Multifaceted MRGPRX2: New Insight into the Role of Mast Cells in Health and Disease

Abstract

Cutaneous mast cells (MCs) express Mas-related G protein-coupled receptor-X2 (MRGPRX2; mouse ortholog MrgprB2), which is activated by an ever-increasing number of cationic ligands. Antimicrobial host defense peptides (HDPs) generated by keratinocytes contribute to host defense likely by two mechanisms; one involving direct killing of microbes and the other via MC activation through MRGPRX2. However, its inappropriate activation may cause pseudoallergy and likely contribute to the pathogenesis of rosacea, atopic dermatitis, allergic contact dermatitis, urticaria and mastocytosis. Gain- and loss-of-function missense single nucleotide polymorphisms in MRGPRX2 have been identified. The ability of certain ligands to serve as balanced or G protein-biased agonists have been defined. Small molecule HDP mimetics have been developed that display both direct antimicrobial activity and activate MCs via MRGPRX2. In addition, antibodies and reagents have been generated that modulate MRGPRX2 expression and signaling. In this article, we provide a comprehensive update on MrgprB2 and MRGPRX2 biology. We propose that harnessing MRGPRX2’s host defense function by small molecule HDP mimetics may provide a novel approach for the treatment of antibiotic-resistant cutaneous infections. By contrast, MRGPRX2-specific antibodies and inhibitors could be utilized for the modulation of allergic and inflammatory diseases that are mediated via this receptor.

INTRODUCTION

Mast cells (MCs) are granulated tissue-resident cells of hematopoietic lineage that are best known for their roles in allergic reactions, but they also play an important role in host defense and wound healing.^1–3 In human, most MCs found in the skin contain tryptase, chymase, carboxypeptidase and cathepsin, and are known as MC_TC. By contrast, majority of MCs that are found in the lung and gut express only tryptase, and are known as MC_T.⁴ In rodents, connective tissue MCs (CTMCs) resemble MC_TC and are found in the nasopharynx, skin, peritoneal cavity, and muscularis propria, whereas mucosal MCs (MMCs) resemble MC_T and populate the mucosa of the gastrointestinal tract. While both subtypes express a high affinity IgE receptor (FcεRI), MC_TC and CTMCs undergo degranulation via an IgE/FcεRI-independent pathway by a diverse group of basic molecules including neuropeptides, antimicrobial host defense peptides (HDPs), FDA-approved drugs and the cationic polymer compound 48/80 (C48/80). Until recently, these secretagogues were thought to activate CTMCs and MC_TC via direct activation of G proteins through a receptor-independent mechanism.^{5, 6}

MRGPRX2 is an acronym for Mas-related G protein-coupled receptor X2, which is one of nine members of the human MRGPR family.⁷ In 2006, Tatemoto et al.,⁸ provided the first demonstration that MRGPRX2 is expressed in MCs. Using transfected cell lines, it was demonstrated that basic secretagogues that induce degranulation in MC_TC selectively activate this receptor. Recent transcriptome analysis revealed that while most MRGPR family members are expressed in peripheral neurons, MRGPRX2 is found predominantly in skin MCs.^9–11 Human basophils and eosinophils have recently been shown to express MRGPRX2.^12–14 However, functional consequence of this expression is not clear as basophils from healthy individuals do not respond to opiates, vancomycin and fluoroquinolones, which activate MCs via MRGPRX2.^15–19

A major complication in studying MRGPRX2 function in vivo is that the gene cluster containing the nine human MRGPRX members is dramatically expanded in mice, consisting of 22 potential Mrg coding genes. In 2015, McNeil et al.,²⁰ made the seminal observation that murine CTMCs express the transcript of a single family member, MrgprB2, and demonstrated that it is the mouse ortholog of human MRGPRX2.²⁰ It is noteworthy that there is only ~53% overall sequence similarity between MrgprB2 and MRGPRX2.²¹ This difference is reflected in the concentrations of ligands that are required to activate these receptors.^{20, 22} Despite these differences, MrgprB2 promotes host defense against bacterial infection, but also contributes to neurogenic inflammation, pain, atopic dermatitis (AD), allergic contact dermatitis (ACD), non-histaminergic itch and pseudoallergy in mice.^{20, 23–27}

Most of the studies performed in the last 5 years with MrgprB2 and MRGPRX2 focused on cutaneous health and disease. In this article, we first review the potential roles of MrgprB2 and MRGPRX2 on cutaneous host defense and then describe how their inappropriate activation may contribute to the pathogenesis of rosacea, AD, ACD, non-histaminergic itch, pseudoallergy and mastocytosis. Clinical implications of naturally occurring missense MRGPRX2 mutations in health and disease are also discussed. Furthermore, we propose that harnessing MRGPRX2’s host defense properties could provide a novel therapeutic approach for the treatment of antibiotic-resistant cutaneous infections. By contrast, inhibiting its inappropriate activation may lead to the development of new treatment options for a number of allergic and inflammatory disorders.

MrgprB2 AND MRGPRX2 IN MC-MEDIATED HOST DEFENSE AGAINST BACTERIAL INFECTION

Cells related to MCs (test cells) have been recognized in the invertebrate urochordate, which appeared approximately 500 million years ago.²⁸ These cells contain heparin and histamine and undergo degranulation in response to C48/80, a potent basic secretagogue for mouse CTMCs and human MC_TC via MrgprB2 and MRGPRX2, respectively.^{29, 30} It has been proposed that test cells represent ancient effector cells of the innate immunity in primitive chordates and raises the interesting possibility that MrgprB2 and MRGPRX2 contribute to host defense in mice and humans, respectively. Interestingly, MRGPRX2 gene has undergone positive selection in human evolution supporting its possible beneficial role.³¹

Quorum-sensing molecules (QSMs) including competence-stimulating peptides (CSPs) are secreted by bacteria to signal population density. Pundir et al.,²⁵ made the remarkable observation that CSP-1 causes degranulation, TNF-α secretion, reactive oxygen species (ROS) and prostaglandin D₂ (PGD₂) generation in mouse PMCs via MrgprB2, resulting in the inhibition of bacterial growth and prevention of biofilm formation. Nasopharynx, peritoneum and skin are enriched with MrgprB2-expressing CTMCs and are the major portal of entry for bacteria.²⁵ There is now evidence to show that MrgprB2 confers protective immunity against bacterial infection at all three sites. In a pneumococcal nasopharyngeal colonization model, MrgprB2^MUT mice exhibit impaired bacterial clearance when compared to wild-type (WT) mice. In a peritoneal infection model, MrgprB2^MUT mice display higher bacterial load after infection with vancomycin-resistant Enterococcus faecium (E. faecium) than WT mice. Similarly, in a dermal infection model, subcutaneous infection with Streptococcus pyogenes (S. pyogenes) or Pseudomonas aeruginosa (P. aeruginosa) results in the development of footpad edema that is significantly larger and display a prolonged progression in MrgprB2^MUT mice when compared with WT mice. In these infection models, WT mice display increased MC density, higher TNF-α level and elevated neutrophil count when compared to MrgprB2^MUT mice. In addition, pharmacological activation of MrgprB2 with either C48/80 or CSP-1 eliminates bacteria and reduces disease score. These findings suggest that activation of CTMCs via MrgprB2 provides an important mechanism for host defense by sensing bacteria-associated QSMs and secreting antibacterial mediators.²⁵

In addition to mouse PMCs, CSP-1 also causes degranulation, ROS, PGD₂ and TNF-α production in a human MC line Laboratory of allergic diseases 2 (LAD2) that endogenously expresses MRGPRX2, and these responses are substantially reduced in MRGPRX2-silenced cells.²⁵ Co-culture of LAD2 cells with clinical isolates of Streptococcus pneumoniae (S. pneumoniae), which produce CSP-1, results in the suppression of bacterial growth but this suppression is attenuated in MRGPRX2-silenced cells. It is well known that biofilms allow bacteria to subvert immune responses and establish chronic infections.³² Interestingly, LAD2 cells reduce biofilm viability, and this response is decreased in MRGPRX2-silenced cells.²⁵ These findings suggest that activation of MRGPRX2 by QSMs plays an important role for antibacterial defense in humans through MC degranulation, ROS production and subsequent recruitment of neutrophils (Fig. 1A).

FIG 1.

Contribution of MRGPRX2 in host defense to bacterial infection and cutaneous wound healing. (A) QSMs released from bacteria activate MCs via MRGPRX2, resulting in degranulation, reactive oxygen species (ROS) generation and recruitment of neutrophils which collectively participate in bacterial killing (red arrows). In addition, skin infection with pathogen such as P. aeruginosa induces human β defensin (hBD) from keratinocytes, which kill bacteria directly and activate MCs via MRGPRX2. MC mediators (histamine, PGD₂) cause additional hBD secretion from keratinocytes, which causes further MC activation and recruitment of neutrophils, resulting in resolution of infection. (B) Based on studies with mice, we propose that topical application of mastoparan activates cutaneous MCs via MRGPRX2 to control S. aureus skin infection. Activated MCs release chemokines and recruit CD301b⁺ dermal dendritic cells to mediate regenerative healing. Mastoparan treatment also elicits S. aureus–specific IgG to induce antibacterial adaptive immunity to control reinfection.

The emergence of antibiotic-resistant infections poses a tremendous public health concern globally and requires urgent need to develop novel therapy for their treatment.^{33, 34} P. aeruginosa, a Gram-negative bacterium, is a common cause of nosocomial skin wound infection and presents a rising therapeutic challenge due to its ability to form biofilms and increase antibiotic resistance.^{35, 36} Zimmerman et al.,³ recently showed that in a topical P. aeruginosa infection model in mice, MCs contribute to both bacterial clearance and wound healing. The authors found that MCs infected with P. aeruginosa in vitro are unable to decrease bacterial load unless they are co-cultured with keratinocytes. Interestingly, P. aeruginosa infection of MC/keratinocyte co-culture results in the secretion of MC-derived mediators, which cause the release of mouse β-defensin (mBD)-14 (Defb14, ortholog of human β-defensin (hBD)-3) from keratinocytes.³ Based on these findings, it has been proposed that in P. aeruginosa skin infection, MC-derived mediators promote the induction and secretion of HDPs, which control infection through their direct antimicrobial activities.³

Zhang et al.,³⁷ showed that Defb3 and Defb14 activate MrgprB2 to induce degranulation in mouse PMCs. Furthermore, activation of MRGPRX2 in human MCs results in histamine and PGD₂ release, which further promotes hBD production from keratinocytes.^38–42 Moreover, infection of human keratinocytes with P. aeruginosa results in the induction of hBD2.⁴³ It now appears that most of the HDPs that display direct antimicrobial activity also activate MCs via MRGPRX2.^{41, 42, 44} Thus, it is likely that during microbial infection, both QSMs produced from bacteria and HDPs generated from infected keratinocytes activate MRGPRX2 to generate mediators that promote the recruitment of neutrophils, resulting in the clearance of microbial infection (Fig. 1A).

The amphipathic peptide mastoparan found in Hymenoptera venom activates murine and human MCs via MrgprB2 and MRGPRX2, respectively, and displays direct antimicrobial activity.^{20, 45} In a mouse model of Staphylococcus aureus (S. aureus)-induced skin infection, topical application of mastoparan promotes neutrophil recruitment and clears infection,⁴⁵ thus functioning similar to endogenously derived HDPs such as hBDs. In addition to degranulation, activation of LAD2 cells by mastoparan results in the generation of chemokines; CCL2, CCL3 and CCL4, which contribute to the mobilization of CD301b⁺ dermal dendritic cells (DDCs) and antigen presenting dendritic cells (DCs). DDCs play critical role in promoting re-epithelialization of sterile wounds.⁴⁶ Additionally, topical application of mastoparan to S. aureus-infected wound results in significantly higher level of S. aureus–specific total IgG when compared to vehicle treated control. This increased IgG is associated with reduced lesion size following reinfection in a mouse model of S. aureus-induced skin infection.⁴⁵ Thus, it is likely that activation of MRGPRX2 by mastoparan not only accelerates bacterial clearance and promotes regenerative healing, but also enhances adaptive immunity, which provides protection against reinfection (Fig. 1B).

ROSACEA

MCs are found beneath the epithelia in close proximity to nerve endings and blood vessels, and likely serve as a functional homeostatic regulatory unit.⁴⁷ Rosacea is a chronic inflammatory skin condition that is provoked by several triggers such as an imbalance in commensal skin microbiota, ultraviolet (UV) radiation and is exacerbated by psychosocial stress.⁴⁸ Although the pathogenesis of rosacea is not fully determined, emerging evidence suggests that it results from dysregulated host defense, excessive production of the HDP LL-37, aberrant activation of cutaneous MCs and abnormal neurovascular signaling (Fig. 2).^{49, 50} Patients with rosacea display increased baseline expression of LL-37, which likely results from the activation of Toll-like receptors (TLRs) on keratinocytes following microbial infection.⁵¹ The number and activity of MCs are increased in the skin of rosacea patients when compared to control subjects.^{52, 53} Although no animal models are available that fully reproduce rosacea in humans, “rosacea-like” features are observed following intradermal administration of LL-37 in WT mice but not in MC-deficient W^sh/W^sh mice.⁵³ Neutrophils that are recruited following MC activation also serve as an important source of LL-37.⁵⁴ Thus, the cross-talk between activated MCs, keratinocytes and neutrophils, that amplify the generation of LL-37, provides a vicious cycle for the continuous activation of MCs to sustain chronic cutaneous inflammation in rosacea (Fig. 2).

FIG 2.

Upregulation of LL-37, induction of SP and aberrant activation of cutaneous MCs via MRGPRX2 contributes to rosacea. Microbial assault induces excessive release of LL-37 from keratinocytes which activates MCs via MRGPRX2. Activated MCs release mediators to promote neutrophil recruitment, which generate additional LL-37 to cause chronic inflammation and rosacea pathogenesis via further activation of MCs via MRGPRX2. Moreover, stress induces the release of SP from the sensory neurons, which also activates MCs by MRGPRX2, resulting in the release of chemokines and cytokines. Recruitment of inflammatory cells leads to neurogenic inflammation and exacerbation of rosacea.

Neurogenic inflammation is a physiological process in the skin by which mediators are released directly from the cutaneous nerves to initiate an inflammatory reaction, resulting in erythema, swelling, and pain. Classic experiments performed in the late 19^th and early 20^th century showed that electrically stimulated dorsal roots induce vasodilation, resulting in neurogenic inflammation. It was thought that this response resulted from the release of the neuropeptides (NPs) SP and calcitonin gene-related peptide (CGRP) from nociceptors, which directly act on vascular endothelial cells to cause vasodilation and increased capillary permeability.⁵⁵ As discussed below, it is now realized that NPs released from sensory neurons not only act on the vasculature but also activate cutaneous MCs via MRGPRX2 to induce degranulation and chemokine/cytokine generation, which may contribute to rosacea exacerbation and the pathogenesis of AD.⁵⁶

The expression of NPs; SP, pituitary adenylate cyclase-activating polypeptide (PACAP), vasoactive intestinal peptide (VIP), are increased in rosacea, which activate MCs via MRGPRX2.^{8, 49, 57, 58} In turn, histamine, tryptase and other MC-derived mediators can lead to the release of NPs from sensory nerve endings, which creates a bidirectional loop between the MCs and sensory nerves. Moreover, SP causes degranulation in human skin MCs via MRGPRX2⁵⁷ and induces chemokines, IL-8, CCL2, CCL3 and CCL4 in LAD2 cells.²⁷ Thus, it is likely that activation of MCs by LL-37 through MRGPRX2 initiates the pathogenesis of rosacea. Furthermore, release of chemokines by NPs via MRGPRX2 promotes neurogenic inflammation and rosacea exacerbation leading to itching, flushing, erythema, and/or burning sensations (Fig. 2).^{48, 59}

ATOPIC DERMATITIS

Atopic dermatitis (AD) or atopic eczema is a chronic inflammatory skin disease clinically characterized as intense pruritus and T-helper-2 (T_H2)-associated hypersensitivity to wide spectrum of allergens including those derived from house dust mites (HDMs).⁶⁰ The skin of AD patients display altered barrier function and are colonized with exotoxin-producing bacteria such as staphylococcal enterotoxin B (SEB)-secreting S. aureus.⁶¹ Additionally, loss-of-function mutation in filaggrin, a component of skin barrier function, has been shown to be risk factor for AD.⁶² Cutaneous nerve fibres of patients with AD display greater expression of SP and protease activated receptor 2 (PAR2) when compared to healthy skin.^{63, 64} Furthermore, AD skin shows increased MC numbers that are found within or in close proximity to epidermis and interact with SP containing nerve bundles displaying signs of degranulation.^65–67 Nattkemper et al.,⁶⁴ used RNA sequencing to analyze the complete transcriptome of skin samples of AD patients and healthy controls. They found that genes encoding SP (Tac1) and its receptor MRGPRX2 are upregulated in AD when compared to healthy controls. Thymic stromal lymphopoietin (TSLP) is a cytokine that is released by epithelial cells and its aberrant activation is a hallmark of atopic diseases, including AD.⁶⁸ Babina et al.,⁶⁹ recently showed that TSLP serves as a novel priming factor for SP-induced degranulation in human skin MCs. These findings suggest that MC-nociceptor interaction contributes to the pathogenesis of AD.

The exact mechanism via which MC-nociceptor interaction contributes to the pathogenesis of AD is not known. However, using a well-established clinically relevant mouse model of AD, Serhan et al.,²⁴ showed that repeated epicutaneous exposure to HDM strain Dermatophagoides farinae (D. farinae) and SEB results in the development of moderate to severe AD like skin lesions characterized by a dysregulation of skin barrier architecture and an exacerbated pathogenic type 2 immune response. Although a variety of immune cells can generate SP,^70–72 its expression in mouse skin is restricted to a subpopulation of transient receptor potential vanilloid type 1 (TRPV1⁺) sensory neurons.²⁴ Furthermore, mice deficient in Tac1 or WT mice treated with resiniferatoxin (RTX), which selectively ablates TRPV1⁺ nociceptors, are protected from the development of AD- like pathology. Application of D. farinae + SEB causes the release of SP from cultured dorsal root ganglia (DRG) neurons but not from DRG neurons treated with RTX. In addition, cysteine but not serine protease activity of D. farinae is required for inducing Ca²⁺ mobilization in TRPV1⁺ DRG neurons ex vivo. Furthermore, unlike the response in WT mice, D. farinae + SEB does not induce AD-like features in MC-deficient or MrgprB2^MUT mice.²⁴ Interestingly, most of activated sensory neurons in the skin are found either in direct contact or in the close proximity with dermal MCs. These findings suggest that cysteine-protease-dependent release of SP from TRPV1⁺ nociceptors and the subsequent activation of MCs via MrgprB2 contribute to the early stages in the development of AD in mice.

Given that the expression of Tac1 and MRGPRX2 genes are upregulated in the skin of AD patients when compared to control skin,⁶⁴ it is likely that studies reported in mice have translational relevance to the human disease (Fig. 3A). More than 90% of AD patients are colonized with S. aureus in the lesional skin, whereas most healthy individuals do not harbor this pathogen. Furthermore, S. aureus δ-toxin induces degranulation in human MCs via MRGPRX2.⁷³ S. aureus also induces the expression and release of hBD2 from human keratinocytes.⁷⁴ Harder et al.,⁷⁵ showed that expression and secretion of these HDPs (hBD2, hBD3) are increased in the skin of AD patients when compared to healthy skin. Furthermore, MCs are abundantly localized close to and within the epidermis in AD.⁷⁶ These HDPs cause MC chemotaxis, degranulation, PGD₂ and IL-31 production via the activation of MRGPRX2.^{41, 42, 77, 78} These findings suggest that in addition to SP and S. aureus δ-toxin, increased expression and secretion of HDPs can contribute to the pathogenesis of AD.⁷⁹ Thus, targeting MRGPRX2 and its signaling may provide an attractive therapeutic approach for the treatment of AD (Fig. 3A).

FIG 3.

MRGPRX2 contributes to AD, ACD and non-histaminergic itch. (A) Dermatophagoides farinae (D. farinae) + staphylococcal enterotoxin B (SEB) activate TRPV1⁺Tac1⁺ sensory neurons to release SP, which activates cutaneous MCs via MRGPRX2. S. aureus secreted δ-toxin and hBDs released from keratinocytes activate human MCs via MRGPRX2. Activated MCs release mediators and cytokines resulting in itch sensations associated with AD. (B) Based on studies performed with mice, we propose that haptens such as SADBE, DNCB and oxazolone cause the release of PAMP12 from keratinocytes, which activate cutaneous MCs via MRGPRX2 to release tryptase and 5HT. Tryptase cleaves proteinase-activated receptors (PARs) present on MrgprD and 5-HT-1F expressing itch sensory neurons to induce itch sensation.

ALLERGIC CONTACT DERMATITIS AND NON-HISTAMINERGIC ITCH

Allergic contact dermatitis (ACD) is an intensely pruritic skin disease; caused by repeated exposure to haptens that penetrate the skin barrier to induce a type IV hypersensitivity reaction and does not respond to antihistamine treatment.^{80, 81} Itch in ACD is thought to be mediated by nociceptors that have their cellular bodies in DRG and projections in the skin where they are found in close proximity to MCs. Compared to a healthy control skin, lesional skin from ACD patients display increased number of MCs as well as the MRGPRX2 ligand pro-adrenomedullin peptide 12 (PAMP12).²⁶ However, unlike IgE-mediated itch, the response in ACD is not modulated by anti-histamines.^{80, 81} This could reflect differences in the activation programs of human MCs via MRGPRX2 and FcεRI that diverge temporally and spatially for granule-derived mediator release.⁸² It is noteworthy that subcutaneous injection of PAMP12 or application of the haptens, squaric acetyl dibutyl acid (SADBE), oxalozone, or dinitrochlorobenzene (DNCB) induces itch behaviour in WT mice but this response is significantly reduced in MrgprB2^MUT mice.²⁶ While both histamine (H₁ and H₄) and serotonin (5HT₂ and 5HT₇) receptor antagonists inhibit IgE-mediated histaminergic itch, they have no significant effect on PAMP12-induced response. Moreover, both scratching bouts and immune cells recruitment, two pathological traits classically associated with ACD, are significantly reduced in MrgprB2^MUT mice. It is noteworthy that activation of human MCs via MRGPRX2 and murine MCs via MrgprB2 elicit similar spatial and temporal MC degranulation pattern indicating that in vitro studies with mouse PMCs and in vivo studies with MrgprB2^MUT mice could provide insight on the mechanism of non-histaminergic itch in ACD.²⁶

At concentrations of antigen and PAMP12 that similarly activate mouse PMCs, as measured by β-hexosaminidase release, there is a remarkable difference in the release of pruritogens. Unlike IgE, PAMP12 is a weak stimulant for histamine and serotonin release but strongly induces the release of tryptase β2.²⁶ It now appears that this differential release of MC mediators results in the excitation of different sensory neuron subpopulations. Thus, majority of the neurons that respond to antigen/IgE also respond to histamine. By contrast, neurons that respond to PAMP12 also respond to the pruritogenic ligands chloroquine (MrgprA3 agonist), serotonin (5-HT-1F receptor agonist) and β-alanine (MrgprD agonist). It is therefore possible that PAMP12-mediated activation of MCs via MrgprB2 results in the release of tryptase, which activates PARs on sensory neurons to induce itch. Xing et al.,⁸³ recently showed that histamine H₁ and H₂ receptors are enriched in MrgprA3⁺ neurons but the possibility that these histamine receptors contribute to MrgprB2-mediated itch has not been determined. The demonstration that ACD skin displays increased numbers of MCs as well as the MRGPRX2 ligand PAMP12 suggests that mouse models for ACD likely reflect the situation in humans and provides an explanation why anti-histamines are ineffective for its treatment (Fig. 3B).²⁶ However, one report showed that intradermal injection of PAMP12 in human subjects provokes transient itch and erythema and that these responses are blocked by an H₁ receptor antagonist.⁸⁴ Future studies with larger human subject populations is needed to resolve this discrepancy.

Compared to human skin MCs, colon MCs express MRGPRX2 at a low level.⁸⁵ However, a recent report indicated that PAMP12-MRGPRX2 interaction contributes to the pathogenesis of ulcerative colitis (UC). In the colon, MCs are found mostly in the mucosa and connective tissue, generally clustered at epithelial surfaces.⁸⁶ It has been shown that adrenomedullin (ADM, proteolytic precursor of PAMP12) is upregulated in activated fibroblasts and secretory epithelial cells in inflamed UC when compared to uninflamed tissue. Furthermore, exposure of lamina propria cells obtained from inflamed UC biopsies to an MRGPRX2 agonist results in the release of MC protease carboxypeptidase-3 (CPA-3). This response is absent in cells obtained from control biopsies. Based on these findings, it has been proposed that ADM generated from inflamed epithelial cells and activated fibroblasts in UC is proteolyzed by CPA-3 secreted from MCs to generate PAMP12, resulting in continuous MC activation through MRGPRX2. Thus, it is possible that MRGPRX2 contributes to both ACD and UC via the activation MCs through the same ligand, PAMP12.^{26, 86}

PSEUDOALLERGY

Allergic reactions that occur during perioperative procedures and anaesthesia in response to neuromuscular blocking drugs (NMBDs; such as atracurium, mivacurium, tubocurarine, rocuronium cisatracurium), opiates, antibiotics, iodinated contrast media, plasma expanders and dyes may be severe and life-threatening.^87–89 These reactions represent a diagnostic challenge for allergists as many of these drugs are administered simultaneously and reactions are often mediated independently of IgE.^{90, 91} The mechanism of their actions, however, remained unknown until the seminal observation by McNeil et al.,²⁰ that many of these drugs activate human MC_TC and murine CTMCs via MRGPRX2 and MrgprB2, respectively. Moreover, in murine models of anaphylaxis, fluoroquinolones and the NMBD cisatracurium cause increased vascular permeability but this response is substantially attenuated in MC-deficient and MrgprB2^MUT mice.^{20, 92, 93} As these drugs also induce degranulation in human MCs via MRGPRX2, it has been proposed that activation of this receptor is responsible for non-IgE-mediated drug-induced anaphylactoid reactions.^{20, 94, 95}

McNeil et al.,²⁰ showed that rocuronium induces Ca²⁺ mobilization in transfected HEK293 cells expressing MrgprB2 and MRGPRX2 with EC₅₀ values of 22.2 μg/mL and 263 μg/mL, respectively. Consistently, rocuronium at a concentration of 20 μg/mL induces significant degranulation in WT mouse PMCs but this response is abolished in PMCs from MrgprB2⁻/⁻ mice.²² However, the possibility that rocuronium activates human MCs via MRGPRX2 has been a subject of controversy. For example, initial studies with rocuronium at a concentration of up to 2 mg/mL failed to demonstrate degranulation in LAD2 cells and transfected RBL-2H3 cells.^{94, 96, 97} Although rocuronium (1640 μM; ~1 mg/ml) induces transient Ca²⁺ mobilization in human CD34⁺-derived MCs, it does not provide sufficient signal for degranulation.⁹⁸ However, a more recent study with transfected RBL-2H3 cells and LAD2 cells showed that rocuronium causes significant degranulation at 500 μg/mL and reaching a maximal at 2 mg/mL. Interestingly, human skin MCs also respond to rocuronium (2 mg/mL) for robust degranulation, as measured by β-hexosaminidase release and cell surface expression of lysosomal-associated membrane protein 1 (LAMP-1).²² The reason for the discrepancy is not clear but could reflect differences in the batches of rocuronium used or different assays used to measure MC activation.²²

Spoerl et al.,⁹⁹ reported 3 cases of rocuronium-induced hypersensitivity; these patients had no increase in serum IgE and basophil activation test (BAT) was negative. Sugammadex is a γ-cyclodextrin that encapsulates aminosteroid NMBDs such as rocuronium, preventing its interaction with the nicotinic receptor, and thereby reversing neuromuscular blockade.^{99, 100} Sugammadex inhibits both rocuronium-induced MRGPRX2-mediated signaling in MCs in vitro and irritative skin reactions in vivo.^{96, 99} Based on some of these findings, it has been proposed that MRGPRX2 plays an important role in drug-induced pseudoallergy and that this reaction should be re-classified as type A adverse reaction.^{99, 101} Furthermore, Navinés-Ferrer et al.,⁹⁴ showed that sera obtained from patients who had experienced anaphylactoid reactions to NMBDs induce degranulation in LAD2 cells via MRGPRX2.

Suzuki et al., ⁹⁷ recently reported a case of perioperative anaphylactic reaction to rocuronium. In this patient intradermal skin testing with undiluted rocuronium (10 mg/mL) resulted in a positive reaction. However, total IgE and specific IgE to rocuronium were negative. Sequence analysis of genomic DNA of this patient revealed three amino acid mutations (M196I, L226P and L237P) in MRGPRX2.⁹⁷ These mutations are located in MRGPRX2’s 5^th and 6^th transmembrane (TM) domains in close proximity to its ligand binding pocket.^102–104 It was proposed that these mutations enhance the affinity of MRGPRX2 for rocuronium, thus providing genetic evidence for the role of MRGPRX2 on rocuronium-induced hypersensitivity. However, Chompunud Na Ayudhya et al.,²² showed that while rocuronium induces degranulation in RBL-2H3 cells expressing WT MRGPRX2, cells expressing the mutated receptors (M196I, L226P and L237P) either respond normally to rocuronium or display loss of-function phenotype. These findings suggest that, at least in this patient, MRGPRX2 does not contribute to rocuronium hypersensitivity. Furthermore, based on the results of triple testing (skin testing, quantification of specific IgE and BAT) in a clinical study conducted with 140 patients suspected of perioperative hypersensitivity to rocuronium, it was concluded that in most patients, the hypersensitivity reaction mainly resulted from IgE-mediated MC activation.^{15, 105}

It is possible that MRGPRX2 contributes to drug-induced pseudoallergy in patients with other cutaneous disorders in which MRGPRX2 expression is upregulated. Expression of MRGPRX2 is increased in cutaneous MCs of patients with chronic spontaneous urticaria (CSU) when compared to control subjects.⁵⁷ Furthermore, icatibant and atracurium induce significantly greater wheal reaction in CSU patients when compared to control subjects.¹⁹ Thus, relative contribution of FcεRI and MRGPRX2 on pseudoallergy is likely to be complex and additional studies are required before adopting a paradigm shift regarding IgE/FcεRI versus MRGPRX2 for drug-induced hypersensitivity reactions.¹⁰¹

MASTOCYTOSIS

Mastocytosis is characterized by a pathologic increase of MCs in tissues and is often associated with mutations in KIT, the receptor for stem cell factor.¹⁰⁶ Given that skin MCs express MRGPRX2, it is possible that symptoms associated with cutaneous mastocytosis involve the activation of this receptor through HDPs and NPs generated from epithelium and nociceptors. However, this possibility has yet to be tested.

Systemic mastocytosis is associated with proliferation and accumulation of MCs in different tissues with approximately 50% risk of developing anaphylaxis.^{106, 107} Giavina-Bianchi et al.,¹⁰⁸ recently reported a case of a patient with systemic mastocytosis who developed 5 episodes of anaphylaxis in response to insect venom and ciprofloxacin used for a urinary tract infection treatment. Serum specific IgE levels to bee, wasp, fire ant and allergen components were negative.¹⁰⁸ Skin testing to Hymenoptera venom and ciprofloxacin was negative. Given that mastoparan found in Hymenoptera venom and ciprofloxacin activate human MCs via MRGPRX2, it was proposed that activation of this receptor contributed to the patient’s life-threatening reaction. However, absence of positive skin reaction to Hymenoptera venom and ciprofloxacin is surprising.¹⁰⁹ It is noteworthy that this particular patient was diagnosed with systemic mastocytosis with increased numbers of MCs in the duodenal and gastric mucosa and the presence of spindle-shaped MCs in the bone marrow without cutaneous involvement.¹⁰⁸ This may explain why intradermal test concentrations of ciprofloxacin and Hymenoptera extracts used did not induce cutaneous MC activation.^{108, 110}

AGONIST-MRGPRX2 INTERACTION AND LOSS-OF-FUNCTION SINGLE NUCLEOTIDE POLYMORPHISMS (SNPs)

MRGPRX2 is a member of class A GPCRs, which shares a common structure of seven transmembrane (TM) α-helices.¹¹¹ Ligand binding on the extracellular region of GPCRs leads to conformational rearrangements of the intracellular region to initiate downstream signaling cascades.¹¹² In the GPCR database (GPCRdb) nomenclature for class A GPCRs, the first number denotes the helix (1–7) and the second number indicates the residue position relative to the most conserved position, which is assigned the number 50. Thus, 4×55 denotes a residue in TM4, which is at 5 positions after the most conserved residue (4×50) and 4×46 denotes a residue in TM4 at 4 positions before the most conserved residue.

A unique feature of MRGPRX2 is that it is activated by a diverse range of cationic ligands, resulting in a multitude of biological responses. Using molecular modeling and docking studies, Lansu et al.,¹⁰² identified negatively charged Glu164 (4×60) and Asp184 (5×38) in MRGPRX2 making ionic contacts with cationic opioid ligands. Accordingly, mutations E164Q and D184N that retain the steric property of the WT residues but remove the negative charges resulted in loss of receptor activation by dextromethorphan, morphine, and related opioid ligands. Reddy et al.,¹⁰³ also predicted that SP-binding pocket in MRGPRX2 consists of a number of structurally conserved hydrophilic residues along with a buried glutamic acid residue Glu164. Accordingly, replacement of Glu164 with a positively charged Arg (E164R) results in loss of MRGPRX2 activation by SP. However, this mutant responds normally to the HDP, LL-37. Based on these findings, it has been proposed that different ligands interact with different amino acid residues on MRGPRX2’s predicted ligand-binding pocket to induce MC degranulation and biological responses.¹⁰³

Single nucleotide polymorphisms (SNPs) are one of the most common genetic variants found among individuals. Missense SNPs in GPCRs may affect their structure leading to variation in their expression and functional properties. This may contribute to individual differences in responses to receptor activation.^{113, 114} Analysis of GPCRdb reveals the presence of 107 missense SNPs in MRGPRX2 with 72 mutations predicted to be deleterious (Fig. 4A). Alkanfari et al.,¹⁰⁴ found that transfection of cDNA encoding naturally occurring missense variants in MRGPRX2’s potential ligand binding region, G165E (4×61, rs141744602) and D184H (5×38, rs372988289) in a MC line (RBL2H3) results in loss-of-function phenotype for degranulation in response to NPs (SP and hemokinin-1), HDP (hBD3), and a cationic drug (bradykinin B2 receptor antagonist, icatibant). Two additional SNPs found within the receptor’s 6^th and 7^th TM domains, W243R (6×55, rs150365137) and H259Y (7×32, rs140862085) also render the receptor unresponsive to all the ligands tested (Fig. 4B). Although cells expressing W243R and H259Y variants did not survive the stable transfection procedure, they expressed normally in transiently transfected cells. It is therefore possible that replacement of the bulky Trp with a positively charged Arg in variant W243R and His with Tyr in variant H259Y influences both the receptor stability and its ability to interact with diverse ligands.

FIG 4.

MRGPRX2 single nucleotide polymorphisms (SNPs). (A) Snake diagram of secondary structure of MRGPRX2 with each circle representing amino acid residue with one letter code. Solid background in red denotes deleterious and green denotes tolerated SNPs in MRGPRX2. (B) SNPs for ligand binding (purple), G protein coupling (green) and β-arrestin binding (blue) are shown. Conserved sequence on MRGPRX2 for β-arrestin binding (β-arrestin recruitment code) and mutations identified within this code are also shown.

MRGPRX2-G PROTEIN COUPLING AND LOSS-OF-FUNCTION SNPs

Despite diversity of ligands, most class A GPCRs including MRGPRX2 interact with a relatively small family of G proteins in order to transduce signals for cell activation, suggesting the presence of conserved activation patterns among the different receptors. Venkatakrishnan et al.,¹¹⁵ analysed the pattern of contact between structurally equivalent residues from the crystal structures of 27 class A GPCRs. From this analysis, it became clear that in the inactive state of class A GPCRs, the residue at position 6×37 is in contact with a conserved hydrophobic residue at position 3×46. Upon receptor activation, this interaction is rearranged so that residue at 3×46 breaks contact from residue 6×37 and forms a new contact with a tyrosine residue at 7×53 within the highly conserved NPXXY motif of TM7. This rearrangement results in the activation of G proteins.

The residues at positions 3×46, 6×37 and 7×53 in MRGPRX2 are Val, Ile, and Tyr, respectively. These residues are either large hydrophobic or aromatic, which likely facilitate contact formation during the receptor conformational rearrangement.¹¹⁵ Using Ala substitution mutations and transfection studies in RBL-2H3 cells, Chompunud Na Ayudhya et al.,¹¹⁶ showed that V123A (3×46) had no effect of Ca²⁺ mobilization but partially inhibited SP-induced degranulation whereas I225A (6×37) and Y279A (7×53) are essential for both SP-induced Ca²⁺ mobilization and degranulation. These findings suggest that MRGPRX2 utilizes a “barcode” similar to those found in other class A GPCRs for transducing signal from ligand binding to G protein activation.

Chompunud Na Ayudhya et al.,¹¹⁶ then searched for SNPs within or close to these conserved residues in MRGPRX2 and found three missense variants; V123F (3×46), T224A (6×36) and V282M (7×56). Using site directed mutagenesis and transfection approaches, the authors reported that SNPs V123F (3×46) and V282M (7×56) display loss-of-function phenotype for SP-induced Ca²⁺ mobilization and degranulation (Fig. 4B). In addition to the conformational changes in TM helices, recent crystallography and spectroscopy studies have shown that intracellular loops of GPCRs also interact with G proteins.^{117, 118} Not surprisingly, two additional SNPs in MRGPRX2’s second intracellular loop (ICL2), R138C (34×54) and R141C (34×57) were also identified to display loss-of-function phenotype for SP-induced Ca²⁺ mobilization and degranulation. Thus, it appears that combination of residues in MRGPRX2’s TM and ICL2 domains transmit signal following MRGPRX2 activation by SP for Ca²⁺ mobilization and degranulation (Fig. 4B). It is noteworthy that the nature of the G protein to which MRGPRX2 couples to depends on the cell type on which the receptor is expressed.^{20, 116} There is a conserved sequence of NPXXY motif, termed as G protein coupling code present in 7^th TM helix (Fig. 4B). It also remains to be determined whether other MRGPRX2 ligands utilize the same residues on the receptor to couple to the same or different G proteins in transfected HEK293, RBL-2H3 and cells that endogenously expressing the receptor such as LAD2 cells, human skin MCs and human CD34⁺ cell-derived MCs.

MRGPRX2 DESENSITIZATION; GAIN AND LOSS OF-FUNCTION SNPs

For most GPCRs, agonist-induced receptor phosphorylation by G protein receptor kinases (GRKs) and the subsequent recruitment of the adapter proteins β-arrestin leads to their desensitization.¹¹⁹ MRGPRX2 possesses a “β-arrestin binding code” in its carboxyl terminus containing Ser325 (Fig. 4B).¹²⁰ Despite this, MRGPRX2 is resistant to phosphorylation and desensitization in response to LL-37.⁴¹ However, Chompunud Na Ayudhya et al.,¹¹⁶ found the naturally occurring variant (S325L, rs779903448) responds to SP for enhanced Ca²⁺ mobilization and degranulation when compared to the WT receptor. It is therefore possible that differences in the susceptibility of MRGPRX2 to undergo desensitization reflects the nature of the agonist used for receptor activation.^{41, 116}

Chen et al.,⁸⁶ recently found that a missense N62S mutation (rs10833049) in MRGPRX2’s first intracellular domain, which results in the creation of a “β-arrestin binding code”, serves as a protective genetic variant for UC. The authors showed that activation of MRGPRX2 by a synthetic high affinity agonist ZINC3573 and the endogenous ligand PAMP12 results in β-arrestin recruitment, which is associated with receptor desensitization. However, ZINC3573 and PAMP12-induced β-arrestin recruitment and receptor desensitization are significantly enhanced in cells expressing the MRGPRX2 N62S variant when compared to the WT receptor. Given that PAMP12 precursor is upregulated in the colon of patients with UC, it has been proposed that individuals harboring N62S mutation would be protected from developing this disease because of loss-of-function phenotype for MC activation due to enhanced receptor phosphorylation, β-arrestin recruitment and desensitization (Fig. 4B). This finding has important implication for ACD, which may be mediated via PAMP12-MRGPRX2 interaction.²⁶ Thus, individuals harboring this mutation may also be protected from developing ACD. Of note, LL-37, which does not promote MRGPRX2 desensitization, only weakly activates β-arrestin and this response is not regulated by the MRGPRX2 N62S mutation.^{41, 86} These findings raise the interesting possibility that unlike ACD and UC, individuals harboring this mutation may not be protected from developing rosacea, which is thought to be mediated via MRGPRX2 activation by LL-37. Future studies will needed to determine impact of MRGPRX2 polymorphisms on a variety of cutaneous and other MC-mediated disorders.

BALANCED AND G PROTEIN-BIASED SIGNALING BY MRGPRX2: ROLE OF β-ARRESTIN-1

In addition to G proteins, most GPCR agonists activate an additional signaling pathway that involves the recruitment of β-arrestins.¹²¹ This pathway was initially characterized for its role in GPCR desensitization and internalization¹²² but subsequent studies showed that it also serves an important role in G protein-independent downstream signaling for cell migration, growth, and differentiation.^{86, 123} GPCR agonists that preferentially activate either G proteins or β-arrestins are known as G protein-biased and β-arrestin-biased, respectively. However, agonists that activate both pathways are known as balanced agonists. Thus, while an angiogenic HDP AG-30/5C, icatibant, codeine and C48/80 induced Ca²⁺ mobilization and degranulation through MRGPRX2 via G proteindependent pathway, there is a remarkable difference in their ability to cause β-arrestin recruitment.^{124, 125} Thus, while C48/80 and codeine induce β-arrestin recruitment and cause substantial MRGPRX2 internalization, AG-30/5C does not.^{124, 125} MCs express both β-arrestin1 and β-arrestin2.¹²⁶ However, β-arrestin1, but not β-arrestin2 contributes to MRGPRX2 internalization.¹²⁵ Chen et al.,⁸⁶ recently showed that cells expressing the gain of function MRGPRX2 phosphorylation variant N62S responds to PAMP12 for enhanced extracellular signal-regulated kinase (ERK) phosphorylation when compared to the WT receptor. Given that ERK phosphorylation is required MRGPRX2-mediated chemokine generation,¹²⁷ it raises the interesting possibility that agonists that serve as balanced ligands could provide a greater signal for chemokine generation than G protein biased agonists. This possibility remains to be determined.

β-ARRESTIN2 IS A NOVEL REGULATOR FOR CIPROFLOXACIN-INDUCED PSEUDOALLERGY AND IgE-MEDIATED ANAPHYLAXIS

Emerging evidence suggests that while β-arrestin-1 contributes to MRGPRX2 desensitization, internalization and down signaling for ERK phosphorylation by balanced agonists, β-arrestin2 regulates MC degranulation via the modification of signaling pathways downstream of Ca²⁺ mobilization.^{86, 125, 128} For example, ciprofloxacin does not cause MRGPRX2-mediated β-arrestin recruitment as determined by transcriptional activation following arrestin translocation (Tango) assay.¹⁰² Despite this, ciprofloxacin-induced degranulation in mouse PMCs in vitro and vascular permeability in vivo are enhanced in mice with MC-specific deletion of β-arrestin2.¹²⁸ Furthermore, absence of β-arrestin2 in mouse bone marrow-derived MCs (BMMCs) has no effect on early F cε RI signaling such as antigen-induced Syk phosphorylation or Ca²⁺ mobilization but results in enhanced degranulation. Consistent with the in vitro findings, MC-specific deletion of β-arrestin2 enhanced IgE-mediated passive cutaneous anaphylaxis. This interesting finding indicates that β-arrestin2 negatively regulates both MrgprB2 and FcεRI-mediated MC degranulation to attenuate drug-induced pseudoallergy and IgE-mediated anaphylaxis, suggesting that this adapter protein could be targeted for the modulation of a diverse array of MC-mediated disorders.

PRECLINICAL ANIMAL MODELS TO STUDY MRGPRX2 FUNCTION IN VIVO

Although MrgprB2 is regarded as the murine ortholog of human MRGPRX2, there is only ~53% sequence homology between the two receptors, resulting in important differences in their interaction with agonists and antagonists. Thus, while ZINC3573 activates MRGPRX2, it is inactive against MrgprB2.¹²⁹ By contrast, rocuronium at a concentration of 20 μg/ml causes significant degranulation in murine PMCs but requires ~100-fold higher concentration to induce similar level of response in human skin MCs.²² The reason for this difference is not clear. However, C48/80, a polymer with a strong positive charge and multiple bulky hydrophobic moieties, is one of the most potent MRGPRX2 agonists known with a potency of ~130-fold higher than for MrgprB2. Although rocuronium has one quaternary amine and another amine that is mostly positively charged at pH 7.4, its hydrophobic steroidal backbone lacks the aromatic ring present in higher affinity MRGPRX2 agonists such as ZINC3573 and C48/80.^{7, 20, 102} Thus, in addition to positive charges, the hydrophobic moieties may be required for MRGPRX2 activation but may interfere with MrgprB2 activation. This may explain why rocuronium has higher potency for MrgprB2 when compared to MRGPRX2.

With the exception of rocuronium, most agonists tested have lower affinity for MrgprB2 than MRGPRX2.²⁰ Despite SP’s lower affinity for MrgprB2, it utilizes this receptor to promote neurogenic inflammation and AD in mice.^{20, 24, 27} Because of the close anatomical co-localization of MCs and TRPV1⁺ nociceptors in the mouse dermis, it has been proposed that local concentration of SP may reach the high activation threshold to activate MrgprB2 in vivo.²⁴ Even if this occurs, it is highly unlikely that potential MRGPRX2 antagonists at therapeutic concentrations will block MrgprB2 response in vivo. Indeed, Ogasawara et al.,¹³⁰ recently identified two small molecule MRGPRX2 antagonists that specifically block MRGPRX2-mediated activation of human MCs but have no effect on MrgprB2-mediated responses. Thus, from a therapeutic standpoint, mice expressing MrgprB2 are unlikely to serve as an appropriate in vivo model to screen novel MRGPRX2 antagonists.

Dogs, minipigs and non-human primates are generally used for preclinical drug safety studies. Histamine-like symptoms or anaphylaxis are sometimes observed in these animal studies, indicating the involvement of MRGPRs. Predicted open-reading frames for canine and porcine MRGPRX2 orthologs display distinctive elongated N-terminal domains, which are not present in primates (human, Rhesus, Cynomolgus) or rodents (Fig. 5A–B).^{131, 132} Interestingly, truncation of the excess N-terminal domain results in comparable Ca²⁺ mobilization response to human MRGPRX2 to the archetypal ligand, C48/80. Although MRGPRX2 is sometimes referred to as “primate-exclusive”, it is not possible to predict candidate MRGPRX2 of non-human primates using available genomic sequences.¹³¹ However, human MRGPRX2 is the only ortholog that requires Gα16 for optimal C48/80-induced Ca²⁺ mobilization in transfected HEK293 cells, indicating the unique properties from related species.¹³¹ This species-disparity restricts our ability to reliably survey the toxicity of newly developed drugs and to screen for novel MRGPRX2 antagonists in vivo.

FIG 5.

Sequence disparity in the N-terminus of Mas-related GPCRs in different species. (A) Snake diagram of secondary structure of MRGPRX2 with solid blue representing first 10 N-terminal amino acids. (B) Comparison of N-terminal sequence of human MRGPRX2 with the corresponding rhesus, cynomolgus (Cyno), minipig, pig, beagle, rat and mouse receptors.

To overcome the translational challenges discussed above, Suzuki et al.,¹³³ recently showed that an antagonistic DNA aptamer that targets MRGPRX2, inhibits anaphylactic shock to SP in MC-deficient rats engrafted with RBL-2H3 cells stably expressing MRGPRX2. Similar MC knock-in procedure can be developed in MC-deficient mice by engrafting MrgprB2⁻/⁻ BMMCs transfected with MRGPRX2 to study the function of this human receptor function in vivo.¹³⁴ The recently reported humanized mice that develop MRGPRX2-expressing human MCs with a partially functioning human immune system may provide an appropriate model to study the function of this receptor in vivo.¹²⁹

HDPs AND SMALL MOLECULE HDP MIMETICS (smHDPMs) AS POTENTIAL THERAPEUTICS FOR THE TREATMENT OF ANTIBIOTIC-RESISTANT MICROBIAL SKIN INFECTIONS

Antibiotics have been used for the treatment of microbial infections since the early 1900s but emergence of multidrug-resistant strains of microbes possess a tremendous public health concern globally.^{135, 136} A large number of HDPs are in clinical trials reflecting their therapeutic potentials.¹³⁶ Their mechanisms of action include (a) direct antimicrobial activity via the cell membrane (b) indirect antimicrobial activity via immune modulation and (c) inhibition of intracellular functions.¹³⁷ However, it now appears that most of the HDPs that display direct antimicrobial activity also activate MCs via MRGPRX2.¹³⁸ Based on these findings, it has been proposed that antimicrobial activity of HDPs reflect their ability to harness MRGPRX2’s immunomodulatory activity.^{34, 45} From a therapeutic view point, HDPs have clear advantages over conventional antibiotics, which include slower emergence of resistance, broad-spectrum activity, and the ability to favorably modulate MRGPRX2-MC-mediated host immune response.^{44,136, 138,139}

Most studies evaluating clinical potentials of HDPs for antimicrobial activity involve their topical application at the site of infection.¹³⁶ At these sites, it is desirable for a therapeutic agent to display antimicrobial activity and to harness MC’s immunomodulatory and wound healing properties.^{2, 3} Arifuzzaman et al.,⁴⁵ recently showed that multiple application of an MRGPRX2 and MrgprB2 agonist on infected mouse skin clears bacterial infection and promotes healing without any local or systemic side effects. The lack of systemic reactions likely reflects the small volume of MRGPRX2 agonist applied locally, which activates predominantly in cutaneous MCs, resulting in responses that are less intense and more transient than IgE-mediated MC activation.⁸² Furthermore, MrgprB2-mediated induction of adaptive immunity results in increased IgG and IgA but not IgE.¹⁴⁰

Although HDPs have important therapeutic potential for topical administration, their metabolic instability and cellular cytotoxicity have limited their utility.¹³⁶ To overcome these limitations, a series of small molecule HDP mimetics (smHDPMs) have been developed.^{141, 142} These compounds are relatively inexpensive to synthesize and have distinct advantages over HDPs in terms of stability, bioavailability, and low toxicity. Furthermore, smHDPMs exhibit potent antibacterial and antifungal activities and thus these compounds could be used as potential therapeutic agents for the treatment of drug-resistant microbial infections in general.¹⁴³ In addition, smHDPMs, which are non-cytotoxic and display direct antimicrobial activity, also activate murine and human MCs via MrgprB2 and MRGPRX2, respectively.³⁴ Thus, harnessing this novel feature of HDPs/smHDPMs to activate MC’s host defense properties in addition to their antimicrobial activities expands their clinical potential.

Despite the difference in sequences between MrgprB2 and MRGPRX2, smHDPMs tested thus far appear to activate human and murine MCs with similar efficacy.³⁴ This important finding suggests that outcome of preclinical studies with smHDPMs in mouse model of skin infection is likely to be relevant in human. As discussed above, smHDPMs have tremendous advantages over HDPs mainly due to ease of synthesis and lack of toxicity. Moreover, several approaches including lipidation has been used to increase hydrophobicity and membrane activity.¹⁴² Protegrin-1 is a HDP that activates human MCs via MRGPRX2⁴⁴ and its lipidated peptidomimetic, murepavadin (also known as POL7080), is an antibiotic with high activity against a broad panel of clinical isolates including multidrug-resistant Pseudomonas bacteria. Compared to 13 antibiotics and their combination tested, murepavadin demonstrated highest activity against P. aeruginosa recovered from cystic fibrosis patients, raising possibility that inhalation route of administration can be utilized for such conditions.¹⁴⁴ A topical formulation of the drug is also being developed for the treatment of antibiotic-resistant P. aeruginosa skin infection.^{145, 146} Thus, if murepavadin and other smHDPMs kill microbes directly and harness MC’s host defense properties via the activation of MRGPRX2, they could serve as novel “antibiotics” for the treatment of antibiotic-resistant skin infections, including diabetic foot ulcer infections.^{136, 137}

INHIBITION OF MRGPRX2 AND ITS SIGNALING AS POTENTIAL THERAPEUTIC STRATEGIES FOR THE TREATMENT OF ALLERGIC AND INFLAMMATORY SKIN DISEASES

MRGPRX2 likely functions as a double-edged sword, contributing to host defense to microbial infection but its inappropriate activation leading to inflammatory skin diseases. Thus, inhibition of MRGPRX2 and its signaling could provide novel strategies for the modulation of cutaneous diseases such as rosacea, AD, ACD and chronic urticaria. MRGPRX2-based photoimmunotherapy to selectively deplete MCs in the skin has been developed.⁸⁵ Injection of photosensitizer-conjugated MRGPRX2 antibody in the skin and subsequent irradiation with near infrared light could selectively deplete MRGPRX2 expressing skin MCs.⁸⁵ Dondalska et al.,¹⁴⁷ recently showed that an immunomodulatory single-stranded oligonucleotide (ssON) inhibited LL-37/MRGPRX2 signaling and MC degranulation without affecting IgE-mediated response. Additionally, ssON treatment ameliorates LL-37-induced inflammation in a mouse model of rosacea. Callahan et al.,¹²⁷ showed that osthole, a natural plant coumarin attenuates SP and LL-37-induced Ca²⁺ mobilization, degranulation and chemokine/cytokine production in LAD2 cells and human primary skin MCs in vitro. It also inhibits LL-37-mediated rosacea-like features in vivo. Molecular docking analysis suggests that osthole does not compete with MRGPRX2 agonist, but rather regulates MRGPRX2 activation via allosteric modification. Furthermore, osthole reduces the expression of MRGPRX2 in human MCs indicating that it can be utilized for the treatment of MRGPRX2-mediated skin disorders (Fig. 6).

FIG 6.

Modulation of MRGPRX2 and its signaling in cutaneous MCs. Injection of photosensitizer conjugated with MRGPRX2 antibody can selectively destroy targeted skin MCs. Single-stranded oligonucleotide (ssON) inhibits LL-37-induced degranulation in human MCs and reduces LL-37-induced experimental rosacea in mice. Osthole, TRPV4 inhibitor HC-067047 and Botulinum toxin (BTX) inhibit different component of MRGPRX2 and its signaling in human MCs to block LL-37induced rosacea in mice.

Increased Ca²⁺ mobilization that occurs almost immediately after MRGPRX2 activation is essential for the release of inflammatory mediators.⁴¹ Occhiuto et al.,¹⁴⁸ showed that store-operated calcium entry (SOCE) via the calcium sensor, stromal interaction molecule 1 (STIM1), regulates MC response to LL-37. Furthermore, a SOCE blocker prevents the development of LL-37-induced rosacea-like features in mice. Orai channels are activated upon the depletion of internal calcium stores and likely contribute to Ca²⁺ increased in response to MRGPRX2 actvation.⁴² By contrast, Mascarenhas et al.,¹⁴⁹ showed that in LL-37-induced rosacea-like features in mice, there is a 672 substantial upregulation of the Ca²⁺-permeable transient receptor potential vanilloid-type 4 (TRPV4) ion channel, which colocalizes with murine MCs. TRPV4 expression is also upregulated in human CD34+ cell-derived primary MCs treated with LL-37 for 24h. In addition, treatment of primary human MCs with pertussis toxin (inhibitor of Gαi/Go family of G proteins) or CRISPR/Cas9-mediated knockdown of MRGPRX2 results in the inhibition of LL-37-induced TRPV4 expression and MC degranulation.¹⁴⁹ Furthermore, pharmacological inhibition of TRPV4 by HC-067047 results in inhibition of MRGPRX2-mediated MC degranulation. Based on these findings, it has been proposed that TRPV4 expressed in cutaneous MCs contributes to the pathogenesis of rosacea.¹⁴⁹,¹⁵⁰ It is therefore possible that both SOCE/Orai and TRPV4 activation in response to MRGPRX2 contribute to rosacea and other MC-mediated inflammatory skin diseases (Fig. 6). As such, inhibitors of these Ca²⁺ entry pathways could be targeted for the modulation of MRGPRX2-mediated disorders.

Botulinum toxin (BTX) is a neurotoxin produced by the bacterium Clostridium botulinum that inhibits the release of acetylcholine at the neuromuscular junction level and has been used to treat a variety of muscular/neuromuscular conditions.¹⁵¹ Intradermal injection of BTX has been shown to improve rosacea symptoms via an unknown mechanism.^151–153 The fusion between vesicles and the plasma membrane requires a specific set of proteins including soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs), including synaptosomal associated protein-25 (SNAP-25) and vesicle-associated membrane protein (VAMP).¹⁵⁴ Choi et al.,¹⁵² recently showed that BTX inhibits C48/80 (agonist for MRGPRX2 and MrgprB2)-induced degranulation in human and murine MCs and that this is associated with decreased cleavage of SNAP-25 and VAMP2. Furthermore, in a mouse model of LL-37-induced rosacea, mice pre-treated with BTX showed significantly less dermal MC degranulation and erythema compared to control. Although direct proof that MRGPRX2 expressed in MCs plays a major role in the pathogenesis of rosacea is lacking, there is now growing body of evidence that LL-37 and SP activate MC_TC via this receptor.^{8, 41, 57, 138} Thus, a variety of approaches targeting MRGPRX2 and its signaling (Ca²⁺ mobilization, SNARE, VAMP) may provide novel approaches for the modulation of rosacea and other MRGPRX2-mediated inflammatory skin diseases (Fig. 6).

CONCLUDING REMARKS AND FUTURE DIRECTIONS

In our last review published in this journal in 2016, we postulated that MRGPRX2 contributes to host defense, neurogenic inflammation, pain itch and chronic inflammatory skin diseases.²¹ Since then, tremendous progress has been made and studies with WT and MrgprB2^MUT mice led to the realization of many of these predictions.^{24–27, 45} Furthermore, a number of reviews and opinion articles have been published summarizing different aspects of MRGPRX2 and MrgprB2 biology.^{7, 155–161} However, this current perspective article provides a comprehensive analysis of MRGPRX2/MrgprB2 biology with particular emphasis on how it could be targeted for MC-mediated health and disease.

While MRGPRX2 was originally thought to be expressed predominantly in cutaneous MCs, recent studies indicate that it is also expressed in colon MCs and contributes to the pathogenesis of UC.⁸⁶ Although human lung MCs express little to no MRGPRX2, its expression is upregulated in lung MCs of individuals who died from asthma when compared to individuals who died from other causes and has been proposed as a new biomarker for allergic asthma.^{57, 162–164} Furthermore, the NP hemokinin-1 released from bronchial cells and major basic protein (MBP) and eosinophil cationic protein (ECP) activate human MCs via MRGPRX2.^{57, 165} It now appears that human basophils and eosinophils express MRGPRX2 but whether or not they respond to agonists that activate the receptor in MCs is the subject of controversy.^{12, 13, 19} Nevertheless, it is clear that expression of MRGPRX2 is not limited to cutaneous MCs as previously thought. Thus, we anticipate that the next few years will herald a new era when basic research with MRGPRX2 and in vivo studies with MrgprB2 will be translated into disease processes and it will be expanded beyond its current cutaneous boundary into other systems such as the airways, synovium and the gut.^{72, 85, 114, 162} Furthermore, we anticipate that MRGPRX2 agonists and antagonists will be further developed for the modulation of MRGPRX2 in health and disease.

What is currently known:

• Unlike FcεRI, which is expressed in all MCs, MRGPRX2 is found predominantly in human skin MCs with low levels in lung and gut MCs.

• Bacterial quorum sensing peptides and HDPs activate MCs via MRGPRX2 and likely contribute to host defense and cutaneous wound healing.

• In mice, MrgprB2 promotes host defense and contributes to pseudoallergy, neurogenic inflammation, pain, atopic dermatitis, non-histaminergic itch and allergic contact dermatitis.

• Expression of MRGPRX2 is upregulated in skin MCs of patients with chronic spontaneous urticaria and lung MCs of individuals who died from asthma-related causes.

• MC-specific mediators and a precursor of MRGPRX2 agonist (PAMP-12) are upregulated in ulcerative colitis.

• Agonist binding sites, G protein coupling domains, phosphorylation sites and β-arrestin binding codes on MRGPRX2 have been partially identified.

• Certain MRGPRX2 ligands act as balanced agonists whereas others serve as G protein-biased agonists.

• Individuals harboring a missense MRGPRX2 variant (N62S) are protected from the development of ulcerative colitis due to enhanced β-arrestin recruitment and receptor desensitization.

What more do we need to know:

• Why MRGPRX2 is predominantly expressed in skin MCs and what local and epigenetic factors regulate its expression?

• Does MRGPRX2 contribute to mastocytosis and mast cell activation syndrome?

• What factors regulate the expression of MRGPRX2 in basophils and eosinophils and do they contribute to the pathogenesis of asthma and asthma exacerbation?

• Are the expression of MRGPRX2 agonists or their proteolytic precursors increased in inflammatory diseases of the skin or other organs?

• Do individuals harboring missense MRGPRX2 mutations display resistance or susceptibility to bacterial infection and inflammatory diseases?

• Can the current pipeline of smHDPMs, which are being developed for direct antibacterial activities, be utilized to harness MRGPRX2’s immunomodulatory properties for the treatment of antibiotic-resistant skin infections?

• Can small molecule antagonists or antibodies targeting MRGPRX2 be utilized for the treatment of rosacea, pain, atopic dermatitis, allergic contact dermatitis, itch, ulcerative colitis and other diseases that depend on MCs?

Abbreviations used

ACD

Allergic contact dermatitis

AD

Atopic dermatitis

C48/80

Compound 48/80

CSP

Competence-stimulating peptide

DCs

Dendritic cells

DDCs

Dermal dendritic cells

DRG

Dorsal root ganglia

GPCR

G protein-coupled receptor

hBD

Human β-defensin

HDM

House dust mite

HDP

Host defense peptide

LAD2

Laboratory of allergic diseases 2

MCs

Mast cells

MRGPRX2

Mas-related G protein–coupled receptor-X2

MrgprB2

Mas-related G protein–coupled receptor-B2

NPs

Neuropeptides

NMBDs

Neuromuscular blocking drugs

PAMP

Pro-adrenomedullin peptide

PAR

Protease activated receptor

PMCs

Peritoneal mast cells

PGD₂

Prostaglandin D₂

QSMs

Quorum-sensing molecules

ROS

Reactive oxygen species

SEB

Staphylococcal enterotoxin B

SNPs

Single nucleotide polymorphisms

SP

Substance P

smHDPMs

Small molecule host defense peptide mimetics

SOCE

Store-operated calcium entry

TM

Transmembrane

TRPV

Transient receptor potential vanilloid

UC

Ulcerative colitis

WT

Wild-type