Keloid Disorder: Genetic Basis, Gene Expression Profiles, and Immunological Modulation of the Fibrotic Processes in the Skin

Abstract

Keloid disorder, a group of fibroproliferative skin disorders, is clinically comprised of keloids, hypertrophic scars, keloidalis nuchae, and acne keloidalis. The prototype of these disorders is keloids, which manifest as cutaneous lesions with excessive deposition of collagen following an initiating trauma of varying degrees. The principal cell type responsible for collagen accumulation is the myofibroblast, and its gene expression is modulated by a network of regulatory factors, including cytokines, growth factors, and noncoding RNA species. In addition, keloids harbor a number of inflammatory cells, including macrophages and mast cells, that interact with fibroblastic cells by direct contact or by paracrine actions. Transforming growth factor-β1/Smad signaling regulates the expression of genes encoding extracellular matrix proteins and also controls cell proliferation and apoptosis. A key profibrotic molecule is the fibronectin splice variant cellular fibronectin extracellular domain A (cFN-EDA), which interacts with a number of cell-surface integrins and TLR4, contributing to the modulation of gene expression by lesional fibroblasts. Collectively, these complex cellular interactions result in accumulation of collagen with clinical development and growth of keloid lesions. Understanding of the precise pathomechanistic details of keloid formation will provide targets for pharmacological interference toward treatment of the keloid disorder, a group of currently difficult to treat skin diseases.

PHENOTYPIC SPECTRUM OF FIBROTIC DISEASES

Fibrotic diseases, a highly heterogeneous group of disorders manifesting with tissue fibrosis, are exceedingly common with extensive morbidity and mortality. In fact, it has been estimated that as much as 40% of the mortality in the Western developed countries, including the United States, is due to fibrotic diseases, collectively more frequent than cancer and vascular disease (Wynn 2008). The prevalence and the clinical impact of these conditions are often unrecognized due to the complexity and extensive phenotypic variability of these disorders. Fibrotic diseases can be divided, in a broad sense, into two general categories: (1) those with systemic, multiorgan fibrosis, and (2) those with fibrotic lesions being limited to an individual organ. An example of systemic, multiorgan fibrotic diseases is systemic sclerosis and scleroderma, which can affect many organs in the body with serious consequences, often leading to early demise due to pulmonary and/or renal fibrosis (Piera-Velazquez et al. 2020). Examples of instances in which an individual organ is affected with limited clinical consequences are those with fibrotic reactions affecting either the lung, liver, kidneys, heart, or skin (see, e.g., Fig. 1; Mei et al. 2022; Moeller et al. 2022; Travers et al. 2022; Zuñiga-Aguilar and Ramírez-Fernández 2022). Collectively, given the wide variety of affected organs, the chronic nature of the fibrotic processes, and the large number of individuals suffering from these devastating conditions, fibrotic diseases pose one of the most serious health problems in contemporary medicine and inflict a serious economic burden on the global health care system. Despite these considerations, there are currently no effective treatments to prevent or counteract these diseases.

Figure 1.

Keloid disorder as the prototype of skin fibrosis within the broad spectrum of fibrotic diseases.

(Figure adapted from data in Piera-Velazquez et al. 2016 and graphic created with BioRender.com.)

This overview summarizes the current review of keloids as a prototype of fibrotic cutaneous lesions with characteristic cutaneous findings. While the details of the overall pathomechanisms of keloids are fairly well understood, there are several pathological alterations, some being pathomechanistic and critical for keloid formation, while some changes are secondarily consequential to the primary lesions and their pathomechanistic consequences. Clearly, keloid is a multifactional disorder manifesting with numerous changes in diverse cells, expression of a number of genes, and numerous cellular and paracrine pathways. While understanding of such pathways will help to appreciate the higher level of the keloid network, the concept of details of the underlying factors is critical to be understood, before the overall unified concept of keloid pathogenesis can be fully understood. Consequently, this review details different mechanisms associated with keloid pathogenesis, eventually leading to the keloid network concept.

KELOIDS AS THE PARADIGM OF FIBROTIC SKIN DISORDERS

As indicated above, systemic fibrotic diseases, such as scleroderma, can affect different organs, including the skin. However, there are a number of disorders where the fibrotic process is limited to the skin, either in the form of localized lesions or with more extensive, diffuse cutaneous involvement. Examples of such fibrotic diseases with fibrosis being largely limited to the skin include morphea (localized scleroderma), connective tissue nevi of the collagen type (collagenoma), hypertrophic scars, and keloid disorder, with little, if any, extracutaneous involvement (Uitto et al. 1980; Macarak et al. 2021).

Among these diseases, morphea can present clinically either as a localized, relatively well-defined cutaneous lesion or as generalized, more diffuse skin involvement, associated with a strong inflammatory component at the early stages of the fibrotic process (Snarskaya and Vasileva 2022). While mostly limited to the skin, generalized morphea can be associated with extracutaneous tissue involvement, such as muscle, joints, as well as bone, and serum antibody profiling can reveal the presence of antinuclear antibodies, implying the critical role of autoimmunity in its pathogenesis. While morphea is also referred to as localized scleroderma, it does not progress into systemic sclerosis, and the lesions typically are self-limited leading to indolent fibrotic plaques. While being occasionally highly disfiguring, especially in the variants known as linear morphea or “en coup de sabre,” typically manifesting with indentations on the frontoparietal region of the face, morphea does not affect the longevity of the patients.

Collagenomas manifest with fibrotic, plaque-like lesions with the deposition of collagen (Uitto et al. 1979). Collagenomas can be present in the skin either as isolated lesions or multiple, together with elastomas, clinically similar cutaneous lesions, yet depicting histologically accumulation of elastic fibers. These lesions can be a component of the Buschke–Ollendorff syndrome, a heritable connective tissue disorder manifesting in addition to skin lesions with osteopoikilosis that manifests with numerous radiologically detectable hyperostotic bone lesions in the periarticular regions (“spotty bones”) (Uitto et al. 1981; Salik et al. 2022). This autosomal-dominant condition is caused by mutations in the LEMD3 gene encoding LEMD3 protein, which critically controls the bone morphogenic protein (BMP) and transforming growth factor β (TGF-β) signaling pathways (Hellemans et al. 2004). When the levels of LEMD3 are reduced, the TGF-β signaling is activated leading to connective tissue accumulation in the skin and the development of osteopoikilotic bone lesions.

The group of fibrotic diseases, collectively designated as keloid disorder, is clinically comprised of keloids, hypertrophic scars, keloidalis nuchae, and acne keloidalis. These conditions manifest with localized cutaneous lesions with extensive accumulation of extracellular matrix (ECM) of connective tissue, particularly collagens, as a result of external trauma. The prototypic fibroproliferative keloid lesions serve as the paradigms of fibrotic skin diseases of varying degree. The early stages of keloid development are characterized by the presence of inflammatory cell infiltrates implying the critical role of immunity in the pathogenesis of the fibrotic processes. Enmeshed in the ECM are also cells with fibroblastic characteristics differentiating into myofibroblasts, the principal cell type responsible for ECM synthesis and deposition in the keloid lesions (Macarak et al. 2021). Keloidalis nuchae manifests as small papules in hair-bearing areas of skin, particularly in the neck of individuals of African ancestral background. It has been postulated that the curly structure of the hair penetrates the epidermis and elicits an inflammatory reaction that then propagates the fibrotic process in genetically predisposed individuals (Ogunbiyi and Adedokun 2015). The role of inflammation in the fibrotic lesions is also suggested by the presence of acne keloidalis, which manifests as relatively small fibrotic papules at the sites of acne lesions. Collectively, these clinical examples show the spectrum of fibrotic lesions in the skin, and they attest to the complex pathomechanisms involving immunological reactions and alterations in the cellular profiles culminating in the deposition of ECM molecules and leading to the development of the cutaneous lesions.

CONNECTIVE TISSUE COMPOSITION AND COLLAGEN ACCUMULATION IN KELOIDS

Keloids are fibroproliferative tumors that in their fully developed and mature form are composed of the ECM of connective tissue, primarily of collagen. The collagen molecules, as examined by histopathology and electron microscopy, are tightly packed into fibers that resist deformation, providing the lesions with firm clinical characteristics (Fig. 2). Biochemical analysis of the genetically distinct collagens in keloids has revealed that type I and type III, the major fibrillar collagens in the dermis, are also the predominant collagen types in keloids (Abergel et al. 1985). In addition, the presence of type VI collagen has been documented, particularly in the early stages of keloid development, suggesting that expression of the type VI collagen may serve as an early biomarker of the activation of the fibrotic process (Peltonen et al. 1991).

Figure 2.

Clinical and histopathological features of keloids. Demonstration of keloids in predilection sites on the earlobes, chest, and jawlines as a result of piercing, trauma, and shaving, respectively (upper panel). The lesions are histopathologically characterized by accumulation of tightly packed collagen (lower left panel, asterisk) and increased cellularity (lower middle panel, asterisk). Trichrome stain reveals high density of collagen fibers (lower right panel, asterisk). Hematoxylin and eosin (H&E) stain, 4.0× and 9.0×, lower right and middle panels, respectively; trichrome stain, 3.0×, lower right panel.

Examination of collagen gene expression in fibroblasts cultured from keloids has demonstrated elevation of the α2(I) and α1(III) mRNA levels, reflecting enhanced synthesis of type I and type III collagen, and indicating pretranslational control of gene expression (Abergel et al. 1987). The synthesis of these genetically distinct collagens in keloids is predicated upon the presence of fibroblastic cells, particularly those with phenotypic features of myofibroblasts as defined by the expression of α-smooth muscle actin (α-SMA) and vimentin. These cells depict prominent rough endoplasmic reticulum and Golgi apparatus, indicating their high biosynthetic activity primarily directed at the production of the ECM of connective tissue, particularly collagens. In addition to fully developed myofibroblasts with characteristic ultrastructural features, keloids have been shown to contain a number of myofibroblast-like subpopulations of cells with a lesser degree of differentiation (Zhao et al. 2017). These observations of the spectrum of fibroblastic cells attest to the plasticity of the cells of this lineage, suggesting a controlled progression from relatively poorly differentiated fibroblasts to fully matured myofibroblasts with high biosynthetic capacity. In addition to fibroblasts, mast cells have been identified in keloids, often in close association with fibroblasts in a direct cell–cell contact. Keloids also harbor increased numbers of endothelial cells representing the vascular component of these lesions, as demonstrated by the presence of factor VIII–positive cells, and endothelial dysfunction has been suggested to promote keloid growth (Sollberg et al. 1991; Noishiki et al. 2017; Wang et al. 2018). Thus, the keloid lesions contain a repertoire of cells, some being part of the immune system and others representing biosynthetically active fibroblastic cells. The interactions of these cells, either by direct contacts or through paracrine effects, can explain the proliferation of cells with a high level of ECM production, thus leading to the accumulation of connective tissue in keloid lesions.

In addition to collagens, the presence of a number of proteoglycan/glycosaminoglycan macromolecules has been documented in keloid lesions (Carrino et al. 2012; Sidgwick and Bayat 2012). These macromolecules, particularly hyaluronic acid, play a critical role in providing hydration to the tissues through their large water-binding capacity. In keloids, however, collagen is the overwhelmingly abundant macromolecular component, and the synthesis of genetically distinct collagens is critical for the early development and growth of keloid lesions.

A specific ECM component, the cellular fibronectin (cFN), has been suggested to play a critical role in the activation of fibrotic processes. Fibronectin is a family of high-molecular-weight proteins that are present in the extracellular matrices as well as in extracellular fluids, including plasma. The human fibronectin gene comprises 45 exons, and as a result of alternative splicing, it can direct the synthesis of as many as 20 different translational protein variants (Schwarzbauer and DeSimone 2011). The fibronectins interact with other matrix components, including collagens and proteoglycans, and they recognize a number of cell-surface receptors, particularly integrins α9β1, α5β1, α4β1, and αVβ3, as well as a Toll-like receptor 4 (TLR4) (Schnittert et al. 2018). One of the splicing variants of fibronectin contains an extra domain (extracellular domain A [EDA]), encoded by a distinct exon. EDA is present in cFN, a gene product of a variety of cells but particularly of fibroblasts and epithelial cells. cFN-EDA is deposited as fibrillar matrices in the ECM, and it has been shown to be essential for normal wound healing, particularly with respect to reepithelialization (Lenselink 2015). cFN-EDA also binds to cell-surface receptors, importantly to TLR4. Several observations on cFN-EDA have shown that it is not only critical for normal wound healing but also plays a role in fibroproliferative disorders, including keloids. These observations have demonstrated that cFN-EDA is an endogenous TLR4 ligand, and this interaction is markedly increased in TGF-β-treated cells (Bhattacharyya et al. 2014). Similar observations in both scleroderma skin and in keloids have suggested that the cFN-EDA/TLR4 interaction-dependent pathways are critical as a unifying pathomechanistic feature for the fibrotic process. In keloids, an abnormal, trauma-related early response can result in activation of this pathway, as demonstrated by a dramatic, up to 70-fold increase in cFN-EDA expression in keloids (Andrews et al. 2015). It should be noted that this specific molecular interaction between cFN-EDA and its endogenous receptors, particularly TLR4, may provide a potential target for treatment to abrogate the development and growth of fibrotic lesions.

PATHOMECHANISMS OF KELOIDS

Genetic Basis of Keloids

The early evidence of a strong genetic component for keloid formation was derived from the observation that keloids often occur in familial clustering, particularly in individuals of certain ethnic backgrounds (Belie et al. 2019; Kouotou et al. 2019). In particular, keloids are relatively common in individuals of skin of color, and the estimated prevalence of keloids in populations of African ancestry varies globally from 4% to 16% (i.e., about 15-fold higher than in Caucasian individuals of European ancestry) (Brown et al. 2008). Also, Asian individuals have an increased risk for keloid formation, although their keloids are clinically somewhat different, being less exophytic than those in individuals of African ancestry. Keloids have also been reported to occur in a familial pattern suggestive of autosomal-dominant inheritance with incomplete penetrance (Marneros et al. 2001; Clark et al. 2009).

A number of genome-wide association studies (GWAS) have identified keloid susceptibility loci, particularly in the Japanese and Chinese Han populations. One of the potential candidate genes identified by GWAS encodes NDD4, a molecule that enhances cellular proliferation and invasiveness of fibroblasts and can activate the TGF-β/catenin signaling pathway (Cusack 1988; Zhao et al. 2016; Fujita et al. 2019; Farag et al. 2020). At the same time, whole-exome sequencing-based gene analysis in African-American keloid patients did not support these previous findings (Hellwege et al. 2018). In addition, several noncoding RNAs have been suggested to contribute to the etiology of keloids by modulating a number of signaling pathways relevant to wound healing and scar formation (see below). Nevertheless, no Mendelian candidate genes underlying keloid formation have been consistently identified, and, specifically, a number of genes, including those encoding pro-α2(I) collagen as well as TGF-β1-3 and TGF-β receptors I, II, and III, have been excluded in the pathogenesis of keloids (Bayat et al. 2004). Finally, while research has increasingly addressed the epigenetic modulation of gene expression with focus on DNA methylation and histone modification, little information directly applicable to keloids is currently available (Stevenson et al. 2021; Nyika et al. 2022).

Altered Cellularity and Gene Expression Profiling in Keloids

Wound healing is a dynamic process that is composed of four main time-dependent phases: hemostasis, inflammation, proliferation, and remodeling. When the mechanisms or machinery of this precise process are impaired, pathologic scar formation can occur. Keloids result from changes in cellularity and dysregulation of cytokines involved in the wound healing process. Here we will discuss the aberrant cellular profiles and altered cytokine expressions found in keloid disorders.

Fibroblasts

Fibroblasts are ubiquitously present in the connective tissues and are critical to the wound healing process. Fibroblasts are mobilized to sites of injury by various growth factors and cytokines, including TGF-β, platelet-derived growth factor (PDGF), and insulin-like growth factor I (IGF-I) (Niessen et al. 1999). Fibroblasts then release these fibrogenic growth factors that play a primary role in wound healing (Fig. 3; Niessen et al. 1999). There are also various fibroblast growth factors (FGFs) that play a role in wound healing. FGFs 7, 10, and 22 are expressed by both fibroblasts and keratinocytes and induce the production of matrix metalloproteinases (MMPs). In addition, FGF-7 has been shown to increase the production of vascular endothelial growth factor (VEGF) and MMP-9 (Demidova-Rice et al. 2012). The heterogeneity among fibroblasts contributes to the variable functions observed between different fibroblast subpopulations. These functions include the breakdown of the fibrin clot, collagen synthesis, ECM remodeling, and wound contraction.

Figure 3.

Cell-signaling and immunomodulatory pathways in keloid formation. Fibroblasts are the principal cell type involved in the cell-to-cell signaling and profibrotic pathways that ultimately lead to keloid formation. Primary interactions involved in keloid genesis are depicted. (IL) Interleukin, (EMT) epithelial-to-mesenchymal transition, (HIF-1α) hypoxia-inducible factor 1α, (PDGF) platelet-derived growth factor, (ECM) extracellular matrix, (TGF-β) transforming growth factor β, (IGF-I) insulin-like growth factor, (VEGF) vascular endothelial growth factor, (KGF) keratinocyte growth factor, (TNF-α) tumor necrosis factor α, (JAK) Janus kinase, (IKK) IκB kinase, (CTGF) connective tissue growth factor. (Graphic created with BioRender.com.)

Fibroblasts are one of the main cells involved in keloid formation. Not only are high concentrations of fibroblasts present in keloid compared to normal tissues (Shaker et al. 2011), but evidence has shown that keloid fibroblasts (KFs) differ significantly from normal skin fibroblasts (NSFs) in their cellular characteristics and responses. KFs are more sensitive to growth factors, particularly TGF-β1 and β2 as well as PDGF, leading to amplified collagen and ECM synthesis and deposition (Ashcroft et al. 2013). Furthermore, KFs produce excessive amounts of disorganized collagen. Specifically, the ratio of type I/III collagen is significantly elevated in keloids compared to that in normal skin or hypertrophic scars (Wang et al. 2021). In addition, KFs have a higher proliferation rate compared to NSFs, further contributing to the high density of fibroblasts in keloid tissues (Zhang et al. 2017). KFs also exhibit enhanced antiapoptotic properties through the effects of tenascin C, a matrix protein that has been shown to be more highly expressed in KFs compared with NSFs (Dalkowski et al. 1999).

Fibroblasts derived from different keloid areas have also been shown to display distinct behaviors that reflect the keloid phenotype. Clinically, keloids are often described as having a raised peripheral margin that invades the surrounding skin and a more depressed, lighter-colored center. Multiple studies have shown that centrally derived fibroblasts have reduced activity and proliferation, and increased levels of apoptosis and senescence (Limandjaja et al. 2020).

Recent studies suggest that the hypoxic microenvironment of keloid tissues, which may be explained by excessive collagen deposition and occluded microvessels, promotes KF proliferation and inhibits apoptosis (Wang et al. 2021). All the above phenomena ultimately contribute to enhanced fibroblast proliferation and survival in keloid tissues.

Myofibroblasts

Myofibroblasts play a key role in wound healing and restoring tissue integrity. Myofibroblasts were first identified by Giulio Gabbiani in 1971 in the granulation tissue of healing skin wounds (Gabbiani et al. 1971). In this work, the authors described myofibroblasts as a state between a fibroblast and a smooth muscle cell, as they express many of the morphological and structural features of both these cell types (Powell et al. 1999). However, they can be distinguished by their abundant expression of α-SMA within their cytoplasmic stress fibers, positive staining for the intermediate filament vimentin, and presence of cFN-EDA. The current scientific consensus is that the differentiation of tissue-resident fibroblasts is the major source of activated myofibroblasts (Macarak et al. 2021). One of the most studied pathways of fibroblasts to myofibroblast differentiation is the TGF-β1-dependent Smad2/Smad3 pathway. Under the influence of mechanical tension, TGF-β1, and cFN-EDA, fibroblasts ultimately differentiate into myofibroblasts (Fig. 3; Zent and Guo 2018). In wound healing, myofibroblasts are primarily involved in extracellular deposition of collagens type I and III and wound contraction. By contracting the edges of the wound, myofibroblasts accelerate the wound repair process. After healing is complete, these cells can be lost through apoptosis. It should be noted, however, that there is an evolving concept that rather than being “bad” players in scarring, the role of myofibroblasts has been recently suggested to maintain junction by terminating contracted state and converting back to post-contractile fibroblast state. It has been proposed as well that myofibroblasts can convert in keloids to adipocytes when treated with BMP (Plikus et al. 2017). Thus, myofibroblasts can be identified as a plastic cell type with a potential role during wound healing.

Several reports have confirmed the presence of α-SMA⁺ myofibroblasts within keloid tissue, suggesting that myofibroblasts play a role in the pathogenesis of keloid formation. In fact, the presence of α-SMA⁺ myofibroblasts in 33%–81% of keloids has been reported (Limandjaja et al. 2020). It has been proposed that in fibrotic diseases, such as keloid disorder, myofibroblasts may fail to undergo apoptosis, resulting in persistent activation of these cells. This ultimately leads to excessive ECM accumulation, abnormal tissue architecture, and pathologic scarring (Van De Water et al. 2013).

Keratinocytes

Keratinocytes are the major cellular component of the epidermis and have several roles in wound healing. Importantly, keratinocytes and fibroblasts communicate with each other via paracrine signaling cross talk and coordinate with each other to restore normal tissue. In response to skin barrier disruption, keratinocytes release prestored interleukin 1 (IL-1), which increases keratinocyte migration and proliferation (Barrientos et al. 2008). IL-1 also acts as a double paracrine signal by up-regulating fibroblast production of keratinocyte growth factor (KGF), which in turn stimulates keratinocyte proliferation and migration (Barrientos et al. 2008). Keratinocytes also secrete VEGF and PDGF, which induce endothelial cell migration and angiogenesis in the wound bed (Barrientos et al. 2008). PDGF additionally promotes fibroblast proliferation and production of ECM (Heldin and Westermark 1999). Finally, keratinocytes may undergo a partial epithelial-to-mesenchymal transition (EMT) in skin wounds. In this process, keratinocytes lose their polarity and cell–cell adhesion and gain mesenchymal properties, which allows them to migrate from the wound edges to the center of the wound and close wounds rapidly (Fig. 3).

The signaling between keratinocytes and fibroblasts is disrupted in keloid pathology. Studies have shown that when keloid keratinocytes are cocultured with NSFs, the expression of collagen is significantly up-regulated (Alghamdi et al. 2020). In addition, when keloid keratinocytes are cocultured with normal or keloid-derived fibroblasts, these cells exhibit increased proliferation (Lim et al. 2001). Cytokines released from keratinocytes have also been implicated in decreasing apoptosis in KFs through paracrine signaling (Funayama et al. 2003). Abnormalities in keratinocytes and the EMT have also been associated with keloid genesis. Keloid keratinocytes have been shown to have a high level of expression of mesenchymal markers, including vimentin and fibronectin (Ma et al. 2015). Further in vitro studies have also demonstrated that keloid keratinocytes undergo a transition from an epithelial to a mesenchymal phenotype in response to hypoxia-inducible factor 1α (HIF-1α) (Ma et al. 2015). The hypoxic microenvironment in keloid tissue causes keloid keratinocytes to adopt a fibroblast‐like appearance through EMT, which may contribute to enhanced invasiveness and the extension of keloids beyond the wound margin (Ma et al. 2015). The abnormalities found in the keloid keratinocytes with respect to fibroblast paracrine signaling and the EMT support the role of keratinocytes in keloid formation.

Mast Cells

Mast cells are resident inflammatory cells abundant in connective tissue and play an active role during wound healing (Ng 2010). They can be stimulated to degranulate by allergens through cross-linking with immunoglobulin E receptors (e.g., FcεRI), physical injury through damage-associated molecular patterns (DAMPs), and various other compounds and mechanisms. Histamine, one of the primary mediators released by mast cells, contributes to inflammation by dilating post-capillary venules and increasing blood vessel permeability.

Mast cells have been implicated in the pathogenesis of keloid development, and, specifically, keloids have been associated with increased numbers of mast cells compared to normal skin and scar tissue (Bagabir et al. 2012). The relationship between mast cells and keloid disorders is also supported by the fact that the incidence of keloids is positively correlated with serum IgE levels as well as atopic disorders (Lu et al. 2018). Additionally, immunophenotyping studies show that degranulated mast cells are greatly increased in intralesional and perilesional keloid sites (Bagabir et al. 2012). The intimate contact between nerve endings and mast cells in conjunction with the high levels of histamine released by degranulated mast cells may explain the significant pain and pruritis associated with keloid lesions. In addition, via close contact and paracrine signaling to fibroblasts, mast cells in keloids can increase fibroblast proliferation and collagen synthesis (Fig. 3; Wang et al. 2011). Overall, mast cells play a role in the development of keloids through IgE-mediated release of histamine and direct stimulation of KFs.

Regulatory T Cells

Regulatory T cells (Tregs) are CD4⁺ T cells that have a critical role in dampening immune responses and maintaining immunological self-tolerance. Tregs express high levels of FOXP3, a transcription factor involved in the regulation of Treg cell development and suppressive activity (Deng et al. 2019), the expression of which can be peripherally induced by TGF-β. They help to maintain self-tolerance by suppressing the proliferation and cytokine production of Th1, Th2, and Th17 cells (Clark 2010). An imbalance of these cells has been implicated in the pathogenesis of cutaneous inflammatory and autoimmune diseases. Studies have demonstrated that keloid tissue exhibits a higher frequency of Tregs (Chen et al. 2019). In addition, cytokines produced by Tregs are present in significantly higher concentrations as well, including TGF-β and IL-10 (Fig. 3; Chen et al. 2019). Finally, the expression levels of FOXP3 are higher in keloid tissue and are associated with a greater amount of collagen III (Chen et al. 2019). These findings demonstrate a connection between Treg dysregulation and keloid formation.

Macrophages

Macrophages are critical to normal wound healing and tissue regeneration. In the earlier stages of wound healing, macrophages are primarily of the inflammatory phenotype (i.e., the M1-like macrophages). They are largely involved in removing cellular debris and foreign materials from the injured site. As inflammation resolves, M1-like macrophages are replaced by regenerative macrophages, also knowns as M2-like macrophages, which can also be produced and polarized from monocytes in the presence IL-4, IL-13, or apoptotic neutrophils (Xu et al. 2020).

Aberrations in macrophages have been found in fibrotic diseases and keloid formation. First, it has been demonstrated that the number of M2-like macrophages in keloids markedly outweighs the number of M1-like macrophages (Li et al. 2017). Recent studies have also demonstrated an increase in IL-4/IL-13 receptors in lesional keloid skin specimens, suggesting an increased IL-4/IL-13 signaling and subsequent M2-like macrophage polarization in keloid tissues (Diaz et al. 2020). Furthermore, skin sites prone to keloid formation, such as the earlobes, mandible, upper back, and shoulders, have been shown to possess altered resident macrophage populations compared to skin sites less prone to keloid formation. Specifically, significantly lower numbers of M1-like macrophages were present in keloid predilection sites compared to non-predilection sites of skin; however, the number of M2-like macrophages was equal in keloid predilection and non-predilection sites (Butzelaar et al. 2017). Furthermore, a significantly increased number of M2-like macrophages have been observed in intralesional and perilesional keloid sites as compared to that of normal skin and scar tissue (Bagabir et al. 2012). The high expression of these cells at both the margins and the intralesional areas of keloids is consistent with the invasive nature of keloids beyond the initial perimeter boundaries and advancement into the surrounding tissue.

M2-like macrophages can contribute to keloid formation by producing cytokines that stimulate the proliferation of keratinocytes and fibroblasts, including PDGF, TGF-β, and IGF-1 (Verreck et al. 2006). Among the M2-like macrophages, there are four different subgroups that vary in their cytokines secretion and their functions. Among them, M2a macrophages are referred to as wound healing macrophages, as they are involved in the stimulation of fibroblasts, ECM synthesis, and angiogenesis (Krzyszczyk et al. 2018). M2a macrophages produce high levels of arginase-1 (Arg-1), PDGF, IGF-1 (Fig. 3; Ogle et al. 2016).

Overall, these findings suggest that in individuals susceptible to keloids, M2-like macrophages may alter the immune microenvironment to suppress adaptive immune responses, including monocyte infiltration and differentiation into M1-like macrophages, contributing to keloid formation (Xu et al. 2020). Although the role of macrophages in keloid predisposition has not been fully elucidated, it seems that an increased number of M2-like macrophages and reduced presence of M1-like macrophages play a fundamental role in keloid formation and susceptibility.

Single-Cell Analysis in Cellular Heterogeneity

Recently, several studies have explored the cellular heterogeneity of keloids, established by single-cell RNA sequencing analysis. One of them explored single-cell RNA sequencing, which revealed lineage-specific regulatory changes of fibroblasts and vascular endothelial cells in keloids (Liu et al. 2022). In this study, unbiased clustering of cells in keloid tissue by single-cell RNA sequencing revealed substantial cellular heterogeneity, including 21 clusters assigned to 11 cell lineages. A significant expansion of fibroblast and vascular endothelial cell subpopulations were detected in keloid subpopulations, reflecting their strong association with keloid pathogenesis. These results highlighted the roles of TGF-β and Eph-ephrin signaling pathways in keloids, adding insights into pathogenesis of keloids and providing potential targets for medical therapy of these fibrotic lesions (Liu et al. 2022).

The heterogeneity of fibroblastic populations has also been documented in keloids by single-cell RNA-seq analyses (Deng et al. 2021). Characterization of KFs allowed division of these cells to four subpopulations: secretory-papillary, secretory-reticular, mesenchymal, and proinflammatory. The percentage of mesenchymal fibroblast subpopulation was significantly increased in keloids compared to wound scar, emphasizing the importance of mesenchymal cells in the pathogenesis of keloid lesions. These findings potentially identify targets for treatment of keloids and possibly other fibrotic diseases (Deng et al. 2021).

The pathomechanisms of keloids have also been examined by RNA sequencing of keloid transcriptome in lesional and non-lesional skin of African-American keloids compared to healthy skin of ethnically matched controls (Wu et al. 2020). The lesional skin of keloids was shown to up-regulate the markers of T-cell activation/migrated pathways, including Th2, Th1, Th17/Th22, and Janus kinase (JAK)/signal transducer and activator of transcription (STAT) signaling. Keloid tissues also showed increased cellular infiltration of T cells, dendritic cells, and mast cells, demonstrating significant immune alterations, particularly Th2 and JAK3, with potential for specific targets for keloids (Wu et al. 2020).

Differential Gene Expression Profiles

With the advancement of cutting-edge technologies, such as single-cell RNA sequencing, microarray analyses, and GWAS, the regulation of genes and other genetic components involved in keloid disorder has been able to be further elucidated. Recent studies have revealed considerable and consistent regulated expression of various genes in the fibrotic processes. The modulation of the gene expression is complex and a number of molecules, including various RNA species, have been shown to play a role in the regulation of gene expression in fibrotic diseases.

Noncoding RNAs

MicroRNAs (miRNAs) are small single-stranded noncoding RNA molecules that function in RNA silencing and post-transcriptional regulation of gene expression, and they have been implicated in the regulation of skin fibrosis. Liu et al. performed an miRNA microarray analysis, which identified a total of 32 differentially expressed miRNAs in keloid tissue compared with normal skin controls (Liu et al. 2012). In these studies, miRNA-21 had the highest fold change among the 23 miRNAs that were up-regulated, and miRNA-203 had the lowest expression level among the nine that were down-regulated. miRNA-21 has been demonstrated to express at significantly higher levels in KFs compared to NSFs, and it inhibits the negative regulation of the TGF-β/Smad signaling pathway by targeting Smad7 mRNA and ultimately promoting fibrosis (Wu et al. 2019). miRNA-203 overexpression in vitro has been shown to lead to a significant decrease in proliferation, invasion, and ECM production of KFs (Shi et al. 2018). In another study, Li et al. found that the miRNA profiles in KFs were different from NSFs by miRNA microarray analysis (Li et al. 2013). Six miRNAs were significantly up-regulated, including miR-152, miR-23b-3p, miR-31-5p, miR-320c, miR-30a-5p, and hsv1-miR-H7, and three miRNAs were significantly down-regulated, including miR-4328, miR-145-5p, and miR-143-3p. Of significance, miRNA-152-3p has been shown to regulate cell proliferation and ECM expression, including type III collagen and fibronectin, by targeting FOXF1 in KFs (Wang et al. 2019). miRNA-31 has also been shown to regulate proliferation and apoptosis of keloid-derived fibroblasts by mediating the HIF-1α/VEGF signaling pathway (Kee et al. 1988). Bioinformatic analyses predict that these other differentially expressed miRNAs also participate in signaling pathways involved in keloid genesis. These data suggest that aberrant miRNA expression in fibroblasts contributes to keloid formation.

Long noncoding RNAs (lncRNAs) are RNA molecules larger than 200 nucleotides that regulate gene expression and have been implicated in keloid formation. In 2015, Liang et al. identified 1731 constantly up-regulated and 782 down-regulated lncRNAs in keloids by microarray (Liang et al. 2015). In particular, they found that the lncRNA CACNA1G-AS1, an antisense RNA to CACNA1G that encodes a subtype of T-type Ca²⁺ channels, had high expression in keloid tissues and thus may play a crucial role in keloid development. Another study by Sun et al. identified four skin-related lncRNAs (CACNA1G-AS1, LINC00312, HOXA11-AS, and RP11-91I11.1) involved in Wnt signaling pathways, which coordinate proliferation and regeneration in keloids (Sun et al. 2017).

CYTOKINE MODULATION OF GENE EXPRESSION

Immune Dysregulation

Immune dysregulation plays a significant role in the pathophysiology of keloids evidenced by the aberrant profile of inflammatory cells as well as increased activation of distinct immune signaling pathways in keloid tissue. However, the overall state of the immune activity in keloids is complex and has not been fully elucidated. A major source of complexity is the observation that both pro- and anti-inflammatory pathways are simultaneously active in keloid tissue. While the presence of Tregs and M2-like macrophages in keloids can foster an anti-inflammatory shift, there is evidence for activation of proinflammatory signaling pathways of IL-6 and nuclear factor κB (NF-κB) in keloids. In parallel, the anti-inflammatory TGF-β1 signaling pathway has been evidenced to have increased activity in keloids. Although it may be oversimplistic to classify the activity of a particular pathway or cell type as pro- versus anti-inflammatory, the coexistence of opposing immune modulating signals in keloids has contributed to our difficulty in understanding the immune pathogenesis of keloids. Here we will discuss three immune signaling pathways involved in keloid formation.

TGF-β1/Smad Signaling Pathway

The TGF-β1/Smad signaling pathway is well established to play a role in the formation of collagen in the fibroblasts (Fig. 3). TGF-β1, overexpressed in keloid tissue, binds to a serine/threonine kinase type II receptor, which recruits and phosphorylates a type I receptor. Following phosphorylation, the type I receptor kinases are activated and they phosphorylate intracellular receptor–regulated Smads, including Smad2 and Smad3. Smad2 and Smad3 bind to the co-Smad Smad4, and this complex subsequently translocates from the cytoplasm into the nucleus, where it regulates the transcriptional activity of various genes.

Studies have found that mRNA expression levels of the collagen genes COL1A1 and COL3A1 are higher in KFs compared to NSFs (Nagar et al. 2021). In addition, altered expression of a number of growth factors, genes, and proteins has been documented in keloid tissues. For example, the expression of connective tissue growth factor (CTGF), which plays a critical role in keloid pathogenesis by promoting collagen synthesis and deposition, has been shown to be highly expressed in keloid tissues (Chen et al. 2013). Human PAI-1, another gene that is potently induced by TGF-β, has been implicated in tissue fibrosis and has been correlated with KF collagen overproduction (Tuan et al. 2003). Activin-A, a dimeric protein that is a member of the TGF-β superfamily, has been observed to be overexpressed in keloid scar tissue (Mukhopadhyay et al. 2007). Moreover, the activin antagonist follistatin has been suggested for keloid therapy, because treatment with follistatin has been shown to reverse the up-regulated expression of CTGF and PAI-1 in KFs (Ham et al. 2021).

Furthermore, the inhibitory Smads, Smad6 and Smad7, play a role in the regulation of TGF-β signaling and are involved in the negative feedback of this pathway. Smad6 and Smad7 are induced by BMPs and in turn inhibit BMP signaling pathways, creating a negative feedback loop. While Smad7 inhibits both TGF-β and BMP signaling, Smad6 preferentially inhibits BMP signaling (Goto et al. 2007). While the inhibitory Smads have several mechanisms of action, some of their major mechanisms involved in keloid genesis are that they bind to TGF-β type I receptors and inhibit Smad2 and Smad3 activation. The expression of Smad6 and Smad7 mRNA has been shown to be significantly decreased in KFs compared to normal scar and NSFs (Yu et al. 2006). The decreased expression of these inhibitory Smads may explain the persistent fibroblast proliferation and collagen deposition in keloid scarring. Activation of the TGF-β1/Smad signaling pathway ultimately results in uncontrolled fibroblast proliferation, decreased apoptosis, increased ECM production, and increased cytokine production.

IL-6-Mediated JAK/STAT3 Signaling Pathway

Up-regulation of the IL-6-mediated JAK/STAT3 signaling pathway has been observed in KFs. IL-6 is mainly produced by infiltrating immune cells within the local microenvironment, including macrophages, T cells, and fibroblasts. IL-6 can bind to either the IL-6 receptor (mIL-6R) on the cell membrane or a soluble type of the IL-6 receptor (sIL-6R). When the complex of IL-6 and either mIL-6R or sIL-6R interacts with the shared signal transducer receptor (IL-6ST or gp130), activation of JAK and subsequently STAT3 occurs.

Not only do KFs exhibit up-regulation of IL-6 and its receptors, but enhanced expression and phosphorylation of STAT3 have also been observed in keloid tissue and in cultured KFs in vitro as well (Lim et al. 2006). Furthermore, inhibition of STAT3 expression has resulted in the loss of collagen production, impaired fibroblast proliferation, and delayed cell migration in KFs.

NF-κB Signaling Pathway

NF-κB is a ubiquitously expressed transcription factor. It is sequestrated in the cytoplasm in an inactive state bound to its inhibitors, IκBs. TNF-α produced by macrophages, lymphocytes, keratinocytes, and fibroblasts in response to injury modulates the activity of NF-κB. When TNF-α binds to its cell-surface receptors, TNFR1 and TNFR2, the IκBs are phosphorylated by specific IκB kinases (IKKs). Subsequently, the IκBs are ubiquitinated and degraded, and NF-κB is then free to translocate into the nucleus. Thus, NF-κB regulates genes involved in inflammation, immune responses, and apoptosis.

Activation of the NF-κB pathway has been shown to be elevated in KFs and is thought to be associated with their abnormal cell proliferation and ECM production (Makino et al. 2008). It has been shown that levels of NF-κB pathway-related proteins are constitutively expressed and elevated in keloid lesions compared to normal skin (Messadi et al. 2004). Genes downstream of NF-κB signaling were up-regulated in KFs compared to NSFs after TNF-α treatment, including genes for proinflammatory cytokines and antiapoptotic genes (Messadi et al. 2004). Furthermore, inhibition of the NF-κB signaling pathway markedly reduces cell proliferation and type I collagen accumulation in KFs, indicating that this pathway is involved in keloid pathogenesis (Makino et al. 2008).

UNIFYING CONCEPTS OF PATHOMECHANISMS OF FIBROTIC DISEASES

Keloids, the paradigm of fibroproliferative skin disorders, represent examples of the spectrum of fibrotic diseases with different phenotypic manifestations. At the same time, the characteristic histopathologic feature of all these conditions is accumulation of the ECM of connective tissue, particularly collagen, and the presence of myofibroblasts, which appear to be the principal cell type responsible for collagen production. Extensive research on several fibrotic skin diseases in general, combined with analysis of wound healing and scar formation processes, have identified a number of pathomechanistic features that may be common with most, if not all, fibrotic diseases. A key molecule proposed to modulate the synthesis of the ECM by myofibroblasts is cFN-EDA, which interacts with a number of cell-surface receptors, including various integrins and TLR4, and thereby modulates the differential expression of genes involved in the matrix production and regulating cell proliferation, migration, and apoptosis. Understanding the details of the pathomechanistic features in fibrotic diseases may identify potential pharmacological targets that could be geared toward development of efficient and specific therapies for this group of disorders, which are currently exceedingly difficult to treat.

Therapeutic Prospective

Despite years of research, the treatment of keloids remains a significant challenge. Common physical treatment modalities for keloid scars include surgical excision, cryotherapy, laser ablation, radiotherapy, and silicone gel sheeting (Table 1). Surgical excision may be indicated if other treatments are unsuccessful; however, surgery alone has been associated with recurrence rates as high as 100% (Berman and Bieley 1996). Cryotherapy is most useful in combination with other treatments. Keloid tissue exposed to cryotherapy has been found to have reduced numbers of mast cells and myofibroblasts as well as reduced TGF-β1 expression (Har-Shai et al. 2011; Awad et al. 2017). Fractional carbon dioxide (CO₂) laser ablation is a relatively new treatment for keloids that has been shown to down-regulate both type I and type III procollagen mRNA levels and TGFs, as well as up-regulate MMP-1 (Qu et al. 2012). Multiple CO₂ laser treatments have been shown to reduce keloid bulk (Nicoletti et al. 2013). Radiation is often used as adjuvant therapy in combination with surgical excision and acts by suppressing angiogenesis and inhibiting fibroblast activity (Lee and Jang 2018). Studies have shown that postoperative radiotherapy is associated with a lower recurrence rate than radiotherapy alone (Mankowski et al. 2017). While the exact mechanism of silicone gel sheeting is not known, it is believed to work by enhancing hydration and creating an occlusive local tissue environment (Puri and Talwar 2009).

Table 1.

Physical and medical treatment modalities for keloids

Category	Treatment	Description	Mechanism of action	Recurrence rate	Source
Physical	Surgical excision	Removal of keloid scar via reconstructive techniques	Reduction of keloid size and improvement of the appearance of the keloid scar	45%–100%	Berman and Bieley 1996
Physical	Cryotherapy	Liquid nitrogen, 10–30 sec freeze–thaw cycle repeated for ∼3 cycles per session	Local destruction of keloid tissue; reduction in numbers of mast cells and myofibroblasts as well as reduced TGF-β1 expression (Har-Shai et al. 2011; Awad et al. 2017)	0%–24%	van Leeuwen et al. 2015
Physical	Laser ablation	Fractional CO2, 10,600 nm	Down-regulation of types I/III procollagen levels and TGFs; up-regulation of MMP-1 (Qu et al. 2012)	73.9%–92.3% within 2 years	Mamalis et al. 2014
Physical	Radiation therapy	Variety of techniques including X-ray, electron beam, or brachytherapy	Suppression of angiogenesis and inhibition of fibroblast activity (Lee and Jang 2018)	37% ± 12%	Mankowski et al. 2017
Physical	Silicone gel sheeting	Soft, flexible, adhesive dressing containing silicone	Enhanced hydration; creation of occlusive environment (Reish and Eriksson 2008)	Unknown	Not applicable
Medical	Intralesional corticosteroids	Triamcinolone, 10–40 mg/mL intralesional injection	Reduction of collagen synthesis and inhibition of fibroblast proliferation (Hochman et al. 2008)	50% after 5 years post-treatment	Morelli Coppola et al. 2018
Medical	Intralesional fluorouracil	5-Fluorouracil, 50 mg/mL intralesional injection	Inhibition of TGF-β-induced collagen type I expression and fibroblast proliferation; increased fibroblast apoptosis (Bijlard et al. 2015)	25%–47%	Bijlard et al. 2015
Medical	Antihypertensives	Angiotensin receptor blockers (ARBs) and angiotensin-converting enzyme inhibitors (ACEIs)	Modulation of IL-6 signaling pathway (Zhang et al. 2021)	Insufficient evidence	Danielsen et al. 2016
Medical	Calcium channel blockers	Intralesional verapamil injections	Modulation of IL-6 signaling pathway (Zhang et al. 2021)	1.4%–48%	Not applicable
Medical	Losartan	5% Topical losartan ointment	Inhibition of TGF-β pathway (Hedayatyanfard et al. 2018)	Insufficient evidence	Not applicable
Medical	Stem cells	Various delivery methods, including systemic or local injection	Inhibition of IL-6 and α-SMA secretion, and keloid fibroblast proliferation (Liu et al. 2018)	Insufficient evidence	Not applicable

Medical treatment modalities for keloids primarily include intralesional corticosteroids and intralesional fluorouracil. Corticosteroids are one of the most commonly used treatments for keloids, and they act by down-regulating collagen synthesis and increasing fibroblast apoptosis (Hochman et al. 2008). Corticosteroids can lead to softening of keloids; however, complete resolution is uncommon and failure to receive routine injections frequently results in regrowth of the lesions. 5-Fluorouracil (5-FU) works by inhibiting TGF-β-induced collagen type I expression and fibroblast proliferation, as well as increasing fibroblast apoptosis (Bijlard et al. 2015). Unfortunately, the therapeutic approaches for keloids are less than optimal with incomplete resolution, poor cosmetic results, and high reoccurrence. Therefore, new innovative treatments are needed.

RNA interference (RNAi) is a regulatory biological process that leads to sequence-specific suppression of gene expression by targeting mRNA for degradation (Han 2018). Its high specificity and selectivity make RNAi technology a powerful tool for gene silencing, and importantly, provide opportunities to uncover novel treatment strategies to treat fibrotic keloid disorders. For example, β-catenin is remarkably up-regulated in keloid tissue, and studies suggest that the Wnt/β-catenin signaling pathway regulates cell apoptosis and proliferation in keloid tissue by transcriptional targeting of telomerase (Yu et al. 2016). Moreover, studies have shown that targeting of β-catenin with small interfering RNA (siRNA) results in sustained inhibition of profibrotic phenotypes (Cai et al. 2017). Autocrine motility factor (AMF) is a tumor-associated cytokine that is also overexpressed in keloid tissue and can promote the proliferation and migration of KFs. Notably, studies have demonstrated that injection of AMF siRNA into subcutaneous tissue inhibited keloid growth, specifically through inhibition of the RhoA/ROCK1 signaling pathway (Tian et al. 2019). Previous studies have also been conducted to examine siRNA knockdown of tissue-type inhibitor of matrix metalloproteinases-1 and -2 (TIMP-1 and -2), which consequently led to increased MMP-1 and MMP-2 activity and ultimately increased the degradation of collagen type I (Aoki et al. 2014).

Other reports have described siRNA targeting in KFs. Expression levels of Runt‐related transcription factor 2 (Runx2) are significantly up-regulated in KFs and keloid tissues as compared to their normal counterparts. Significantly, Runx2 knockdown with siRNA was found to inhibit KF proliferation, migration, and ECM deposition, as well as to increase apoptosis of KFs. Additionally, RNAi of VEGF has been shown to down-regulate plasminogen activator inhibitor 1 (PAI-1) expression in KFs, leading to inhibited KF growth (Zhang et al. 2008). Several previous studies investigated siRNA targeting of Smad2 and Smad 3, intracellular effectors of TGF-β signaling, and found decreased collagen synthesis in these KFs (Gao et al. 2006; Wang et al. 2007).

RNAi's high specificity, efficiency, and transmissibility make this technology an attractive prospect for therapeutic techniques. Specifically, siRNA silencing of targeted mechanisms known to be involved in keloid genesis holds great promise in bringing advancement to current treatment options for keloid disorders.

Currently, other therapies are being investigated that treat keloids by regulating cytokines involved in keloid genesis. Antihypertensive pharmaceuticals, such as angiotensin receptor blockers (ARBs), angiotensin-converting enzyme inhibitors (ACEIs), and calcium channel blockers are thought to treat keloids by modulating the IL-6 signaling pathway (Zhang et al. 2021). Topical losartan oinment has also been found to reduce vascularity, stiffness, pain, and pruritus of keloid scars by inhibiting the TGF-β pathway (Hedayatyanfard et al. 2018). Other studies have found that adipose stem cells (ASCs) may be a potential therapeutic target for keloid disorders. Liu et al. (2018) found that ASC-conditioned media inhibited the secretion of IL-6, α-SMA, and proliferation of KFs. Other emerging therapies include collagenases, interferons, and genetic and epigenetic therapies (Ojeh et al. 2020). Given the lack of successful treatments, a deeper understanding of the pathomechanisms of keloid disorder and further research investigating treatment efficacy are needed to develop innovative and potentially more effective treatments.