IL-13Rα2 Regulates the IL-13/IFN-γ Balance during Innate Lymphoid Cell and Dendritic Cell Responses to Pox Viral Vector-Based Vaccination

We have shown that manipulation of IL-13 and STAT6 signaling at the vaccination site can lead to different innate lymphoid cell (ILC)/dendritic cell (DC) recruitment, resulting in high avidity/poly-functional T cells and effective antibody differentiation. Here we show that permanent versus transient blockage of IL-13 and STAT6 at the vaccination site can lead to unique ILC-derived IL-13 and IFN-γ profiles, and differential IL-13Rα2, type I and II IL-4 receptor regulation on ILC. Specifically, STAT6−/− BALB/c mice given fowl pox virus (FPV) expressing HIV antigens induced elevated ST2/IL-33R+ ILC2-derived IL-13 and reduced NKp46+/− ILC1/ILC3-derived IFN-γ expression, whilst the opposite (reduced IL-13 and elevated IFN-γ expression) was observed during transient inhibition of STAT6 signaling in wild type BALB/c mice given FPV-HIV-IL-4R antagonist vaccination. Interestingly, disruption/inhibition of STAT6 signaling considerably impacted IL-13Rα2 expression by ST2/IL-33R+ ILC2 and NKp46− ILC1/ILC3, unlike direct IL-13 inhibition. Consistently with our previous findings, this further indicated that inhibition of STAT6 most likely promoted IL-13 regulation via IL-13Rα2. Moreover, the elevated ST2/IL-33R+ IL-13Rα2+ lung ILC2, 24 h post FPV-HIV-IL-4R antagonist vaccination was also suggestive of an autocrine regulation of ILC2-derived IL-13 and IL-13Rα2, under certain conditions. Knowing that IL-13 can modulate IFN-γ expression, the elevated expression of IFN-γR on lung ST2/IL-33R+ ILC2 provoked the notion that there could also be inter-regulation of lung ILC2-derived IL-13 and NKp46− ILC1/ILC3-derived IFN-γ via their respective receptors (IFN-γR and IL-13Rα2) at the lung mucosae early stages of vaccination. Intriguingly, under different IL-13 conditions differential regulation of IL-13/IL-13Rα2 on lung DC was also observed. Collectively these findings further substantiated that IL-13 is the master regulator of, not only DC, but also different ILC subsets at early stages of viral vector vaccination, and responsible for shaping the downstream adaptive immune outcomes. Thus, thoughtful selection of vaccine strategies/adjuvants that can manipulate IL-13Rα2, and STAT6 signaling at the ILC/DC level may prove useful in designing more efficacious vaccines against different/chronic pathogens.

The ILC are lineage negative cytokine-producing cells, which neither express lymphoid differentiation lineage markers nor T or B cell receptors. ILCs are generally divided into three distinct subsets, ILC1, ILC2, and ILC3, based on expression of cytokines, phenotypic markers, and transcription factors. Specifically, ILC2 are characterized by tissue specific surface expression of ST2/IL-33R (lung), IL-25R (muscle), or thymic stromal lymphopoietin receptor (TSLPR) (skin), and cytokines IL-4, IL-5, and IL-13, plus transcription factor GATA3 [24][25][26][27]. ILC1 and ILC3 are defined by the expression of NKp46, and their IFN-γ, IL-22, and IL-17A production capacity and linked to transcription factors T-bet and RORγt [27]. However, several studies have shown that the ILC populations can be highly plastic according to different cell/tissue milieus [24,25,28,29]. Recent studies in our laboratory have shown that following viral vector-based vaccination, ILC2 was the major source of IL-13 at the vaccination site 24 h post-delivery [24,30]. This was also linked to differential DC recruitment and downstream adaptive immune outcomes, which were also route dependent [24,[31][32][33][34][35][36], suggesting that the optimal balance/regulation of IL-13 at the first line of defense was likely crucial for cell homeostasis and immune regulation.
Recently, we have designed two poxviral vector-based HIV vaccines that transiently manipulate IL-13 and IL-4 activity at the vaccination site to improve vaccine efficacy [32,33]. The IL-13Rα2 adjuvanted vaccine co-expressed HIV antigens together with soluble IL-13Rα2, which transiently sequestered IL-13 at the vaccination site [24,32]. The IL-4R antagonist adjuvanted vaccine co-expressed HIV antigens together with C-terminal deletion mutant of the mouse IL-4, lacking the essential tyrosine required for signaling. This antagonist was able to bind to both type I and type II IL-4 receptor complexes and transiently block both IL-4 and IL-13 signaling via the STAT6 pathway [33]. In a prime-boost modality, these vaccines were able to induce high avidity/poly-functional HIV specific mucosal and systemic CD4/CD8 T cells with improved protective efficacy, both in mice and macaques [31][32][33][34][35]. Moreover, in the context of humoral immunity, the IL-4R antagonist adjuvant vaccine was able to induce effective HIV gag-specific IgG1 and IgG2a differentiation in mice, unlike IL-13Rα2 adjuvanted vaccine [33]. To further confirm the role of IL-13 in antibody differentiation, when a cohort of knockout mice were vaccinated with unadjuvanted prime-boost strategy, although IL-4 −/− and STAT6 −/− animals showed enhanced IgG2a, IL-13 −/− mice showed extremely low IgG2a antibody responses [37]. More interestingly, when STAT6 −/− mice were given the IL-13Rα2 adjuvanted vaccine, elevated IgG1 and low IgG2a antibody responses were observed, similarly to the IL-13 −/− mice given the unadjuvanted vaccine [37]. These observations clearly indicated that the presence of IL-13 at the vaccination site was critical for effective antibody differentiation, and an STAT6 independent pathway was involved in this process, likely associated with IL-13Rα2 [37]. Interestingly, IFN-γ is also known to play an important role in antibody differentiation [38][39][40][41][42]. Moreover, during inflammation, IFN-γ has shown to inhibit ILC2 activation and IL-5 production [43], and similarly, under airway hyperreaction and asthma conditions IFN-γ has also shown to directly inhibit ILC2 function [44,45]. Recently we have also shown an interesting association of IL-13Rα2 and IFN-γR with different DC subsets under different IL-13 conditions [14]. However, the relationship between IL-13 and IFN-γ at the ILC/DC level in the context of viral vector-based vaccination, and the molecular mechanism by which ILC2-derived IL-13 regulates ILC1/ILC3 or DC activity, remain elusive. Therefore, in this study wild type (WT) BALB/c, IL-13, and STAT6 gene knockout mice on BALB/C background were vaccinated with unadjuvanted (FPV-HIV) and WT BALB/c mice with IL-13Rα2 or IL-4R antagonist adjuvanted viral vector-based vaccines (two models which cause enduring and transient inhibition of IL-4/IL-13/STAT6, respectively), to unravel the IL-4/IL-13 receptor regulation mechanisms on ILC and DC under different IL-13 conditions.

Materials and Methods
Mice: 5-6 week old female wild type (WT) BALB/c, IL-13 −/− , and STAT6 −/− mice on BALB/c background were obtained from the Australian Phenomics Facility, the Australian National University.
Ethics Statement: All animals were maintained and experiments performed in accordance with the Australian National Health and Medical Research Council (NHMRC) guidelines, within the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. The animal ethics were approved by The Australian National University's Animal Experimentation and Ethics Committee (AEEC). Protocol numbers A2014/14, and A2017/15.
Preparation of lung lymphocytes: The mice were euthanized using cervical dislocation according to the approved AEEC guidelines. Lung tissues were removed and kept in complete RPMI medium (Sigma) on ice until processing. Single cell lung suspensions were prepared as described previously [24,32]. Specifically, the lung tissues were first cut into small pieces, and then enzymatically digested in 1 mL of digestion buffer containing 1 mg/mL collagenase (Sigma-Aldrich, St Louis, MO, USA), 1.2 mg/mL Dispase (Gibco, Auckland, New Zealand), and 5 Units/mL DNase (Calbiochem, La Jolla, CA, USA) in complete RPMI. During digestion, samples were gently vortexed every 10 min and incubated in a 37 • C water bath for 45 min. The digested lung tissues were mashed and passed through a 100 µm Falcon cell strainer and the resulting lung cell suspensions were centrifuged for 15 min at 1500 RPM (524× g) at 4 • C using a Beckman ALLEGRA X-12R centrifuge. Next, the supernatants were removed, cells were resuspended in 5 mL red blood cell lysis buffer (at room temperature) containing 0.16 mM NH4Cl and 0.17 M Tris HCl (pH 7.65) for 3 min at room temperature, 30 mL of complete RPMI medium was added, and centrifuged at 1500 RPM (524× g) for 5 min at 4 • C. Cells were washed once more with complete RPMI and passed through sterile gauze to remove any remaining debris, followed by two washes in complete RPMI medium, and the cell pellets were then resuspended in 0.5 mL complete RPMI medium, counted using a hemocytometer (Tiefe Depth Profondeur 0.100 mm), and stored on ice until use.
Statistical analysis: In this study, cell numbers were calculated using the formula (number of receptor or cytokine expressing cells/number of CD45 + cells) × 10 6 . Results are represented as a percentage of CD45 + cells, and can also be found in the supplementary figures ( Figure S7). IL-4 and IL-13 receptor proportions on lung DCs were calculated as a percentage of parent MHC-II + CD11c + CD11b + CD103 − cDC and MHC-II + CD11c + CD11b − B220 + pDC population. Note that less than 10 cells expressing the receptor was set as a cut-off. Statistical analysis was performed using GraphPad Prism software (version 6.05 for Windows). One-way ANOVA using Tukey's multiple comparisons test and unpaired t-test were used. The p-values are denoted as: ns-p ≥ 0.05, *-p < 0.05, **-p < 0.01. ***-p < 0.001, ****-p < 0.0001. All experiments were repeated at least three times.

Vaccination under Transient or Permanent Inhibition of IL-13 or STAT6 Signaling Did Not
Modulate IL-13Rα2 Expression on NKp46 + ILC1/ILC3, Unlike NKp46 − ILC1/ILC3 When IL-4/IL-13 receptors on NKp46 + ILC1/ILC3 were accessed under transient versus permanent inhibition of IL-13 or STAT6 signaling, surprisingly there were no major differences in the number of NKp46 + ILC1/ILC3 expressing IL-13Rα2 (even though the number of cells expressing the receptor was lower in the KO mice compared to the BALB/c given the unadjuvanted vaccine, p < 0.5) (Figure 6a). In contrast, the number of NKp46 + ILC1/ILC3 expressing γC, IL-4Rα, and IL-13Rα1 (Figure 6b,c) was found to be significantly lower under permanent STAT6 or IL-13 inhibition (KO mice) compared to transient inhibition. Interestingly, both transient blockage of STAT6 (FPV-HIV-IL-4R antagonist vaccination of wild type mice) and IL-13 (FPV-HIV-IL-13Rα2 vaccination of wild type mice) showed a significantly elevated number of NKp46 + ILC1/ILC3 expressing γC, IL-4Rα, and IL-13Rα1 compared to BALB/c mice given the unadjuvanted FPV-HIV vaccine (Figure 6b,c). (For data presented in the form of percentage of CD45 + cells, please see Figure S7f).  (s.d.). The p values were calculated using one-way ANOVA using Tukey's multiple comparisons test and unpaired t-test. * p < 0.05, ** p < 0.01, **** p < 0.0001. Experiments were repeated a minimum of 3 times.

rFPV and rVV Vaccinated Lung cDCs and pDC Exhibited Uniquely Differential IL-4/IL-13 Receptor Expression Profiles 24-72 h Post-Delivery
We have previously shown that the nature and replication status of a viral vector can significantly alter the ILC2-derived IL-13 level at the vaccination and can modulate lung DC recruitment [30]. Moreover, a low IL-13 environment at the vaccination site can recruit enhanced cDC leading to CD8 + T cells of higher avidity [31,32], whist IL-13 was necessary for effective antibody differentiation [33,51]. Knowing that pDCs play a role in effective antibody differentiation [52,53], in this study IL-4/IL-13 receptor regulation on cDCs and pDCs were also assessed 24-72 h post rFPV and rVV vaccination (high and low IL-13 conditions), as described in Figure S8. The results indicated that compared to rFPV, known to induce low ILC2-derived IL-13 [24,30], rVV induced considerably elevated ILC2-derived IL-13 at the lung mucosae by an ST2/IL-33R − ILC subset, 24 h post vaccination (p < 0.0001) (Figure 7a). There was significant regulation of the different IL-4/IL-13 receptors on both cDC and pDC at early stages of vaccination. Interestingly, although the percentage of cDCs expressing IL-13Rα2 was much greater 24-48 h (90%) compared to 72 h post rFPV delivery (~80%) (p < 0.0001) (Figure 7b), the IL-4Rα and IL-13Rα1 on cDC were significantly up-regulated only after 48 h (24 vs. 48 h and 24 vs. 72 h p < 0.0001) (Figure 7b). In contrast, post rVV vaccination, significantly elevated and sustained IL-13Rα2 expression (99%) was detected throughout the time course (Figure 7c), whilst the IL-13Rα1/IL-4Rα expression trends were very similar to rFPV vaccination (Figure 7c). Notably, at these time points γC receptor, which forms the IL-4 type I receptor complex (IL-4Rα and γC), was not expressed or regulated at 72 h post vaccination ( Figure S9).
IL-13Rα2 densities, 24 to 72 h post rFPV vaccination, were down regulated ( Figure S10a), whilst the opposite was observed with IL-13Rα1 and IL-4Rα ( Figure S10b,c). In contrast, post rVV vaccination down-regulation of both IL-13Rα2 and IL-13Rα1 densities at 48 h (24 vs. 48 h p < 0.0001), followed by an up-regulation at 72 h, comparable to 24 h ( Figure S10d,e) and a gradual but significant increase in the IL-4Rα densities, were detected over time (24 vs. 48 h p = 0.0127, 48 vs. 72 h and 24 vs. 72 h p < 0.0001) ( Figure S10f). Interestingly, on cDC the IL-13Rα2 densities were approximately ten times greater than IL-13Rα1 and IL-4Rα.
The IL-13Rα2 expression on pDCs post rFPV vaccination was found to be in the order of (24 > 48 < 72 h) ( Figure S12b). Interestingly, the density of IL-13Rα2 on rVV vaccinated pDC was approximately 10 times greater than that of IL-13Rα1 and IL-4Rα. , and p values were calculated using one-way ANOVA followed by Tukey's multiple comparison test (black lines) and unpaired non-parametric student's t test (grey lines). * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Experiments with each vector were repeated minimum a 2-3 times.
ties, under IL-13 −/− conditions, IL-4 expressed by other cells at the lung mucosae may interact with type I and/or II IL-4 receptor complexes on ILC2 to compensate for the loss of IL-13 in the milieu to regulate the ILC/cytokine balance at the vaccination site (e.g., IL-13/IFN-γ). The above findings further corroborated our notion that at the early stages of vaccination, IL-13 is likely the master sensor/regulator of lung ST2/IL-33R + ILC2 and NKp46 − ILC1/ILC3 activity/function. IL-13Rα2 were co-regulated. Moreover, the ILC2-derived IL-13 and NKp46 − ILC1/ILC3-derived IFN-γ were also inter-regulated by their respective receptors (IFN-γR and IL-13Rα2) present on these ILCs. Taken together our findings indicate that at the early stages of vaccination, IL-13 is the master regulator of different ILC subsets. However, whether there is regulation of IFN-γ via IFN-γR on ILC1/ILC3 warrants further investigation. Moreover, see Figure S8 for STAT6 and IL-13 transient versus permanent inhibition scenarios. (b) Under low (rFPV) and high (rVV) IL-13 conditions DCs are differentially regulated. Under low IL-13 conditions, IL-13 signals via the high affinity receptor IL-13Rα2 on pDCs and cDCs, inducing effective B cell (e.g., differentiated antibody responses) as well as high avidity/polyfunctional T cells responses [32,33,36,37]. Under high IL-13 conditions, IL-13 signals via the low affinity Type II IL-4 receptor IL-13Rα1/IL-4Rα complex, on pDCs and cDCs, while IL-13Rα2 sequesters/regulates excess IL-13 at the vaccination site maintaining homeostasis, resulting in effective B cell but not T cell outcomes.
On NKp46 + ILC1/ILC3s the expression patterns of IL-4/IL-13 receptors were vastly different compared to the other two ILC subsets. The stable IL-13Rα2 expression on NKp46 + ILC1/ILC3 under the transient and control vaccination conditions indicated that these cells were not involved in regulation of ILC2 and NKp46 − ILCs, but were likely associated with maintenance of IL-13 homeostasis at the lung mucosae, similar to what has been observed in lung DC under high IL-13 conditions [14], and under IL-13 mediated chronic inflammatory conditions [60,61]. Furthermore, the low IFN-γ expression and minimal regulation of IL-13Rα2 at the early stages of intranasal viral vector vaccination was suggestive of the noninvolvement NKp46 + ILC1/ILC3s in the regulation of lung ST2/IL-33R + ILC2.
Unlike rFPV associated with low ILC2-derived, rVV vaccination, which induced significantly elevated ILC2-derived IL-13 (the highest compared to all previously tested viral vectors) [30], showed elevated expression of the high affinity IL-13 receptor IL-13Rα2 on lung DCs 24-72 h post-delivery, including significant up-regulation of the low affinity Type II IL-4Rα/IL-13Rα1 complex at 48-72 h. These receptor regulation patterns once again provoked the notion that under high IL-13, IL-13Rα2 likely sequestered excess IL-13 (noting that rVV is a replicating vector), whilst signaling took place via the low affinity IL-4Rα/IL-13Rα1 complex (which works under high IL-13 conditions) (Figure 8). These findings are highly consistent with our recent observations, where low and high IL-13 conditions showed differential regulation of IL-13Rα2 on DC [14]. These uniquely different early events in the innate immune compartment may explain "how and why" (i) in a primeboost vaccination modality, rFPV prime can generate high avidity T cells, unlike rVV [62]; and (ii) the order of vector delivery significantly impacts vaccine-specific adaptive immune outcomes [63,64]. Moreover, the observed IL-13/IL-13Rα2 regulation patterns on ILC and DC at the vaccination site may explain why a more attenuated and unrelated viral vector to the host may help induce a higher quality vaccine-specific T cell immunity. Specifically, why rFPV and its relative, canarypox virus prime modalities, may have the capacity to induce more effective immune outcomes than other pox viral vectors [34,[63][64][65], given that priming creates the initial antigen-specific T cell population, which gets expanded during the booster vaccination [32,33].

Conclusions
Collectively, our findings reveal that 24 h post intranasal viral vector vaccination, lung ILC2-derived IL-13 is regulated in an autocrine fashion via IL-13Rα2, and that most likely there is inter-regulation of ILC2-derived IL-13 and NKp46 − ILC1/ILC3-derived IFN-γ by their respective receptors (IFN-γR and IL-13Rα2) present on ILC ( Figure 8). Specifically, at the early stages of viral vector vaccination: (i) IL-13 is likely the master regulator of ILC2 and NKp46 − ILC1/ILC3, as well as DC, and responsible for shaping the downstream adaptive immune outcomes (Table 1); and (ii) IL-13Rα2 is the key IL-13 regulator of both lung ILC and DCs. Thus, taken together with our previous findings, we propose that the IL-13Rα2 and IFN-γR receptor regulation process at the ILC and DC level may play an important role in shaping not only the T cell but also B cell immune outcomes (Table 1), in a vaccine vector, adjuvant, and a route dependent manner; which warrants further investigation.

Data Availability Statement:
The authors declare that all data supporting the findings of this study are available within the paper and supplementary files.