Elevated, FcεRI‐dependent MRGPRX2 expression on basophils in chronic urticaria

Abstract Background Chronic urticaria (CU) is a skin condition driven by mast cells and basophils. The exact responsiveness profile of these cells, especially regarding the anti‐IgE treatment, Omalizumab, is not fully investigated. We sought to characterize the surface activation profile of basophils in CU during Omalizumab treatment and their responsiveness to IgE and non‐IgE stimulation. Methods Whole blood basophils from 11 CU patients and 10 healthy controls were stimulated with either medium, anti‐IgE, fMLP, C5a, or Substance P for 30 min and characterized by flow cytometry. Results CU patients showed a broad range of basophil count as opposed to healthy subjects. An increased number of unstimulated CD69+ (p = 0.05), but not CD63+ basophils was observed in CU groups in comparison to healthy. The expression of CD203c and CD200R were comparable between all groups, whilst the FcεRI was reduced with the treatment. Both IgE and non‐IgE mediated stimulations upregulated CD63, CD203c and CD200R, but not CD69 in all groups, however, no difference between the groups was observed. Among unstimulated basophils, expression of MRGPRX2 was higher in CU patients after Omalizumab treatment than in the healthy group (2.4% vs. 1.5%, p = 0.01). The anti‐IgE stimulation increased the number of MRGPRX2‐expressing basophils in the CU group before and after omalizumab as compared to the healthy (p = 0.003; p = 0.005). The fMLP and C5a stimulations showed a similar effect to the IgE‐mediated stimulation. The MRGPRX2 ligand, Substance P did not activate basophils. Conclusion CU basophils show increased expression of MRGPRX2 after IgE and non‐IgE stimulation.


| INTRODUCTION
Chronic urticaria (CU) is a group of skin diseases characterized by the appearance of transient pruritic wheals remaining on the skin for 6 weeks or longer. 1 Two types of CU can be distinguished, spontaneous (CSU) and inducible (CIndU). While the key cellular players of the disease, mast cells (MCs) and basophils, are known, their responsiveness and phenotype require further investigation, especially in the context of the current treatment strategy, which includes the anti-IgE biological drug, Omalizumab (OMZ). Although OMZ This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. prevents the activation of MCs and basophils by binding IgE, it is not equally effective in all cases of CSU. 2 Thereby, patients with no response to OMZ treatment are categorized as non-responders (NR). In contrast, within the responder group, quick (QR) and slow responders (SR) are determined based on the time between the first injection and symptom release.
The current understanding of basophils' role in CSU highlights their altered responsiveness towards IgEmediated stimulation, lowered peripheral cell number (basopenia), and transmigration to the skin wheals. [3][4][5] While stimulation of basophils in CSU with Substance P (SP) induces marked histamine release, 1,6 the anaphylatoxin, complement component C5a results in reduced mediator release. 1 The expression of SP receptor-NK1R has been found elevated on basophils in CSU patients, supporting the effect of SP. 6 Additionally, another SP receptor, Mas-related G-protein coupled receptor member X2 (MRGPRX2), showed higher expression on MCs in lesioned skin of CSU patients, 7 but to our knowledge, this has not been investigated in CU basophils.
Characterization of basophil activation status either by determining the released mediators or by detecting molecules that appear on the plasma membrane upon fusion of secretory vesicles allows for a better understanding of basophil phenotype in a disease. Thus, the expression of activation receptors, for example, CD63, CD203c, CD69 has been investigated in several chronic urticaria studies. [8][9][10][11][12] The increased baseline levels of CD63 and CD69 and comparable CD203c values contrary to non-allergic controls were defined by Vasagar et al., 11 whereas Christensen et al. showed augmented expression of CD203c and CD63. 10 In the present study, we aimed to investigate the expression of CD63, CD69, CD203c, CD200R, MRGPRX2, and FcεRI on resting and IgE or non-IgEactivated basophils from CU patients. Additionally, we sought to determine an effect of OMZ treatment on these markers.

| MATERIALS AND METHODS
For a detailed description of the methods, please refer to the Supporting information.

| Study design
Eleven CU patients and 10 healthy subjects were enroled in the study. Whole blood samples were collected before the first OMZ administration (BO) and 12 weeks (3 dosages) after initiation of OMZ treatment (AO). Healthy participants were subjected to one blood sampling. Patients were classified as either QR with an improvement < 1 month from the first injection, as SR with improvement ≥ 1 month from the first injection, or NR without improvement ≤ 3 months from the first injection.

| Flow cytometric analysis of basophils
Heparinized whole blood samples were used for the investigation of surface receptors. The blood was analyzed within 4 h after collection. In brief, whole blood was simultaneously activated and stained with antibodies for 30 min at 37°C, followed by erythrolysis, fixation, and a readout on the flow cytometer.

| Patients' characteristics
The healthy control group (n = 10) almost entirely matched the CU group (n = 11) ( Table 1). Nine out of 11 CU patients were diagnosed with CSU and CSU with concomitant CIndU, while two CIndU patients were diagnosed with urticaria factitia and cholinergic urticaria (Table S1). Three patients were classified as SR, whereas the remaining seven, except for one CIndU NR, were classified as QR (Table S1).

What is already known about this topic?
� Basophils are one of the effector cells in chronic urticaria (CU). � Omalizumab changes basophils responsiveness in CU.

What does this study add?
� Basophils in CU patients showed a partially preactivated profile that Omalizumab treatment reduces. � Basophils in CU were characterized by lower sensitivity towards IgE-mediated stimulation. � Increased expression of MRGPRX2 was found on resting and IgE and non-IgE stimulated basophils from CU patients.

| Basophil cell count
Initially, we focussed on the basopenic theory of CU patients. CU patients showed a broad range of basophil count ranging from 0.004% to 1.44% of single cells ( Figure 1a). Two basopenic patients were identified, one in QR and one in SR/NR group.

| Expression of activation, inhibitory and FcεRI receptors on resting, IgE, and non-IgE stimulated basophils
We next investigated if OMZ treatment modulated the basophil activation profile. The resting expression profile of basophils (CD123 + CRTH2 + , Figure S1) showed no expression (<2%) of CD63 in either of the groups and higher expression of CD69 in the BO group compared to AO (p = 0.005) and healthy (p = 0.05) ( Figure 1b). Moreover, fewer basophils expressing CD69 were observed in the AO than in healthy individuals (p = 0.02) ( Figure 1b). Interestingly, comparable levels of the wellrecognized basophil activation markers CD203c and CD200R were measured in all groups ( Figure 1b). As expected, FcεRI expression on resting basophils was reduced by the OMZ treatment (p = 0.01). Furthermore, a trend of elevated expression of FcεRI was found in the BO group as opposed to healthy individuals (p = 0.09) ( Figure 1c). IgE-mediated stimulation with serial dilution of anti-IgE resulted in a dose-dependent increase in the percentage of CD63 + basophils and elevated expression of CD203c and CD200R ( Figure S2, A-D). However, no changes in the expression of receptors were observed between the groups. Interestingly, OMZ altered the expression pattern for all receptors from scattered in the BO group to stacked, indicating an effect of the treatment on receptor expression (Figure 2a). fMLP and C5a stimulations increased the percentage of CD63 + basophils and induced upregulation of CD203c and CD200R in all groups ( Figure S3, A,C,D). OMZ tended to reduce the number of CD63 expressing basophils induced by fMLP (p = 0.075) and by C5a (p = 0.14) (Figure 2b). The upregulated expressions of CD203c and CD200R were comparable for both stimulants between all groups, and there were no changes in the expression of CD69 (Figure 2c-e). SP failed to trigger basophil activation ( Figure 2).

| Expression of MRGPRX2 on resting, IgE, and non-IgE activated basophils
We then evaluated the expression of MRGPRX2 on resting basophils. Contrary to healthy, we found a higher percentage of MRGPRX2 + basophils in the OMZ-treated group (p = 0.01) (Figure 3a). Interestingly, OMZ altered the resting expression of MRGPRX2 on basophils in CU patients (p = 0.08) similarly to the flow values of the activation and inhibitory receptors. As we expected, no expression (x̄= 1.5%) of MRGPRX2 was found on resting healthy basophils (Figure 3a).
IgE-mediated stimulation induced a marked, dosedependent increase in the percentage of MRGPRX2 + basophils in the BO group (Figure 3b). The expression of MRGPRX2 on basophils was higher in the BO and AO groups compared to healthy (p = 0.003; p = 0.005). Moreover, OMZ tended to reduce the percentage of MRGPPRX2 + basophils (p = 0.08) (Figure 3c).
The fMLP stimulation triggered the highest increase in the percentage of MRGPRX2 + basophils in all groups, followed by C5a and SP, with the latter triggering no changes. The most pronounced difference was observed for the BO group in comparison to the healthy ( Figure S3,E). The fMLP stimulation resulted in a higher

| DISCUSSION
In this study, we sought to evaluate the effect of OMZ treatment on basophils' responsiveness and activation status. We confirmed the constitutive expression of CD203c and CD200R and no expression of CD63 and CD69 on resting basophils of healthy individuals. However, contrary to Christensen et al. 10 and Chen et al., 9 we did not observe the expression of CD63 in resting basophils of CU patients, whereas the CD203c expression showed comparable levels to healthy. Differences between the studies could stem from staining procedures, where basophils were stained with either fluorophore-labelled anti-FcεRI or anti-IgE that could potentially activate cells. Additionally, in the study by Chen et al., enrichment of basophils was utilized, known to preactivate basophils. 9 The increased number of basophils expressing CD69 in CU suggests an in vivo stimulation which could be induced either by dysregulated intracellular pathways governing an expression of the receptor or by increased concentrations of molecules priming basophil responsiveness such as IL-3. The lower expression of CD69 on basophils in CU induced by OMZ might indicate a link between the activation marker and the FcERI-IgE pathway, as OMZ reduced levels of both in CU patients. Additionally, it implies that OMZ decreases the activation profile of basophils. Although the CD200R was not reduced, a tendency of decreased expression was observed in a few patients indicating ameliorated inhibitory properties of basophils in these CU patients, which could also be a reason for the elevated expression of CD69 on resting basophils. The increased CD63 expression induced by lower anti-IgE concentrations and the higher CD63 flow values suggests a more responsive profile of IgE-mediated activation of basophils in CU. This could be explained by a disrupted intracellular signalling pathway with increased phosphorylation of SYK needed for the propagation of the signal and/or reduced function of inhibitory molecules such as SHIP1/2. 13,14 As the CD63 upregulation pathway differs from CD203c, 15,16 the CD203cbased sensitivity profile was also investigated. No difference between basophils in CU and healthy in regards to CD203c expression was found, implying that two different modes of basophil activation are independently affected in CU. Additionally, a tendency of higher CD200R expression in CU basophils implies that not only are inhibitory mechanisms still effective, but they also are more sensitive in contrary to healthy basophils. The GPCR-mediated activation with fMLP and C5a upregulated CD63, CD203c, and CD200R in all groups. As fMLP stimulation requires a more robust cytosolic response than IgE-mediated stimulation, 17 the amplified response of CU basophils via this pathway might suggest either higher reserves of Ca 2+ , their faster disposition, or both. In contrast to Luquin et al., 1 we did not observe reduced responsiveness of basophils in urticaria patients F I G U R E 2 IgE and non-IgE mediated activation of basophils-an expression of activation and inhibitory receptors. Basophils were activated with serial dilution of anti-IgE (4-4000 ng/ml), fMLP, C5a, and SP. (a) Flow values of percentage (CD63 + and CD69 + ) and GeoMean (CD203c and CD200R) of basophils. (b-e) Non-IgE induced expression of CD63 (b) and CD69 (c) and GeoMean of CD203c (d) and CD200R (e). BO and AO refer to patients' groups before and after Omalizumab treatment, respectively; CIndU patients are marked in orange; BO (n = 10), AO (n = 9), and healthy (n = 10); Statistics applied-unpaired t-test; ns, non-significant; Mean � SD; CIndU, chronic inducible urticaria; SD, standard deviation. BARTKO ET AL. to C5a stimulation, possibly due to a 10 times higher concentration of C5a in this study. Correspondingly to Deza et al., 2 we measured higher FcεRI expression on resting basophils in CU, with QR presenting increased values contrary to SR. Even though NR basophils showed FcεRI levels similar to the values of QR, they did not respond to the OMZ treatment, suggesting an alternative mechanism driving the disease.
To our knowledge, we demonstrated for the first time a trend of elevated expression of MRGPRX2 on resting CU basophils compared to the healthy. This, however, did not correspond with the SP activation of basophils, which failed to induce degranulation and was in contrast to the study by Zheng et al. 6 We hypothesized that even though the number of MRGPRX2 + basophils was higher in CU, the expression of the receptor must have been insufficient to result in cell degranulation. Additionally, this might indirectly suggest an absence of an alternative pathway for SP activation, for example, NK1R. IgE and non-IgE mediated stimulations further expanded the resting expression of MRGPRX2 on basophils in CU. The anti-IgE activation resulted in the MRGPRX2 upregulation, which profile followed the pattern of CD63 upregulation. However, contrary to CD63, the pattern of the dose-dependent curves of MRGPRX2 differed drastically in the healthy subjects. This marked difference between MRGPRX2 and CD63 highlights the role of the former in activating basophils in CU. Moreover, intracellular components responsible for the upregulation of MRGPRX2 might be dysregulated in CU basophils since in healthy subjects, the upregulation of CD63 and MRGPRX2, though showing comparable sensitivity towards a stimulant, resulted in the marked upregulation of the former, however marginal of the latter. Another explanation for the increased number of MRGPRX2 + basophils in CU could be the matrix surrounding basophils, that is, serum, which components might prime cells to respond to the IgE-mediated stimulation in a certain way. Thus, the mechanism linking the IgE-FcERI pathway with MRGPRX2 in CU depicts basophils with a decreased threshold for the activation via the IgE pathway leading to higher upregulation of MRGPRX2. In contrast to Wedi et al., 18 we demonstrated no (x̄= 1.5%) expression of MRGPRX2 on healthy basophils. This, however, partially supports results from Sabato et al., 19,20 showing only a minority of basophils in healthy subjects expressing MRGPRX2. Moreover, our results align with the study by Fujisawa et al.,7 in which elevated MRGPRX2 expression was found on MCs in lesioned skin of CU patients. In addition, an increased serum level of soluble MRGPRX2 correlated with CSU severity showing strong evidence and AO refer to patients' groups before and after Omalizumab treatment, respectively; CIndU patients are marked in orange; BO (n = 10), AO (n = 9) and healthy (n = 10); Statistics applied-unpaired t-test with a two-sided α-level, <0.05 considered as significant; ns, non-significant, p > 0.05, *p ≤ 0.05;**p ≤ 0.01 Mean � SD; CIndU, chronic inducible urticaria; SD, standard deviation. as a marker for disease severity. 21 These findings and our data suggest an involvement of the pseudo-allergen pathway in the regulation of effector cell activation in CU patients and the disease profile.
The major limitation of this study was the low number of participants and the heterogeneity of patients. The latter justifies the broad range of basophil count in patients, which is known to depend on disease severity. Thus, this study should be considered a foundation for future projects focussing on MRGPRX2 signalling and the sensitivity profile of basophils in urticaria patients.
In summary, CU patients presented a broad spectrum of basophil count and a higher anti-IgE-induced sensitivity profile of basophils. Furthermore, the resting expression of MRGPRX2 on CU basophils was significantly upregulated upon IgE and non-IgE mediated stimulations.