The Missing Link between Itch and Inflammation in Atopic Dermatitis

Abstract

Atopic dermatitis is a common skin disease with high morbidity and is associated with severe itch and chronic skin inflammation. In this issue of Cell, Wilson et al. demonstrate that epithelial cells communicate directly with cutaneous sensory neurons via a cytokine to induce itch.

Atopic dermatitis (AD) is a cutaneous disorder that affects 10%–30% of pediatric and 2%–10% of adult populations in developed countries and it’s prevalence is rising (Eichenfield et al., 2012). It is characterized by intensely and incessantly itchy, dry, inflamed skin that begins in infancy and often persists throughout adulthood. AD often heralds the onset of other allergic IgE-mediated diseases that may affect skin or different epithelial surfaces, including food sensitivities, asthma, and allergic rhinitis in what is known as the “atopic march” (Jansén et al., 1973). Although the pathogenesis of AD is still unknown, it is established that epidermal barrier disturbance and immune dysfunction resulting in IgE sensitization are critical factors in the development of cutaneous inflammation. Thymic stromal lymphopoietin (TSLP), an epithelial-derived cytokine that is upregulated in the setting of barrier disruption, has been implicated in the pathogenesis of AD and asthma (Leyva-Castillo et al., 2013). Levels of TSLP expression in skin correlate with symptoms and severity in AD patients and in animal models (Leyva-Castillo et al., 2013). TSLP acts as a potent stimulator of Th2 cytokines including interleukins (IL) 4, 5, and 13, that, in turn, trigger IgE production and release from plasma cells (Brandt and Sivaprasad, 2011). Although TSLP’s actions on immune cells have been extensively studied and assumed by some to be the indirect source of pruritogens that cause itch in AD, its effects on cutaneous sensory nerves has remained unclear. In this issue of Cell, Wilson et al. uncover a mechanism by which TSLP acts as a primary pruritogen that directly activates peripheral sensory nerves to induce itch.

Previous studies have shown that constitutive overexpression of TSLP in keratinocytes triggers vigorous scratching and the development of eczematous skin lesions in mice (Ziegler, 2012). Here, Wilson et al. (2013) provide evidence for direct communication between keratinocytes and somatosensory nerves via TSLP. The authors confirm neural expression of the TSLP receptor (TSLPR), a heterodimer that consists of an IL7 receptor α chain and a TSLP-specific receptor chain, using RT-PCR, in situ hybridization, and immunohistochemistry. Building on earlier work that demonstrated itch-evoking properties of chronic exposure to TSLP (Jessup et al., 2008; Leyva-Castillo et al., 2013; Yoo et al., 2005), the authors show that injection of TSLP elicits acute bouts of scratching in wild-type mice. Importantly, scratching behavior in response to TSLP delivery is preserved in transgenic mice lacking T, B, or mast cells, suggesting that immune recruitment is not required for TSLP-evoked neural activation. Consistent with direct neural activation, TSLP application to cultured murine DRG neurons resulted in increased intracellular calcium in a defined subset of nociceptive fibers. TSLP-responsive fibers also responded to capsaicin and allyl isothiocyanate, agonists for members of the transient receptor potential (TRP) family of ion channels, TRPV1 and TRPA1, respectively. Utilizing a series of pharmacologic inhibitors, the authors determine that TSLP-induced neuronal excitability and scratching behavior requires PLC-mediated activation of TRPA1 activity (Figure 1). TSLP-sensitive neurons appear to represent a novel neural subset distinct from previously identified DRG populations that express TRPA1 and/or TRPV1 in that these fibers do not respond to other pruritogens such as histamine, chloroquine, or BAM8-22.

Figure 1. Keratinocyte-Derived TSLP Directly Activates Primary Sensory Afferents to Evoke Itch.

(A) Endogenous and exogenous proteases activate PAR2 signaling in keratinocytes. PAR2 activation triggers Ca²⁺ release from and depletion within the ER, which subsequently stimulates STIM1-mediated opening of ORAI1 channels on the plasma membrane. Ca²⁺ influx into keratinocytes triggers calcineurin-dependent dephosphorylation and nuclear translocation of NFAT. NFAT activation induces transcription of TSLP. Secreted TSLP acts directly on immune cells and nerves. (B) Keratinocyte-derived TSLP binds to the heterodimeric TSLP receptor on peripheral afferent C-fibers, activating the PLC signaling pathway and resulting in the opening of TRPA1 ion channels. TRPA1 activation stimulates membrane depolarization and initiates neural signaling cascades that result in the sensation of itch.

Given the importance of TSLP in stimulating both the itch and inflammation associated with AD, Wilson et al. next identify critical upstream molecular mechanisms influencing the expression and release of TSLP by keratinocytes. Previous work has established a role for protease signaling via protease-activated receptor 2 (PAR2) in regulating TSLP expression. Tryptase, an endogenous PAR2 ligand, and PAR2 activity are elevated in the skin of AD patients (Steinhoff et al., 2003). PAR2 activation by tryptase or a synthetic hexapeptide agonist SLIGRL triggers increased expression of TSLP in cultured epithelial cells (Moniaga et al., 2013). The authors confirm and extend these findings by showing that tryptase and SLIGRL evoke keratinocyte secretion of TSLP via a calcium-dependent mechanism that requires Ca²⁺ release from intracellular stores and influx via the store-operated Ca²⁺ entry (SOCE) pathway. Utilizing pharmacologic manipulation, electrophysiology, and siRNA knockdown strategies, they demonstrate that ORAI1 and the calcium-dependent ORAI regulator, stromal interaction molecule 1 (STIM1), are required for tryptase- and SLIGRL-activated, PAR2-evoked SOCE and TSLP secretion by human keratinocytes. Similar to its role in immune cells, ORAI1/ STIM1 signaling activates nuclear translocation of NFAT in cultured keratinocytes, thereby triggering TSLP expression and secretion (Figure 1). Inhibition of NFAT-mediated transcription using cyclosporine reduced TSLP release from keratinocytes and airway epithelial cells in vitro and attenuated epithelial expression of TSLP in vivo.

Although the importance of TSLP in the pathogenesis of atopic disease has been appreciated for over a decade, several aspects of this study make it exciting and provocative: First, it defines a novel and direct role for TSLP in keratinocyte-neuron communication that does not require immune stimulation. Direct neural activation by TSLP may explain why individuals suffering from atopic dermatitis typically develop itchy skin prior to the onset of eczematous skin lesions, the development of which requires recruitment of immune cells and further cytokine elaboration. Second, the observation that TSLP signaling in nerves requires TRPA1 activation raises the possibility that epithelial-neural crosstalk, in addition to epithelial-immune signaling, contributes to the development of elements of atopic disease and the atopic march. The authors aptly mention that TRPA1 afferents innervate cutaneous, bronchial, and gastrointestinal epithelia. Moreover, TRPA1 activation triggers inflammation in mouse models of asthma and inflammatory bowel disease (Bautista et al., 2013). The findings of this study then beg the question of whether TSLP-mediated activation of TRPA1, or other epithelial-nerve interactions, may also play a disease-inciting role in the lung and gut. Finally, Wilson et al. provide new insights into plausible mechanisms by which cyclosporine and other calcineurin inhibitors effectively treat chronic pruritic and/or inflammatory disorders, including atopic dermatitis, prurigo nodularis, lichen planus, and others. Nonetheless, the relative impact of immune cell inhibition and reduction in epithelial-evoked TSLP release and neural activation will require further scrutiny.

Several key questions remain unanswered by this study. First, what are the relative contributions of TSLP-mediated keratinocyte-neuron and immune-cell-neuron communication to the induction and maintenance of atopic disease? Additional studies using tissue-specific TSLPR-deficient mice will be required to resolve this issue. Second, what role might TSLP- and TRPA1-sensitive nerve fibers play in routine somatosensation? Given the lack of response to other known pruritogens, could it be that these fibers act as an innate monitor for epithelial barrier compromise? Furthermore, whether there are distinct neural populations that respond to epithelial-derived cytokines other than TSLP is yet unknown. Answering these questions will be crucial to our evolving understanding of the pathogenesis of atopic disease and other inflammatory cutaneous disorders.