R‐Ras deficiency does not affect papain‐induced IgE production in mice

Abstract Introduction R‐Ras GTPase has recently been implicated in the regulation of immune functions, particularly in dendritic cell (DC) maturation, immune synapse formation, and subsequent T cell responses. Methods Here, we investigated the role of R‐Ras in allergen‐induced immune response (type 2 immune response) in Rras deficient (R‐Ras KO) and wild type (WT) mice. Results Initially, we found that the number of conventional DC's in the lymph nodes (LNs) was reduced in R‐Ras KO mice. The expression of co‐stimulatory CD80 and CD86 molecules on these cells was also reduced on DC's from the R‐Ras KO mice. However, there was no difference in papain‐induced immune response between the R‐Ras WT and KO as measured by serum IgE levels after the immunization. Interestingly, neither the DC number nor co‐stimulatory molecule expression was different between WT and R‐Ras KO animals after the immunization. Conclusions Taken together, despite having reduced number of conventional DC's in the R‐Ras KO mice and low expression of CD80 on DC's, the R‐Ras KO mice are capable of mounting papain‐induced IgE responses comparable to that of the WT mice. To our knowledge, this is the first report addressing potential differences in in vivo allergen responses regulated by the R‐Ras GTPase.

R-RAS can regulate inflammation [12][13][14][15]. Dendritic cells (DC's) from R-Ras deficient animals have reduced ability to prime T cell responses due to unstable immune synapse formation [14]. In experimental autoimmune encephalomyelitis model (EAE), Rras À/À mice showed attenuated disease course likely due to elevated numbers of peripheral tolerogenic and regulatory T cells (Tregs) [15]. Recently we showed that R-Ras is required for tumor development and progression in inflammationdependent skin cancer model [13]. Although R-Ras was present only in the blood vessels and in skin, it was required both for the induction of pro-inflammatory cytokine production and for the extravasation of inflammatory cells to the skin [13].
Given the key role of DC's in adaptive immune responses and the role of R-Ras in regulating DC functions [14,15] we wanted to know whether type2 immune response is regulated by R-Ras in vivo. We found that despite the fact that untreated R-Ras KO had decreased number of conventional DC's and the DC's show reduced expression of CD80/CD86 costimulatory molecules, these mice were capable of mounting a normal allergen-induced response after papain immunization.

Methods
Mice, immunizations, ELISA, cell culture, flow cytometry R-Ras KO (¼ Rras Gt(OST24882)Lex ) mice (C57BL/6) have been described [16]. Animal experiment protocol was approved by Finnish National Animal Experiment Board (Permit: ESAVI/4738/04. 10.07/2014). Administration of papain was performed as a modified version of the protocol described earlier by Sokol et al. [17]. Mice were injected subcutaneously either with PBS alone or PBS containing 0.5 mg papain (Merck Millipore, Darmstadt, Germany) on days 0 and 14. Blood samples were collected from the tail vein on day 0. Mice were euthanized on day 16 or day 21, and spleens and LNs were harvested. Blood was collected by cardiac puncture into serum blood tubes (BD, Franklin Lakes, NJ). IgE ELISA was from eBioscience (Santa Clara, CA).

Results and Discussion
R-Ras KO mice lacking functional R-Ras protein are phenotypically similar to WT counterparts [13,16,20]. However, DC maturation and immunological synapse formation with T cells is impaired in R-Ras KO mice and the capability of Rras À/À DC's to induce allogeneic or antigen-specific autologous T cell proliferation is reduced [14]. As described [14], we found that B-and T cell numbers in spleen of untreated R-Ras KO mice (n ¼ 3) were normal (Fig. 1A, gating strategy in Supplementary Fig. S1A).
The less severe EAE in R-Ras KO mice is due to a local increase in the number of Tregs and tolerogenic DC's [15]. This led us to study the number and phenotype of DC's in WT and R-Ras KO mice. Analysis of magnetically enriched conventional and migratory DC's from the LN (gating strategy in Supplementary Fig. S1C) indicated that R-Ras KO mice had less conventional DC's than WT mice (Fig. 1B), (p ¼ 0.0126), while the number of migratory DC's was similar between the two genotypes (Fig. 1B). As previously shown [14], the analysis of the enriched splenic DC's from spleen showed no difference between WT or R-Ras KO mice (Fig. 1C, gating strategy in Supplementary Fig. S1D). We chose to exclude plasmacytoid DC's from our analysis, as they are currently considered less relevant to allergic responses: they are known producers of IFNg upon TLR7/ TLR9 activation during viral infections and autoimmune disorders [21].
The strength of T cell receptor(TCR)-engagement in cytokine-free environment regulates T cell differentiation. Weak TCR-signal leads to spontaneous, GATA3-and IL-2dependent, IL-4 production while strong TCR signal causes, ERK-dependent, IFN-g production [22]. The DC surface markers that take part in the DC/T-cell interaction include MHCII (antigen presentation to TCR), CD80/CD86 (costimulatory molecules associated with TCR) and OX40L (enhancement of Th2 differentiation at least in humans) [23]. MHCII expression was identical in both conventional and migratory LN DC's between WT and R-Ras KO mice ( Fig. 2A). CD80 expression, in turn, was reduced in both conventional (p ¼ 0.0078) and migratory (p ¼ 0.0176) subsets from R-Ras KO mice ( Fig. 2A), while CD86 expression was decreased in conventional DC subset of R-Ras KO mice (p ¼ 0.0241), but not in migratory DC subset ( Fig. 2A). OX40L expression was identical in conventional or migratory DC's between KO and WT mice ( Fig. 2A). In spleen, MHCII expression was higher in the KO DC's (p ¼ 0.0129) while CD80 expression was lower in the KO DC's (p ¼ 0.0207) and, CD86 or OX40L expression were identical (Fig. 2B).
To better understand the role of R-Ras in tuning DC activation, we analyzed the response of splenic DC's from R-Ras deficient and WT mice to IL-4, LPS, and poly(I:C) during overnight plate culture, which ''spontaneously'' activates DC's [24]. Although LPS and poly(I:C) are not considered driving agents of Th2 polarization, we were interested to examine whether there was some general impairment of DC responses in Rras À/À mice and thus included these stimulants into the comparison.
Collectively, in splenic DC's MHCII expression is higher on DC's from R-Ras KO mice under all tested culture conditions. The differences observed in CD80 and CD86 expression between R-Ras KO and WT cells after overnight culture were abolished when the cells were further activated with poly(I:C), LPS or IL-4, while OX40L was induced by poly(I:C) and LPS, but R-Ras expression plays no role in the regulation of OX40L expression as no difference was identified between KO and WT cells ( Supplementary Fig. S2A-D).
Taken together, the main difference in DC's between WT and R-Ras KO mice is that there is significantly less conventional DC's in R-Ras KO mice and these cells express less co-stimulatory molecules, particularly CD80. These results suggest, that antigen presentation via MHCII is not impaired in R-Ras DC's but the formation of immunological synapse due to decreased CD80/CD86 expression is impaired, which in turn might result in impaired CD28 signalling in T-cells.
Since EAE in R-Ras KO mice is less severe than in WT mice, we wanted to study if type 2 responses are affected in these mice. To analyze a possible ''skew'' in the steady-state CD4 T cells in R-Ras KO mice, we studied the mRNA expression of critical transcription factors related to helper T cell differentiation from the total splenocytes of WT and R-Ras KO mice, but found no difference in the mRNA expression of Tbet, GATA3, Foxp3, or Rorg (Fig. 2C, upper  panel). Also, the mRNA expression of IFNg, IL-13, and IL-10 was identical between WT and R-Ras KO mice (Fig. 2C lower panel).
To study the allergen-induced response in R-Ras KO mice, we utilized papain-immunization as an in vivo model. Papain-induced responses are mediated via migratory DC's in skin [25] and we had already observed a decrease in CD80 expression in migratory DC's ( Fig. 2A) in R-Ras KO mice. We subjected R-Ras KO mice and WT animals (n ¼ 8 for each genotype) to papain injections twice: first immunization was performed on day 0 and re-immunization on day 14. We measured the serum level of IgE on day 0 and again when the mice were euthanized on day 21. We found no significant difference in the serum IgE levels between the two genotypes ( Fig. 3A) 21 days after the first immunization. Next we wanted to rule out the possibility that the initiation phase of the type 2 response to papain might be delayed due to the impaired immune synapse formation [14] in R-Ras KO mice. For this, we measured the IgE levels of papainimmunized WT and KO (n ¼ 17 and 18, respectively) mice only two days after the second immunization (i.e., 16 days after the first immunization). However, no difference in the serum level of IgE was found between R-Ras KO mice and WT (Fig. 3B). The concentration of IgE was approximately 10 mg/ml on day 16, while it was $50 mg/ml on day 21, indicating that the allergen response was still being induced after day 16 ( Fig. 3A and B).
Next, we determined the number of DC's on day 16 after the papain immunization. Unlike in unmanipulated mice, which showed a difference in the number of conventional DC's between WT and R-Ras KO animals (Fig. 1B), we found no difference in their number (Fig. 3C) in papaintreated animals. Similarly, the number of splenic DC's was alike in WT and R-Ras KO mice (Fig. 3D). We also determined Th2-associated surface markers on CD4 T cells after papain immunization. The expression of the cytokine-binding receptor chain for IL-4 (IL-4Ra) is critical for CD4 T cell response to IL-4 and is up-regulated on CD4 cells during Th2 response. We found no difference in IL-4Ra expression between WT or R-Ras KO mice (n ¼ 5) either in CD4 or CD8 cells (Fig. 3E). IL-33R is another cytokine receptor chain up-regulated in CD4 T-cells in Th2 responses. IL-33 can regulate ''effector cytokine,'' such as IL-5 and IL-13, expression in Th2 cells in concert with Stat5 activating cytokines, independently of antigens [26]. IL-33 has been shown to be indispensable for papaininduced IgE production in murine lung and genital tract [27,28]. We found that the percent of IL-33R positive cells was reduced in R-Ras KO CD4 T cells (Fig. 3F, the gating strategy for IL-33R is shown is supplementary Fig. S1B). This might suggest that while IL-4 mediated events (IgE production, IL-4Ra expression) are intact in R-Ras KO mice, IL-4 independent (but GATA3-dependent) events, such as IL-33R expression can be affected by the deletion of R-Ras. The requirement of IL-4 for Th2 differentiation in vivo is discussed in detail in reference [29]. Then we analyzed the expression of DC activation markers from papain-immunized mice. Unlike in the healthy mice, the conventional and migratory DC's from the LN showed no difference between the two genotypes in the expression of MHCII, CD80, CD86, or OX40L (Fig. 4A). In splenic DC's the expression of MHCII and CD80 were similar between WT and R-Ras KO animals, but CD86 expression was reduced after papain-immunization in R-Ras KO mice (Fig. 4B). Papain has been shown to induce Th2 polarization in multiple studies, determined by the IL-4 production of T cells [17,28,30] and T cell recall responses upon in vitro restimulation with papain [17,28,31,32]. We measured the mRNA expression of transcription factors associated with T cell differentiation from splenocytes by qPCR after papain-immunization. The mRNA expression levels of Tbet, GATA3, and Foxp3 were identical between WT and R-Ras KO mice but for Rorg, the master regulator of Th17differentiation, the mRNA expression was significantly lower (p ¼ 0.0492) in the R-Ras KO mice (Fig. 4C). Accordingly, hallmark cytokine mRNA analysis from these samples indicated no difference between IFN-g, IL-13, or IL-10 expression (Fig. 4C). The notion that RORg-transcription factor was expressed at reduced levels in R-Ras KO mice compared to WT is logical as IL-17 production is decreased in CNS of Rras À/À mice [15] undergoing EAE or carcinogen induced tumor development [13]. RORg, on the other hand, plays a critical role in the development of IL-17 producing CD4þ T cells [33]. Also, a recent study showed that  epicutaneous papain administration was able to induce the differentiation of allergen-specific Th17 cells [31], suggesting that papain induces not only Th2 but also Th17 responses.
To our knowledge, this is the first study that addresses the effect of R-Ras on allergen-induced immune response. Although we could not demonstrate any role for R-Ras in regulating the immune response toward papainimmunization, future work utilizing IL-4-dependent, and IL-4-independent Th2 polarizing models (such as Trichurus muris and Nippostrongylus brasiliensis infections) could specify a more defined role for R-Ras in immune response

SUPPORTING INFORMATION
Additional supporting information may be found in the online version of this article at the publisher's web-site. Figure S1. Gating strategies used in flow cytometry. (A) Spleens were harvested from euthanized mice. Single cell suspension was created as described in Materials and Methods and red blood cells lysed before staining the cells for surface markers and applying them onto the flow cytometer. Cell debris was excluded and remaining cells were studied for CD3 and B220 expression, and T cells were gated as CD3 þ population (T cell) and B cells as B220 þ group (B cell). T cells were further analyzed for CD4 and CD8 surface marker expression, and CD4 þ CD8 À and CD8 þ CD4 À subpopulation were identified as CD4þ T cells (CD4þ) and CD8þ T cells (CD8þ), respectively. (B) Strategy for IL-33R gating in CD4 cells. (C) For LN DC's, single cell suspensions of LN's were prepared as in Figure 1. CD11c þ MHCII þ dendritic cells involve two subpopulations: conventional dendritic cells (convDC) show higher CD11c and lower MHCII expression while migratory dendritic cells express less CD11c but more MHCII. (D) For splenic DC's, spleens were prepared as described in Materials and Methods. Cell debris was excluded and cells were investigated for the expression of CD3 (T cell marker), B220 (B cell marker) and F480 (macrophage marker). CD3 À B220 À F480 À cells (lin.neg.) were then studied further for the expression of CD11c and MHCII. The figure shows the gating of one representative sample for each case. Figure S2. Expression of DC surface markers in WT and R-Ras KO splenic DC's in response to various stimulants. Spleens were prepared as described in materials and methods from WT (N ¼ 3) an R-Ras KO (n ¼ 3) mice. The cells were either stained directly (ex vivo) or after o/n culture with medium alone or with poly(I:C), LPS, or IL-4 as indicated. The expression of MHCII (A), CD80 (B), OX40L (C), and CD86 (D) was measured with flow cytometer.