Takahashi and Yamanaka 2016). Indeed, SOX2 is critical for the maintenance of epithelial stem cells in oral mucosa but is only found in the dermal papilla and Merkel cells of the skin (Jones and Klein 2013; Lesko et al. 2013; Ohmoto et al. 2020; Sharma et al. 2020). Furthermore, expression of Sox2 in murine keratinocytes enhanced keratinocyte proliferation (Uchiyama et al. 2019). PITX1 and PITX2 are critical for the specification of oral epithelia from embryonic ectoderm and the anterior pituitary gland (derived from the developing hard palatal oral ectoderm) and for delineating left–right asymmetry and lower limb posteriorization (Tran and Kioussi 2021). Like PITX1, PAX9 is essential to the formation of various craniofacial structures such as the filiform papilla and taste buds of the tongue and has maintained expression in oral mucosa into adulthood (Peters et al. 1998; Jonker et al. 2004). Loss of Pax9 in vivo promoted epithelial hyperplasia and the formation of oral squamous cell carcinomas (SCCs) (Xiong et al. 2018). Beyond the expression of these factors in the adult oral mucosa and during tooth development, it is virtually unknown what contribution they play to oral biology (Zhang et al. 2005; Duverger and Morasso 2008). Recent insight from a study of cutaneous SCCs demonstrated that PITX1, along with SOX2 and TP63, were critical for the transcriptional maintenance of tumor propagating cells within tumors (Sastre-Perona et al. 2019). Although these transcription factors regulate distinct transcriptional networks, their combinatorial effect in the oral keratinocyte may be vital for the innate healing ability of these cells. Knockdown of either PITX1 or SOX2 in vitro abrogated oral keratinocyte migration, whereas overexpression of those factors in vitro in cutaneous keratinocytes enhanced migration (Iglesias-Bartolome et al. 2018). Additional studies will need to capitalize on our growing understanding of the panel of effectors of an oral keratinocyte transcriptional program.

The prevalence of advanced sequencing technologies for transcriptomics such as single-cell RNA-seq (scRNA-seq) and epigenomics such as single-cell chromatin immunoprecipitation sequencing (scChIP-seq), assay for transposase-accessible chromatin sequencing (ATAC-seq), or chromosome confirmation capture techniques (Hi-C) in future work will be paramount to ascertaining a tissue-scale view of wound healing. Several studies have already used scRNA-seq to identify the multitude of cell types in the healthy and wounded skin and oral mucosa (Joost et al. 2016, 2018; Philippeos et al. 2018; Byrd et al. 2019; Chen et al. 2019; Guerrero-Juarez et al. 2019; Haensel et al. 2020; Phan et al. 2020). scRNA-sequencing of oral tissues in health and periodontal disease has permitted the generation of a single-cell atlas of the oral mucosa, including buccal mucosa and gingiva (Williams et al. 2021). In addition to unveiling cell populations present in these tissues, these analyses also provide insight into intercellular communication and demonstrate enrichment in stromal–immune cell interactions in the oral mucosa. These interactions were found to be mediated by pattern recognition receptors and damage associated molecular patterns including Toll-like receptors, C-type lectin receptors, and RIG-I like receptors. Expression of these factors were found prominently expressed in epithelial, endothelial, fibroblast, and immune cells within the oral mucosa. The presence of neutrophils was found to be significantly enriched in oral mucosa and many of the immune-related pathways found up-regulated in gingiva included the neutrophil-specific genes CSF3, CXCL1, CXCL2, and CXCL8. These findings indicate a heightened microbe and damage sensing response linked to increased neutrophil recruitment. Altogether, ascertaining the separate in vivo contribution of each cellular constituent of the skin or oral mucosa to wound healing is evolving into establishing the combinatorial effort of each of these tissues in increasingly finite detail to develop efficacious therapeutics for wound healing.

Deriving Insights from Mouse Models

Mouse models have been an invaluable asset for wound healing research, allowing for high-throughput transcriptomic studies to be performed in vivo. Yet the notable physiologic and anatomical differences between mice and humans have limited more rapid progress in the field. In brief, the process of cutaneous wound healing in mice is significantly different at the structural level as mice have a thin muscular panniculus carnosus layer at the base of the hypodermis that contracts following injury (Zomer and Trentin 2018). This has led to multiple protocols for skin splinting to mitigate the effect of contraction in cutaneous wound healing in mice to faithfully recreate the human healing process (Davidson et al. 2013; Wang et al. 2013). In the oral mucosa, the progression of healing is more similar between both species but has its own challenges in the limited space available to conduct wounding protocols in the murine oral cavity. Byrd et al. (2019) could only induce ∼250-µm oblong wounds in the interrugae space of the hard palate of mice with blunted needle tips. Nevertheless, they were able to identify and characterize unique niches of both slow-cycling/self-renewing and fast dividing/differentiation-prone stem cells that dynamically respond to wounding. Importantly, they connected the biologic function of the heterogenous stem cell populations to the masticatory stressor placed upon the hard palate by hard rodent chow. Other model species such as the Duroc pig present a closer morphology and function to that of the human skin and have contributed critical insights applicable to human biology (Mak et al. 2009; Wong et al. 2009). Wong and Gallant-Behm performed 2-mm experimental wounds on the hard palates of humans and Duroc pigs as well as similar wounds to pig skin. Results showed that human and Duroc pig hard palate wounds healed faster and with less scar tissue than the skin wounds on pigs. Primarily because of experimental cost, there has been a conscious effort to adeptly align mouse wounding healing in the skin and oral mucosa as a human analog, and mice remain as the model system of choice for the time being.

The power of mouse models to the wound healing field can be highlighted through the paired tissue sample studies of murine skin and oral wounds. Studies by Chen et al. (2010) and Simões et al. (2019) elucidated the transcriptome of the healing oral and skin wounds, the diversity of miRNA expression between the two tissue types, and the applicability of miRNAs as a potential therapeutic modality. The longitudinal study conducted by Iglesias-Bartolome et al. had the strength of paired human oral and skin wounds that delineated biologic insights that could then be taken to the mouse to verify hypotheses on how they contribute to the wound healing cascade. As an example, the study conducted by Uchiyama et al. (2019) expanded upon the identification of SOX2 as an oral signature gene. Transgenic expression of Sox2 in murine cutaneous keratinocytes via a tamoxifen-inducible keratin 14 (K14)-CreERT construct confirmed the hypothesis that Sox2 would enhance cutaneous wound healing. Importantly, by transcriptionally up-regulating the expression of various epidermal growth factor receptor (EGFR) ligands such as heparin binding EGF-like growth factor (HB-EGF) and amphiregulin, Sox2 enhanced cutaneous keratinocyte proliferation and migration and potentiated their activation in response to wounding.

Applying the Lessons of Transcriptomics: From Oral Acute Wounds to Chronic Nonhealing Wounds

The question remains: How can we translate our understanding of the regenerative capabilities of the oral mucosa to cutaneous wound healing? The limited availability and efficacy of small molecule compounds to treat acute wound underscores the need for better knowledge of the biology of the healing skin (Öhnstedt et al. 2019; Sen 2019). The partial efficacy of cellular- or protein-targeted therapeutics for cutaneous wound healing mimics what is often seen in cancer therapeutics (Alsahafi et al. 2019; Choonoo et al. 2019). Combinatorial regimens and advanced pharmaceutical design yield better outcomes in cancer treatment, and similar strategies have been suggested for the treatment of wounds (Lin et al. 2014; Song et al. 2018; O'Rourke et al. 2019). The lack of “targeted” therapeutics stands in contrast to the ample array of advanced wound dressings and bioengineered modalities that have been cleared for a variety of wounds and burns (Madaghiele et al. 2014; Sood et al. 2014; Savoji et al. 2018). Yet these systems ultimately rely upon the capacity of the surrounding skin to heal the impacted tissue, and subsequent scarring and its related comorbidities lead to lasting social and financial impact (Sen 2019). Mobilizing the keratinocytes in the epidermis to cover the site of injury is a protracted affair in comparison to the oral mucosa (Turabelidze et al. 2014). Potentiated activation and migration/proliferation of epidermal keratinocytes into the wound bed may be a potential therapeutic strategy in concert with advanced dressings. Cautious consideration to the side effects of such a stratagem must be given—the balance between wound healing and malignant transformation has long been a concern (Dvorak 1986). Oral SCCs, as an example, are far more likely to malignantly progress and metastasize into surrounding tissues and are more lethal overall compared to cutaneous SCCs (Yan et al. 2011). Advanced, targeted wound repair modalities must keep this key balance in mind.

Transcriptomics has also been used to dissect gene signatures, biomarkers, and their aberrant expression contributing to chronic nonhealing wounds (Charles et al. 2008; Stojadinovic et al. 2008, 2014; Liang et al. 2016; Stone et al. 2017; Januszyk et al. 2020; Sawaya et al. 2020; Theocharidis et al. 2020). Studies in venous leg ulcers (VLUs) have connected findings at the bench to the clinic by establishing distinct gene expression patterns that can be used to guide surgical debridement (up-regulation of SPRR3, KLK12, CEACAM6, and MMP1 and down-regulation of EGFR and Krt2a) (Brem et al. 2007). In addition, gene expression profiling has been used to identify miRNAs that are deregulated in VLU and diabetic foot ulcer (DFU) tissue samples (Pastar et al. 2012; Banerjee and Sen 2015; Liang et al. 2016; Ramirez et al. 2018). Results showed that miR-21, -16, -132, and -15b-5p were aberrantly regulated in VLUs and DFUs, and this deregulation was demonstrated to target and suppress leptin receptor, TGF-β, TNF-α, EGF, and DNA repair pathways.

Identifying key pathways and regulators involved in the various stages of wound healing in the oral mucosa would provide the basis to translate findings to the clinic for patients with chronic wounds (Fig. 4). The FOXM1 transcription factor was recently identified in the oral mucosa that functions to modulate the activation, recruitment, and survival of immune cells (Sawaya et al. 2020). Pharmacological inhibition of FOXM1 in diabetic mouse models resulted in inhibition of healing and decreased immune cell recruitment. Target genes of FOXM1 that include STAT3, SOD2, IL6, and cell cycle genes CCNB1 and CCNA2 are all deregulated in chronic wounds. Therefore, targeting the FOXM1 pathway and downstream target genes may provide new therapeutic avenues aimed at reprogramming chronic nonhealing wounds into acute healing wounds.

Figure 4.

The intrinsic networks responsible for rapid oral wound healing can be translated to elucidating the pathophysiology of chronic nonhealing wounds for the development of novel therapeutics. Comparison analyses of rapid (oral mucosa) and normal (skin) wounds to nonhealing diabetic foot ulcers (DFUs) can be used to determine chronic wound-specific gene signatures and wound cell profiles responsible for poor healing outcomes in patients with DFUs. Analysis of wounds with different healing capacities provides a tool to identify molecular mechanisms and therapeutic targets aimed at reprogramming chronic wounds into healing-competent wounds.

CONCLUDING REMARKS

Comprehensive comparative transcriptomic analyses have been invaluable in elucidating important molecular and cellular mechanisms underlying the wound healing process in different tissues. The unique environment of the oral cavity represents a wound healing program geared toward effective wound resolution. Understanding the regulatory networks that determine the “ideal” oral repair will highlight fundamental mechanisms of inflammation and repair in humans and provide insights into therapeutic targeting of chronic and nonhealing wounds.