Bona Fide Th17 Cells without Th1 Functional Plasticity Protect against Influenza

Abstract

Optimal transcriptional programming needed for CD4 T cells to protect against influenza A virus (IAV) is unclear. Most IAV-primed CD4 T cells fit Th1 criteria. However, cells deficient for the Th1 "master regulator," T-bet, although marked by reduced Th1 identity, retain robust protective capacity. In this study, we show that T-bet's paralog, Eomesodermin (Eomes), is largely redundant in the presence of T-bet but is essential for the residual Th1 attributes of T-bet-deficient cells. Cells lacking both T-bet and Eomes instead develop concurrent Th17 and Th2 responses driven by specific inflammatory signals in the infected lung. Furthermore, the transfer of T-bet- and Eomes-deficient Th17, but not Th2, effector cells protects mice from lethal IAV infection. Importantly, these polyfunctional Th17 effectors do not display functional plasticity in vivo promoting gain of Th1 attributes seen in wild-type Th17 cells, which has clouded evaluation of the protective nature of Th17 programming in many studies. Finally, we show that primary and heterosubtypic IAV challenge is efficiently cleared in T-bet- and Eomes double-deficient mice without enhanced morbidity despite a strongly Th17-biased inflammatory response. Our studies thus demonstrate unexpectedly potent antiviral capacity of unadulterated Th17 responses against IAV, with important implications for vaccine design.

Figure 1:. Eomes expression by CD4 T cells does not impact the magnitude of effector responses primed by IAV.

1x10⁶ Naive OT-II cells obtained from WT or Eomes^−/− mice were adoptively transferred to unprimed congenic hosts that were then infected with PR8-OVA_II. (A) The number of donor cells in stated organs at 7 dpi is shown for 9 mice/group (summary of 2 experiments). In separate experiments, WT or Eomes^−/− mice were challenged with PR8 and (B) IAV-specific CD4 T cells were detected at 8 dpi using NP_311–325 tetramer, with (C) the number of NP_311–325⁺ CD4 T cells shown for stated organs for 7–8 mice/group. (D) Representative staining of stated activation markers from WT (black) and Eomes^−/− (grey) OT-II cells at 7 dpi, with control staining (dotted). (E) Representative staining for Eomes from WT (black) and Eomes^−/− OT-II cells (grey) in stated organs at 7 dpi (left) with summary analysis (right) for the frequency of Eomes^high WT cells from 4 mice/group. Similar analysis was performed for T-bet expression in WT OT-II cells with (F) representative staining, including control staining from T-bet^−/− OT-II cells, and summary analysis shown for 4 mice/group. All results representative of at least 3 independent experiments.

Figure 2:. Eomes impacts Th1 cytokine production by IAV-primed CD4 T cells but not protective potential.

1x10⁶ WT or Eomes^−/− OT-II cells were transferred to congenic WT hosts that were then primed with IAV. (A) Representative staining of donor cells isolated from stated organs for IFNγ and IL-2 after restimulation. (B) Summary analysis for IFNγ production from 6 mice/group. In separate experiments, polyclonal WT and Eomes^−/− CD4 T cells in the lungs at 8 dpi were analyzed for cytokine production with (C) representative staining (left) and summary analysis (right) from 4 mice/group shown. (D) WT and Eomes^−/− OT-II cells responding in the lung were analyzed for IL-10 production at 7 dpi. (E) Specific killing of OT-II peptide loaded target cells in the lung was determined at 8 dpi from 8 mice/group receiving either 1x10⁶ WT or Eomes^−/− OT-II cells (summary of 2 independent experiments). (F) Weight loss from mice receiving 3x10⁶ WT (black circle) or Eomes^−/− (open circle) Th1-polarized OT-II effector cells, and from control mice not receiving cells (open triangle), is shown following lethal IAV challenge (4 mice/group; 1 of 2 experiments). (G) Viral copies as determined by RT-PCR analysis is shown from lungs collected from infected mice at the stated timepoints as described in (F) (4–7 mice/group per day, pooled from 2 separate experiments. Dotted line represents the limit of detection).

Figure 3:. DKO CD4 T cells primed by IAV lose Th1 attributes, gain Th17 hallmarks, and retain killing capacity.

1x10⁶ T-bet^−/− or DKO OT-II cells were transferred to congenic WT hosts that were then primed with IAV. (A) The number of donor cells in the lungs at 7 dpi is shown for 6 mice/group with (B) representative staining for ki-67 with the percentage of ki67 high cells shown compared to host splenic T cells (grey) and (C) donor cell expression of CXCR3 as determined by MFI (1 of 3 independent experiments). (D) In separate experiments the number of WT or DKO OT-II cells was determined in stated organs at 7 dpi (10 mice/group; summary of 2 independent experiments). (E) Representative staining for IFNγ and IL-17 production by T-bet^−/− or DKO OT-II cells responding in the lung at 7 dpi (left) with summary analysis for individual mice (right). (F) IL-10 production, (G) NKG2A/C/E expression, and (H) Granzyme B expression by WT or DKO OT-II cells responding in the lung at 7 dpi (summary of 2 separate experiments). (I) Specific killing of OT-II peptide-loaded target cells in the lungs was determined from mice receiving 1x10⁶ WT or DKO OT-II cells (summary of 2 independent experiments).

Figure 4:. Infection-induced signals act on primed DKO cells to induce Th17 identity.

1x10⁶ DKO OT-II cells were transferred to congenic WT hosts that were then primed with IAV. (A) Representative staining (left) and summary analysis from 3 mice/group (right) of DKO OT-II cell IFNγ and IL-17 production at 7 dpi in stated organs, as well as (B) similar analysis of DKO OT-II Rorγt expression. (C) In separate experiments, naive WT or DKO OT-II cells were plated in Th17 or Th1 conditions for 5 days, or plated in Th1 conditions for 2 days then washed thoroughly and plated in Th17 conditions for 3 days. Shown is representative staining of IFNγ and IL-17 from all conditions with (D) summary analysis from 3 wells per condition (1 of 3 independent experiments). (E) Donor cell Rorγt expression at 5 days from stated conditions. (F) In separate experiments, DKO OT-II cells were plated in Th17 conditions, Th1 for 2 days then moved to full Th17 conditions, or Th17 conditions lacking TGFβ and IL-6. Shown is the frequency of IL-17+ cells in each condition (triplicate wells, 1 of 2 independent experiments). In separate experiments, 1x10⁶ WT or DKO OT-II cells were transferred to host mice that were untreated or treated with neutralizing antibodies against TGFβ and IL-6. At 7 dpi, (G) donor cell Rorγt expression was determined in the lung, with representative staining (left) and summary analysis shown, and (H) donor cells expression of IL-17 and IL-22 was determined after restimulation. Results from individual mice pooled from 2 independent experiments.

Figure 5:. Non-functionally plastic Th17 cells can protect against IAV.

3x10⁶ effector cells generated from naive DKO OT-II precursors primed in Th1 or Th17 conditions were transferred to naive mice that were then challenged with lethal IAV. (A) Weight loss patterns from groups of 6 mice/condition (one of 2 independent experiments), with survival of mice in parentheses in the legend (summary of 2 independent experiments). (B) Pulmonary viral titer analysis from individual mice either receiving effector cells or not (one of two independent experiments). (C) Representative cytokine staining from restimulated WT or DKO effector cells responding in the lung at 5 dpi and (D) donor cell Rorγt staining (one of three experiments). (E) Representative cytokine staining from Th2 DKO effector cells in the lung at 5 dpi and (F) GATA3 expression from stated populations (one of two experiments). (G) Survival of host mice receiving 3x10⁶ Th2 or Th17 DKO effector cells versus mice not receiving cells after lethal IAV challenge (summary of 2 experiments, 10 mice per group) and (H) viral titers at 8 dpi from individual mice (summary of 2 independent experiments).

Figure 6:. Th17 responses in DKO mice are highly protective against IAV infection.

WT or DKO mice were challenged with PR8 and shown is (A) weight loss from 5 mice/group and (B) viral titer analysis at 8 dpi from mice treated with either CD4 and CD8 depleting or control antibodies. Groups of WT or DKO PR8-primed mice were challenged at 45 dpi with A/Philippines with (C) weight loss and (D) viral titer shown for individual mice (results from one of three similar experiments). (E) Survival of PR8-primed mice depleted of either CD4⁺ or CD8⁺ cells prior to A/Philippines challenge, with unprimed WT mice as controls (4 mice per group; one of two experiments) (F) Histological analysis was performed on lung sections from PR8-primed WT and DKO mice and unprimed WT mice at 4 dpi with A/Philippines, with representative images and pathology score from 3 mice/group shown as indicated. Separate mice were analyzed for (G) IL-17, IFNγ, and IL-4 from lung homogenates taken at 4 dpi with A/Philippines (summary of individual mice from two experiments). (H) CD4⁺ and (I) CD8⁺ CD44^high T cells from lungs of primed mice at 4 dpi with A/Philippines were analyzed for stated Th1 and Th17 hallmarks (5 mice/group; one of two separate experiments).