Innate immune regulates cutaneous sensory IL-13 receptor alpha 2 to promote atopic dermatitis

The clinical significance and regulators of IL-13Rα2 in itch and atopic dermatitis (AD) remain unclear. To identify disease-driven regulatory circuits of IL-13Rα2, transcriptomic/pathological analysis was performed in skin from patients with AD, psoriasis, healthy subjects, and murine AD model. Functionality was investigated in sensory neurons, keratinocytes and animal model, by using knockdown (KD), calcium imaging, RNA-seq, cytokine arrays, pharmacological assays, and behavioural investigations. In our study, an upregulated IL-13Rα2 expression was revealed in skin of AD patients, but not psoriasis, in a disease activity-dependent manner. In cultured human keratinocytes, IL-13 increased IL-13Rα2 transcription levels, and this were downregulated by IL-13Rα1KD. IL-13Rα2KD reduced transcription levels of EDNRA, CCL20, CCL26. In contrast, sensory neuron-derived IL-13Rα2 was upregulated by TLR2 heterodimer agonists, Pam3CSK4 and FSL-1. In a mouse cheek model, pre-administration of Pam3CSK4 and FSL-1 enhanced IL-13-elicited scratching behaviour. Consistently, in cultured sensory neurons Pam3CSK4 enhanced IL-13-elicted calcium transients, increased number of responders, and orchestrated chemerin, CCL17 and CCL22 release. These release was inhibited by IL-13Rα2KD. Collectively, IL-13 regulates keratinocyte-derived IL-13Rα2 and TLR2 to modulate neuronal IL-13Rα2, thereby promoting neurogenic inflammation and exacerbating AD and itch. Thus, the cutaneous IL-13-IL-13Rα2 and neuronal TLR2-IL-13Rα2 pathway represent important targets to treat AD and itch.

IL-13, Cytokine IL-13Rα1 and 2 cytokine arrays, pharmacological assays, and behavioural investigations. In our study, an upregulated IL-13Rα2 expression was revealed in skin of AD patients, but not psoriasis, in a disease activity-dependent manner. In cultured human keratinocytes, IL-13 increased IL-13Rα2 transcription levels, and this were downregulated by IL-13Rα1KD. IL-13Rα2KD reduced transcription levels of EDNRA, CCL20, CCL26. In contrast, sensory neuronderived IL-13Rα2 was upregulated by TLR2 heterodimer agonists, Pam3CSK4 and FSL-1. In a mouse cheek model, pre-administration of Pam3CSK4 and FSL-1 enhanced IL-13-elicited scratching behaviour. Consistently, in cultured sensory neurons Pam3CSK4 enhanced IL-13-elicted calcium transients, increased number of responders, and orchestrated chemerin, CCL17 and CCL22 release. These release was inhibited by IL-13Rα2KD. Collectively, IL-13 regulates keratinocyte-derived IL-13Rα2 and TLR2 to modulate neuronal IL-13Rα2, thereby promoting neurogenic inflammation and exacerbating AD and itch. Thus, the cutaneous IL-13-IL-13Rα2 and neuronal TLR2-IL-13Rα2 pathway represent important targets to treat AD and itch.
IL-13 has emerged as a prime target in AD (Guttman-Yassky et al., 2019; Stott et al., 2013). It is a key driver of type-2 T-helper (Th2) inflammation and is elevated in the lesional skin of AD patients (Leyva-Castillo et al., 2020;Meng et al., 2021). Upon release in peripheral skin, IL-13 activates neighbouring cells expressing its receptors, recruits inflammatory cells into the skin, and alters the skin microbiome. IL-13 also decreases epidermal barrier function and activates sensory nerves (Leyva-Castillo et al., 2020). The IL-13 receptor alpha-1 (IL-13Rα1) pairs with IL-4R (IL-4Rα) to form a high-affinity ternary receptor complex thereby propagating a Th-2-type inflammatory response and itch signalling (Zurawski et al., 1993). Antagonising this complex is remarkable in treating AD (Brunner et al., 2017). However, the functional contribution of IL-13Rα2 in the pathophysiology of AD still remains elusive (Bitton et al., 2020;Sivaprasad et al., 2010;Ulzii et al., 2019), particularly concerning its exact receptor pathway activating skin cells and sensory neurons.
Thus, the aim of the current study was to investigate (i) the transcription of IL-13Rα2 (in comparison with IL-13Rα1) in skin samples from patients with lesional and non-lesional AD, lesional and non-lesional psoriasis, healthy controls, animal models, primary human keratinocytes (phKCs), as well as sensory neurons; (ii) the functional contribution of IL-13Rα2 (relative to IL-13Rα1), to the IL-13-mediated activation and alterations of itch, inflammatory and neurogenic responses; (iii) the modulatory mechanism of innate immune inflammation to IL-13Rα transcriptional and functional alteration. This study sheds new light on the pathogenesis and progression of AD, in particular onto the neuro-immune circuits involved in this pruritic inflammatory disease condition, aiming at developing new targets to optimize current treatment options.

Animal rights
All animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals of Henan University and approved by the Animal Ethics Committee of Henan University, China.

RNA-seq
Human skin original RNA-seq datasets and methods are available from our previous paper (Larkin et al., 2021), in which punch biopsies were taken from AD, both lesional (AL) and non-lesional AD (ANL), and psoriasis, both lesional (PL) and non-lesional (PNL), and healthy control (HC).

Real-time reverse transcription PCR
Lesional and non-lesional skin of IL-31-overexpressing (OE) mice (Dillon et al., 2004) and age-matched wild-type control skin were quantified for mRNA level of TLR2. Primers were purchased from Origene (China).

Culture of phKCs and murine primary sensory neurons
phKCs were cultured in KBM-Gold medium with KBM-Gold Single-Quot KC supplement (Lonza) as described in . Cells were maintained in the above medium for 3 days before use.
Murine dorsal root ganglion were isolated from postnatal d5 C57BL/ 6 mice, and cultured for 7 days in vitro before use, as previously described (Meng et al., 2018).

Knockdown of IL-13Rα1 and IL-13Rα2 and cytokine array
Cells were incubated in culture medium containing shRNA lentiviral particles that specifically target IL-13Rα1, IL-13Rα2 (OBiO Technology Corp.), or non-targeted scrambled lentiviral particles, as previously described (Meng et al., 2018). Cells were lysed in LDS sample buffer for Western blotting to confirm the knockdown effect. Rabbit anti-IL-13Rα1 (Abcam) or goat anti-IL-13Rα2 (R&D Systems) and mouse monoclonal SNAP-25 (BioLeged) or SNAP-23 (Synaptic Systems) were used for immunoblot.
For cytokine array, cells incubated with or without Pam3CSK4 (1 µg/ ml) or vehicle for 6 h at 37 • C were washed and stimulated by IL-13 (100 ng/ml, Abcam) at 37 • C for 24 h. Cell-culture supernatants were analysed using proteome profiler mouse XL cytokine array kit (R&D Systems) before being analysed with ImageJ software. Resultant densitometry values from treated samples were calculated relative to nontreated control values to determine the fold change.

Behavioural experiments
To establish a prolonged model of AD, MC903 (2 nmol/20 μl in ethanol, Sigma) or vehicle (ethanol) was topically applied to the left ear of C57BL/6 female mice. This was repeated daily for 12 consecutive days. The ear was harvested, paraffinized and embedded before being sectioned for immunohistochemical staining (Meng et al., 2018).

Intracellular Ca 2+ measurements
mDRGs incubated with Pam3CSK4 (1 µg/ml) or vehicle for 6 h were loaded with Fluo-4 AM. Cells were then video-imaged with ImageXpress Micro 4 Automated Cell Imaging System (Molecular Devices) using MateXpress6 software at 8 s intervals. After a baseline period, IL-13 (100 ng/ml) was applied. For each cell, intracellular calcium increases were normalised to F/F0, with F denoting the fluorescence and F0 the baseline fluorescence, and graphed relative to time. Responding cells were analysed and presented as % total cells (Steinhoff et al., 2000). The chi-square statistic with Yates correction is applied for p-value significance evaluation.

Immunofluorescence staining
Paraffin sections were deparaffinised, rehydrated before permeabilization and incubation in block reagent, as described previously (Meng et al., 2018); cultured phKCs were permeabilized and blocked. Specimens were then incubated with rabbit anti-IL-13Rα1 or goat anti-IL-13Rα2 in blocking solution (4 • C, overnight). The specimens were washed in PBS and incubated with donkey anti-rabbit Alexa 594 and donkey anti-goat Alexa 488. After the final wash of the secondary antibody, specimens were mounted onto slides using anti-fade reagents containing DAPI (4 ′ ,6-diamidino-2-phenylindole). Images were taken by IX73 Olympus inverted microscope using CellSens Dimension Imaging software (Larkin et al., 2021).
Other data are presented as mean ± SEM with dot plots or bar graphs. We made two-group comparisons with a two-tailed Student's ttest followed by Welch's correction, except that datasets in Fig. 4J, K and Fig. 5C-F were analyzed using 2-way ANOVA followed by Bonferroni's post hoc analysis to determine where those differences occurred among multiple groups. The criterion for statistical significance is ns P > 0.05, *P < 0.05; **P < 0.01; ***P < 0.001 (Prism 7, GraphPad Software, La Jolla, CA, USA).

Transcription of IL-13Rα2 in skin is correlated with disease activity of AD, but not psoriasis
To examine the clinical relevance of IL-13Rα2 with respect to disease activity in human AD and psoriasis, we analysed skin samples from lesional AD (AL), non-lesional AD (ANL), lesional psoriasis (PL) and nonlesional psoriasis (PNL), and compared to healthy controls (HC). IL-13Rα2 mRNA levels were found to be upregulated in AL vs. HC (Fig. 1A) and in AL vs. ANL (Fig. 1B). IL-13 mRNA levels were also increased in AL vs. HC (Fig. 1A), and in AL vs. ANL (Fig. 1B). In contrast, IL-13Rα1 mRNA levels were not altered in AL vs. HC or in AL vs. ANL.
In contrast to AD, in psoriasis IL-13α1 showed small but significant increment in PL vs. HC (Fig. 1C), based on the thresholds: (log2FCH ≥ 1.0 and FDR < 0.05). However, neither IL-13 nor IL-13Rα2 transcription was altered in PL vs. PNL (Fig. 1D). Thus, IL-13Rα2 mRNA levels are not changed in skin lesions of human psoriasis compared to controls. Moreover, when AL is compared directly to PL, both IL-13 mRNA and IL-13Rα2 mRNA levels were significantly upregulated (Fig. 1E). The close association of IL-13Rα2 expression with disease activity of AD, rather than psoriasis, may represent an important disease-specific role of IL-13 Rα2 in AD, but not in psoriasis, for inflammation and itch.
Further, we analysed IL-13 and IL-13Rα1/2 expression in skin from a MC903-induced murine AD model (Larkin et al., 2021). RNA-seq revealed that the transcription levels of IL-13 and IL-13Rα2 (but not IL-13Rα1) were increased in ears of MC903-treated, when compared to vehicle-treated mice (Fig. 1F), based on log2FCH ≥ 1.0 and FDR < 0.05. Thus, similar to human AD, IL-13Rα2 represents an AD diseaseregulated gene in mice.
Immunohistochemically, the protein levels of IL-13Rα1 and IL-13Rα2 within the epidermis in MC903-treated ears are increased when compared to controls (Fig. 1G, H). MC903 induced ear skin model has been deemed to display key features associated with AD, including inflammation, skin thickening, transepidermal water loss, hypervascularization, spongiosis, strong immune cell infiltration, epidermal hyperplasia and intensive pruritic behaviours (Li et al., 2006;Oetjen et al., 2017). But the MC903 mouse model still scratches normally in the IL31Rα1 knockout mice, suggesting the difference existing between this model and human AD (Oetjen et al., 2017). Nevertheless, our quantitative fluorescence intensity measurements confirmed increased epidermal fluorescence staining of IL-13Rα2 over IL-13Rα1 in MC903vs. vehicle-treated ear skin (Fig. 1I). Thus, IL-13Rα2 protein levels are also increased in mouse AD skin.

IL-13 Upregulates IL-13Rα2 transcription levels in Keratinocytes, and contributes to modulation of neuro-immune circuits
To address the underlying mechanism for the disease-driven (caption on next page) upregulation of IL-13Rα2 in the skin including epidermis, we investigated its location in cultured human primary keratinocytes (phKCs). Robust expression levels of IL-13Rα1 and IL-13Rα2 in phKCs was confirmed, showing IL-13Rα1 resided closer to the cell surface than IL-13Rα2 ( Fig. 2A). Live cell calcium imaging revealed that a fraction (22.8%) of phKCs responded to 100 ng/ml IL-13 (Fig. 2B). RNA-seq revealed that transcription of IL-13Rα2, but not IL-13Rα1, was upregulated when treated with IL-13 (Fig. 2C), which is in consistent with previous report (Furue et al., 2020b). To gain insight into the influence of IL-13Rα1 and IL-13Rα2 on each other, knockdown (KD) using specific shRNA against each receptor was performed, and compared to scrambled shRNA-treated control cells (Fig. 2D, E). Western blot analysis confirmed that KD of IL-13Rα2 expression to ~ 56% was achieved in comparison to scrambled shRNA-treated cells (Fig. 2D). Subsequently, in a separate experiment, the IL-13Rα2KD phKCs were treated with or without IL-13 before RNA-seq. Sequencing results revealed that in IL-13Rα2KD cells IL-13Rα1 expression was not altered regardless of IL-13 treatment (Fig. 2F), suggesting IL-13Rα2 has no impact on IL-13Rα1 expression. In a similar experiment, knockdown of IL-13Rα1 by its specific shRNA achieved ~ 70% (Fig. 2E). Interestingly, IL-13Rα1KD significantly reduced IL-13Rα2 transcription when treated with IL-13 (Fig. 2G). This effect was not observed without IL-13 treatment (Fig. 2G). This result suggests that IL-13Rα1 influences IL-13Rα2 transcription in the presence of IL-13.
To study the influence of KD of these receptors on the IL-13-induced alterations of genes in phKCs, we then analysed RNA-seq on each KD cells in comparison to the scrambled shRNA-treated controls.
Taken together, IL-13-driven upregulation of IL-13Rα2 in keratinocytes may explain the observed enhanced level of epidermal IL-13Rα2 in human AD skin. IL-13Rα2 independently participated in regulation of certain genes (i.e. EDNRA, CCL20, CCL26, CXCL6, SERPIN family, as well as TNF family members).
AD skin is frequently associated with skin barrier dysfunction that involves downregulation of skin barrier-related proteins filaggrin (FLG), involucrin (IVL), FLG2 and loricrin (LOR). This downregulation is induced by IL-13, IL-4, IL-22 and IL-17A (Furue, 2020;Kim et al., 2008). To understand the regulatory impact of IL-13 to these genes after receptor knockdown, we then analysed FLG, FLG2, LOR, and IVL gene transcription levels in phKCs. Our results showed that in the scrambled control-treated phKCs FLG and IVL were significantly upregulated after 6 h treatment with IL-13. Downregulation of FLG2 and LOR were also did not reach significance (DEG threshold: 2-fold up-or down-regulation with FDR < 0.05) (Fig. 3C). Interestingly, IL-13Rα2KD increased the transcription levels of FLG and FLG2 in the presence of IL-13. However, IL-13Rα1KD did not upregulate FLG, FLG2, IVL and LOR transcriptions. This finding may suggest IL-13Rα2 is important for regulating skin barrier dysfunction.

Innate immune response upregulates IL-13Rα2 transcription in sensory Neurons, but not in keratinocytes
IL-13 activates intra-epidermal nerve terminals of sensory neurons to propagate itch, and these terminals are abundantly present in skin (Feld et al., 2016). Thus, it is necessary to assay sensory neuronal IL-13Rα2 transcription. TLR innate immune response has been associated with the pathogenesis of AD (Jin et al., 2009;Niebuhr et al., 2009). Therefore, attention was then paid to dysfunctional TLR in the regulation of IL-13 receptors.
Indeed, our data confirmed an increased TLR2 transcription in ears from MC903-vs. vehicle-treated mice by RNA-seq (Fig. 4A). In skin from IL-31-overexpressing AD model mice (Meng et al., 2018), increased TLR2 transcription were also detected when compared with their age matched wild-type mice (Fig. 4B). We thus used TLR2 heterodimer agonists Pam3CSK4 (1 µg/ml) or FSL-1 (1 µg/ml) to treat dorsal root ganglionic neurons (mDRGs) and phKCs for 6 h before RNA-seq. Interestingly, this revealed an upregulated transcription of IL-13Rα2 in mDRGs, but not IL-13Rα1 ( Fig. 4C and D). Moreover, these agonists did not significantly influence IL-13Rα1 and Rα2 in phKCs ( Fig. 4E and F). The different results between these cell types suggest a tissue-specific regulation of IL-13Rα2. Our findings highlight disease driven-TLR2 activation in AD may promote neuronal IL-13Rα2 transcription.

Pam3CSK4 enhances IL-13-induced calcium transient in sensory neurons and elevates IL-13-induced Itch-like behaviours in mice
As a cutaneous regulator, IL-13 activates sensory neurons and participates in the initiation of AD and itch (Erickson et al., 2021). Hence, we investigated whether TLR2 facilitates IL-13 signalling in sensory neurons. Cultured mDRGs were then pre-incubated with Pam3CSK4 (1 µg/ml) or vehicle for 6 h before washing, and application of IL-13 (100 ng/ml) for measurement of calcium imaging. In comparison to IL-13 alone that elicited calcium spikes in a small fraction (3%) of neurons (Fig. 4G), Pam3CSK4 increased IL-13-elicited calcium influx in 5.67% neurons (Fig. 4H). The increment in the number of IL-13 responders are significant. (For G and H, the chi-square statistic with Yates correction is 5.1983. The p-value is 0.022609). The changes of the area under the curve (AUC) also confirmed the enhanced intracellular calcium by Pam3CSK4 almost doubled that of IL-13 alone (Fig. 4I).
To assay the biological consequence of TLR2-enhanced IL-13 signaling in itch, we used mouse cheek model. Pam3CSK4 (20 μg/100 μl) or its vehicle (sterilized water) was orally administered to mice three times by gavage for over 7 days. Next day after the final gavage, IL-13 (1 μg) or vehicle was intradermally injected into one cheek. IL-13 alone did not induce a significant increase in scratching bouts compared to a vehicle injection, and this is consistent with previous findings (Oetjen et al., 2017). In mice pre-administered with Pam3CSK4, IL-13 injection caused about a doubling of scratching compared to vehicle injection (Fig. 4J).
Thus, these results strengthen the notion that TLR2 activation promoted the itch-like response of mice to IL-13, an outcome likely to be attributed to enhanced IL-13Rα2 activity in sensory neurons.

TLR2-regulated IL-13 signalling orchestrates specific cytokine and chemokine release from sensory neurons through IL-13Rα2-dependent pathway
To investigate the mechanism underlying the Pam3CSK4 caused itch sensitization to IL-13, the functional contribution of IL-13Rα2 to TLR2 regulated IL-13 response in sensory neurons was evaluated. KD of each IL-13Rα was carried out in cultured mDRGs. Western blot analysis confirmed that expression levels of IL-13Rα1 and IL-13Rα2 were reduced by their respective shRNA compared to the scrambled shRNA treatment (Fig. 5A, B). Cytokine array demonstrated that in mDRGs treated with scrambled shRNA, Pam3CSK4 pre-treatment (for 6 h, followed by washing) enhanced IL-13-mediated release of cytokines including chemerin, TNFα, CCL17 and CCL20 (Fig. 5C-F). These cytokines are known to be related to neurogenic inflammation and AD (Deftu et al., 2018;Stojek, 2016). Importantly, Pam3CSK4-enhanced IL-13elicited release of cytokines chemerin (Fig. 5C), CCL17 (Fig. 5E) and CCL20 (Fig. 5F) was significantly reduced in the IL-13Rα2KD mDRGs compared to the scrambled shRNA-treated cells. In contrast, TNFα was not affected by IL-13Rα2KD (Fig. 5D). Thus, these results highlight that TLR2 modulates IL-13Rα2 downstream mediators involved in human dermatitis and neurogenic inflammation.
This finding provides a previously unknown immune-modulatory mechanism of pruritic signaling in AD by revealing an innate TLR2 immune-based exaggeration of IL-13-mediated itch. IL-13-induced IL-13Rα2 signaling in keratinocytes and TLR2 upregulated peripheral IL-13Rα2 signalling can drive itching sensation and facilitate neurogenic inflammation (Fig. 6). Taken together, peripheral TLR2-IL13Rα2 might be a new therapeutic target to control AD and chronic itch.

Discussion
Our study highlights the clinical relevance of the IL13Rα2 signaling pathway in controlling itch responses, inflammatory cascades, and neurogenic inflammation in AD. Cutaneous IL-13Rα2 upregulation is observed in lesional AD, but not psoriasis, and is functionally confirmed in AD murine models. In AD, this upregulation occurs in a disease activity-dependent manner.
Previous findings suggested IL-13, IL-4, and TNF-α enhance expression of IL-13Rα2 in human keratinocytes (Furue et al., 2020a;Sivaprasad et al., 2010). These studies indicate that IL-13Rα2 acts as the decoy receptor to antagonize and inhibit IL-13Rα1 function. Our findings confirmed that IL-13Rα1KD reduced IL-13-induced IL-13Rα2 levels in keratinocytes, a feature not observed under normal condition. Thus, we add on new information that IL-13Rα2 expression is dependent on IL-13Rα1 (Bitton et al., 2020), but only if IL-13 is present.
The observed influence of IL-13Rα1KD on IL-13-induced IL-13Rα2 transcription somehow masked dissecting IL-13Rα2 function. To overcome this, we reduced the IL-13Rα2 expression in keratinocytes using shRNA thereby down-regulating IL-13-induced transcriptomic changes of certain itch receptors and inflammatory mediators (c.f. Fig. 2H-J). These include EDNRA, CCL20, CCL26, CXCL6 and SERPINs. Thus, we conclude that keratinocyte-derived IL-13Rα2 may mediate IL-13induced signaling independently from IL-13Rα1. Our study revealed a direct disease-driven modulation and engagement of IL-13Rα2 in AD condition. Our findings also argue against IL-13Rα2 being solely a decoy receptor in keratinocytes because IL13Rα2 actively participates in signal transduction pathways. In fact, in other biological process such as the malignant melanoma and glioblastoma mutiforme, IL-13Rα2 is implicated in inducing angiogenesis and promoting signal transduction (Fichtner-Feigl et al., 2007;Fichtner-Feigl et al., 2006;Okamoto et al., 2019).
Apart from epidermal keratinocytes, IL-13 directly activates sensory neurons to promote itch sensation enriched in AD skin. However, IL13Rα2 function in sensory nerves has been ignored long-time. Here, for the first time we demonstrate that IL13Rα2 is upregulated in TLR2 agonist-treated peripheral sensory neurons. Disturbances of innate and adaptive immune responses were implicated in pathogenesis of AD (Boguniewicz and Leung, 2011;Weidinger et al., 2018). Innate TLR2 signaling not only promotes itch and pain but convert transient TH2 cellmediated dermatitis into persistent inflammation which is linked to chronic human AD Liu et al., 2012;Niebuhr et al., 2009;Wang et al., 2020). Thus, our findings herein provide an  but not in phKCs;TLR2 agonist, scratching behaviour in mice. A. RNA-seq based heatmap of TLR transcription in ear skin of MC903-vs. vehicle-treated mice. B. Real-time reverse transcription PCR for TLR2 in lesional and non-lesional skin of IL-31-overexpressing (OE) mouse vs. age matched wild-type mice. RNA-seq of IL-13Rα1/2 in Pam3CSK4-treated or FSL-1-treated vs. vehicle-treated mDRGs (C, D) and phKCs (E, F). Representative traces for calcium spikes elicited by IL-13 in mDRGs without (G) or with (H) Pam3CSK4 pre-treatment for 6 h; Pie chart showing the fraction of IL-13 responders. (I) Area under curve of IL-13 elicited calcium spikes in mDRGs pretreated with or without Pam3CSK4. (J) Orally administrated Pam3CSK4 or (K) intraperitoneal injection of FSL-1 enhanced IL-13-induced scratch bouts in a mouse cheek model (2-way ANOVA followed by Bonferroni's post hoc analysis). Data are presented as mean ± SEM; n ≥ 3. For B and I, Student 2-tailed t-test: *P < 0.05.
Previous finding has shown that activation of TLR2/1 heterodimers by Pam3CSK4 mediates pain and itch, whereas activation of TLR2/6 heterodimers by lipoteichoic acid (LTA) or zymosan drives itch (Wang et al., 2020). Together, TLR2 is required for both histamine-dependent, histamine acute itch, dry skin itch as well as chronic allergic contact dermatitis (ACD) itch (Wang et al., 2020). However, the underlying mechanism still remains unclear. In our study, upregulation of sensory neuronal IL13Rα2 by TLR2/1 agonist Pam3CSK4 and TLR2/6 by FSL-1 also resulted in an enhanced itch-like behaviour when compared to IL-13-injection alone in mice. KD experiments in sensory neurons confirmed that IL13Rα2 promotes itch and inflammatory cytokine release from keratinocytes. The enhanced inflammatory mediator release mediated by IL-13Rα2 through TLR2 activation can be independent from IL13Rα1. Downstream-released mediators induced by IL-13Ra2 include chemerin, and CCL17, CCL22, known to be implicated in skin lesion or AD (Getschman et al., 2017;Homey et al., 2006;Renert-Yuval and Guttman-Yassky, 2020). Thus, this brand new pathway that employs innate immune TLR2 to potentiate IL-13Rα2 effects in itch underlies a disease-driven innate immune-boosted neuronal IL-13 signalling mechanism in pruritic AD.
Here, we report tissue-specific mechanisms involved in regulation of IL-13Rα2 in keratinocytes and sensory neurons in AD, highlighting differential immune-neuronal modulatory mechanisms of IL-13 signaling in skin inflammation and itch in AD pathogenesis, strengthening the concept that the innate and adaptive immune system both represent major itch regulators by modulating itch, scratching, cutaneous sensation, and sensory perception beside their role in inflammation and skin barrier .
Indeed, there is an ongoing debate about the role of IL-13 in inducing itch in mice. One previous report showed IL-13 intradermal injection could not directly induce scratching behaviours in mouse cheek, instead, it potentiates other pruritogens to induce itch (Oetjen et al., 2017). A later report showed IL-13 was able to directly induce cheek scratching at a certain window of concentration (Campion et al., 2019). Moreover, the IL-13 transgenic mice display TRPA1-dependent itch-like behaviour (Oh et al., 2013). Thus, fine-tuned concentration-dependent mechanisms may lead to different effects on receptors and subsequent signalling systems, as known for many receptors (Green et al., 2006;Green and Dong, 2016). In this study, IL-13 alone induced a slight increment in scratching bouts in mice received vehicle for Pam3CSK4, but this effect did not reach statistical significance. However, oral administration of Pam3CSK4 or intraperitoneal injection of FSL-1 significantly induced itch-like behaviours when compared with IL-13 alone. Because these agonists of TLR2 upregulate IL-13Ra2 in sensory neurons and the direct activation by IL-13 is dose-dependent, we conclude that IL-13Ra2 functionally activated by Pam3CSK4 or FSL-1 contributed to this elevated itch response. However, the possibility of TLR2-independent events induced by Pam3CSK or FSL-1 stimulation cannot be completely excluded from regulation of IL-13Ra2 expression in sensory neurons.
Overall, our findings reveal a new mechanism how IL13Rα2 contributes to neuroimmune circuits by releasing mediators from sensory nerves involved in immune regulation and skin barrier function, and from keratinocytes involved in itch and inflammation, thereby pinpointing a distinct regulatory mechanism of IL-13Rα2 in AD. We propose that the TLR2-IL-13Rα2 pathway represents a new target for the treatment of chronic itch and neuroinflammation in AD.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Fig. 6. IL-13 promotes IL-13Rα2 signaling in keratinocytes, whereas TLR2 activation elevates IL-13-mediated itch and skin sensitisation in AD via potentiating IL-13Rα2. Activation of TLR2 enhances IL-13Rα2 mediated chemerin, CCL17, CCL22 release from sensory neurons. IL-13 upregulates IL-13Rα2 transcription in keratinocytes, and promotes the synthesis of itch receptor (i.e. EDNRA), and a number of itch and inflammatory mediators in skin. Breaking the link of TLR2 and IL-13Rα2 will block the excess release of these cytokines leading to attenuation of the hypersensitization and pruritic condition in AD.