One Size Does NOT Fit All: Diversifying Immune Function in the Skin

Abstract

Our bodies most outward facing epithelial barrier, the skin, serves as the frontline defense against myriad environmental assailants. To combat these motley treats, the skin has evolved a sophisticated immunological arsenal. Herein, I provide an overview of the skin’s complex architecture and the distinct micro niches in which immune cells reside and function. I review burgeoning literature on the synchronized immune, stromal, epithelial, and neuronal cell responses in healthy and inflamed skin. Next, I delve into the distinct requirement and mechanisms of long-term immune surveillance and tissue adaptation at the cutaneous frontier. Finally, by discussing the contributions of immune cells in maintaining and restoring tissue integrity, I underscore the constellation of non-canonical functions undertaken by the skin immune system. Just as our skin’s immune system benefits from embracing diverse defense strategies, so too must we in the immunology research community support disparate perspectives and peoples from all walks of life.

Introduction

The skin, our body’s outermost and largest barrier, interfaces with the terrestrial environment and is tasked with protecting our internal organs from external threats. This topologically heterogenous organ not only serves as a physical barrier, but has evolved sophisticated sensory mechanisms to detect extrinsic stimuli that perturb the steady state and mount tailored responses to restore homeostasis(1). The skin’s extraordinary ability to rebound after assaults impinges upon the development and maintenance of diverse physical, chemical, and immunological defense mechanisms that incorporate and adapt to cues from the extrinsic environment(2–4). Over the last decade, technological advances have led to an explosion in our understanding of skin immunology and the remarkable means by which immunity is integrated into skin structure and function(5). Here I discuss the context specific function of the cutaneous immune system and highlight the myriad strategies employed to ensure robust host responses in the face of danger. I conclude by drawing parallels between cutaneous immune heterogeneity and that of researchers engaging in immunological research to emphasize the importance of diversity for the advancement of basic discovery (6, 7).

The skin’s structure

The outermost layer of the skin, the epidermis is comprised of tightly stacked epithelial cells, which form a semipermeable barrier (Figure 1). The epidermis can be subdivided into four distinct layers: stratum basale, stratum spinosum, stratum granulosum and stratum corneum. Proliferative epithelial stem and progenitor cells (ESPCs) reside in the lowermost or basal layer of the epidermis, where they self-renew and differentiate upwards to give rise to the stratified tissue layers(8). Upon exiting the basal layer, epithelial cells lose their mitotic potential and adopt a polygonal shape to intercalate in the spinous layer(9). The next layer is lipid filled barrier made of granules and lamellar bodies and aptly named the granular layer (10). Finally, the lion’s share of the mechanical and chemical protection is provided by the cornified layer(11). Cells anucleate and flatten as they move into this superficial layer. These flat dead cells combined with proteolytic-resistant proteins, hydrophobic lipids, corneodesmosomes, and tight-junctions generate a “Teflon-like” hydrophobic coating to seal the skin. The epidermal layers act in unison to physically restrict the movement of water and noxious agents.

Figure 1: Immune niches in the skin.

Schematic of distinct immune niches in the skin including: the interfollicular epidermis, pilosebaceous unit (hair follicle and sebaceous gland), papillary, reticular and hypo dermis, vascular and neuronal niches. Each of these locations are home to highly specialized immune populations that crosstalk with adjacent parenchyma to maintain barrier function.

Beneath the epidermis lies the dermis which can be fractioned into three structurally and functionally distinct layers (Figure 1). Directly below the epidermis is the papillary dermis. It is connected to the epidermis via the basement membrane, which provides a supportive niche for epithelial progenitors. Epidermal invaginations such as hair follicles and sebaceous glands (collectively called the pilosebaceous unit) and sweat glands penetrate into the underlying papillary dermis(12–14). These structures are essential for production of surface lipids that act as emollients, thermoregulation by controlling water loss, and sensory and protective functions of hair. The keratinocytes of the hair follicle are contiguous with the epidermis and also form stratified layers with a water tight barrier (15).

The reticular dermis lies underneath and is thicker than the papillary dermis. It is composed of densely packed collagen and elastin fibrils, and extra cellular matrix (ECM) (16). The lowermost layer of the dermis, the hypodermis, is also known as the subcutaneous fat layer and is rich is adipocytes (17). The hypodermis insulates the body from cold and aides in shock absorption. Each of the dermal layers house distinct fibroblast populations with unique functional properties(16, 18). Additionally, the dermis and epidermis are innervated and have both lymphatic and vascular endothelial vessels(19, 20). Below I discuss the distinct immune features of these cutaneous microcosms and highlight the supportive and direct contribution of their parenchyma to immune function.

Local Government

Emerging literature has revealed the remarkable regional segregation of immunity in the skin and the mechanisms by which a unique immune milieu is sustained in each of the skin’s micro niches. Organization of immune networks and maintenance in distinct areas of the skin requires discrete expression of recruitment, retention, survival and activation factors in specific locations to draw in and sustain immune cells with corresponding receptors. In turn, specialized immune cells supply their surrounding parenchyma with context-specific instructive cues that are essential for tissue function.

Epidermal patrolling immune cells

The epidermis is constantly surveyed by Langerhans cells (LCs), CD8⁺ resident memory T cells (T_RM), innate lymphoid cells (ILCs), and in mouse skin, dendritic epidermal γδ T cells (DETCs) (Figure 1) (13, 21, 22). LCs, ILCs and DETCs seed the skin during embryonic development and maintain residence throughout life. Intriguingly, these populations localize to the skin at long before birth and yet their function in development remains largely unexplored. Arising from primitive CSFR1⁺ yolk sac progenitors and fetal liver monocytes, LC arrive in the epidermis at embryonic day (E) 12.5 (22). Shortly thereafter, around E16, DETCs precursors establish themselves the directly above the basal layer and project their dendrites into the suprabasal epidermal layers where they can sense barrier disruptions(23). T_RMS also localize to the basal epidermis and were first identified as persistent epithelial residents following clearance viral infections in mice(24, 25). Thus, their accumulation in the skin was largely thought to be following antigen encounters postnatally(26). However, recent work characterizing prenatal human skin have identified T_RMs even before birth and linked their induction and functional responsiveness to microbiota derived signals found in fetal tissue (27, 28). These early studies underscore the role of the fetal microenvironment in establishing skin immunity and raise the tantalizing possibility that maternal health may be a key determinant of early life immunity and susceptibility to childhood skin diseases such as atopic dermatitis(29).

Epidermal resident immune populations rely on local chemokine and cytokine cues for recruitment and survival. For instance, the absence of LCs in fetal dermis and identification of cycling LCs in the epidermis suggests that LC differentiation is largely restricted to the epidermal environment(30, 31). Indeed, keratinocytes robustly express a LC survival factor, interleukin (IL)-34 (32). Similarly, epithelia also express chemokines CCL27 and CCL20 to recruit epidermal lymphocytes and survival factors IL-7 and IL-15 that sustain these cells locally(33–36). Local competition for these factors dictates composition of the epithelial lymphocyte pool. With age and antigen exposure, CD8⁺ T_RMs displace DETCs in the epidermis(37). Thus, the frontline troupes that survey the epidermis for threats, are recruited there even before birth and their homeostatic pool is continually refined throughout life based on microbial and other inflammatory encounters.

Hair follicles are focal centers of immunity

Penetrating into the papillary and reticular dermis, the pilosebaceous unit is a dynamic neighborhood for immunity. This region is enriched in both epidermal and dermal immune cells separated by a basement membrane(36, 38). These cells are recruited and retained in perifollicular region by unique combinations chemokines and cytokines expressed in upper region of the follicle (infundibulum) toward the skin surface(36, 39). Hair follicles (HFs) are also densely colonized with commensal microbes which are known to augment expression of follicular immune modulators(35). While follicle epithelia are contiguous with the intrafollicular epithelium (IFE), the tight junctions are expressed at a lower level than in the IFE likely resulting in a more permeable barrier(40, 41). Dermal antigen presenting cells (APCs), including CD103 dendritic cells (DCs) and dermal DCs and have been observed survey the follicle (42). Indeed, these cells are known to prime homeostatic commensal-specific Tc17 cells following topical microbial colonization(42, 43). Permissive HFs are often considered a portal for allergen and pathogen entry, and rely on commensal signals to maintain robust immune surveillance. Intriguingly, follicular stem cells have evolved unique mechanism to shield themselves from immune mediators by expressing key immune dampening molecules and evade immune detection by lowering their MHCII expression(2, 44). Thus, this mini-organ must weigh active immunity with tissue preservation and the mechanism by which epithelia achieve this balance are actively being investigated.

Endothelial vessels draw immune attention in the reticular and hypodermis

Immune cells in the deep dermis line the vascular endothelium to survey the blood system for internal threats and to efficiently recruit troops during inflammation. Innate immune cells including macrophages and mast cells, are particularly enriched in the perivascular space. (Figure 1)(45). Both macrophages and mast cells seed the skin neonatally from yolk-sac, fetal liver and/or bone marrow progenitors(46). Thus, anticipatory immunity in the fetal skin is not limited to the epidermal skin but also extends to the lower layers of the dermis.

During skin inflammation, perivascular mast cells and macrophages secrete cytokines and chemokines to activate the endothelium to permit entry of neutrophils, monocytes and lymphocytes from circulation(47). The perivascular space is also an epicenter for effector T cell activation as dermal DCs localize to and cluster with effector T cells to mediate contact hypersensitivity (48). Why do leukocytes gather around vessels? Strategically, positioning cutaneous immune cells in this location likely facilitates rapid communication with systemic immunity, while at the same time monitoring organismal nutritional status and health. Indeed, in the hypodermis, fat associated adipose tissue macrophages have with a high endocytic capacity respond to systemic changes in metabolism and infectious challenge(49). Thus, immune surveillance in the skin not only focuses on environmental threats, but also monitors dangers that loom within our body.

What is and isn’t an immune cell?

Immune supportive functions of non-hematopoietic cells, in particular epithelia, fibroblasts and endothelia, have been long appreciated(2). As discussed above these cells produce an array of factors to regulate the distribution and function of immune cells in healthy and inflamed skin. However, more recently, neurons have emerged as key modulators of skin immunity and the neuroimmune interface is increasingly recognized as the next frontier of immunology (19).

Linking inflammatory status to host behavior, such as scratching, allows for physical removal of noxious stimuli in the skin. Indeed, skin innervating nerve fibers robustly express receptors for a number of cytokines including IL-4 receptor, and remarkably inhibiting JAK1 signaling downstream of IL-4ra dramatically improves clinical pruritis(50). The neuroimmune axis is pathologically co-opted pruritis(itch)-associated inflammatory diseases such as atopic dermatitis, psoriasis and chronic spontaneous urticaria(51). Just as immune stimuli provoke neurons, sensory neurons similarly drive inflammation. Blanco et al found that ablating TRPV1⁺ Nav1.8⁺ nociceptors (noxious stimuli-sensing neurons) dramatically reduced inflammation in a mouse model of psoriasis(52). These nociceptors interact with dermal DCs and stimulated IL-23 production to fuel IL-17 responses in the skin via the neuropeptide CGRP(53). Thus, communications between neurons and immune cells are mediated by both immune derived cytokine and neuronal peptides and the integration of these two systems links organismal behavior to inflammatory status.

Similar to their immune brethren, neurons are exquisite at sensing noxious, thermal, mechanical and other stressors through G coupled protein surface receptors(19). Neurons also directly sense pathogenic stimuli, such as Staphylococcus aureus, leading to pain stimuli. Strikingly, elimination of nociceptors during infection lead to immune hyperactivation suggesting that pathogens may dampen immune function by modulating neuronal activity(54). Cutaneous neurons also have immunoregulatory functions in the absence of pathogenic infections. Nonpeptidergic neurons in the skin constitutively produce glutamate to suppress mast cell hyperactivation(55). Similarly, neuronally supplied TAF4 modulates macrophage IL-10 production and consequently tissue repair following sunburn-like skin injury(56). These early studies illustrate the dynamic balance between pro and anti-inflammatory function of the cutaneous nervous system. The study of neuroimmune communication has focused on nociceptors, leaving the door open for exploration of other types of neurons for example mechanosensory neurons that innervate the hair follicle or accessory cells such as Schwan cells that activate pain sensing neurons in immune modulation(57, 58).

Concomitant immunity, Inflammatory Memory, and Heterologous Protection

Perhaps one of the most powerful mechanisms of immune mediated-barrier function is the ability to remember and learn from inflammatory encounters. In some cases, immunity is ongoing with chronic antigen persistence such as with chronic infections or microbiota derived antigens (59). These persistent immune responses, occurring in the presence or absence of antigen, are leveraged to initiate rapid and informed responses to subsequent stressors both locally and at distal skin sites.

Concomitant and commensal-driven homeostatic immunity

Concomitant immunity or long-term persistence of pathogens in a host that is also able to maintain strong resistance to reinfection was first described in the context of chronic parasites infections such as Leishmania major (60). This balancing act relies on long-term maintenance of pathogen specific effector-memory and regulatory T cells (T_reg) that mitigate tissue pathology but rely on low level pathogen persistence(61, 62). This complex volleying between host and parasite is useful not only to control the pathogen locally, but to maintain resistance to re-infections(63). Thus, the idealized goal of sterile pathogen clearance may not be optimal for coping with chronic and recurring infections that warrant constant immune activity.

Our skin is entrenched in commensals, which are a rich source of antigens throughout our lifetime(64). Commensal signals tune the level of skin immunity by inducing cognate T cell responses and non-specifically by augmenting antimicrobial peptides and other innate immune pathways(35, 42, 65, 66). Within moments of birth, commensals rapidly colonize the skin(67). The neonatal immune system establishes tolerance to these primordial colonizers by inducing commensal specific regulatory T cells (T_regs)(68). Remarkably, our fledgling immune system discriminates colonizing commensals from pathobionts by sensing microbial toxins(69). Early colonization with the α-toxin producing Staphylococcus aureus induces IL-1β production from myeloid cells and inhibits the generation of S. aureus specific T_regs(69). Intriguingly, longitudinal clinical studies uncovered disturbances to commensal microbiota of infants well before the onset of atopic dermis, suggesting that early life commensal-immune interactions may be critical for establishing a healthy barrier and limiting inflammatory diseases(70).

With age and exposure, the composition of the skin microbiota dynamically changes(67). So too, does our immune system’s response to commensals. Engagement of the adult immune system with commensals results in the induction of homeostatic Tc17, Th17, γδ T cell, and MAIT responses. Moreover, while some keystone species are capable of inducing broad immunity, other species and induce unique populations of immune cells. For example, certain strains of Staphylococcus epidermidis uniquely induce a population of Tc17 cells that reside in the epidermis(42). Similarly, Corynebacterium are particularly adept at eliciting Type 17 γδ T cell(71). The details of how the immune system detects and discriminates between these microbes in adulthood to engage unique immune features remains to be seen. Towards this end, DCs, key immune sentinels, have been observed to extend their dendrites through the various epidermal layers to capture antigen and induce commensal specific lymphocytes.(42, 72)

Commensally driven lymphocytes largely reside in the epidermis and upper dermis and constitutively reinforce the barrier by inducing the production of antimicrobial peptides that in turn regulate the microbial ecosystem(42). Indeed, both dysbiosis of skin commensal communities and their translocation to cutaneous lymph nodes have been observed in mice lacking adaptive immunity(73). Thus, the skin barrier uses commensal signals to constitutively tune local immunity. Whether, commensal specific lymphocytes require sustained antigen signal or can form memory in the absence of the inducing microbes remains to be explored.

Inflammatory Memory

Memory of inflammatory encounters endure long after clearance of the initial stimuli or antigen and can be parsed into antigen-specific adaptive immune memory and non-specific innate training(26, 74). Both of these features have been observed in the skin and contribute to local modulation of immunity. More recently, however, we and others have found that non-immune cells also maintain a memory of inflammation(75). Exposure of murine skin to psoriatic-like inflammation results in dramatic and lasting alterations to the chromatin structure of ESPCs that persist for 180 days (~15 human years). Larsen et al. recently found that these retained inflammatory memory domain are induced and maintained by AP-1 family of transcription factors which cooperate with stimulus specific inflammatory transcription factors to index ESPC chromatin(76).

While such memory can be beneficial in boosting barrier immunity, it can also be co-opted to fuel inflammatory disease. Indeed, resident memory lymphocytes have long been thought to drive recurrent psoriatic inflammation(77, 78). However, the resurgence of plaques in the same skin locations suggests that other innate immune and non-immune tissue cells may also retain a memory of disease(79, 80). Accordingly, keratinocytes cultured ex vivo from psoriatic plaques are enriched for progenitor signatures with reduced differentiation signatures compared to non-lesional skin(81). These early studies of patient samples prompt further exploration of tissue memory of inflammation beyond immunity and suggest that achieving long-term clearance of inflammatory disease may require treating both immune dysfunction and broader tissue inflammatory memory.

Heterologous protection

Concomitant immunity and homeostatic commensal-specific responses represent situations in which antigen constitutively stimulates immune activity similar to a “buzzer” whereas, immune memory persists long-after clearance of the initial stimuli much like a “switch”. While these responses can promptly target the initial stimulus, they can also be activated non-specifically via the persistence of local antigen or the production of cytokines and directed towards another secondary stimulus. For instance, co-infection with L. major and Staphylococcus aureus results in augmented immunity against both microbes(82). Similarly, commensal microbes act as natural adjuvants and stimulate both antigen-specific lymphocytes and anti-microbial peptides from keratinocytes to limiting early invasion of Candida albicans and promote clearance of L. major (42, 83).

T_RMs and other tissue resident lymphocytes are able to respond to alarmins and other factors released by damaged tissue to respond to non-specific stimuli(84). These cells also play a vital role in early control of transformed cells and restrain tumor growth long-term. For instance, in an epicutaneous melanoma transplant model, CD8⁺ T_RMs limited the outgrowth of tumors but also did not eliminate melanoma cells altogether, maintaining an “equilibrium” with the transformed melanocytes(85). Just as immune cells can non-specifically respond to secondary threats, epithelial memory is also repurposed to boost tissue repair(75). The ability to easily repurpose various immune modules generated at the cutaneous interface may be the most cost-effective means of rapidly coping with recurrent stressors or those that warrant similar responses.

Tissue supportive roles of immunity

The last decade has seen an explosion of studies on the non-canonical functions of immune cells in the skin. These findings extend the role of immune cells well-beyond host defense into the realm of tissue maintenance and repair. In particular, T_regs and macrophages, both enriched in the skin, regulate epithelial and stromal cells function via diverse mechanisms (86).

The hair follicle undergoes cyclical bouts of rest and regeneration impinging upon follicle stem cell (SC) quiescence and activation. For over two decades groups have observed that the composition of immune cells changes with the stage of the follicle cycle, even at steady state(87). However, the reason for this flux and the contribution of perifollicular immune cells to follicle SC behavior and cycle stage was, until recently, unclear. Single nucleotide polymorphisms in genes associated with SC-immune crosstalk are strongly implicated in Alopecia areata, a hair loss disorder, suggesting a functional interaction between immune cells and follicle SCs may be at play(88). Indeed, Rosenblum and colleagues, recently found that T_regs are specifically enriched around the follicle bulge where SCs reside and supply them with activating notch signals to facilitate SC activation and kick-start the hair cycle(38). On the other hand, a dermal TREM2+ macrophages subset secretes oncostatin M to maintain SC quiescence and keep follicles in the resisting phase(89). How these pro-and-anti-quiescence populations are regulated and if follicle SCs are at the reigns or if other tissue signals dictate perifollicular immune dynamics warrants further investigation.

Sebaceous glands, another component of the pilosebaceous unit are responsible for secreting lipid rich sebum to lubricate the skin. Type 3 innate lymphoid cells (ILC3s), residing in the upper hair follicle, control sebaceous hyperplasia and composition of sebum via TNF ligands and notch(13). Consequently, the composition of skin commensals in animals lacking ILC3s was dramatically altered. In contrast to the homeostatic restrain of sebaceous glands, over production of the alarmin TSLP results in migration of T cells into sebaceous glands and hypersecretion of sebum(90). These animals experienced selective depletion of white adipose tissue and dramatic over all weight loss. Thus, regulation of skin appendages by immune cells have consequences for both the local microenvironment and the organismal macroenvironment.

Perhaps the most heavily studied atypical functions of immune cells are in the context of tissue damage and repair(91). Immune derived signals direct repair of epithelial, stromal and neuronal cells following acute tissue damage. Here an interesting bifurcation emerges, whereby crosstalk with damaged epithelium is relegated to patrolling adaptive immune cells and the tissues dermal constituents are engaged by innate immune cells, in particular macrophages.

Following damage, epithelial progenitors at the wounds edge proliferate and migrate to swiftly seal the exposed surface in a process termed re-epithelialization(92). Havran and colleagues published pioneering studies detailing the function of DETCs in epithelial proliferation via the production of various growth factors(93). Since then, IL-17A emanating from homeostatic Tc17 cells, MAIT cells, dermal γδ T cells, and ILC3s and their supplied factor has been identified as a key regulator of re-epithelialization(43, 65, 94, 95). Notably, many of these studies were performed in immunodeficient animals. If IL-17A from distinct cellular sources is required for repair, its temporal dynamics of activation, and the functional redundancy between various homeostatic lymphocytes remain open questions. Moreover, how repair associated lymphocytes are activated following damage requires elucidation. Decoding the mechanisms of lymphocyte-epithelial crosstalk may enable targeting of this interaction in non-healing wounds that fail to re-epithelialize(96). Indeed, aged skin, which often fails to heal, exhibits a severe impairment in the DETC-epithelial crosstalk(97).

During wound repair, the dermis must re-establish its architecture by depositing ECM, promoting revascularization and innervation(98). Following injury, dermal fibroblasts proliferate and differentiate into ECM-producing myofibroblasts. CD301b⁺ dermal macrophages produce insulin-like growth factor and platelet derived growth factor to directed myofibroblasts differentiation and pro-fibrotic function(99). Loss of CD301b⁺ macrophages and resultant fibroblast defects leads to a profound delay in repair(100). The macrophage-nervous crosstalk is also central to the repair process. CX3CR1+ macrophages are intimately associated with cutaneous sensory nerves contribute to axonal sprouting following injury(101). Sensory neurons cross regulate macrophages by inducing IL-10 production via the neuropeptide TAF4 to limit fibrosis following UV induced tissue damage(56). Macrophages also orchestrate the dynamics of blood vessel formation and regression following injury, where early in repair macrophages promote angiogenesis and after the wound has healed they direct vessel clearance(102). Thus, highly specialized macrophages populations simultaneously interact with and direct the repair of distinct dermal components(103). If these macrophages are developmentally specified to perform these functions or obtain their unique powers once the occupy distinct niches in the skin (for example perivascular vs. perineuronal) is poorly understood. However, harnessing the pro-repair functions of macrophages may transform regenerative therapies in the skin and beyond.

Conclusion and Perspective

Far from a simple physical barrier, the skin has evolved sophisticated mechanisms to cope with myriad threats, while still maintaining complex microbial communities on the its surface. In addition to classical anti-pathogenic functions, coupling tissue regulatory and repair functions with immunity may be advantages to mitigate immunopathology and collateral tissue damage resulting from inflammation. Another unappreciated dimension of skin immunity is the unique regulation of its resident immune populations relative to other barrier tissues. For instance, homeostatic Th17 cells in the skin are reliant on IL-1 for function while gut Th17 cells are controlled by IL-6(83). Similarly, ILC2s in the skin are uniquely responsive to IL-18 compared to ILC2s from other barrier tissues(104). The aforementioned areas of skin immune complexity, immune-tissue crosstalk, inflammatory memory and tissue repair are being actively investigated and may enable a skin-specific therapeutic modulation of immune function. However, a key principle cemented in field of skin immunity is that one size does not fit all. Highly tailored immune retort is vital to tackle unique challenges and defend the host from diverse onslaughts.

Just as the immune system embraces unique and diverse strategies to safeguard the skin, we immunologists must also seek a range of perspectives to forge new paths to discovery. In fact, I argue that the multifaceted nature of biology demands examination from different angles, and the engine of discovery falters without it. Indeed, a wealth of evidence indicates that diversity positively impacts scientific discovery through improved problem-solving, innovation, prediction, evaluation, and strategic thinking(6, 105, 106). Yet, inclusion of underrepresented and marginalized groups in biomedical research remains a major challenge.

The brutal murders of Ahmaud Arbery, Briana Taylor, George Floyd and countless others further highlighted the paucity of diversity in biomedicine much in the same way that the #metoo movement highlighted the profound gender imbalance and barriers to advancement of women in science. Fortunately, key stakeholders, including research institutions, government agencies, and private foundations, are responding by publicly proclaiming their commitment to diversity, equity, and inclusion (DEI). I am hopeful that our institutional and individual commitments to DEI will allow folks from all walks of life to share the scientific spot light and ultimately transform the process of discovery in the field of immunology.