IL-7 is essential for accumulation of antigen-specific CD8 T cells and to generate clonotype-specific effector responses during airway influenza/A infection

Airborne diseases are the leading cause of infectious disease-related deaths in the world. In particular, the influenza virus activates a network of immune cells that leads to clearance or an overzealous response that can be fatal. Tight regulation of the cytokines that enable proper activation and function of immune cells is necessary to clear infections efficiently while minimizing damage to the host. Interleukin-7 (IL-7) is a cytokine known for its importance in T cell development and survival. How IL-7 shapes CD8 T cell responses during an acute viral infection is less understood. We had previously shown that IL-7 signaling deficient mice have reduced accumulation of influenza-specific CD8 T cells following infection. We sought to determine whether IL-7 affects early CD8 T cell expansion in the mediastinal lymph node and effector function in the lungs. Using IL-7Rα signaling deficient mice, we show that IL-7 is required for a normal sized mediastinal lymph node and the early clonal expansion of antigen-specific CD8 T cells therein. Bone marrow chimeric models and adoptive transfer of transgenic TCR CD8 T cells reveal a cell-intrinsic role for IL-7 in the accumulation of NP366–374 and PA224–233-specific CD8 T cells. We also found that IL-7 dictates terminal differentiation, degranulation and cytokine production in PA224–233-specific but not NP366–374-specific CD8 T cells. We further demonstrate that IL-7 is induced in the lung tissue by viral infection and we characterize multiple cellular sources that contribute to IL-7 production. Drugs that manipulate IL-7 signaling are currently under clinical trial for multiple conditions. Our findings on IL-7 and its effects on lower respiratory diseases will be important for expanding the utility of these therapeutics. Author Summary Interleukin-7 plays an important role in development of immune cells such as lymphocytes. In recent years, its role in the immune system has been expanded beyond the development of immune cells to include revitalizing of lymphocytes during tumor and chronic viral response. We show here that IL-7 is required for accumulation and function of specialized lymphocytes in the lungs during an acute influenza infection.


Introduction 62
The influenza virus is an airway pathogen that infects lung epithelial cells and  In addition to TCR-MHC engagement (signal 1) and co-stimulation (signal 2), 79 cytokine cues (signal 3) have great influence in activating and shaping CD8 T cell 80 responses and their terminal differentiation. Once a CD8 T cell receives these signals, it 81 is driven towards a robust clonal expansion phase whereby a single cell expands to 82~10 5 cells (4). The signal 3 cytokines that govern T cells are multifaceted and include 83 interleukin-2 (IL-2), IL-6, IL-10, IL-12, IL-15 and others which dictate their terminal 84 differentiation and inflammatory functions (5, 6). 85 The common gamma chain (γc) cytokine IL-7 is produced mainly by stromal cells 86 in the bone marrow and thymus. At steady state, it plays an indispensable role in the 87 development of both pre-and pro-B cells and T cells (7-10). IL-7 is important in the 88 development and survival of T cells at specific stages of maturation in the thymus as the 89 expression of IL-7Rα (CD127) is dynamically regulated (8,(11)(12)(13). IL-7 shares the IL-90 7Rα with thymic stromal lymphopoietin (TSLP), an alarmin cytokine that plays a major 91 role in mucosal sites. In addition to its role in development, IL-7 also plays a canonical 92 role in maintenance of memory T cells (14). The span of IL-7's function was further 93 expanded in the past decade when it was shown to be able to shape the effector 94 responses of cytotoxic CD8 T cells by enhancing their responses against tumors (15) 95 and bacterial infection (14), and reverse T cell exhaustion caused by chronic LCMV 96 infection, thus, preventing liver pathology (16). However, the extent to which IL-7 97 regulates CD8 T cell response to acute viral infections is unknown. We had previously 98 shown that IL-7 but not TSLP is important for the accumulation of influenza-specific 99 CD8 T cells in the lungs but the mechanism by which this occurs is unclear (17). Since  In this study, we asked: what modulatory effects does IL-7 have on CD8 T cell 104 priming and effector functions during an acute airway influenza infection? Using IL-7Rα 105 knock-in mice, we have shown that in the lung draining mediastinal LNs (mdLNs), IL-7 106 is important for early priming and accumulation of CD8 T cells specific for influenza 107 NP 366-374 and PA 224-233 presented on H2D b in a cell intrinsic manner. We also show that 108 IL-7 is important for the terminal differentiation and cytokine production in CD8 T cells.

109
This study will aid in therapeutic development and vaccine adjuvant studies to design 110 combinatorial therapeutic strategies.

IL-7Rα signaling is required for accumulation of influenza-specific CD8 T cells 127
To assess the importance of IL-7 signaling in CD8 T cells following infection with 128 influenza, we infected WT and IL-7Rα 449F mice with A/PR/8/34 (PR8) influenza virus 129 and measured influenza-specific CD8 T cells by flow cytometry using MHC-I tetramers. 130 We found that IL-7Rα 449F mice have reduced proportions of NP 366-374 and PA 224-233 -131 specific cells within CD8 T cells in the lungs 7 days post-infection (dpi) (Fig. 1a), which 132 phenocopies past observation of this defect at 9 dpi in a new, embryo re-derivation 133 based, specific pathogen free facility (17). Since the majority of pathogen-specific T 134 cells originate from tissue draining lymph nodes, we examined the mediastinal lymph 135 nodes (mdLNs) of infected mice and found that IL-7Rα 449F mice have reduced lymph 136 node sizes, particularly the mdLN, compared to WT mice. Interestingly, unlike WT mice, 137 there was little increase in mdLN size of IL-7Rα 449F mice after influenza infection ( Fig.   138 1b). Additionally, enumeration of the total antigen-specific cells in the mdLN at multiple 139 days post infection revealed a consistent and substantial defect that is not due to a 140 delay in expansion kinetics. (Fig. 1c). Consistent with lack of LN hyperplasia, the 141 proportion of antigen-specific cells in IL-7Rα 449F mdLN at 5 dpi was reduced indicating a 142 defect in early priming of CD8 T cells (Fig. 1d).

144
Intrinsic requirement for IL-7Rα signaling in the accumulation of influenza-145 specific CD8 T cells in the mdLN 146 Previous reports have shown that IL-7 is required for the generation of lymph 147 nodes independent of the peripheral lymphocyte pool (19,20). This likely contributed to 148 the reduced lymph node sizes noted above. Therefore, it is possible that the reduction 149 in influenza-specific CD8 T cell accumulation was due to factors extrinsic to T cells in 150 the lymph node and that IL-7 was indirectly important for shaping the cellular and 151 cytokine environment for optimal T cell activation. To address this, we created bone 152 marrow (BM) chimeric mice whereby we grafted BM cells of wild type (WT) and IL-153 7Rα 449F mice into lethally irradiated RAG-1-deficient hosts (Fig. 2a). Since WT 154 lymphocytes outcompete IL-7Rα 449F lymphocytes during development (21), we 155 delivered a 1 to 10 ratio of WT to IL-7Rα 449F cells respectively. Following engraftment 156 and infection of the hosts, we noted a reversal of this ratio within the CD8 T cell 157 compartment in the mdLN (Fig. 2b). More importantly, IL-7Rα 449F CD8 T cells resulted in 158 reduced NP 366-374 and PA 224-233 -specific cells in proportion despite engraftment in a 159 competent niche (Fig. 2b). These data suggest that IL-7Rα signaling plays an intrinsic 160 role necessary for CD8 T cell expansion during influenza infection.  conventional DCs (22). Furthermore, IL-7 has been shown to regulate CD4 T cell 186 proliferation in conditions of lymphopenia indirectly though DCs (23). We found that in 187 the lungs of IL-7Rα 449F mice, CD11b + CD103 -DCs but not CD11b -CD103 + DCs, 188 accumulate up to 9 dpi, while in WT mice these DCs peak at 7 dpi and decrease in 189 numbers at 9 dpi (Suppl. Fig. 2a). This increased accumulation could be due to a cell-190 extrinsic factors such as increased viral load in the IL-7Rα 449F mice as a result of lack of 191 appropriate T cell response. Another plausible reason could be due to impaired lymph 192 node homing signals from chemokines as a result of reduced draining lymph node size.

193
Alternatively, IL-7 may have a direct effect on CD11b + CD103 -DC maturation or 194 migration. To test these hypotheses, we created 50:50 BM chimeras using WT:IL-   To determine if IL-7 is induced in vivo, we used the IL-7 eGFP/WT mice. We noted 213 an increase in the number of cells expressing eGFP at 6 dpi ( Fig. 4b). Interestingly, the 214 majority of IL-7-eGFP + cells during infection are epithelial cells. The number of IL-7-215 eGFP + stromal cells also significantly increased during infection. However, IL-7-eGFP + 216 lymphatic endothelial cells were reduced following infection (Fig. 4b). Altogether, these 217 in vitro and in vivo experiments suggest that lung epithelial cells are responsive to viral 218 infection, and that during influenza infection, they become the primary source of IL-7. 219 220 221 The cytokine milieu that CD8 T cells are exposed to throughout the course of an WT and IL-7Rα 449F mice was greater in PA 224-233 -specific compared to NP 366-374 -specific 232 cells (Fig. 5). These data suggest that IL-7Rα signaling plays a role in the terminal 233 differentiation of antigen specific CD8 T cells.

248
To determine if IL-7 signaling affects cytokine production, we treated whole lung 249 single cell suspensions from infected mice with NP 366-374 and PA 224-233 peptides ex vivo 250 and stained intracellular cytokines to detect IFNγ and TNFα using flow cytometry. NP 366-251 374 -specific cells generated low proportion of IFNγ + TNFα + cells compared to PA 224-233 -252 specific cells regardless of mouse genotype (Fig. 7a). However, WT PA 224-233 -specific 253 cells generated abundant IFNγ + TNFα + cells, which were largely absent within IL-254 7Rα 449F PA 224-233 -specific cells (Fig. 7b). We used TSLPR -/mice as controls since IL-255 7Rα is required for both IL-7 and TSLP signaling. We found that TSLPR -/-CD8 T cells presented with reduced cytokine production as well, however, this effect did not follow 257 the same pattern as with IL-7Rα 449F mice ( Fig. 7a and b).

258
To determine if IL-7 independently affects cytokine production, we used IL-259 7 eGFP/eGFP mice that have an eGFP gene inserted disruptively into an Il7 exon thus 260 serving as an IL-7 ligand knock-out in homozygotes (30). Using this mouse model, we 261 established that IL-7 is required for accumulation of IFNγ + TNFα + cells within PA 224-233 -262 specific but not NP 366-374 -specific cells ( Fig. 8a and b). showed that IL-7Rα 449F and IL-7 eGFP/eGFP CD8 T cells have higher expression of this 272 molecule ( Fig. 9a and b). Specifically, the increase in PD-1 expression in IL-7Rα 449F and 273 IL-7 eGFP/eGFP CD8 T cells was only evident in PA 224-233 -specific cells but not NP 366-374 -274 specific cells (Fig. 9a and b). In addition, TSLPR -/-CD8 T cells did not present with 275 increased PD-1 expression (Fig. 9a). Together, this suggests that IL-7 plays distinct  (14). In 295 the current study, we found that defective IL-7Rα signaling led to reduced accumulation 296 of influenza-specific CD8 T cells in the secondary lymphoid organ (mdLN) at early 297 priming stages (5 dpi) which ultimately led to reduced accumulation of influenza-specific 298 CD8 T cells in the lungs. Examination of IL-7Rα 449F mdLN revealed a great reduction in 299 its size. This is consistent with the fact that IL-7 is required for the development of  It is known that a population of CD8 T cells specific to a distinct peptide do not 310 originate from a single naïve precursor but rather from 10s to 100s of precursors (39-  Therefore, the accumulation of DCs we noted in IL-7Rα 449F mice was not due to a 335 problem with migration or maturation and was likely due to the fact that viral clearance 336 was impaired which led to continued recruitment of DCs to the lungs. 337 We have demonstrated for the first time that IL-7 is inducible in lung epithelial 338 cells in response to viral infection in vivo. While the increase in total IL-7-eGFP + cells in 339 the lung was modest, we noted a shift in the population that are positive for IL-7-eGFP.

340
In naïve mice, the majority of IL-7-eGFP + cells were lymphatic endothelial cells (LECs) studies suggest that PA 224-233 -specific IL-7 signaling deficient CD8 T cells do not receive the necessary signals to down regulate PD-1. Further studies are required to define the 393 relationship between IL-7 and PD-1 in an acute infection setting and the mechanism by 394 which this specifically affects T cells in a clone-specific manner.

395
In summary, we have found that IL-7 is required for an optimal response to acute      Lungs from mice infected (7-9 days) with PR8 were excised and prepared as above. To 515 measure CD107a, 5x10 6 lung cells were re-stimulated for 4 hours (37 o C, 5% CO 2 ) with 516 1% BSA RPMI containing 50 ng/ml PMA and 500 ng/ml Ionomycin from Sigma-Aldrich  Following re-stimulation, cells were stained for viability and surface markers then 520 analyzed by flow cytometry.

521
To measure IFN-γ and TNF-α, 5x10 6 lung cells were re-stimulated for 3 hours (37 o C, 522 5% CO 2 ) with 10 nM NP 366-374 and PA 224-233 peptides from Anaspec (Fremont,523 California) in the presence of Brefeldin/A from BD Biosciences (Franklin Lakes, New 524 Jersey) in 1% BSA RPMI. Following re-stimulation, cells were stained for viability and 525 surface markers followed by intracellular cytokine staining using the Cytofix/Cytoperm 526 kit from BD Biosciences (Franklin Lakes, New Jersey) then analyzed by flow cytometry.