Influenza Cytokine Storm and Preserved Memory in T Cells With

Altered T-bet Dominance in IFN- −γ Decoupled CD4+ T Cells with Attenuated Cytokine Storm and Preserved Memory in Influenza This information is current as of October 4, 2021. Avijit Dutta, Shi-Chuen Miaw, Jhang-Sian Yu, Tse-Ching Chen, Chun-Yen Lin, Yung-Chang Lin, Chia-Shiang Chang, Yueh-Chia He, Sheng-Hao Chuang, Ming-I Yen and Ching-Tai Huang

J Immunol 2013; 190:4205-4214; Prepublished online 18 Downloaded from March 2013; doi: 10.4049/jimmunol.1202434 http://www.jimmunol.org/content/190/8/4205 http://www.jimmunol.org/ References This article cites 35 articles, 17 of which you can access for free at: http://www.jimmunol.org/content/190/8/4205.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

by guest on October 4, 2021 • No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2013 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Altered T-bet Dominance in IFN-g–Decoupled CD4+ T Cells with Attenuated Cytokine Storm and Preserved Memory in Inﬂuenza

Avijit Dutta,* Shi-Chuen Miaw,† Jhang-Sian Yu,† Tse-Ching Chen,‡ Chun-Yen Lin,x Yung-Chang Lin,{ Chia-Shiang Chang,* Yueh-Chia He,* Sheng-Hao Chuang,* Ming-I Yen,* and Ching-Tai Huang*

Cytokine storm has been postulated as one of the major causes of mortality in patients with severe respiratory viral infections such as influenza. With the help of an influenza Ag- specific mouse experimental system, we report that CD4+ T cells contribute effector cytokines leading to lung inflammation in acute influenza. Although virus can no longer be detected from tissues 14 d postinfection, virus-derived Ag continues to drive a CD4+ T cell response after viral clearance. Ag-specific CD4+ T cells proliferate Downloaded from and evolve into memory CD4+ T cells efficiently, but the production of effector cytokines is seriously hampered during this phase. This decoupling of proliferation and effector cytokine production doesn’t appear in conjunction with increased suppression by regulatory T cells or decreased induction of transcription factors. Rather, GATA-3 and ROR-gt levels are elevated when compared with cells that have effector cytokine production. T-bet dominance over GATA-3 and ROR-gt decreases with the disarmament of effector cytokine production. Importantly, upon reinfection, these decoupled cells produce elevated levels of IFN-g and were effective in virus eradication. These results provide a mechanism through altered T-bet dominance to dampen the cytokine http://www.jimmunol.org/ storm without impeding the generation of memory T cells in influenza virus infection. The Journal of Immunology, 2013, 190: 4205–4214.

nfluenza strikes millions of people worldwide and causes Damage from such insult must be contained quickly for improved significant death every year. The number of deaths in- recovery. I creases in pandemic years (World Health Organization Influenza virus infection induces both Ab and cytotoxic T cell fact sheets, http://www.who.int/mediacentre/factsheets/fs211/en/). (CTL) responses, each of which plays a role in recovery from infection and resistance from reinfection. Ab and CTL responses

The symptoms of influenza are exaggerated by the patient’s im- by guest on October 4, 2021 mune system responding to the virus, rather than by the virus are dependent on help from CD4+ T cells (9, 10). CD4+ T cells itself. The immune system’s prolonged and unbalanced response proliferate extensively and secrete high levels of effector cyto- generates a cytokine storm and causes the lungs to become in- kines, IFN-g in particular, which is critical for viral clearance (10– flamed and significantly damaged (1–8). The fight continues in 12). After the majority of virus has been eradicated, residual virus- the damaged lung even after the virus is cleared from the body. derived Ag can still generate a modest expansion of CD4+ T cell effectors with intermediate phenotypes, some of which are capa- ble of persisting as memory (13). *Division of Infectious Diseases, Department of Medicine, Chang Gung Univer- sity and Chang Gung Memorial Hospital, Kweishan, Taoyuan 33333, Taiwan; Jelley-Gibbs et al. (13) restricted focus on the persistence of †Graduate Institute of Immunology, National Taiwan University College of Medicine, cells, whereas data reveal an unexplained phenomenon of con- ‡ Taipei 10617, Taiwan; Department of Pathology, Chang Gung University and Chang tained IFN-g production by proliferating influenza-specific CD4+ Gung Memorial Hospital, Kweishan, Taoyuan 33333, Taiwan; xDivision of Hepato- gastroenterology, Department of Medicine, Chang Gung University and Chang Gung T cells in response to residual virus-derived Ag. The decoupling of Memorial Hospital, Kweishan, Taoyuan 33333, Taiwan; and {Division of Hematol- proliferation and effector cytokine response implies a mechanism ogy and Oncology, Department of Medicine, Chang Gung University and Chang for restraining damaging acute-phase immune reactions and cy- Gung Memorial Hospital, Kweishan, Taoyuan 33333, Taiwan tokine storm in acute influenza. Received for publication August 31, 2012. Accepted for publication February 14, 2013. In this report, we focus on the decoupling of proliferation This work was supported by National Health Research Institute Project NHRI- and effector cytokine response through the use of an influenza EX-9303SC from the National Health Research Institute, Taiwan; Projects NSC- hemagglutinin (HA)-specific transgenic mouse model. After sub- 95-2745-B-182A-006 and NSC-98-2314-B-182A-049-MY3 from the National lethal infection, the majority of influenza virus is eradicated within Science Council, Taiwan; and CMRPG 350321, 370861, 370862, and 380411 from + the Medical Research Project Fund, Chang Gung Memorial Hospital, Taiwan. A.D. 14 d. We adoptively transferred naive HA-specific CD4 Tcells is receiving a postdoctoral fellowship from the National Science Council, Taiwan into syngeneic recipient mice on day 0 or 14 of influenza virus (NSC-95-2745-B-182A-006 [2007-2009]; NSC-98-2811-B-182A [2009-2012]). infection. On day 4 posttransfer, residual influenza virus-derived Address correspondence and reprint requests to Dr. Ching-Tai Huang, Division Ag generated CD4+ T cell expansion and cytokine production of Infectious Diseases, Department of Medicine, Chang Gung Memorial Hospi- tal, No. 5, Fushing Street, Kweishan, Taoyuan 33333, Taiwan. E-mail address: such as IL-2 and IFN-g, on a level comparable to the response [email protected] driven by live virus at the time of progressive infection. The Abbreviations used in this article: Ct, threshold cycle; HA, hemagglutinin; MDCK, production of IFN-g fell drastically thereafter in response to Madin–Darby canine kidney; qRT-PCR, quantitative real-time PCR; Treg, regulatory residual Ag. TNF-a, IL-10, and IL-17a levels decreased on day T cell. 7, whereas IL-2 and IL-6 were shown to increase. In contrast, + Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 residual Ag-driven CD4 T cells proliferated better and con- www.jimmunol.org/cgi/doi/10.4049/jimmunol.1202434 4206 T-bet DOMINANCE AND IFN-g DECOUPLING tracted less compared with cells responding to acute infection. Intracellular cytokine staining The induction of IFN-g receptor 1 was less in residual Ag-driven + Organs were harvested from the experimental mice on indicated days. CD4 T cells compared with the cells responding to acute infec- Single-cell suspensions (5–10 million cells/well of a 24-well plate) tion, whereas the Foxp3+ regulatory T cell (Treg) levels remained were restimulated for 5 h with 100 mg/ml MHC class II-restricted HA 110 120 + comparable between the two groups. Induction of transcription peptide ( SFERFEIFPKE ) for clonotypic CD4 T cells and PMA factors such as T-bet, GATA-3, and ROR-gtbyresidualAgwas (25 ng/ml) plus ionomycin (500 ng/ml) for endogenous T cells in the presence of 5 mg/ml brefeldin A (Sigma-Aldrich). Concentration of brefeldin not impaired either. However, the relative expressions of T-bet/ A was maintained throughout the intracellular cytokine staining. Re- GATA-3 and T-bet/ROR-gt became altered in the cells exhibiting stimulated cells were surface stained with anti-6.5 and anti-CD4, as the phenomenon of decoupled proliferation and effector cytokine described above, fixed in intracellular fixation buffer (eBiosciences), production. Compared with the cells activated by acute infection, washed, and stained in permeabilization buffer (eBiosciences) containing fluorochrome-conjugated Abs against target cytokines. the decoupled cells were also very efficient for viral eradication FOXP3 and T-bet were stained ex vivo without restimulation. After surface upon reinfection with influenza virus. Interestingly, the decou- staining, cells were fixed and permeabilized in fixation/permeabilization buffer pled cells provided a better memory response during a secondary (eBiosciences), washed, and stained in permeabilization buffer (eBiosciences) influenza infection. Altered T-bet dominance may be a key mecha- containing fluorochrome-conjugated Abs. nism for generating a more balanced immune response, and thereby Virus titer restraining the devastating cytokine storm that is so detrimental for patients infected with influenza virus. We measured live virus in organs of influenza virus-infected mice using a modified Madin–Darby canine kidney (MDCK; American Type Culture Collection) cell plaque assay (14). Organs were collected at the indicated 2

times in 1 ml DMEM, snap frozen in liquid nitrogen, and stored at 80˚C Downloaded from Materials and Methods until they are ready for use. MDCK monolayers were grown in DMEM Mice and infection supplemented with 10% FCS and antibiotic-antimycotic (100 IU/ml peni- cillin, 100 mg/ml streptomycin, and 0.25 mg/ml amphotericin B; Life The TCR-transgenic mouse line 6.5 expresses a TCR recognizing an I- Technologies BRL). The 10-fold dilutions of the tissue homogenates were Ed–restricted HA epitope (110SFERFEIFPKE120; provided by H. Von prepared in DMEM supplemented with 0.00025% trypsin. A total of 500 Boehmer, Harvard University, Boston, MA). The 6.5 and wild-type mice ml of each dilution was added to confluent monolayers of MDCK cells in are of two kinds, Thy 1.1 +ve and Thy 1.2 +ve. All mice are of C57BL/10. 6-well plates in duplicates for 1 h at 35˚C, 5% CO2. Each well received 2 D2 genetic background and are maintained in specific pathogen-free ml of an agar overlay (0.3%) in medium containing DMEM with http://www.jimmunol.org/ condition. All experiments involving the use of mice were performed in 0.00025% trypsin. Cells were incubated for 3 d at 35˚C and were fixed accordance with protocols approved by the Animal Care and Use Com- with 10% formalin for 20 min. The agar overlay was then removed, and mittee of Chang Gung Memorial Hospital, Taiwan. fixed monolayers were stained by adding 1:10 dilution of 2% crystal The A/PR/8/34 (PR8) and WSN (A/WSN/33) strains of H1N1 in- violet prepared in 20% ethanol. The results are presented as PFU/ml = fluenza virus were produced in the allantoic fluid of 10-d-old embry- (mean number of plaques 3 2) 3 (1/dilution factor). onated chicken eggs and characterized by a core facility at the Chang Gung University, Taiwan. Although both the PR8 and WSN strains Influenza-specific Ab response belong to H1N1 influenza virus, the WSN strain does not have the PR8 strain-specific HA epitope 110SFERFEIFPKE120. The PR8-H3N2 (A/ Influenza-specific Ab responses were determined by plaque reduction assay. Taiwan/447/2007) is a recombinant strain with heterologous HA and Sera from influenza-infected mice were collected on the indicated days neuraminidase from H3N2 on a PR8 backbone and was provided by postinfection. They were heat inactivated for 30 min at 56˚C to inactivate by guest on October 4, 2021 S.-R. Shih (Research Center for Emerging Viral Infections, Chang Gung the complements. A total of 50 PFU live PR8 virus was allowed to infect University, Taiwan). Mice were inoculated via intranasal route during a monolayer of DMEM cells in the presence or absence of decomple- light diethyl ether anesthesia (J.T. Baker, Phillipsburg, NJ) with indi- mented sera with specified dilution for 1 h at 35˚C. Infected cells were cated units of virus in 50 mlPBS. cultured for another 72 h in agarose-mixed DMEM at 35˚C. Formation of plaques was counted manually after fixing with formalin and staining with Adoptive transfer crystal violet. Reduction of plaques due to inhibition by anti-influenza Ab in the sera was calculated. Clonotypic 6.5 CD4+ TCR transgenic T cells were prepared from pooled spleen and lymph nodes of 6.5 transgenic mice. Clonotypic percentage Histopathology was determined by flow cytometry analysis. The naive phenotype was confirmed by activation marker CD44 and CD62L. The 6.5 CD4+ Tcells Lungs from experimental mice were harvested at indicated days and fixed (2.5 3 106) were injected through the tail veins of the recipient mice in with 10% neutral buffered formalin solution. Following fixation, lungs were 0.2 ml HBSS. embedded in paraffin and 5-mm sections were cut. Sections were stained For adoptive transfer of CFSE (5, FAM diacetate succinimidyl ester; with H&E and scored blinded. The inflammatory cell infiltrations, in- Molecular Probes, Eugene, OR)-stained cells, clonotypic T cells (1 3 107 cluding lymphocytes, neutrophils, and plasma cells, were separately scored cells/ml) were stained with 5 mM CFSE in HBSS for 10 min. Cells were as negative, 1+,2+,3+ according to the severity of each infiltration. Vascu- washed three times with complete RPMI 1640 medium, and CFSE- litis and fibrosis were also scored as negative, 1+,2+,3+ on the basis of stained 6.5 CD4+ T cells (2.5 3 106) were similarly transferred through their intensity. An average of these scores was represented as overall in- the tail veins of the recipient mice. flammation in the lung.

Flow cytometry In vitro proliferation Cell suspensions were incubated on ice with saturated concentrations of Single-cell suspension was prepared from the spleens. A total of 2 3 105 fluorochrome-labeled mAbs in FACSbuffer(PBSplus0.5%BSAand cells was restimulated for 72 h with class II–restricted HA peptide 0.02% NaN ). Donor T cells were identified using mAbs, as follows: 3 (SFERFEIFPKE) at 37˚C in presence of 5% CO2. Tritiated thymidine biotin-conjugated anti-clonotypic TCR 6.5 (provided by H. Von was added 18 h prior to harvest. Cells were harvested, and the cpm was Boehmer); avidin-PE (Caltag, Burlingame, CA) or avidin-allophyco- measured by a beta counter. cyanin (Caltag); and PerCP- or FITC-conjugated anti-CD4. All the fluorochrome-conjugated Abs were from BD Biosciences, except for T- Quantitative real-time PCR analysis bet and IgG1k (eBiosciences). Cells were acquired on FACSCalibur, and analysis was performed using CellQuest Pro (BD Biosciences) or The sorted HA-specific CD4+ T cells were immediately used for RNA FlowJo (Tree Star) software. extraction by TurboCapture 96 mRNA Kit (Qiagen). Reverse tran- The anti-clonotypic TCR 6.5 Ab was replaced by anti-Thy1.1 Abs in the scription was performed with Moloney murine leukemia virus high case of transferring Thy1.1+ donor cells into Thy1.2 recipient mice or anti- performance reverse transcriptase (EPICENTRE Biotechnologies). Thy1.2 Abs in case of transferring Thy1.2+ donor cells into Thy1.1 re- cDNA levels were analyzed by quantitative real-time PCR (qRT-PCR) cipient mice. with the iCycler Real-Time PCR Detection System (Bio-Rad, Hercu- The Journal of Immunology 4207 les, CA). Each sample was assayed in triplicate for the target gene Results together with b-actin as the internal reference in 25 mlfinalreaction Exaggerated immune response causes lung inflammation upon volume containing 5 ml 10-fold diluted cDNA, 2.5 ml 10-fold PCR influenza virus infection buffer, 2 mldNTP(2.5mM),1.5mlMgCl2 (50 mM), 0.5 mlforward primer (10 mM), 0.5 ml reverse primer (10 mM), 0.1 ml platinum Taq Naive influenza HA-specific naive CD4+ T cells were adoptively DNA polymerase (5 U/ml) (Invitrogen), 0.05 mlFluA(5mM), 0.025 ml 3 transferred into syngeneic mice. The recipients were infected SYBR Green I (100 ), and 12.825 mlddH2O. The primers used for 3 qRT-PCR are as follows: T-bet:forward,59-CAA CAA CCC CTT TGC with 2.5 3 10 PFU influenza virus on the same day. On day 4, CAA AG-39 and reverse, 59-TCC CCC AAG CAG TTG ACA GT-39; HA-specific CD4+ T cells expanded in number and accumulated GATA-3:forward,59-AGA ACC GGC CCC TTA TCA A-39 and rein the lungs of infected mice. CD44 was upregulated and CD62L verse, 59-AGT TCG CGC AGG ATG TCC-39; RORgt:forward,59- was downregulated on the membrane of these clonotypic cells. CGC CTC ACC TGA CCT ACC-39 and reverse, 59-TTG CCT CGT + TCT GGA CTA TAC-39; Ifng:forward,59-CAA GTG GCA TAG ATG HA-specific CD4 T cells produced effector cytokines such as TGG AAG-39 and reverse, 59-GAA GAA GGT AGT AAT CAG GTG IFN-g, IL-2, and TNF-a. The clonotypic percentage, CD44 up- TG-39; Il4:forward,59-TGT CAT CCT GCT CTT CTT TCT C-39 and regulation, CD62L downregulation, and production of effector reverse, 59-TCT GTG GTG TTC TTC GTT GC-39; Il17a:forward,59- cytokines increased further on day 7. The activation status of AGG CAG CAG CGA TCA TCC-39 and reverse, 59-TGG AAC GGT + TGA GGT AGT CTG-39; Ifng R1:forward,59-GATTCTGGTGGTT- HA-specific CD4 T cells waned thereafter. However, in com- GCTCCTCTTAC-39 and reverse, 59-GGTGCGGTGTGACAAGTGAA- parison with the donor cells of uninfected mice, the activation 39; Ifng R2:forward,59-CTATGAGAATGTTACTGTTGGA-39 and status of the donor cells on day 14 was significantly higher in reverse, 59-CAATATACTCTGTATGGCTTCA-39;andb-actin:for- infected mice (Fig. 1A). The gross appearance of the lung changed ward, 59-TGT ATG AAG GCT TTG GTC TCC CT-39 and reverse, 59- postinfection and the tissue was inflamed, with highest congestion AGG TGT GCA CTT TTA TTG GTC TCA A-39. The relative mRNA frequencies were determined by normalization to the b-actin. Briefly, and black patches by day 7 (data not shown). Histological analysis Downloaded from we normalized each set of samples using the difference in the power of lung sections revealed infiltration of immune cells such as value of threshold cycles (Ct) between the target gene and the b-actin, lymphocytes, neutrophils, and plasma cells, in addition to the as follows: ΔCt sample = (Ct sample/Ct b-actin). Relative mRNA fold swelling of the bronchial wall on day 4 postinfection. The degree changes were calculated as 2ΔΔCt,whereΔΔCt = (ΔCt sample/ΔCt naive sample). of inflammation increased further on day 7. Inflammation decreased thereafter, although the persistence of infiltration was

Statistical analyses evident in lung sections prepared from mice sacrificed on day http://www.jimmunol.org/ 6 14 postinfection (Fig. 1B). Data represented as means SD. We used GraphPad Prism version 5 for 3 2 2 Student t test analyses. We considered all p values .0.05 not to be sig- Postinfection with 2.5 3 10 ,53 10 ,or13 10 PFU in- nificant. fluenza virus, the mice successfully cleared the virus from the by guest on October 4, 2021

FIGURE 1. Exaggerated host immune response causes lung inflammation upon influenza virus infection. Wild-type C57/B10.D2 mice were infected with 2.5 3 103 PFU influenza A virus and adoptively transferred with 2.5 3 106 influenza HA-specific naive CD4+ T cells. (A) Activation status of HA-specific CD4+ T cells in the lung was analyzed for clonotypic percentage, surface expression of activation markers CD44 and CD62L, and production of cytokines IL-2, IFN-g,andTNF-a. The activation of HA-specific naive CD4+ T cells was compared between infected (O) and uninfected (N)mice.(B) The infected mice were sacrificed on the stated days, and their lungs were analyzed for histopathology from H&E-stained sections. An average of histopathological score (negative, 1+,2+,and3+) was calculated as mentioned in Materials and Methods.(C) Live virus was determined in the lung by plaque assay on the stated days postinfection with mentioned infection doses. All data are represented as mean 6 SD and representative of at least three similar experiments (n = 6/group). 4208 T-bet DOMINANCE AND IFN-g DECOUPLING body within 14 d of infection. Live virus could be detected in CCR7 was comparable between late and early transferred cells the lungs on day 4 and 7 postinfection with all doses; however, (Fig. 3A). In addition to IFN-g, reduced amounts of TNF-a, virus could no longer be detected by day 14 (Fig. 1C). Although IL-10, and IL-17a were produced by the late transferred cells virus was detected in the lung, plaque assays showed the lymph (Fig. 3B). IL-2 and IL-6 production did not differ between the nodes and spleen to be virus free, even at all three doses (data two groups (Fig. 3B). not shown). Better proliferation and less contraction of HA-specific CD4+ Decoupling of proliferation and IFN-g response hampers the T cells responding with decoupled proliferation and effector production of effector cytokines cytokine production + On day 14 postinfection, naive HA-specific naive CD4 T cells On day 4 posttransfer, clonotypic percentages did not differ be- were adoptively transferred. From this point, these cells will tween late and early transferred HA-specific CD4+ T cells in the be referred to as late transferred cells. In parallel, naive HA- lung and spleen. Upon restimulation with cognate HA peptide, the + specific CD4 T cells were similarly transferred into mice in- in vitro proliferation of the late and early transferred cells was fected on the same day as transfer. These cells are the early trans- similar as well (Fig. 4A). On day 7 posttransfer, clonotypic per- ferred cells. centages were lower for late transferred cells than early transferred The production of IFN-g andIL-2bythelateandearly cells. However, when stimulated in vitro, the late transferred cells transferredcellswascomparableonday4afteradoptive proliferated better than early transferred cells (Fig. 4A). On day 14 transfer (Fig. 2A, 2B). IFN-g production by the late transferred posttransfer, clonotypic percentages for late transferred cells ex- cells was greatly reduced thereafter, and production level was ceeded the early transferred cells. The late transferred cells also almost equivalent to the level of naive cells on days 7 and 14. showed greater proliferation in vitro than early transferred cells Downloaded from However, IFN-g production by the early transferred cells in- (Fig. 4A). creased further on day 7 and remained elevated on day 14 (Fig. The late and early transferred HA-specific CD4+ T cells were 2A). For IL-2 production, both late and early transferred cells sorted from the spleens and lungs on day 7 posttransfer and an- were equally potent on days 4 and 7. In fact, late transferred alyzed for IFN-g receptor 1 and 2. Relative to the level in naive cells produced more IL-2 in the lungs on day 14 (Fig. 2B). cells, the induction of IFN-g receptor 1 was decreased in the late Expansion of the late transferred cells was marginally lower transferred cells when compared with the early transferred cells http://www.jimmunol.org/ than early transferred cells on day 7, as evidenced by the CFSE (Fig. 4B). The induction of IFN-g receptor 2 in the late and early profile. However, there was a significant expansion of the late transferred cells, however, was similar (Fig. 4B). transferred cells compared with the control naive cells that were + similarly transferred into uninfected mice (Fig. 2C). On day 7, An increase in Foxp3 Treg is not responsible for the the late transferred cells displayed an upregulation of activation decoupling of proliferation and effector cytokine response markers such as CD44 and CD69; there was a downregulation Both the late and early transferred HA-specific CD4+ T cells re- of CD62L, although not to the level of early transferred cells sponded with coupled proliferation and IFN-g production on day 4 (Fig. 3A). The expression of chemokine receptors CCR5 and posttransfer (Figs. 2, 4A). Production of IFN-g and other effector by guest on October 4, 2021

FIGURE 2. Disarmament of IFN-g production in activation of influenza-specific CD4+ T cells after viral eradication. Naive HA-specific CD4+ T cells were adoptively transferred into syngeneic C57/B10.D2 recipient mice on day 14 after influenza virus infection (Late). Similarly, HA-specific naive CD4+ T cells were transferred on day 0 with infection (Early). (A) IFN-g– and (B) IL-2–producing late (d) and early (:) transferred HA-specific CD4+ T cells from the stated organ were compared on mentioned days after transfer by intracellular staining (mean 6 SD; *p , 0.01, ***p , 0.0001, n = 6/group). (C) In vivo expansion of late and early transferred HA-specific CD4+ T cells from the stated organ was compared on day 7 by their CFSE profile. The Journal of Immunology 4209

FIGURE 3. Activation profiles of naive T cells at different time points during infection. Naive HA-specific CD4+ T cells were adoptively transferred into syngeneic C57/B10. D2 recipient mice on day 14 after influenza virus infection (Late). Similarly, HA-specific naive CD4+ T cells were transferred on day 0 with infection (Early). (A) The late (n) and early (N) transferred cells were compared for their ex vivo

surface expression of the stated Downloaded from molecules on day 7. (B) Cytokine production on day 7 was compared between the late (n) and early (N) transferred cells by intracellular staining. The ex vivo surface expression of the stated molecules and

cytokine production by naive cells http://www.jimmunol.org/ ( ) transferred into uninfected mice is presented as control. All data are represented as mean 6 SD and representative of at least three similar experiments (*p , 0.01, **p , 0.001, ***p , 0.0001, n = 6/group). by guest on October 4, 2021

cytokines was reduced thereafter in late transferred cells. This could Altered relative expression, but not impeded induction of be caused by the active suppression of Treg. We analyzed the natural transcription factors in the phenomenon of decoupling Treg population by definition of Foxp3 expression in CD4+ T cells + We analyzed the intracellular expression of T-bet and IFN-g on days 4 and 7 after adoptive transfer. Foxp3 adoptively trans- in the HA-specific CD4+ T cells. The percentage of T-bet– ferred cells were comparable between naive, late, and early trans- expressing late transferred cells was more than T-bet+ early + ferred HA-specific CD4+ T cells. The native Foxp3 HA-specific transferred cells on day 4 posttransfer. However, IFN-g pro- + CD4 T cells were analyzed from donor cells transferred in unin- duction was comparable between late and early transferred + + fected mice (Fig. 5A, 5B). Foxp3 endogenous CD4 T cells were cells at this time point (Fig. 6A). T-bet induction was increased + elevated in infected mice in general, but the percentages of FOXP3 with time, as .90% of the late and early transferred cells + CD4 T cells were comparable between the groups with late or stained positive for T-bet on day 7 posttransfer (Fig. 6A). De- early transferred cells (Fig. 5A, 5B). spite comparable T-bet induction, IFN-g production was almost On day 7 after adoptive transfer, the clonotypic and endogenous silenced in the late transferred HA-specific CD4+ T cells on + CD4 T cells were analyzed for subsets expressing CD25, day 7 posttransfer (Fig. 6A). The naive HA-specific CD4+ T cells CTLA4, LAG3, or GITR, as these subsets have shown regulatory expressed basal levels of T-bet and IFN-g at these time points + activity in several experimental systems. The CD4 T cell subsets after being adoptively transferred into uninfected mice (dotted with expression of CD25, CTLA4, or LAG3 were not increased in line, Fig. 6A). late transfer group when compared with the early transfer group We sorted late and early transferred cells on day 7 posttransfer (Fig. 5C, 5D). Compared to early transfer group, the GITR+ clono- and analyzed the mRNA levels of T-bet and IFN-g to confirm typic CD4+ T cells were not increased in late transfer group as the findings at the protein level. The induction of T bet in the well. The GITR+ endogenous CD4+ T cells were fewer in lungs of late and early transferred cells, relative to the level in naive cells, late transfer group too. However, in the spleen, the percentage of was similar as well (Fig. 6B). IFN-g mRNA was not detected GITR+ endogenous CD4+ T cells was higher in the late transfer in the late transferred cells, whereas a significant increase in group than the early transfer group (Fig. 5C, 5D). IFN-g transcript was detected in the early transferred cells (Fig. 4210 T-bet DOMINANCE AND IFN-g DECOUPLING Downloaded from

FIGURE 5. Decoupling of proliferation and cytokine production was not associated with increased Treg. Naive HA-specific CD4+ T cells were adoptively transferred into syngeneic C57/B10.D2 recipient mice on day 14 after influenza virus infection (decoupled; late). Similarly, HA-specific

naive CD4+ T cells were transferred on day 0 with infection (coupled; http://www.jimmunol.org/ early). (A) FOXP3 expression in HA-specific CD4+ T cells was compared between late (n) and early (N) transferred cell on day 7 by flow cytometry. (B) At the same time, FOXP3-expressing endogenous CD4+ T cells were + FIGURE 4. The second wave of naive influenza-specific CD4 T cells compared between the groups of mice (late, n and early, N) (mean 6 SD; proliferates better with less contraction than the cells of the first wave. n = 6/group). Naive HA-specific CD4+ T cells were transferred into un- + Naive HA-specific CD4 T cells were adoptively transferred into synge- infected mice, and their FOXP3 expression (-----) was presented as control. neic C57/B10.D2 recipient mice on day 14 after influenza virus infection (C) The percentages of CD25+,LAG3+, CTLA4+, and GITR+ HA-specific + (second wave; late). Similarly, HA-specific naive CD4 T cells were CD4+ T cells were compared between late (n) and early (N) transferred cell transferred on day 0 with infection (first wave; early). (A) Clonotypic on day 7 by flow cytometry. (D) At the same time, CD25+, LAG3+, + by guest on October 4, 2021 percentage of HA-specific CD4 T cells was compared between the late CTLA4+, and GITR+ endogenous CD4+ T cells were compared between (n) and early (N) transferred cells on days 4, 7, and 14. In vitro prolifer- the groups of mice with late (n) and early (N) transferred cells (mean 6 ation of these late (n) and early (N) transferred cells upon restimulation of SD; ***p , 0.0001; n = 6/group). cognate peptide was compared by thymidine incorporation assay. Prolif- eration of the naive clonotypic cells transferred into uninfected mice is presented as control. (B) On day 7 after transfer, RNA was isolated from n N + the sorted late ( ) and early ( ) transferred HA-specific CD4 T cells The phenomenon of decoupling is a process that provides retrieved from the spleen and lung. Genomic expression of the mentioned better memory response receptors was measured by qRT-PCR. Fold changes in the transcripts of these genes compared with that in the naive cells are presented as men- Our studies show that the late transferred cells responded with the tioned in Materials and Methods (***p , 0.0001, n = 6/group). All data phenomenon of decoupling. To explore this further, we studied are represented as mean 6 SD and representative of at least three similar the impact of decoupling on memory response. Upon reinfection experiments. with influenza virus, both the late and early transferred cells facilitated the eradication of virus when compared with their naive counterparts. When compared with the early cells, the late transferred cells conferred better viral clearance. No virus was 6B). According to transcript levels, induction of GATA-3 was detectable in the lungs of mice with late transferred cells on day 2 elevated in the late transferred cells compared with the early postreinfection, whereas there was virus in the lungs of mice with transferredcells.IL-4mRNAwasincreasedintheearly early transferred cells at the same time. We could not spot any transferred cells, but not detected in the later transferred cells virus on day 7 postreinfection in the lungs of mice with either late or the naive cells (Fig. 6B). The induction of ROR-gtwasel- or early transferred cells. However, a significant amount of virus evated in the late transferred cells. Finally, IL-17a transcripts could be measured in the lungs of mice with adoptively trans- were comparable between late and early transferred cells ferred naive cells at these time points (Fig. 7A). The clonotypic (Fig. 6B). percentage of late transferred cells was higher when compared For the early transferred cells, expression of T-bet was dominant with early transferred cells on day 7 postreinfection in the lungs, over GATA-3. For the late transferred cells, the dominance of mediastinal lymph nodes, and spleens (Fig. 7B). The production T-bet was decreased and the relative expression between T-bet of both IFN-g and IL-2 was greater by these late transferred cells and GATA-3 was significantly less (Fig. 6C). The ratio of rela- after the reinfection (Fig. 7C). The late transferred cells pro- tive expression between T-bet and ROR-gt was altered as well duced significant amounts of IFN-g upon reinfection despite (Fig.6C).TheGATA-3/ROR-gt ratio did not change between their previous disarmament of IFN-g production (Fig. 7C). The late and early transferred cells (data not shown). IFN-g production by late transferred cells upon reinfection was The Journal of Immunology 4211

FIGURE 6. Decoupled proliferation and cytokine production were associated with an altered ratio between T-bet and GATA-3. Naive HA-specific CD4+ T cells were adoptively transferred into syngeneic C57/ B10.D2 recipient mice on day 14 after influenza virus infection (decoupled; late). Similarly, HA-specific naive CD4+ T cells were transferred on day 0 with infection (coupled; early). (A) T-bet expression of late (n) and early (N) transferred cells was compared on stated days by flow cytometry with intracellular staining. IFN-g production by late (n) and early (N) transferred cells was measured by intracellular staining and was compared on stated days. Dotted line represents basal levels of T-bet expression or IFN-g production by naive HA-specific CD4+ T cells into uninfected mice. (B)On day 7 after transfer, RNA was isolated from the sorted late (n) and early (N) transferred HA-specific CD4+ Downloaded from T cells retrieved from the spleen and lung. Genomic expression of mentioned transcription factors and cytokines was measured by qRT-PCR. Fold changes in the transcripts of these genes compared with that in the naive cells are presented as mentioned in Materials and Methods (*p , 0.01, **p , 0.001, ***p , 0.0001, n = http://www.jimmunol.org/ 6/group. UD, Undetectable). (C) Relative expression of transcription factors in late (n) and early (N) transferred HA-specific CD4+ T cells from spleens and lungs is presented. Dotted line represents relative expression of the mentioned transcription factors in naive HA-specific CD4+ T cells into uninfected mice. by guest on October 4, 2021

less than the IFN-g production by early transferred cells upon clearance, as evidenced by plaque assay on day 7 after second primary infection (Fig. 7C). challenge (Fig. 8B). On day 7 after the second challenge, the Decoupled proliferation/effector function of the late transferred endogenous CD4+ T cell response was comparable among all HA-specific CD4+ T cells provided better memory response than three groups of mice (Fig. 8B). The proportion of IFN-g–pro- the early transferred cells. It may have been caused by the more ducing endogenous CD8+ T cells was similar, but the percentage mature influenza immunity resulting from the primary infection, of endogenous CD8+ T cells was less in mice that received the first instead of the decoupled clonotypic CD4+ T cells themselves. challenge of PR8 than mice first challenged with the other two Additional control experiments were performed with heterologous strains of influenza virus (Fig. 8B). The adoptively transferred influenza virus strains for the primary infection; these resulted in HA-specific CD4+ T cells responded to the second challenge with mature influenza immunity in general, but not the decoupled PR8 in all three groups of mice, and the clonotypic percentages clonotypic CD4+ T cell response. Naive 6.5 CD4+ T cells were were comparable. However, the production of IFN-g by clono- adoptively transferred 14 d after the first challenge with PR8, PR8- typic CD4+ T cells was impeded in the mice with first infection of H3N2, or WSN. On day 7 after adoptive transfer, endogenous PR8, but not of PR8-H3N2 or WSN (Fig. 8B). Anti-influenza Ab CD4+ and CD8+ T cell responses to the challenges with all three responses were primed by the first infection with all three strains strains of influenza virus were comparable in the lungs in terms of of influenza virus, as the sera from these mice reduced plaque percentage and IFN-g or IL-2 production (Fig. 8A). The adop- formation by PR8 virus in a MDCK monolayer in vitro (Fig. 8B). tively transferred HA-specific 6.5 CD4+ T cells responded with Ab response upon second challenge with PR8 was comparable decoupled proliferation and IFN-g production only in PR8- between mice with first challenge of PR8 or PR8-H3N2, slightly challenged mice (Fig. 8A). The clonotypic CD4+ T cells did not higher than that of WSN (Fig. 8B). respond significantly to the challenge of PR8-H3N2 or WSN as these strains do not have the clonotypic HA epitope 110SFER- Discussion FEIFPKE120, to which the 6.5 CD4+ T cells respond (Fig. 8A). Our data show that influenza HA-specific CD4+ T cell responses On day 21 after the first challenge (7 d after adoptive transfer of contribute effector cytokines that lead to lung inflammation in naive 6.5 CD4+ T cells), all mice were challenged with the PR8 acute influenza. The effector cytokine production is restrained by strain of influenza virus. Eradication of the PR8 virus was en- HA-specific CD4+ T cells when they are activated by residual hanced by the first infection with all three strains of influenza influenza virus-derived Ag. However, the proliferation of these virus. Mice with PR8 as the first challenge revealed the best viral cells remains preserved in the response to residual Ag. Impairment 4212 T-bet DOMINANCE AND IFN-g DECOUPLING

FIGURE 7. Decoupled HA-specific CD4+ T cells provided a better memory response upon reinfection of influenza virus. Naive HA-specific CD4+ T cells were adoptively transferred into syngeneic C57/ B10.D2 recipient mice on day 14 after influenza virus infection (decoupled; late). Similarly, HA-specific naive CD4+ T cells were transferred on day 0 with infection (coupled; early). Half of the mice from every group was sacrificed on day 7 after adoptive transfer, and late or early transferred cells were compared with naive for (B) clonotypic percentage and (C) cytokine production. Other infected mice with late Downloaded from or early transferred cells were reinfected (2˚ Inf) with influenza virus and were sacrificed after another 7 d. (A) The virus titer in the lungs on mentioned time after reinfection and the responses of the late and early transferred cells to second in- B fection were compared for ( ) clonotypic http://www.jimmunol.org/ percentage and (C) cytokine production. Data represented as mean 6 SD. All data are representative of three similar experiments (**p , 0.001, ***p , 0.0001, n =6/ group). by guest on October 4, 2021

of T cell function, especially the effector cytokine IFN-g response well (21, 22). These cells produce more IFN-g, TNF-a, and IL-2 in proliferating T cells, has also been shown in systems of trans- than IL-4, with dominant expression of T-bet over GATA-3. T-bet plantation and autoimmunity (15–17). We previously observed dominance is known to repress the binding of GATA-3 to its target that T cell division is not strictly linked to effector function (18, DNA and block IL-4 production (23). In contrast, cytokine pro- 19). Preserved T cell proliferation despite disarmament of effector duction is disarmed in proliferating HA-specific CD4+ T cells in cytokine secretion implies a possible mechanism for restrained response to residual influenza virus-derived Ag. The cells with cytokine response in autoimmunity (20). The same can be implied decoupled proliferation and effector cytokine response show in- as a mechanism for controlling the cytokine storm in influenza creased expression of GATA-3. Overexpression of GATA-3 alters virus infection. the relative expression between T-bet and GATA-3 and perhaps Preserved T cell proliferation despite disarmament of effector suppresses cytokine production by decoupled HA-specific CD4+ cytokine response has been described as an intermediate stage of T cells, in ways similar to the known suppression of IFN-g pro- development for Treg in systems of autoimmunity (16–19). duction in T-bet–expressing CD4+ T cells by GATA-3 overex- However, it appears that decoupling proliferation and effector pression (24, 25). However, the reason for suppression of IL-4 cytokine expression help memory T cell generation in viral in- production in GATA-3–overexpressing decoupled HA-specific fection, particularly in influenza (our data) (13). With disarmed CD4+ T cells is not clear. effector function, the level of exhaustion decreases in decoupled The Th1/Th2 paradigm has recently shifted to the Th1/Th2/ influenza-specific CD4+ T cells. Preserved proliferation and less Th17/Treg hypothesis, a multilineage commitment from the contraction enable a higher frequency of Ag-experienced cells same Th precursor cells (26, 27). Consequently, the relationship to survive. Compared with naive, these cells become trained for between the transcription factors T-bet, GATA-3, ROR-gt, and better eradication of virus to provide resistance to reinfection. Foxp3 has become important in determining the commitment Upon reinfection with influenza virus, these cells also respond of the cells. In our experiments, more IFN-g production than with higher IFN-g production; this is in comparison with the re- IL-17a by HA-specific CD4+ T cells is with T-bet dominance sponse driven by cells with coupled proliferation and effector over ROR-gt in coupled proliferation and cytokine response to cytokine production during an initial infection. acute influenza infection. Suppression of ROR-gt–mediated Upon acute infection with influenza virus, proliferation and Th17 cell differentiation by T-bet has been shown in different cytokine expression by HA-specific CD4+ T cells have been linked experimental systems, although the mechanism remains to be together in our experiments and shown in several other systems as characterized (28–32). Altered dominance of T-bet over ROR-gt The Journal of Immunology 4213

FIGURE 8. Efficient viral clearance with restrained IFN-g production by decoupled HA-specific CD4+ T cells provides better memory response during reinfection with influenza virus. Wild-type C57/B10.D2 mice were infected with 2.5 3 103 PFU PR8 or PR8- H3N2 or WSN strains of influenza A virus and adop- 3 6 tively transferred with 2.5 10 influenza HA-specific Downloaded from naive CD4+ T cells on day 14 postinfection. (A)On day 7 after adoptive transfer, the percentages of clonotypic and endogenous T cells in the lungs were compared ex vivo (Naive: uninfected mice with syn- chronized adoptive transfer). The IFN-g– and IL-2– producing clonotypic and endogenous T cells were

compared by intracellular staining. (B) On day 7 after http://www.jimmunol.org/ adoptive transfer, all mice were infected with 2.5 3 103 PFU PR8 strain of inﬂuenza virus. Naive mice receiving an adoptive transfer of 2.5 3 106 naive HA- speciﬁc CD4+ T cells and infection of 2.5 3 103 PFU PR8 virus served as the control. On day 7 after secondary infection, viral titer in the lungs, Ab response, percentages of T cells, and the IFN-g– and IL-2–producing T cells were compared. Data represented as mean 6 SD (***p , 0.0001, n = 6/group). by guest on October 4, 2021

coincides with the decoupling of proliferation and cytokine re- with decreased IFN-g receptor 1 expression, not decreased ex- sponse in our experiments. The alteration does not affect IL-17a pression of T-bet, perhaps facilitated survival of a higher fre- production, but a serious impairment of IFN-g production has quency of decoupled influenza-specific CD4+ T cells. been observed. The relationship between ROR-gt and T-bet is The thymus releases mature naive T cells to the periphery perhaps similar to the relationship between GATA-3 and T-bet. through a continuous process. Various waves of naive T cells For the fourth factor of the Th1/Th2/Th17/Treg hypothesis, encounter Ag differentially during the course of an influenza vi- Foxp3+ and other Treg are comparable between decoupled and rus infection. Our results describe the response by naive T cells that coupled proliferation and effector cytokine response in our appears in the later phase of an influenza infection. Whereas there is experiments. a decoupling of proliferation and effector cytokine response, these Strong T bet expression leads to high expression of the receptor cells significantly contribute to the generation of memory T cells. for IFN-g (33), a factor that may be important in T cell contraction Mechanistically, the disarmament of effector cytokines in prolif- (34, 35). Compared with the cells responding with effector cyto- erating influenza-specific CD4+ T cells appears with an alteration kines, T-bet dominance over GATA-3 and ROR-gt was altered of relative expressions, but not impeded induction, of transcription in decoupled influenza-specific CD4+ T cells in our experiments. factors. These results provide a mechanism for restraining over- The expression of IFN-g receptor 1 was less in the decoupled whelming acute-phase reactions without impeding the generation influenza-specific CD4+ T cells as well. Loss of T-bet dominance of memory T cells. Perhaps methods to alter T-bet dominance may 4214 T-bet DOMINANCE AND IFN-g DECOUPLING be a solution for controlling the damaging cytokine storm seen in 16. Chen, T. C., S. P. Cobbold, P. J. Fairchild, and H. Waldmann. 2004. Generation of anergic and regulatory T cells following prolonged exposure to a harmless the clinical setting of severe influenza infections. antigen. J. Immunol. 172: 5900–5907. 17. Sarween, N., A. Chodos, C. Raykundalia, M. Khan, A. K. Abbas, and Acknowledgments L. S. Walker. 2004. CD4+CD25+ cells controlling a pathogenic CD4 response inhibit cytokine differentiation, CXCR-3 expression, and tissue invasion. J. We thank Dr. Aimee S. Marson for advice and help in the preparation of the Immunol. 173: 2942–2951. manuscript. 18. Huang, C. T., C. J. Workman, D. Flies, X. Pan, A. L. Marson, G. Zhou, E. L. Hipkiss, S. Ravi, J. Kowalski, H. I. Levitsky, et al. 2004. Role of LAG-3 in regulatory T cells. Immunity 21: 503–513. Disclosures 19. Huang, C. T., D. L. Huso, Z. Lu, T. Wang, G. Zhou, E. P. Kennedy, C. G. Drake, The authors have no financial conflicts of interest. D. J. Morgan, L. A. Sherman, A. D. Higgins, et al. 2003. CD4+ T cells pass through an effector phase during the process of in vivo tolerance induction. J. Immunol. 170: 3945–3953. 20. Qin, S., S. P. Cobbold, H. Pope, J. Elliott, D. Kioussis, J. Davies, and References H. Waldmann. 1993. “Infectious” transplantation tolerance. Science 259: 974– 1. Simmons, C., and J. Farrar. 2008. Insights into inflammation and influenza. N. 977. Engl. J. Med. 359: 1621–1623. 21. Avni, O., and A. Rao. 2000. T cell differentiation: a mechanistic view. Curr. 2. To, K. F., P. K. Chan, K. F. Chan, W. K. Lee, W. Y. Lam, K. F. Wong, N. L. Tang, Opin. Immunol. 12: 654–659. D. N. Tsang, R. Y. Sung, T. A. Buckley, et al. 2001. Pathology of fatal human in- 22. Bird, J. J., D. R. Brown, A. C. Mullen, N. H. Moskowitz, M. A. Mahowald, fection associated with avian influenza A H5N1 virus. J. Med. Virol. 63: 242–246. J. R. Sider, T. F. Gajewski, C. R. Wang, and S. L. Reiner. 1998. Helper T cell 3. Cheung, C. Y., L. L. Poon, A. S. Lau, W. Luk, Y. L. Lau, K. F. Shortridge, differentiation is controlled by the cell cycle. Immunity 9: 229–237. S. Gordon, Y. Guan, and J. S. Peiris. 2002. Induction of proinflammatory 23. Hwang, E. S., S. J. Szabo, P. L. Schwartzberg, and L. H. Glimcher. 2005. T cytokines in human macrophages by influenza A (H5N1) viruses: a mechanism helper cell fate specified by kinase-mediated interaction of T-bet with GATA-3. for the unusual severity of human disease? Lancet 360: 1831–1837. Science 307: 430–433. 4. de Jong, M. D., C. P. Simmons, T. T. Thanh, V. M. Hien, G. J. Smith, T. N. Chau, 24. Usui, T., R. Nishikomori, A. Kitani, and W. Strober. 2003. GATA-3 suppresses D. M. Hoang, N. V. Chau, T. H. Khanh, V. C. Dong, et al. 2006. Fatal outcome of Th1 development by downregulation of Stat4 and not through effects on IL- Downloaded from human influenza A (H5N1) is associated with high viral load and hyper- 12Rbeta2 chain or T-bet. Immunity 18: 415–428. cytokinemia. Nat. Med. 12: 1203–1207. 25. Yagi, R., I. S. Junttila, G. Wei, J. F. Urban, Jr., K. Zhao, W. E. Paul, and J. Zhu. 5.Kobasa,D.,S.M.Jones,K.Shinya,J.C.Kash,J.Copps,H.Ebihara, 2010. The transcription factor GATA3 actively represses RUNX3 protein- Y. Hatta, J. H. Kim, P. Halfmann, M. Hatta, et al. 2007. Aberrant innate regulated production of interferon-gamma. Immunity 32: 507–517. immune response in lethal infection of macaques with the 1918 influenza 26. Weaver, C. T., R. D. Hatton, P. R. Mangan, and L. E. Harrington. 2007. IL-17 virus. Nature 445: 319–323. family cytokines and the expanding diversity of effector T cell lineages. Annu. 6. Peiris, J. S., W. C. Yu, C. W. Leung, C. Y. Cheung, W. F. Ng, J. M. Nicholls, Rev. Immunol. 25: 821–852.

T. K. Ng, K. H. Chan, S. T. Lai, W. L. Lim, et al. 2004. Re-emergence of fatal 27. Bettelli, E., T. Korn, and V. K. Kuchroo. 2007. Th17: the third member of the http://www.jimmunol.org/ human influenza A subtype H5N1 disease. Lancet 363: 617–619. effector T cell trilogy. Curr. Opin. Immunol. 19: 652–657. 7. Shen, H. H., J. Hou, W. W. Chen, B. K. Bai, H. B. Wang, T. S. Guo, A. X. Liu, 28. Mathur, A. N., H. C. Chang, D. G. Zisoulis, R. Kapur, M. L. Belladonna, Y. L. Li, M. Zhao, P. Y. Mao, et al. 2011. Immunologic changes during Pandemic G. S. Kansas, and M. H. Kaplan. 2006. T-bet is a critical determinant in the (H1N1) 2009, China. Emerg. Infect. Dis. 17: 1053–1055. instability of the IL-17-secreting T-helper phenotype. Blood 108: 1595– 8. Sun, J., R. Madan, C. L. Karp, and T. J. Braciale. 2009. Effector T cells control 1601. lung inflammation during acute influenza virus infection by producing IL-10. 29. Gocke, A. R., P. D. Cravens, L. H. Ben, R. Z. Hussain, S. C. Northrop, Nat. Med. 15: 277–284. M. K. Racke, and A. E. Lovett-Racke. 2007. T-bet regulates the fate of Th1 and 9. Lucas, S. J., D. W. Barry, and P. Kind. 1978. Antibody production and protection Th17 lymphocytes in autoimmunity. J. Immunol. 178: 1341–1348. against influenza virus in immunodeficient mice. Infect. Immun. 20: 115–119. 30. Abromson-Leeman, S., R. T. Bronson, and M. E. Dorf. 2009. Encephalitogenic 10. Sarawar, S. R., M. Sangster, R. L. Coffman, and P. C. Doherty. 1994. Admin- T cells that stably express both T-bet and RORgt consistently produce IFN- istration of anti-IFN-gamma antibody to beta 2-microglobulin-deficient mice gamma but have a spectrum of IL-17 profiles. J. Neuroimmunol. 215: 10–24.

delays influenza virus clearance but does not switch the response to a T helper 31. Guo, S., D. Cobb, and R. B. Smeltz. 2009. T-bet inhibits the in vivo differen- by guest on October 4, 2021 cell 2 phenotype. J. Immunol. 153: 1246–1253. tiation of parasite-specific CD4+ Th17 cells in a T cell-intrinsic manner. J. 11. Topham, D. J., R. A. Tripp, S. R. Sarawar, M. Y. Sangster, and P. C. Doherty. 1996. Immunol. 182: 6179–6186. Immune CD4+ T cells promote the clearance of influenza virus from major histo- 32. Rivera, A., T. M. Hohl, N. Collins, I. Leiner, A. Gallegos, S. Saijo, J. W. Coward, compatibility complex class II2/2 respiratory epithelium. J. Virol. 70: 1288–1291. Y. Iwakura, and E. G. Pamer. 2011. Dectin-1 diversifies Aspergillus fumigatus- 12. Romań, E., E. Miller, A. Harmsen, J. Wiley, U. H. Von Andrian, G. Huston, and specific T cell responses by inhibiting T helper type 1 CD4 T cell differentiation. S. L. Swain. 2002. CD4 effector T cell subsets in the response to influenza: J. Exp. Med. 208: 369–381. heterogeneity, migration, and function. J. Exp. Med. 196: 957–968. 33. Harris, D. P., S. Goodrich, A. J. Gerth, S. L. Peng, and F. E. Lund. 2005. 13. Jelley-Gibbs, D. M., D. M. Brown, J. P. Dibble, L. Haynes, S. M. Eaton, and Regulation of IFN-gamma production by B effector 1 cells: essential roles for T- S. L. Swain. 2005. Unexpected prolonged presentation of influenza antigens bet and the IFN-gamma receptor. J. Immunol. 174: 6781–6790. promotes CD4 T cell memory generation. J. Exp. Med. 202: 697–706. 34. Haring, J. S., and J. T. Harty. 2006. Aberrant contraction of antigen-specific CD4 14. Lukacher, A. E., V. L. Braciale, and T. J. Braciale. 1984. In vivo effector function T cells after infection in the absence of gamma interferon or its receptor. Infect. of influenza virus-specific cytotoxic T lymphocyte clones is highly specific. J. Immun. 74: 6252–6263. Exp. Med. 160: 814–826. 35. Joshi, N. S., W. Cui, A. Chandele, H. K. Lee, D. R. Urso, J. Hagman, L. Gapin, 15. Lin, C. Y., L. Graca, S. P. Cobbold, and H. Waldmann. 2002. Dominant trans- and S. M. Kaech. 2007. Inflammation directs memory precursor and short-lived plantation tolerance impairs CD8+ T cell function but not expansion. Nat. effector CD8(+) T cell fates via the graded expression of T-bet transcription Immunol. 3: 1208–1213. factor. Immunity 27: 281–295.