IgE-mediated regulation of IL-10 and type I interferon enhances rhinovirus-induced Th2 differentiation by primary human monocytes

Abstract

Rhinovirus (RV) infections are linked to the development and exacerbation of allergic diseases including allergic asthma. IgE, another contributor to atopic disease pathogenesis, has been shown to regulate dendritic cell antiviral functions and influence T cell priming by monocytes. We previously demonstrated that IgE-mediated stimulation of monocytes alters multiple cellular functions including cytokine secretion, phagocytosis, and influenza-induced Th1 development. In this study, we investigate the effects of IgE-mediated stimulation on monocyte-driven, RV-induced T cell development utilizing primary human monocyte-T cell co-cultures. We demonstrate that IgE crosslinking of RV-exposed monocytes enhances monocyte-driven Th2 differentiation. This increase in RV-induced Th2 development was regulated by IgE-mediated inhibition of virus-induced type I interferon and induction of IL-10. These findings suggest an additional mechanism by which two clinically significant risk factors for allergic disease exacerbations – IgE-mediated stimulation and rhinovirus infection, may synergistically promote Th2 differentiation and allergic inflammation.

Introduction:

The link between viruses and allergic diseases has long been appreciated, and is best represented by the coinciding increase in allergic asthma exacerbations with peaks in fall respiratory virus epidemics [1]. Rhinoviruses (RV) are closely associated with allergic disease, representing the most frequently detected viruses in children with asthma exacerbations [1, 2]. Early RV infection in allergic children also increases the risk of asthma development by up to 8-fold [3]. The degree of atopy is similarly linked to disease severity; for example, serum allergen-specific IgE levels in allergic asthmatics directly correlate with virus-induced exacerbation severity [4]. Clinically, decreasing serum IgE with omalizumab therapy reduces atopic disease exacerbations [5-7]. In a recent study, children with allergic asthma treated with omalizumab had decreased RV-induced upper respiratory symptoms, shortened duration of virus shedding, and decreased peak virus titers [8], suggesting that IgE-mediated processes modulate RV infection in vivo.

Monocytes are antigen presenting cells (APCs) recruited to the airway during respiratory virus infections [9, 10] and following allergen challenge [11], whose functions are modulated by the allergic environment [12-15]. Studies indicate roles for monocytes and monocyte-derived cells in RV-induced allergic disease pathogenesis. Experimental RV-infection of asthmatic subjects, as compared to controls, resulted in exaggerated recruitment of monocytes/macrophages to airways which correlated with virus load [9]. Our group has shown that IgE crosslinking induces monocyte secretion of regulatory and pro-inflammatory cytokines, including IL-10, IL-6, and tumor necrosis factor (TNF) [12, 13]. IL-10 has been linked to allergic disease pathogenesis and virus-induced wheezing [16-18], although its role in IgE-mediated monocyte antiviral responses remains unexplored. We recently demonstrated that critical monocyte antiviral functions, including influenza-induced upregulation of antigen-presenting molecules and CD4 Th1 T cell priming are inhibited by IgE crosslinking [12]. Thus, IgE-mediated regulation of monocyte functions during RV infection may play a role in promoting allergic inflammation.

Th2-mediated responses, a hallmark of allergic disease, are implicated in RV-induced inflammation. Enhanced Th2 inflammatory responses correlate with increased disease severity in allergic asthmatics experimentally infected with RV [19]. While multiple cell types and factors likely contribute, studies support roles for both APCs and IgE in promoting Th2 responses. In cat allergic subjects, omalizumab treatment decreased ex vivo allergen-induced pDC-driven Th2 stimulation [20]; similarly, monocyte-derived DCs from allergen-exposed atopic subjects induced CD4 T cell Th2 cytokine responses [21]. These studies suggest that IgE-mediated APC stimulation may regulate Th2 priming, however the role of IgE-mediated processes in RV-induced T cell responses is largely undefined.

Type I interferons (IFN) are key mediators of antiviral responses and also negatively regulate human Th2 development and effector functions in vitro [22, 23]. IgE-mediated inhibition of virus-induced APC IFN secretion is one proposed mechanism involved in RV-associated allergic exacerbations [5, 24, 25]. IgE crosslinking inhibits influenza- and RV-induced IFN secretion from plasmacytoid dendritic cells (pDCs) and PBMCs [24-26], and omalizumab treatment in children with allergic asthma restores ex vivo antiviral IFN responses [5, 25]. Within this study, the group with the greatest post-omalizumab increases in IFN had fewer asthma exacerbations [5], suggesting that virus-induced type I IFN plays an important role in regulating allergic inflammation in vivo.

It is currently unknown how IgE regulates antiviral responses to RV and promotes Th2-mediated allergic inflammation. Given the critical role of monocytes and monocyte-derived cells in antigen presentation and T cell priming, we investigated the impact of IgE-mediated stimulation on RV-induced monocyte-driven T cell differentiation. We show that IgE-driven effects on monocytes, including inhibition of virus-induced IFN and induction of IL-10, enhance RV-induced Th2 development. These data provide evidence that IgE-mediated alteration of monocyte antiviral functions during RV infection may promote Th2 priming and enhance allergic inflammation.

Results:

IgE-mediated stimulation enhances RV-induced monocyte-driven CD4 Th2 cell development

Our previous work demonstrated multiple effects of IgE crosslinking on monocyte functions [12, 13], including inhibition of influenza-driven Th1 cell development with no observed impact on Th2 responses [12]. Therefore, we wanted to utilize our model of monocyte-driven T cell development in other contexts to investigate a role for IgE-mediated processes in monocyte-driven Th2 differentiation. We first evaluated the effect of IgE crosslinking on monocyte-driven Th development in the absence of virus-exposure. IgE crosslinking did promote a small increase in monocyte-driven Th2 (IL-4+IFNγ− cells; Fig. 1A) and Th1 (IFNγ+ cells; Fig. 1B) differentiation and proliferation (Fig. 1C), without any significant change in the final Th1/Th2 ratio (Fig. 1D). Because of the epidemiological link between RV exposure and allergic disease, we then determined how IgE-mediated stimulation impacted RV-induced naïve CD4 T cell development by primary human monocytes (Fig. 2 and Supp. Fig. 1 & 2). RV-exposure alone increased monocyte-driven Th2 (Fig. 2A) and Th1 (Fig. 2B) differentiation. More notable was the impact of IgE crosslinking plus RV exposure, which significantly enhanced Th2 differentiation (Fig. 2A). Similar to our previous findings with influenza [12], the strong Th1 response induced by RV-exposed monocytes was inhibited by IgE crosslinking (Fig. 2B, Supp. Fig. 2A-C). This effect on Th1 cell differentiation included the entire IFNγ expressing population, significantly decreasing both IFNγ single positive and IFNγ+IL-4+ co-expressing cell populations (Supp. Fig. 2A-C), similar to our prior observations with the Th1-dominant virus, influenza. In a subset of experiments, we also measured expression of another Th2-related cytokine, IL-13 (Supp. Fig. 2 D-F). IL-13 expressing T cells (IL-13+ IFNγ− and total IL-13+ population) were not significantly increased by rhinovirus-exposed monocytes in the presence or absence of IgE crosslinking (Supp. Fig. 2 E-F). Due to the overall low number of IL-13 positive cells, we may have been unable to detect small differences in this population. However, the lack of significant impact on IL-13 expression suggests a potential difference in the regulation of IL-4 and IL-13 expression in our system which is consistent with previous studies [27, 28].

Figure 1. IgE crosslinking enhances monocyte-driven T cell differentiation.

Primary human monocytes were either mock- (media alone) or treated in the presence or absence of IgE crosslinking (αIgE) or isotype control (IgG) antibodies and co-cultured with naïve allogeneic CD4+ T cells to evaluate monocyte-induced (A) Th2 (IL-4+IFNγ−) and (B) Th1 (IFNγ+) differentiation. (C) In a subset of experiments (n=5) T cells were labeled with cell proliferation dye VPD450 prior to co-culture and evaluated for cell division. (D) Th1/Th2 ratios for experiments in (A & B). (A, B, & D) Compiled data for n=15 independent donor pairs. Violin plots show mean as black line and individual donor pairs as open circles. Bar graphs show SEM. P values obtained by two-way ANOVA with post-hoc comparison and denoted as *p<0.05, **p<0.01, and “ns” p>0.05.

Figure 2. IgE crosslinking enhances RV-induced monocyte driven Th2 differentiation.

Primary human monocytes were mock- (media alone) or RV-exposed in the presence or absence of IgE crosslinking (αIgE) or isotype control (IgG) antibodies, then co-cultured with naïve CD4+ T cells to evaluate monocyte-induced T cell differentiation. RV-induced monocyte-driven (A) Th2 (IL-4+IFNγ−) and (B) Th1 (IFNγ+) development was measured in an allogeneic co-culture system, n=17 independent donor pairs. (C) Th2 development was compared between IgE crosslinking in the presence or absence of RV by student’s 2-tailed paired t-test; n=10 independent donor pairs. (D) Th1/Th2 ratios for experiments in (A & B). (E) % T cell division for a subset (n=10) donor pairs. (F) Linear regression analysis of monocyte surface FcεR1α expression versus %IL-4+IFNγ-population induced by RV-exposed (blue triangles) or RV+αIgE crosslinked (orange circles) monocytes. Pearson correlation coefficient and two-tailed p value shown, n=11 experiments. (G) RV-specific Th2 priming of autologous naïve CD4 T cells, expressed as IL-4+IFNγ− cells per 1000 CD4+ T cell events, box plots show 95% confidence intervals and mean for n=3 biological replicates. (A-C) Violin plots show the mean as a black line and independent donor pairs with open circles. (D-E) Bar graphs show SEM of compiled donor pairs. Unless noted, p values obtained by two-way ANOVA with post-hoc comparison; *p<0.05, **p<0.01, and ***p<0.001.

Given that IgE-crosslinking alone was capable of enhancing Th2 development (Fig. 1A), in a subset of experiments with sufficient cell numbers we directly compared IgE-mediated Th2 development in the presence or absence of RV exposure (Fig. 2C). Monocytes exposed to IgE crosslinking plus RV promoted greater Th2 priming than monocytes exposed to IgE crosslinking alone (Fig. 2C), indicating a role for both the virus and IgE-crosslinking in this response. The IgE-mediated enhancement of RV-driven Th2 and inhibition of Th1 differentiation significantly reduced the Th1/Th2 ratio (Fig. 2D). RV-driven T cell proliferation was also decreased by IgE crosslinking (Fig. 2E). Both Th1 and Th2 development were dependent on the presence of monocytes as naïve T cells cultured alone showed minimal differentiation (Supp. Fig. 3). Furthermore, MHC II-TCR interactions were required for monocyte-driven T cell differentiation, as MHC II blocking antibodies (αHLA-DR) significantly decreased monocyte-driven Th1 and Th2 differentiation (Supp. Fig. 4). To further support IgE- and virus-mediated effects on the monocyte, the percentage of Th2 differentiation induced by RV combined with IgE crosslinking positively correlated with baseline monocyte surface expression of the high affinity IgE receptor (FcεRIα); no correlation was observed with RV exposure alone (Fig. 2F). To investigate whether IgE crosslinking similarly impacts virus-specific Th2 priming, we co-cultured RV-exposed monocytes with autologous naïve CD4 T cells, followed by restimulation with irradiated RV-exposed autologous PBMCs. Consistent with the allogeneic system, combined exposure of monocytes to IgE crosslinking and RV enhanced CD4 T cell Th2 development (increased IL-4 expressing T cells; Fig. 2G), providing evidence of IgE-mediated effects on antigen-specific Th2 development. These data show that the combination of IgE-mediated allergic stimulation with RV-exposure significantly enhances monocyte-driven Th2 differentiation, shifting the Th1/Th2 balance towards allergic inflammation. This effect is related to monocyte IgE receptor (FcεRIα) expression, supporting a role for both the virus and IgE in monocyte-driven Th2 differentiation.

Type I interferon (IFN) negatively regulates RV-induced Th2 development

To investigate potential regulatory mechanisms underlying IgE-induced RV-driven Th2 development, we evaluated the role of type I IFN, a known negative regulator of human Th2 development [23]. IgE crosslinking significantly impaired RV-induced monocyte IFNα secretion (Fig. 3A). To specifically evaluate the effect of impaired IFN on Th2 priming, we blocked type I IFN receptor signaling (αIFNAR) which increased RV-induced monocyte-driven Th2 development (Fig. 3B). Consistent with previous findings, exogenous recombinant human IFNα reduced monocyte-driven Th2 differentiation in the absence of virus exposure [23] (Supp. Fig. 5A), and conversely enhanced Th1 priming [29, 30] (Supp. Fig. 5B) of allogeneic naïve CD4 T cells. Similar to the effects on mock-infected monocytes (Supp. Fig. 5A), addition of IFNα also decreased RV-driven Th2 differentiation (Fig. 3C). Finally, addition of IFNα reversed IgE-mediated enhancement of RV-induced Th2 cells (Fig. 3C), and partially restored IgE-mediated inhibition of the Th1 population (Fig. 3D). These data indicate that IgE-mediated inhibition of RV-induced type I IFN regulates monocyte-driven Th2 and Th1 development.

Figure 3. Type I interferon reverses IgE-mediated enhancement of RV-induced Th2 development by primary monocytes.

Monocytes were exposed to either mock or RV conditions in the presence or absence of IgE crosslinking (αIgE) or isotype control (IgG) antibody. (A) IFNα was measured in monocyte cultures at 24 h post infection (SEM for n=6 biological replicates). (B-D) Monocytes were co-cultured with allogeneic naïve CD4 T cells and (B) anti-human IFNAR (αIFNAR) or isotype control antibodies (10μg/mL) or (C-D) rhIFNα (10,000 IU/mL), added at time of co-culture. (B-C) Th2 (%IL-4+IFNγ−) or (D) Th1 (%IFNγ+) differentiation was measured. SEM is shown for n=3 independent donor pairs. All p values determined by two-way ANOVA with post-hoc comparison; *p<0.05 or **p<0.01.

IgE-induced IL-10 regulates RV-induced Th2 development

Our findings suggested an IgE-specific factor involved in monocyte-driven Th2 development given the observed increase in the Th2 population by IgE crosslinking both alone and combined with RV (Fig. 1 and 2). IgE-mediated stimulation of monocytes induces the production of IL-10 [13, 15], a cytokine known to regulate pro-inflammatory processes including virus-induced type I IFN [31] and T cell development [32-34]. To investigate the role of IL-10, we confirmed that monocytes produced IL-10 after IgE crosslinking, which was not altered by RV-exposure (Fig. 4A). Addition of exogenous rhIL-10 was sufficient to inhibit RV-induced IFNα secretion in monocyte cultures (Fig. 4B). In monocyte-T cell cocultures, rhIL-10 enhanced monocyte-driven Th2 development in both the presence and absence of virus (Fig. 4C), and suppressed RV-driven Th1 priming (Fig. 4D). Blocking IL-10 and its receptor (IL-10Rα) reversed IgE-induced enhancement of RV-driven Th2 differentiation (Fig. 4E), without affecting Th1 differentiation (Fig. 4F). IL-10 blockade in monocyte-T cell co-cultures also partially reversed IgE-mediated inhibition of IFNα (Fig. 4G). These results indicate that IgE-mediated IL-10 secretion regulates RV-induced type I IFN to modulate RV-driven Th2 cell development.

Figure 4. IgE-induced IL-10 is required and sufficient for IgE-mediated enhancement of RV-driven Th2 development.

(A) Primary human monocytes were exposed to mock or RV in the presence or absence of IgE crosslinking antibody, and IL-10 measured by ELISA analysis at 24 hours, mean and SD of technical replicates from a representative donor are shown. (B-D) rhIL-10 (10ng/mL) was added to monocyte cultures in the presence or absence of RV-exposure and (B) IFNα was measured in monocyte supernatants at 24 hours post infection by ELISA analysis, mean with SD of technical replicates for a representative donor are shown. (C-D) At 24 hours post infection monocytes were co-cultured with allogeneic naïve CD4 T cells to measure either (C) Th2 (%IL-4+IFNγ−) or (D) Th1 (%IFNγ+) T cell differentiation; SEM shown for n=4 independent donor pairs. (E-G) IL-10/IL-10Rα blocking (αIL-10) or isotype control antibodies (5μg/mL each) were added to RV-exposed monocyte-T cell co-cultures and T cells evaluated for (E) Th2 or (F) Th1 differentiation; SEM shown for n=4 independent donor pairs. (G) IFNα was measured in co-culture supernatants at day 3 by ELISA, mean and SD of technical replicates from a representative donor are shown. p values by (A, B, G) one-way or (C-F) two-way ANOVA with post-hoc comparison; *p<0.05, **p<0.01 and “ns” p>0.05. All experimental results were reproduced with a minimum of 3 independent experiments.

Discussion:

This study shows that IgE-mediated stimulation of monocytes enhances RV-driven Th2 differentiation of naïve CD4 T cells, which is regulated by IgE-induced IL-10 production and inhibition of virus-induced type I IFN. In our proposed model (Figure 5), the combination of IgE-mediated allergic stimulation and RV exposure leads to enhanced Th2 differentiation via IgE-induced effects on monocytes. In the absence of IgE-mediated signals, Th2 differentiation is regulated by viral-induced type I IFN, resulting in a Th1-predominant antiviral response. However, in the setting of IgE-mediated stimulation, IgE-induced IL-10 production inhibits monocyte IFN responses to RV infection which combined with other unknown IgE-mediated factors, such as altered expression of co-stimulatory molecules, cytokines, chemokines, etc., allows for increased Th2 differentiation. In vivo, allergic inflammation could be enhanced during infection due to the loss of IFN-induced negative regulation on Th2 cell development.

Figure 5. Model of IgE-mediated enhancement of RV-induced monocyte-driven Th2 development.

IgE-mediated allergic stimulation combined with RV exposure promotes Th2 priming by monocytes. IgE-induced effects including IL-10 production inhibit RV-driven type I IFN production by monocytes. The reduction of type I IFN-mediated negative regulation of Th2 development combined with other unknown Th2-promoting factors, such as co-stimulatory receptors or cytokines, drive increased Th2 development. This altered antiviral response may contribute to enhanced allergic inflammation.

A limitation of this study is that we did not characterize the clinical atopic status of the subjects. Our previous work has shown that monocytes isolated from both healthy and atopic donors with a wide range of serum IgE will respond to the IgE crosslinking treatment utilized in this experimental system [12, 13]. Although the impact of clinical allergic disease and levels of serum or monocyte surface-bound IgE were not evaluated, we did measure monocyte FcεRIα expression which is known to positively correlate with serum IgE levels and is an established surrogate marker for atopy [35]. The correlation between monocyte surface FcεRIα expression and IgE-enhanced RV-driven Th2 differentiation (Fig. 2F), and lack of similar correlations with either RV alone (Fig. 2F) or IgE crosslinking in the absence of virus (data not shown), suggests that atopic status may influence IgE-induced antiviral Th2 priming in vivo. Our observation that IgE crosslinking enhanced RV-specific Th2 priming in an autologous model suggests that exposure to relevant allergens in atopic individuals could promote increased RV-induced Th2 inflammatory responses during infection. Further investigation is needed to determine how this pathway is impacted in healthy versus atopic subjects and by age – children versus adults.

These data have implications regarding RV-induced development and exacerbation of allergic diseases. Monocytes recruited to the airway epithelium likely encounter allergen-induced IgE crosslinking and virus exposure simultaneously. IgE-mediated local monocyte IL-10 production and inhibition of type I IFN via feedback mechanisms on monocytes and paracrine effects on airway epithelial cells and other APCs could significantly impact local inflammatory responses, potentially enhancing virus replication, virus-induced tissue destruction, and virus-induced Th2 inflammation. Our study begins to address the synergistic role of IgE and RV, two signals linked to allergic disease development and pathogenesis. The differences induced by IgE crosslinking and virus exposure of monocytes, alone versus in concert, demonstrate the importance of evaluating the combined effects of these stimuli on cellular functions. While IgE crosslinking alone promoted Th2 differentiation, an increase in Th1 development and T cell proliferation was also observed in our study. Consistent with this, others have reported that IgE-mediated effects on dendritic cells promote antigen-specific Th2 priming and T cell proliferation [36], and supports previous work demonstrating a role for Th1 cells in allergic airway inflammation [37, 38]. By evaluating Th1 priming in both the presence and absence of RV, we have established a possible role for IgE-mediated inhibition of virus-induced type I IFN in regulating both Th1 and Th2 responses (Fig. 3).

Deficient antiviral responses, specifically type I IFN, have been repeatedly implicated in the pathogenesis of allergic disease. Enhancement of antiviral IFN responses has a potential therapeutic role in allergic diseases with multiple approaches to augment antiviral pathways being investigated [39-41]. In a clinical trial of inhaled IFNβ to prevent virus-induced asthma exacerbations, treatment decreased asthma symptom scores and promoted faster recovery of morning peak expiratory flow in subjects with moderate/severe disease - most of whom were moderate in severity. Interestingly, IFNβ treatment also enhanced antiviral gene transcription in sputum cells obtained from study participants [41]. In vitro, type I IFN is a robust negative regulator of human Th2 function including both naïve Th2 development [23] and memory effector functions [23, 27, 28, 42]. In RV-exposed PBMCs from healthy subjects in the absence of allergic stimulation, blockade of type I IFN increased virus-induced memory Th2 cytokine production while exogenous addition of IFNβ decreased production [42]. Similarly, in our system, complementation of the IgE-induced IFN deficit with rhIFNα treatment fully reversed IgE-enhanced RV-driven Th2 differentiation and partially restored impaired virus-induced Th1 responses by monocytes (Fig. 3). The capacity of IFN to regulate allergic inflammation supports its potential value as a therapeutic target in atopic diseases.

The role for IL-10 in allergic disease is more complex[17, 18]; it has been shown to play a role in allergen tolerance [43] and virus-induced wheezing [16]. Atopic individuals have elevated serum IL-10 [44] and increased IL-10 producing monocytes capable of promoting Th2 cytokine production in T cells ex vivo [17]. In our system, IL-10 was both required and sufficient to enhance RV-driven Th2 development, however IL-10 alone did not increase Th2 priming above IL-10 treated monocytes in the absence of virus (Fig. 4C). This is in contrast to our observations comparing IgE crosslinking in the presence and absence of RV-exposure (Fig 2C) where, despite both conditions containing IL-10 (Fig. 4A), Th2 priming by RV combined with IgE crosslinking was greater than IgE crosslinking alone. This could be due to a temporal relationship between IgE-mediated IL-10 production or another undefined, IgE-mediated effect on monocytes (Fig. 5). Contribution of additional IgE-mediated factor(s) is supported by our finding that although rhIL-10 treatment, like IgE crosslinking, suppressed RV-driven Th1 priming (Fig. 4D), IL-10 blockade was not sufficient to reverse IgE-mediated Th1 inhibition (Fig. 4F) which was consistent with our studies evaluating IgE-mediated inhibition of influenza-driven Th1 priming [12]. Because IL-10 often requires additional factors to modulate APC-driven T cell differentiation [32, 33], IgE-induced effects such as modulation of monocyte antigen-presenting molecules [12], expression of co-stimulatory molecules that enhance Th2 priming [32, 45], or other cytokines [12, 13, 46], combined with the loss of antiviral IFN likely contribute to IgE-mediated Th1 and Th2 regulation. While our data demonstrate a cellular mechanism by which IgE-induced IL-10 inhibits antiviral IFN responses to promote virus-induced Th2 priming, additional investigation is needed to elucidate additional factors involved in the complex interaction between IgE-mediated signals and RV-induced monocyte-driven T cell development.

To our knowledge, this is the first report that IgE-mediated stimulation promotes RV-induced naïve Th2 development by human monocytes and that impaired monocyte IFN responses may specifically contribute to Th2-mediated inflammation. While IL-10 has been shown to inhibit virus-induced type I IFN responses in human PBMCs [31], our results support a previously undescribed role for IL-10 in the inhibition of virus-induced IFN responses to promote Th2 inflammation. Finally, we demonstrate that IFNα can reverse IgE-mediated induction of antiviral Th2 differentiation in human cells. Given the link between RV infection and allergic disease development and exacerbation, and the critical role of monocytes in airway immune responses, this study provides another potential mechanism by which IgE-mediated activation synergizes with RV infection to drive allergic inflammation. Identification of additional molecular mechanisms underlying these processes could lead to potential therapeutic targets for improved treatment and prevention of viral exacerbations of allergic diseases.

Materials and methods:

Human Subjects

Primary human immune cells were isolated from 2 sources: leukocyte-enriched blood samples from unknown donors acquired through a local blood bank and blood samples from healthy volunteers. Unknown donors were deemed healthy enough for routine blood donation by the local blood bank, no demographic or clinical information (i.e. atopic status) were obtained from these unknown donors. Studies were approved by the UT Southwestern institutional review board (Study #STU 122010-139). For known volunteers, written informed consent and assent were obtained.

Reagents and Media

Phosphate Buffered Saline (PBS) with fetal bovine serum (FBS) and EDTA (complete PBS, cPBS) and complete Roswell Park Memorial Institute Medium 1640 (cRPMI) were prepared as previously described [13]. Rabbit anti-human IgE (αIgE) was purchased from Bethyl Laboratories (Montgomery, TX) and rabbit whole IgG control antibody (IgG) from Jackson ImmunoResearch (Westgrove, PA). Unless specified, recombinant human (rh) cytokines, anti-human cytokine and receptor antibodies were purchased from R&D Systems (Minneapolis, MN).

Purification of Immune Cells from Blood

Monocytes and naïve CD4⁺ T cells were purified by negative selection as previously described [12, 13] using the EasySep^™ Human Monocyte Enrichment or Human Naïve CD4+ T cell isolation kits respectively, per manufacturer’s instructions (STEMCELL Technologies, Vancouver). Purity was assessed by flow cytometry; monocytes were identified as CD14+ and naïve CD4+ T cells as CD3+CD4+CD45RA+. Samples with <85% purity were excluded.

Monocyte Culture Conditions

Monocytes were cultured in cRPMI with recombinant human (rh) M-CSF (1 ng/ml) at a concentration of 1 x 10⁶ cells/ml. As a surrogate for IgE-mediated allergic stimulation, we utilized a previously published method to crosslink cell surface bound IgE using a rabbit anti-human IgE polyclonal antibody (αIgE; Bethyl Laboratories, Montgomery, TX) [12, 13, 26, 47]. This takes into account that IgE is bound to surface FcεRI receptor on monocytes (via the FcεRIα subunit). While all individuals have surface bound IgE on monocytes, the level is related to serum IgE concentrations and thus varies across subjects [48-50]. Even with this variation, we have previously shown that this method of direct crosslinking induces reproducible IgE-mediated phenotypes in monocytes such as cytokine release [13] and inhibition of influenza-induced Th1 differentiation [12]. To best simulate the in vivo environment, monocytes were treated immediately ex vivo after isolation to ensure adequate crosslinking of native surface bound IgE and minimize any downregulation of surface receptors. Cells were treated with the following: αIgE (10 μg/ml), rabbit IgG isotype control (10 μg/ml), or IL-10 (10 ng/ml), anti-IL-10 and anti-IL-10 receptor (IL-10Rα) (5μg/mL each), or mouse IgG isotype controls (total of 10μg/mL mouse IgG concentration) were added to monocyte cultures at the start of culture. Purified rhinovirus A serotype 16 (RV) (kindly provided by Jim Gern and Yury Bochkov, U Wisconsin) was added at a multiplicity of infection (MOI) 10; a dose chosen based on published studies of RV infection of monocytes [51-53]. Cultures were incubated at 37°C for indicated times. Cells and supernatants were harvested for further analysis or for T cell co-culture experiments.

Monocyte-T Cell Co-Cultures

Monocytes were cultured as above in the presence or absence of αIgE, control IgG, or RV at 37°C for 18-24 hours and then mixed 1:1 with allogeneic naïve CD4⁺ T cells. Cells were cultured in media containing 50 international units per mL (IU/mL) rhIL-2. For select experiments, mouse anti-human antibodies against IL-10, IL-10Rα, IFNAR, and MHC II (αHLA-DR, BD Biosciences), or matched IgG1 or IgG2b isotype controls were added at 5-10μg/mL each for a total mouse IgG concentration of 10 μg/ml prior to co-culture with T cells. All antibodies were obtained from R&D Biosciences unless otherwise specified. As indicated, recombinant human (rh) IFNα2a (1000-10,000 IU/mL) was added at the initiation of co-culture. After 3 days of co-culture, T cells were expanded 1:5 vol:vol in media with IL-2 and cultured for an additional 3-4 days. Cells were rested overnight in media without IL-2 and restimulated for 6 hours with phorbol 12-myristate 13-acetate (PMA, 50 ng/ml, Sigma, St. Louis, MO), ionomycin (1μM, ThermoFisher, Grand Island, NY), and monensin (1x, BioLegend, San Diego, CA) to evaluate intracellular cytokine expression by flow cytometry analysis as previously described [12]. In a subset of experiments to evaluate cell proliferation, T cells were labeled prior to culture with the cell proliferation dye, VPD450, per manufacturer’s instructions (BD Biosciences, San Jose).

To evaluate virus-specific monocyte-driven Th2 development, autologous PBMCs, monocytes, and naïve CD4 T cells were isolated from known donors. Monocytes were mock treated (media alone) or exposed to RV with and without IgE crosslinking or isotype IgG control. At 24 hours, monocytes were co-cultured at 1:1 ratio with autologous naïve CD4 T cells in U-bottom 96-well plates in media with rhIL-2; cells were fed fresh media with IL-2 on days 6 and 9. On day 12, cells were expanded (1:10 vol:vol; 2mL) into 24-well plates in media with 10 IU/mL IL-2 for an additional 4 days. Cells were then rested in media without IL-2 overnight prior to restimulation. Frozen autologous PBMCs were thawed and exposed to RV (MOI=5) for 48 hours and x-ray irradiated (dose of 13Gy; 450 seconds at 250kV potential, 15mA, XRAD 320 model) prior to T cell restimulation. Irradiated PBMCs were washed, counted and added to primed naïve CD4 T cells at a 1:1 ratio for 24 hours. Monensin (golgi block) was added for the final 6 hours to retain cytokines intracellularly. Cells were fixed in 2% paraformaldehyde and stained for flow cytometry analysis with the following antibodies: CD4 APC, HLA-DR FITC, IL-4 PE, and IFNγ PE-Cy7. At least 40,000 events were acquired and primed T cells were identified as CD4+ HLA-DR low/intermediate and quantified based on cytokine expression (IL-4 or IFNγ single positive). Irradiated PBMCs cultured alone were used as a control to confirm negligible cytokine expression by the PBMCs (data not shown). Cytokine expressing cells were normalized as the total number of cells per 1000 CD4+ events.

Flow Cytometry Analysis of Surface Antigens

The following fluorochrome-conjugated anti-human antibodies or molecules were used: CD3-peridinin chlorophyll protein complex (PerCP), CD4-allophycocyanin (APC), CD14-V450, CD45RA-phycoethrin (PE), and HLA-DR-FITC (BD Pharmingen, San Diego, CA). To evaluate surface IgE receptor expression, monocytes were stained with FcεRIα-fluorescein (FITC, AER-37 (CRA-1), eBioscience). The AER-37 (CRA-1) clone was chosen as it does not compete with IgE binding and thus can detect total FcεRIα receptor levels (manufacturer’s technical information)[54]. Staining and flow cytometry were performed as previously described [13]. Samples were acquired on a BD LSR II flow cytometer (BD Biosciences) and analyzed with the FlowJo 8 software (FLOWJO, LLC, Ashland, OR).

Flow Cytometry Analysis of Intracellular Cytokines

The following antibodies were used: IFNγ-PE-Cy7, IL-4-PE (BD Biosciences, San Jose, CA), and IL-13-APC (Biolegend, San Diego, CA). Cells were fixed in 2% paraformaldehyde and permeablized with 0.1% saponin in PBS to detect intracellular antigens. Samples were acquired and analyzed as above described. Gating of cytokine-expressing cells was guided by staining negative control (unstimulated cells) and positive control T cells differentiated under Th1- or Th2-inducing conditions [23]. Percentages of total cytokine-positive populations were then determined. Th2 cells were identified as IL-4 single positive; Th1 cells were identified as the total IFNγ population. IL-13 expression was detectable and measured in a subset of experiments (Supp. Fig. 2D-F). Due to low numbers of IL-13+ cells in our system, likely due to the short timeframe of differentiation (7 days), we chose to focus on the IL-4 expressing population as representative of Th2 development. To measure cell proliferation, divided cells were identified as VPD450 low, as compared to unstimulated naïve T cells. Representative flow cytometry plots are shown in Supplemental Figures 1 and 2.

Cytokine Secretion Analysis

Supernatants were stored at −80°C until use. The following ELISA kits were used according to manufacturer recommendations: ELISA Max human ELISA for IL-10 (Biolegend, San Diego, CA), anti-pan human IFN alpha (Interferon Source, Cincinnati, OH). Absorbance was measured on a Biorad iMark microplate reader (Biorad, Hercules, CA) per manufacturer’s instructions.

Data Analysis and Statistics

All experiments were performed with a minimum of 3 biological replicates along with technical replicates for each assay. Data are presented as either compiled data of all biological replicates or as representative donor data demonstrating results consistent with those of the other biological replicates. Data are presented as means ± standard error of the mean (SEM) or standard deviation (SD). In experiments containing 3 or more conditions, one-way or two-way (for experiments with multiple donors) repeated measures ANOVA and pairwise Tukey’s post hoc comparisons were performed. For experiments comparing 2 conditions, paired t tests were performed with Holm-Sidak correction where appropriate. p<0.05 was considered significant, pertinent p values are denoted as follows: *p<0.05, **p<0.01, and ***p<0.001, or ns (non-significant) for p>0.05. All statistical analyses were performed using GraphPad Prism version 8.

Abbreviations:

APC

antigen presenting cell

αIgE

Rabbit anti-human IgE antibody

FcεRI

high-affinity IgE receptor

IFN

interferon

IgG

Rabbit whole IgG isotype control antibody

IL

interleukin

PBMCs

peripheral blood mononuclear cells

pDC

plasmacytoid dendritic cells

rh

recombinant human

RV

Rhinovirus

Th

CD4 helper T cell

IFNAR

Interferon α/β receptor

TCR

T cell receptor