Draining Lymph Nodes in Response to Pulmonary Virus Infection

This information is current as of November 18, 2018. Sequential Activation of CD8 + T Cells in the Draining Lymph Nodes in Response to Pulmonary Virus Infection Heesik Yoon, Kevin L. Legge, Sun-sang J. Sung and Thomas J. Braciale J Immunol 2007; 179:391-399; ; doi: 10.4049/jimmunol.179.1.391 http://www.jimmunol.org/content/179/1/391 References Subscription Permissions Email Alerts This article cites 42 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/179/1/391.full#ref-list-1 Why The JI? Submit online. Rapid Reviews! 30 days* from submission to initial decision No Triage! Every submission reviewed by practicing scientists Fast Publication! 4 weeks from acceptance to publication *average Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Submit copyright permission requests at: http://www.aai.org/about/publications/ji/copyright.html Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts Downloaded from http://www.jimmunol.org/ by guest on November 18, 2018 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 2007 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606.

The Journal of Immunology Sequential Activation of CD8 T Cells in the Draining Lymph Nodes in Response to Pulmonary Virus Infection 1 Heesik Yoon,* Kevin L. Legge,* Sun-sang J. Sung, and Thomas J. Braciale 2 * We have used a TCR-transgenic CD8 T cell adoptive transfer model to examine the tempo of T cell activation and proliferation in the draining lymph nodes (DLN) in response to respiratory virus infection. The T cell response in the DLN differed for mice infected with different type A influenza strains with the onset of T cell activation/proliferation to the A/JAPAN virus infection preceding the A/PR8 response by 12 24 h. This difference in T cell activation/proliferation correlated with the tempo of accelerated respiratory DC (RDC) migration from the infected lungs to the DLN in response to influenza virus infection, with the migrant RDC responding to the A/JAPAN infection exhibiting a more rapid accumulation in the lymph nodes (i.e., peak migration for A/JAPAN at 18 h, A/PR8 at 24 36 h). Furthermore, in vivo administration of blocking anti-cd62l Ab at various time points before/after infection revealed that the virus-specific CD8 T cells entered the DLN and activated in a sequential conveyor belt -like fashion. These results indicate that the tempo of CD8 T cell activation/proliferation after viral infection is dependent on the tempo of RDC migration to the DLN and that T cell activation occurs in an ordered sequential fashion. The Journal of Immunology, 2007, 179: 391 399. Naive T cells continually recirculate through the blood and lymph and, during recirculation, they enter the secondary lymphoid organs (SLO) 3 (e.g., lymph nodes (LN), Peyer s patches, and spleen) searching for cognate Ag delivered and displayed in the SLO by professional APC (e.g., dendritic cells (DC)) (1). Naive T cells enter the LN across high endothelial venules (HEV) by multistep adhesion cascades and, whether T cells encounter specific Ag there, these T cells activate/proliferate in the T cell-enriched LN (2, 3). This T cell/ APC encounter is believed to occur primarily in the LN paracortex (where HEV are enriched). Intravital LN microscopy studies suggest that, upon entering the LN draining the peripheral sites of Ag deposition/infection (via the afferent lymphatics), the Ag-bearing DC migrate to and localize in the vicinity of the HEV (4, 5). This HEV localization event facilitates the APC encounter with naive Ag-specific T cells entering the LN from the blood. Once activated in the LN cortex, the responding T cells proliferate over several days, and they are thought to then migrate to the LN medulla and egress from the LN (via the efferent lymphatics) (5 7). Important information on the in vivo *Carter Immunology Center and Department of Pathology, Department of Microbiology, and Department of Internal Medicine, University of Virginia Health Sciences Center, Charlottesville, VA 22908; and Department of Microbiology, University of Iowa, Iowa City, IA 52242 Received for publication January 22, 2007. Accepted for publication April 11, 2007. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by U.S. Public Health Service Grants AI-15608, HL- 33391, and AI-37293 (to T.J.B.) and HL-70065 (to S.J.S.). K.L.L. was supported by the Cancer Research Institute. 2 Address correspondence and reprint requests to Dr. Thomas J. Braciale, Carter Immunology Center, University of Virginia, P.O. Box 801386, Charlottesville, VA 22908. E-mail address: tjb2r@virginia.edu 3 Abbreviations used in this paper: SLO, secondary lymphoid organ; DC, dendritic cell; LN, lymph node; HEV, high endothelial venule; RDC, respiratory DC; DLN, draining LN; Tg, transgenic; RT, room temperature; LYVE-1, lymphatic endothelium hyaluronic acid receptor 1; i.n., intranasal(ly); HA, hemagglutinin; p.i., postinfection; NDLN, nondraining LN; CL-4, Clone-4. Copyright 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00 activation/proliferation tempo of naive T cells has come from studies using the adoptive transfer of T cells labeled with dilutionsensitive dyes (e.g., CFSE) into recipient mice subsequently challenged with Ag (5, 6, 8). Such studies have typically revealed evidence of extensive proliferation of transferred T cells in the SLO after Ag contact with considerable heterogeneity in the apparent tempo of T cell proliferation in situ (9 11). The murine response to type A influenza infection has been extensively used as a model for the analysis of the requirements for CD8 T cell induction in vivo (e.g., Refs. 12 and 13), the role of CD8 T cells in virus clearance, and T cell effector mechanisms used in recovery from infection (14 16). More recently, this model has been used in the analysis of the response of respiratory DC (RDC) to pulmonary infection and of the role of RDC in CD8 T cell induction in the draining LN (DLN) following intranasal infection (17). By using TCR-transgenic (Tg) mouse lines whose CD8 (or CD4 ) T cells express TCR directed to type A influenza epitopes, we (10, 18, 19) and others (20 23) have adapted the strategy of adoptively transferring naive TCR-Tg T cells into influenza-infected recipients to evaluate early events in the naive T cell response to Ag during natural infection. In this current report, we have used this strategy to analyze the in vivo CD8 T cell response to murine infection with two different influenza strains. We demonstrate that the tempo of naive T cell activation/proliferation correlates with the tempo of RDC migration (from the respiratory tract to the DLN), and we demonstrate that the heterogeneity in the T cell proliferative response in the DLN reflects the sequential immigration of circulating naive Ag-specific CD8 T cells into the LN followed by their subsequent activation/proliferation there. Materials and Methods Mice The Clone-4 (CL-4) Tg mice and Thy1.1 BALB/c mice were provided originally by Dr. R. W. Dutton (Trudeau Institute, Saranac Lake, NY) and Dr. R. I. Enelow (Yale University, New Haven, CT), respectively, and maintained under a pathogen-free environment. www.jimmunol.org

392 SEQUENTIAL ACTIVATION/PROLIFERATION OF NAIVE CD8 T CELLS IN VIVO Virus and viral peptides The preparation of influenza virus A/PR/8/34 (H1N1) and mouse-adapted influenza virus A/JAPAN/305/57 (H2N2) were described elsewhere (10). Synthetic peptide HA 529 537 of A/JAPAN (IYATVAGSL) and HA 533 541 of A/PR8 (IYSTVASSL) were synthesized by the University of Virginia Biomolecular Research Facility. Their purity was confirmed by HPLC. Adoptive transfer of clone 4 CD8 T cells Naive CD8 T cells were purified from the spleen by positive magnetic bead selection (MACS; Miltenyi Biotec) as described earlier (18). More than 90% purified CD8 T cells were then washed with serum-free DMEM and labeled with 5 M CFSE (Molecular Probes) for 10 min at RT. After extensive washing, 10 7 labeled cells were transferred into Thy1.1 BALB/c. Mice were rested at least for 24 h, then intranasally infected with a 0.1 LD 50 of influenza virus. CFSE-labeled CL-4 T cells were also used in in vitro cultures. In vitro culture 2 10 6 of purified and CFSE-labeled CL-4 CD8 cells were incubated with 2 10 7 of peptide-loaded or virus-infected Thy1.1 BALB/c splenocytes in IMEM supplemented with 10% FBS, 10 U/ml penicillin G, 10 g/ml streptomycin sulfate, 2 mm L-glutamine, and 0.05% 2-ME without IL-2. Splenocytes were pulsed with 100 nm peptides for 1 h at RT or infected at either a multiplicity of infection of 6.3 for A/PR8 or 0.63 for A/JAPAN. Free unbound peptides or viruses were washed away before adding the splenocytes to the coculture. Virus titers Virus titers were determined as described previously (24). Briefly, homogenated lung suspensions of influenza virus-infected mice were plated on Mardin-Darby canine kidney cells in flat-bottom 96-well plates with multiple dilutions. After 3 days of incubation in a humidified 7% CO 2 incubator at 37 C, the supernatant of each well was mixed with fresh 1% chicken erythrocytes. Hemagglutination results were analyzed after 1 h of incubation at RT and TCID 50 was calculated. Detection of migrating RDC Migrating RDC in the respiratory DLN were enumerated as previously described (17). Briefly, 50 l ofa8mmcfse stock was intranasally administered to mice to label RDC and, at 6 h after CFSE labeling, the mice were then infected with influenza virus. At specified time points after viral infection, the DLN were harvested and the numbers of CD11c and CFSE cells were counted by flow cytometry. Preparation of tissue lymphocytes and flow cytometry At the time of collection, the LN or lungs were removed and placed in ice-cold FACS staining buffer (PBS with 2% FBS and 0.02% NaN 3 ). LN or lung tissues were then subsequently disrupted and passed through a cell strainer (70 m; BD Falcon). Approximately 10 6 cells were incubated with anti-cd16/32 (2.4G2) to block nonspecific FcR binding, stained with anti- CD8 (53-6.7), anti-thy1.2 (53-2.1), and anti-cd69 (H1.2F3; BD Pharmingen, and fixed in FACS lysing solution (BD Biosciences). Flow cytometric acquisitions were done with FACSCalibur or FACSCanto (BD Pharmingen) and analyzed with FlowJo software (Tree Star). Blocking entry of T cells into LN by MEL-14 Abs Anti-CD62L mab (MEL-14) was produced and purified by the University of Virginia Lymphocyte Culture Center. To block further entry of T cells into LN, 100 g of whole IgG 2a molecules was injected i.v. into the tail vein of CFSE CD8 CL-4-transferred Thy1.1 BALB/c mice at various times before or after influenza virus infection. The persistence, specificity, and duration of Ab binding to CD8 CL-4 T cells were confirmed until 4 days after administration by flow cytometry (data not shown). Enrichment of CL-4 T cells for flow cytometry CL-4 T cells of low frequency were enriched by pull-down assay (25). Briefly, cells from homogenated and pooled LN were incubated with anti- CD90.2 magnetic beads (Miltenyi Biotec) for 15 min. After purification, the positively selected fractions were subjected to routine flow cytometry analysis. Immunohistochemistry of LN LN were excised, frozen in OCT and sectioned at 5- m thickness. The LN sections were blocked with 2.4G2 anti-fc-rii/riii mab and serum and stained with Alexa Fluor 555- or Alexa Fluor 647-conjugated anti-lineage mab (Caltag Laboratories), anti-thy1.2 mab, anti-hev mab MECA-79 (BD Pharmingen), or rabbit anti-lymphatic vessel endothelial hyaluronic acid receptor 1 (LYVE-1) Ab (AngioBio). Anti-Thy1.2, MECA-79, and LYVE-1 were conjugated to the Alexa Fluors by mab labeling kits according to the supplier s (Molecular Probes) protocol. Three-color confocal microscopy was performed on a Zeiss LSM 510 assembly and the data were analyzed using the manufacturer s Image Browser software. Results Kinetics of the in vivo CD8 T cell response to influenza virus infection in the DLN We examined the activation kinetics of naive TCR-Tg CD8 T cells in vivo in response to intranasal (i.n.) influenza virus infection in a CD8 T cell/adoptive transfer model (10). CD8 T cells were purified from the influenza hemagglutinin (HA)-specific TCR-Tg, CL-4 mouse line (10). These T cells recognize a site within the transmembrane domain of the A/PR8 (H1N1) HA corresponding to HA residues 533 541 (HA 518 526 in the mature HA). The CL-4 T cells also recognize in a cross-reactive fashion, albeit less efficiently (R. M. Chu and T. J. Braciale, unpublished data), the corresponding HA amino acid sequence in the A/JAPAN (H2N2) HA (i.e., HA residues 529 537 which differs in two amino acids from the corresponding A/PR8 HA site). To examine the early phase of CD8 T cell activation in vivo in response to infection, purified CL-4 T cells (Thy1.2 ) were labeled in vitro with the dilution-sensitive dye CFSE before transfer into naive Thy1.1 -congenic recipients. We routinely find that 98% of CL-4 T cells purified have a naive phenotype; they have uniform expression of the surface markers ex vivo and do not produce effector cytokines at 24 h poststimulation with cognate peptide in vitro (H. Yoon, unpublished data). Twenty-four hours later, recipient mice were infected i.n. with a sublethal dose of either A/PR8 or A/JAPAN virus. Over the ensuing 4 5 days, the spleen, lungs, peribronchial LN and mediastinal LN draining the respiratory tract and pooled nondraining LN (NDLN) were excised from the infected recipient mice and examined for the localization, activation, and proliferation of the transferred CL-4 T cells. CL-4 T cell activation was assessed by the up-regulation of the early activation marker CD69, and T cell proliferation was evaluated by CFSE dye intensity. Fig. 1 shows the results of this analysis for the lung-draining peribronchial and mediastinal LN (DLN). CL-4 activation (i.e., CD69 up-regulation) was minimal in the DLN at day 1 postinfection (p.i.) with either virus (Fig. 1A) and comparable to CL-4 T cells transferred into uninfected recipients (Ref. 10 and data not shown). Over the next 24 h, up-regulation of CD69 was detected in the DLN of both A/PR8- and A/JAPAN-infected mice, with a larger fraction of CD69 CL-4 T cells detected in the DLN of A/JAPAN-infected mice than in A/PR8-infected mice (i.e., 50% vs 25%, see Fig. 1A). Of particular note was the T cell proliferation tempo. In response to A/JAPAN infection, the CL-4 CD8 T cells underwent extensive proliferation in the DLN between days 2 and 3 p.i., while in response to A/PR8, their proliferation was delayed by up to 24 h (i.e., onset between days 3 and 4 p.i.). Also noteworthy is the finding that, after the onset of T cell division in the DLN (i.e., between days 2 and 3 for A/JAPAN infection and days 3 and 4 for A/PR8 infection), the activated CD8 T cells had undergone up to five to six divisions in vivo over a 24-h period. This result suggests that the cell cycle time in vivo for the activated responding T cells may be extremely rapid. Of equal interest is the observation that CD69 expression levels on the responding CD8 T cells decreases incrementally with each successive cell division (i.e., corresponding to the stepwise decrease in CFSE staining intensity). As previously reported (10), there is no significant activation/proliferation of naive CL-4 T cells present

The Journal of Immunology 393 FIGURE 1. Tempo of naive CD8 T cell activation in response to influenza virus infection. CL-4 TCR-Tg CD8 T cells were isolated from the spleen/ln of donor mice, labeled with CFSE, then administered i.v. to recipient mice. The recipient mice were subsequently infected i.n. with either A/PR8 or A/JAPAN virus. A, Time course of CL-4 T cell activation (i.e., CD69 up-regulation) and proliferation (i.e., CFSE dye dilution) in the DLN at the indicated day p.i. B, CD69 expression and CFSE content for labeled CL-4 T cells in the spleen, lungs, and NDLN at day 4 p.i. (A/PR8) or day 3 p.i. (A/ JAPAN). Data are representative of at least six independent experiments. in the spleen, NDLN, or lungs of these infected mice at day 3 p.i. with A/JAPAN infection or at day 4 p.i. with A/PR8 infection (Fig. 1B). Likewise, we do not detect any activation/proliferation of naive CL-4 T cells in the liver or bone marrow of A/PR8-infected mice and no evidence of activated/proliferating cells in the blood before day 4 of infection (H. Yoon, unpublished observations). These findings are consistent with the concept that the DLN are the primary site of CD8 T cell activation in response to pulmonary type A influenza virus infection. Consistent with recently published observations (26, 27), naive Tg CD8 T cells can be transferred into infected recipients up to 5 6 days p.i. with A/PR8 virus and still undergo several rounds of proliferative expansion in the DLN, and low-level activation/ proliferation of a small fraction of transferred Tg T cells can be FIGURE 2. Response of CD8 T cells to Ag-expressing APC in vitro. Purified CFSE dye-labeled CL-4 TCR-Tg CD8 T cells were cocultured with A/PR8 peptide (PHA 529 )- or A/JAPAN peptide (JHA 533 )- pulsed splenocyte APC (A and B) or with A/PR8 or A/JAPAN virus-infected splenocyte APC (C and D). At the indicated times (days) after in vitro culture, the cultured CL-4 T cells were examined for activation marker (i.e., CD69) expression (A and C) and cell division/cell accumulation (i.e., CFSE dilution; B and D). Data are representative of four independent experiments.

394 SEQUENTIAL ACTIVATION/PROLIFERATION OF NAIVE CD8 T CELLS IN VIVO observed when adoptive transfer is conducted up to 3 4 wk p.i. (C. W. Lawrence, H, Yoon, and T. J. Braciale, unpublished observations). Tempo of CD8 T cell activation in vitro The differences in the kinetics of CD8 T cell activation in vivo in response to infection with these two viruses could be due to a difference in the efficiency of processing and presentation of the respective cognate PR8 HA 533 541, and cross-reactive JAPAN HA 529 537 epitope by APC, or a difference in the efficiency of T cell recognition (i.e., TCR engagement) of these two epitopes by CL-4 T cells might also account for these results. To investigate these possibilities, we examined the tempo of CL-4 CD8 T cell activation in vitro after stimulation with splenic APC that had been either pulsed with one of the synthetic peptides (PHA 533 541 or JHA 529 537 ) or infected with virus (A/PR8 or A/JAPAN). As Fig. 2 demonstrates, naive CD8 T cells stimulated with either synthetic HA peptide rapidly activated (i.e., up-regulated CD69, by day 1 of in vitro culture; Fig. 2A) and underwent significant proliferation (i.e., two to three cell divisions; Fig. 2B) by day 2 poststimulation. CD69 up-regulation on the naive T cells exposed to preprocessed peptide in vitro was uniform, suggesting that Ag encounter and the initial activation of the T cells was relatively synchronous. Indeed, the majority of the cultured T cells had undergone two to three divisions by day 2 of in vitro culture (Fig. 2B). The activation/proliferative responses to the preprocessed peptide epitopes was comparable at the peptide dose used to pulse APC (100 nm). At lower peptide doses (down to 1 nm or 100 pm), the cross-reactive JHA 529 537 peptide was recognized with decreased efficiency, i.e., fewer CL-4 T cells activated, but did so with the same kinetics as the higher peptide doses. The lower limit of stimulation of the naive CD8 T cells by the partial agonist A/JAPAN peptide epitope was 1 nm with half-maximal T cell stimulation at a peptide dose of 50 nm, while the corresponding threshold and half-maximal peptide concentrations for the agonist A/PR8 peptide epitope were 10 and 500 pm, respectively. Coculture of purified naive CD8 T cells with control (untreated/irrelevant peptidepulsed) APC did not result in CL-4 T cell activation in vitro (data not shown). Analysis of the in vitro response to the virus-infected APC revealed that the kinetics of T cell activation (i.e., CD69 up-regulation; Fig. 2C) was similar to stimulation by preprocessed peptides. Here again, the tempo of activation/proliferation (Fig. 2, C and D) was comparable for the two virus strains, although overall stimulation by A/JAPAN-infected APC was less efficient by day 3 of stimulation than with A/PR8-infected APC (Fig. 2D). Kinetics of virus replication in the lungs and migration of RDC in vivo The above in vitro analysis revealed minimal if any difference in presentation of the two influenza HA epitopes and recognition by the T cells. We and others (17, 28, 29) have previously suggested that RDC migrating from the respiratory tract to the DLN in response to respiratory viral infection may play a critical role in the delivery/presentation of viral Ag to naive virus-specific CD8 T cells in the DLN. Because inflammatory stimuli such as virus replication in the respiratory tract will drive RDC maturation and subsequent migration, it was of interest to examine the time course of virus replication/accumulation in lungs of mice after i.n. infection with A/JAPAN or A/PR8 and the tempo of migration of the RDC to the DLN in response to infection. Fig. 3A shows the results of the lung virus titer analysis. For each of these mouse-adapted virus strains, virus replication was rapid with A/JAPAN virus reaching steady-state virus titers by 24 h p.i. and A/PR8 virus reaching steady-state titers by 48 h p.i. For each infection, mice FIGURE 3. Kinetics of virus replication in the lungs and RDC migration to the DLN during influenza infection. Time course of virus replication in the lungs of mice after sublethal (0.1 LD 50 ) i.n. dose of A/JAPAN (3.2 10 3 PFU) or A/PR8 (10 PFU) virus. Virus titers at days 1 5 are the mean value for lungs from three donors at each time point. Virus titers ( ) atday 0 indicate the virus inoculum used for i.n. infection (A). Mice received CFSE dye by i.n. instillation 18 24 h before influenza virus infection. At the indicated time points (hours after infection), the DLN were harvested and the number of CFSE migrant RDC within the DLN of infected mice was determined (B). Values are the number of RDC/LN in pooled DLN from three mice per time point. Data are representative of two independent experiments. were inoculated with an i.n. dose of infectious virus corresponding to 0.1 LD 50 units. However, the i.n. inoculum dose for the more virulent A/PR8 virus (10 PFU) was 300-fold lower than that for the A/JAPAN virus (3.2 10 3 PFU). The difference in the initial rate of virus accumulation in the infected lungs between the two virus strains and the time difference before steady-state virus titers are achieved likely reflects this difference in initial inoculum size. When we examined the time course of RDC migration to the DLN in response to infection, a corresponding difference was observed (Fig. 3B). RDC undergo a transient accelerated migration to the DLN (for both A/JAPAN and A/PR8) during the first 48 h p.i. However, consistent with the difference in the accumulation of infectious virus in the lungs, RDC migrating in response to A/JAPAN infection reached peak accumulation in the DLN more rapidly than RDC responding to the A/PR8 virus.

The Journal of Immunology 395 FIGURE 4. Effect of anti-cd62l administration on CD8 T cell proliferation in the DLN. Naive CFSE-labeled CL-4 TCR-Tg CD8 T cells were transferred into recipient mice 24 h before i.n. infection with A/PR8 virus. Mice received no Ab (A) or a single i.v. dose of anti-cd62l mab at the following times p.i.; B, at 24 h before infection ( 24 h); C, at the time of infection (0 h); D, at36h;ore, at 60 h p.i. The pooled DLN from three mice for each treatment group were excised and analyzed for activation marker expression (CD69 level) and proliferation (CFSE dilution) at 3.5 days p.i. Numbers above subpopulations in each dot plot represent the number of cell divisions that the responding T cells have undergone in the DLN. Insets are the corresponding histogram plots of cell numbers (y-axis) vs CFSE dilution (x-axis) for each group at the time of DLN harvest. Data are representative of the following number of independent experiments: A,10; B/C, 2; D, 3; ande, 4. CD8 T cells activate in a synchronous sequential manner in the DLN The more rapid onset of RDC migration to the DLN following i.n. infection with A/JAPAN virus paralleled the higher percentage of CD69 cells and the more rapid onset of cell division by the CD8 T cells responding in the DLN to this strain (Fig. 1A). This finding suggested a link between the time course of RDC migration to the DLN and the response of virus-specific CD8 T cells trafficking through this site. Consistent with this notion, we detected minimal CD69 up-regulation on CL-4 T cells present in the DLN at 24 h p.i. (Fig. 1A). However, by 48 h p.i., after RDC had initiated their accelerated migration to the DLN (Fig. 3B), CD69 was up-regulated on a significant fraction of CL-4 T cells in the DLN of both infected groups with a larger fraction of CD69 CL-4 T cells observed in response to A/JAPAN infection ( 50%) than with A/PR8 infection ( 25%; Fig. 1A). Thus, the timing of the viral Ag delivery to the DLN by migrant RDC appears to dictate the onset of the CD8 T response in the DLN. Because the initial accelerated wave of viral Ag delivery to the DLN by migrant RDC occurs over a narrow time frame (i.e., between 6 and 36 48 h p.i.; Fig. 3B), viral Ag presentation and therefore CD8 T cell activation/proliferation in vivo might be expected to occur in a relatively synchronous fashion in the DLN. However, the observations of in vivo T cell proliferation in this report (Fig. 1A) as well as in other reports (9, 10) suggest an asynchronous (based on CFSE dilution) proliferation pattern. This apparent asynchronous cell division pattern could represent heterogeneity in the response of a single cohort of naive CD8 T cells and their daughter cells in the DLN that have stopped proliferating after undergoing one or more cell divisions. An alternative hypothesis consistent both with the pattern of RDC migration observed here, and with the well-defined pattern of naive CD8 T FIGURE 5. Time course of CL-4 T cell proliferation in the DLN after anti-cd62l Ab administration. Naive CFSE-labeled CL-4 TCR-Tg CD8 T cells (10 7 ) were transferred into recipient mice 24 h before infection with a sublethal i.n. dose of A/PR8 virus (A). At 24 h p.i., mice received a single i.v. dose of anti-cd62l Ab (or a control, no Ab), and the DLN of the infected mice were harvested on the indicated days p.i., then analyzed for cell division (CFSE dilution). Numbers within panels indicate the number of CL-4 cell divisions as reflected in CFSE dye intensity. Data for each time point represent values for pooled DLN from three recipients for Abtreated mice and one recipient at each time point for untreated controls. Data are representative of six independent experiments for each treatment group/time. 5 10 4 purified CL-4 T cells were adoptively transferred into A/PR8-infected control mice (B) or infected mice receiving anti-cd62l Ab at 24 h p.i. (C). DLN were isolated and pooled from control mice (n 10) or Ab-treated mice (n 7) at day 4 p.i. Donor cells were enriched for Thy1.2 cells by pull-down assay for flow cytometry. cell recirculation through the SLO (1, 2), is that circulating naive virus-specific CD8 T cells continually enter the DLN, and, after RDC deposition, successive cohorts of T cell entering the DLN encounter Ag, activate, and then initiate the steps in cell division (all in a sequential manner reminiscent of a conveyor belt). Accordingly, the first T cells to enter the DLN after Ag deposition would also be the first to activate/proliferate; and these cells would undergo the greatest number of cell divisions (i.e., display the lowest CFSE intensity) in the DLN at the time of analysis. Consistent with this concept of circulating naive virus-specific T cells which subsequently enter the DLN would also activate/proliferate; but they would undergo fewer divisions at the time of DLN sampling than the first cohort that had arrived in the DLN. This conveyor belt effect would result in a pattern of apparent asynchronous cell division observed in the DLN in vivo after the onset of cell division (e.g., Fig. 1A). To test this conveyor belt hypothesis of sequential T cell activation in response to infection, we conducted a series of experiments to inhibit the migration of circulating CL-4 T cells from the blood into the DLN through the HEV by blocking CD62L function before/after i.n. infection by i.v. administration of the monoclonal anti-cd62l Ab Mel-14 (30). In the first series of experiments, CFSE-labeled CL-4 T cells were transferred into Thy1.1 -congenic recipients; then these T cell recipients received a single i.v. dose (100 g of mab/recipient) of Mel-14 Ab at 24, 0, 36, or 60 h p.i. with A/PR8 virus. Infected CL-4 T cell recipients receiving no Ab served as controls. Between 3 and 4 days p.i. (i.e., 3.5 days p.i.), when CL-4 T cell proliferation in response to infection has begun in the DLN (Fig. 4A), the DLN were excised and

396 SEQUENTIAL ACTIVATION/PROLIFERATION OF NAIVE CD8 T CELLS IN VIVO FIGURE 6. Localization of responding CL-4 T cells in the DLN of anti-cd62l Ab-treated and control mice. DLN from recipient mice of CFSE-labeled CL-4 Thy1.2 CD8 T cells were excised, sectioned, and subjected to immunofluorescence microscopy at day 1 (A) or day 4 (B E) after i.n. A/PR8 infection. Day 4 DLN donors were untreated (B and C) or treated with a single i.v. dose of anti- CD62L Ab (D and E) at 24 h p.i. Day 1 infected LN shows transferred CFSE /high CL-4 T cells with CFSE (green) and Thy1.2 (blue) colocalization and location of HEV, stained with MECA-79 (red). B, Staining of LN lymphatic endothelium (turquoise, short arrow) and HEV (red) of day 4 untreated DLN with DLN cortex (cortex identified). C, Distribution of CFSE /low CL-4 T cells (blue) in the DLN of day 4 p.i. control mice. D, Staining of lymphatic endothelium (turquoise) and HEV (red) in DLN of day 4 p.i. anti-cd62l-treated mice. Medullary sinuses are identified by MS. E, Localization of CFSE /low Thy1.2 CL-4 cells (blue) to the medullary cords of DLN of day 4 p.i. anti-cd62ltreated mice. analyzed for CL-4 T cell localization/proliferation. Spleens and NDLN of the recipients were excised and analyzed in parallel. As expected, administration of blocking anti-cd62l Ab at 24 h p.i. (Fig. 4B) or 0 h p.i. (Fig. 4C) resulted in almost complete depletion of transferred CL-4 T cells from the DLN with no evidence of significant CL-4 cell proliferation (CFSE dye dilution). The NDLN were likewise depleted of CL-4 T cells; but the transferred T cells were enriched in the spleen (data not shown), consistent with the reported CD62L-independent trafficking of circulating naive T cells to the murine spleen (30). In contrast, Ab administration at 60 h p.i. (before the onset of CL-4 T cell proliferation in the DLN to A/PR8 virus but after CD69 up-regulation on the T cells in the DLN) had minimal effect on the proliferative response of the transferred T cells to infection with at least five CL-4 T cell divisions detected in the DLN (Fig. 4E), a value comparable to untreated control CL-4 recipients infected (Fig. 4A). Ab administration at 36 h p.i. (when at least a portion of the circulating CL-4 T cells had entered the DLN and encountered Ag) resulted in a distinctly different CL-4 T cell proliferation profile (CFSE dilution). Inhibition of further naive CL-4 T cell migration into the DLN at 36 h p.i. produced a population of responding T cells displaying a restricted range of CFSE dye intensity, suggesting that the majority of T cells had undergone three to five divisions at the time of analysis (Fig. 4D). This finding was consistent with the hypothesis that a limited number of naive CL-4 T cells had sequentially entered the DLN and activated/proliferated in response to an Ag in a sequential manner. In this connection, it should be noted that identical results were obtained when F(ab ) 2 of the Mel-14 Ab were administered in vivo, indicating that the effect of the Ab was not due to Ab Fc-dependent sequestration of the transferred CL-4 T cells (data not shown). The above results suggested that, if any additional migration of circulating CL-4 T cells into the DLN were inhibited after the first cohort of T cells had entered the DLN in response to viral infection, then this initial cohort of T cells would proliferate in a synchronous manner, i.e., this early cohort of CL-4 T cells would respond as a single population displaying a single-cell division number (i.e., uniform CFSE dye dilution intensity) at the time of

The Journal of Immunology DLN sampling. To examine this possibility, CL-4 T cell recipients received blocking Mel-14 Ab at 24 h p.i. with A/PR8 virus, and the DLN of both Ab recipients and control (untreated) animals were analyzed at days 3 5 p.i. As Fig. 5A demonstrates, the single cohort of T cells that entered the DLN by 24 h p.i. (i.e., after RDC migration from the lungs to DLN had been initiated) exhibited a remarkably restricted CFSE dye intensity which is consistent with the synchronous division of this first naive T cell cohort. Furthermore, at day 4 p.i., the T cells from the Ab-treated recipients had undergone five divisions (based on CFSE dye intensity), approximately the same number of divisions as the cells that had undergone the most extensive division (lowest dye intensity) in the untreated control CL-4 recipients (Fig. 5A). This finding suggests that the T cells in the control DLN which had undergone five rounds of division at day 4 p.i. represented the first cohort of naive CL-4 cells to respond to viral infection in the DLN. In the above analysis, 10 7 CFSE-labeled TCR-Tg CL-4 T cells were adoptively transferred into recipient mice to allow for ready isolation of sufficient numbers of the responding T cells in the DLN of infected animals for subsequent analysis, particularly after Mel-14 Ab treatment. Both the pattern of sequential activation of the responding T cells in the DLN and the synchronization of the responding T cell cohort in the draining node after Mel-14 Ab treatment was observed when a 200-fold lower number of Tg T cells (i.e., 5 10 4 CL-4 T cells/recipient) was used in the adoptive transfer (Fig. 5, B and C). Localization of synchronized CL-4 T cells in the DLN If administration of anti-cd62l Ab at 24 h p.i. restricted the responding T cells to a single early cohort of circulating CL-4 T cells entering the DLN, then this cell cohort might likewise traffic from the DLN cortex to the medulla in a synchronized manner (i.e., sequentially localize to distinct anatomical sites in the DLN during the course of naive T cell activation and proliferation in response to infection). To explore this possibility, we examined the localization of CFSE-labeled CL-4 T cells in the DLN of infected mice by immunofluorescence microscopy of DLN sections removed at day 1 p.i. when anti-cd62l was administered or at day 4 p.i. of Ab-treated or control (untreated)-infected CL-4 T cell recipients. Day 4 p.i. was chosen as the time for DLN sampling because our earlier studies (10) indicated that this was a time point when responding CL-4 T cells had undergone extensive proliferation in the DLN but had not as yet exited from the DLN and trafficked to the lungs. We used Abs specific for HEV (anti-peripheral LN addressin, MECA-79) and for LYVE (anti-lyve-1) to distinguish the cortex and the medulla of the infected DLN. Dye-labeled CL-4 T cells were identified at day 1 p.i. by the intense CFSE dye fluorescence and at day 4 p.i. (after a loss of CFSE dye staining due to T cell proliferation) by staining with anti-thy1.2 Abs to distinguish the transferred CL-4 T cells from recipient (Thy1.1 ) T cells in the DLN. As Fig. 6A demonstrates, at day 1 p.i., CFSE high CL-4 T cells were readily detected in the cortex of the DLN largely, but not exclusively, in the vicinity of the HEV. The cell distribution pattern likely reflects the localization of both CL-4 T cells that had recently transited from the HEV into the DLN (and encountered viral Ag) as well as CL-4 T cells that had entered the DLN before the in flux of viral Ag bearing migrant RDC. At day 4 p.i. in untreated (control) infected CL-4 T cell recipients, lymphatic endothelial staining of harvested DLN (Fig. 6B) demarcated the cortex (and bordering subcapsular sinus) and the medulla (cords and sinuses). There was extensive accumulation of CFSE low/ Thy1.2 CL-4 T cells uniformly throughout the DLN of these control animals (Fig. 6C), a pattern consistent with sequential influx and 397 activation/proliferation of multiple successive cohorts of responding T cells dispersed throughout the DLN. In the DLN of infected mice receiving anti-cd62l Ab at 24 h p.i., the medullary sinuses were likewise readily demarcated by lymphatic endothelial staining (Fig. 6D), but the Thy1.2 (CFSE low/ ) responding CL-4 T cells were, by contrast, predominately localized to a distinct region of the DLN in apposition to the lymphatic endothelial cells lining the LN medullary sinuses (Fig. 6E). The distinct anatomical localization of these CL-4 T cells within the DLN of treated animals supports the concept that a single initial cohort of CD8 T cells has responded to infection by activating, proliferating, and subsequently migrating to the medullary cords of the DLN during the 4-day course of infection. In these Ab-treated day 4 DLN, the B and T cell regions of the LN cortex were identifiable, but relatively depleted of cells (compared to control DLN), with depletion of CD4 T cells B cells CD8 T cells and with the majority of CD3 Thy1.2 cells in the medullary cords representing resident recipient CD8 T cells (data not shown). Discussion In this report, we have used a TCR-Tg CD8 T cell adoptive transfer model to examine the tempo of T cell activation/proliferation in the DLN in response to respiratory virus infection. We found that the onset of T cell proliferation in the DLN differed for mice infected with two different type A influenza strains. This difference in CD8 T cell activation tempo was not attributable to an intrinsic difference in efficiency of TCR-Tg T cell recognition of the two viruses, since there was no difference in the onset of activation (i.e., CD69 up-regulation) or proliferation in vitro in response to either virus-infected or peptide-pulsed APC expressing the A/PR8 and A/JAPAN viral epitopes. However, the difference in the T cell activation/proliferation tempo (in response to the two viruses) did correlate with the onset and timing of accelerated RDC migration from the infected lung to the DLN, with the migrant RDC responding to the A/JAPAN infection exhibiting a more rapid tempo of accumulation in the DLN (i.e., peak migration of A/JAPAN at 18 h, A/PR8 at 24 36 h). To better define the relationship between T cell migration into the DLN and the T cell proliferation tempo, we inhibited naive T cell influx into the DLN by i.v. administration of blocking anti- CD62L Ab at different times before/after infection. We found that anti-cd62l administration at 24 h p.i. allowed only a single cohort of naive Tg CD8 T cells to enter the DLN and undergo activation there. This cohort of T cells proliferated in a synchronous fashion within the DLN, as evidenced by the homogenous pattern of CFSE dilution (i.e., uniform or single-dye intensity) displayed by responding T cells at days 4 and 5 p.i. Histologic analysis of the DLN of mice treated with anti-cd62l at 24 h p.i. revealed that this single cohort of synchronized cells was localized at day 4 p.i. within the DLN medulla aligned along the medullary lymphatic channels. The induction of the adaptive (CD8 T cell) response to murine respiratory influenza virus infection is primarily restricted to the LN draining the infection site (i.e., the respiratory tract) (31). This restriction of viral Ag presentation to the DLN presumably reflects the fact that most influenza virus strains are restricted in their replication to the respiratory tract, with minimal systemic dissemination of the virus (32). Viral Ag transfer to the site of naive CD8 T cell activation is believed to be mediated by professional APC, primarily DC including RDC, which have activated and migrated from the lungs to the DLN in response to infection (17, 28, 29), although low levels of infectious virus can be transiently detected in the DLN early in infection (19). Our analysis of the response of influenza HA-specific TCR-Tg CD8 T cells to i.n. infection with two different type A influenza strains revealed a difference in the

398 SEQUENTIAL ACTIVATION/PROLIFERATION OF NAIVE CD8 T CELLS IN VIVO activation tempo of naive CD8 T cells in the DLN (Fig. 1A). This difference in the timing of T cell activation/proliferation after infection with two viruses did correlate with the extent of accumulation of infectious virus in the lungs which occurred more rapidly after A/JAPAN inoculation than A/PR8 inoculation during the first 48 h p.i. The difference in infectious virus accumulation likely was not due to a difference in the rate of virus replication in vivo in the lungs but rather to the difference in the size of the initial i.n. inoculum used. T cell activation/proliferation tempo also correlated with the onset and timing of accelerated RDC migration from the infected lungs to the DLN with the migrant RDC responding to the A/JAPAN infection exhibiting a more rapid tempo of RDC accumulation in the DLN and a more rapid onset of the CD8 T cell responses in the DLN. In this connection, we have found that increasing the i.n. inoculum dose of A/PR8 virus (to lethal inoculum dose levels) we can shorten the interval between the onset of infection and the initiation of CD8 T cell activation/proliferation (H. Yoon, unpublished observations). We believe that the tempo of RDC migration from the respiratory tract to the DLN is not dictated by the i.n. inoculum dose of virus per se, but rather is dictated by the time required for replicating virus to reach peak steady-state virus load and this virus load provides the stimulus for RDC migration. Therefore, by increasing the i.n. inoculum dose of A/PR8 virus, more respiratory tract cells are initially infected and peak steady-state virus titers are achieved more rapidly. Overall, these results support the concept that migrating DC are the major (primary) transporter of viral Ag presented to naive CD8 T cells in the DLN. As reported by us (17) and observed here (Fig. 3), accelerated RDC migration in response to respiratory influenza virus infection occurs over a limited time frame, with RDC migration returning to basal levels by 48 h p.i. (with either virus). Thus, if migrant RDC serve as an important APC for activating naive virus-specific CD8 T cells in the DLN, then viral Ag directly presented by RDC may not be available later in infection for presentation to naive CD8 T cells entering the DLN at day 3 p.i. or beyond. In this connection, several lines of evidence suggest that LN-resident DC (LNDC), in particular CD8 LNDC, which have acquired viral Ag (from migrant RDC) by cross-presentation are the major APC for the induction of naive CD8 T cell responses in the DLN (33 35). The APC activity of the CD8 LNDC is reported to be maximal at day 3 p.i (36). Although this time point is after the onset of CD8 T cell activation in the DLN in the studies described here (i.e., day 1 2 p.i.), differences in the CD8 T cell activation tempo may reflect the virus strain used for infection and the inoculum dose of virus in a particular study. However, because migrant RDC were not detected in the DLN after 48 h p.i., naive CD8 T cells that entered the DLN at 48 h p.i. or beyond were most likely stimulated by viral Ag displayed by LNDC, presumably through a cross-presentation mechanism. A determination of the relative contribution of migrant RDC and LNDC as APC in vivo for the CD8 T cell response to influenza virus will require additional analysis. A consistent feature of the in vivo analysis of the response of transferred CFSE-labeled naive CD8 (and CD4 ) T cells to Ag displayed in the SLO is the presence of activated T cells exhibiting heterogeneity in the number of cell divisions that they have undergone in the SLO (9 11). This heterogeneity in the extent of cell division could simply reflect differences in either the efficiency of the T cell/apc encounter, the strength of the signal imparted to the TCR on individual T cell precursors, or both, resulting in different onsets of T cell activation/proliferation or differences among naive T cells in the number of T cell divisions triggered by TCR engagement. Our findings suggest an alternative explanation, i.e., the time point at which naive T cell precursors leave the circulation and enter the Ag localization site (in the DLN) determines the onset of T cell activation and the subsequent programmed proliferation. Accordingly, the naive Ag-specific T cells that enter the DLN conveyer belt first would have undergone the greatest number of divisions (in this report, the lowest CFSE intensity). Therefore, these T cells would have also proceeded farthest along the differentiation pathway toward effector generation. This process of sequential T cell activation would be most readily demonstrated when (as in this report) Ag is localized to a peripheral site (e.g., the respiratory tract) and circulating T cells encounter Ag in the DLN of the site of Ag deposition/infection. As previously reported (10), there is a progressive and incremental decrease in the intensity of CD69 expression on TCR-Tg CL-4 T cells responding to infection in vivo in the DLN, as the cells undergo successive rounds of cell division (Fig. 1). CD69 expression on T cells is controlled in part by Ag (through TCR engagement) and by cytokines (in particular type 1 IFN) produced in response to inflammatory stimuli/infection (37, 38). Because CD69 levels can be rapidly up-regulated on activated CD8 T cells after TCR engagement (11, 39), the progressive (cell division-dependent) loss of CD69 expression on the responding CL-4 T cells in the DLN suggests that these T cells may be no longer in continuous contact with Ag after initial TCR engagement upon entering the DLN from the circulation. Our observations demonstrating the preferential anatomic localization of a single cohort of CL-4 T cells at day 4 p.i. to the medullary region of the DLN after synchronization by anti-cd62l Ab administration at 24 h p.i. (Fig. 6) is consistent with this idea and further suggests that, as the activated/proliferating T cells migrate from the LN cortex to the medulla, they may cease to encounter Ag, but may continue programmed divisions (40). The recent evidence linking CD69 expression with the control of T cell egress from LN also supports that idea (38). It is also noteworthy that the initiation of cell proliferation by the transferred CL-4 T cells in the DLN appeared to diminish significantly by days 4 and 5 p.i., despite the fact that the numbers of naive circulating CL-4 T cells available to enter the DLN and activate (at this time) are not limiting. Several lines of evidence suggest that naive Ag-specific T cells can enter LN draining the site of infection at relatively late times (i.e., days 4 and 5 p.i.), undergo Ag-driven activation, and at least several rounds of proliferation without differentiating into mature effector cells (and may be destined for the memory T cell pool) (26, 27, 41). It is not clear whether this difference is due to a change in the quality of the antigenic stimulus for the T cells in the DLN or a modification in the architecture of the DLN themselves (42). However, the findings reported here reinforce the view that the time frame over which Ag-specific naive CD8 T cells encounter Ag and activate after entering the DLN is likely to be limited, at least for those naive cells giving rise to mature effector T cells. It is also important to note that, according to the conveyor belt model, the first cohort of T cells to initiate cell division in vivo (e.g., at 72 h p.i. for A/PR8; H. Yoon and T. J. Braciale, unpublished observations) will have undergone at least four to six divisions in the DLN over a 24-h period (i.e., by 96 h p.i.). This finding suggests that activated CD8 T cells responding in vivo to an antigenic stimulus like virus infection may replicate with an extremely rapid initial division time (i.e., cell cycle time of 4 h). This would be a much faster division time than observed for naive T cell proliferation in vitro, and, if verified, suggests a novel mechanism for regulating T cell proliferative expansion in situ in the SLO. Furthermore, as suggested by the findings in Figs. 1 and 5, the responding T cells in the DLN appear to proliferate more rapidly (i.e., undergo more cell divisions per unit time at early points)

The Journal of Immunology 399 after the onset of cell division. Indeed, we found that the cell cycle time of the responding T cells increases significantly (more than doubles) after the cells have undergone more than five to six cell divisions (H. Yoon and T. J. Braciale, unpublished observations). The mechanism to explain the basis for the longer cell cycle time of the responding CD8 T cells in vivo is currently unknown and under investigation. Overall, our results demonstrate that naive CD8 T cells responding to Ag deposition at a peripheral mucosal site (e.g., the respiratory tract) encounter Ag in the DLN, then activate/proliferate in an ordered sequential fashion reminiscent of a conveyor belt. The tempo of the T cell response is regulated by Ag availability in the DLN, which is, in turn, dependent on the tempo of Ag delivery to the DLN by migrant DC trafficking from the mucosal site. In the case of influenza infection of the respiratory tract, efficient Ag delivery to the DLN maybe limited. Thus, the continuous stream of circulating naive Ag-specific CD8 T cells in the blood that will give rise to effector most likely enter the DLN, encounter Ag, and activate within a narrow time window. Findings reported here further suggest the division time of activated proliferating CD8 T cells in the DLN is extremely rapid, thereby allowing for the rapid expansion of the number of responding T cells and the rapid generation of effectors. As these activation/differentiation events occur within the architectural constraints of the DLN, a deeper understanding of the interplay between structural and cellular elements of the SLO which regulate the host response to infection will be critical, both for elucidating the mechanism(s) of immune dysregulation mediated by microbes and for optimum vaccine design. Acknowledgments We thank R. M. Chu for access to unpublished data and T. Kim and S. Gill for many helpful discussions. Disclosures The authors have no financial conflict of interest. References 1. von Andrian, U. H., and T. R. Mempel. 2003. Homing and cellular traffic in lymph nodes. Nat. Rev. Immunol. 3: 867 878. 2. Miyasaka, M., and T. Tanaka. 2004. Lymphocyte trafficking across high endothelial venules: dogmas and enigmas. Nat. Rev. Immunol. 4: 360 370. 3. Ley, K., and G. S. Kansas. 2004. Selectins in T-cell recruitment to non-lymphoid tissues and sites of inflammation. Nat. Rev. Immunol. 4: 325 335. 4. Bajenoff, M., S. Granjeaud, and S. Guerder. 2003. The strategy of T cell antigenpresenting cell encounter in antigen-draining lymph nodes revealed by imaging of initial T cell activation. J. Exp. Med. 198: 715 724. 5. Mempel, T. R., S. E. Henrickson, and U. H. Von Andrian. 2004. T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases. Nature 427: 154 159. 6. Bousso, P., and E. Robey. 2003. Dynamics of CD8 T cell priming by dendritic cells in intact lymph nodes. Nat. Immunol. 4: 579 585. 7. Stoll, S., J. Delon, T. M. Brotz, and R. N. Germain. 2002. Dynamic imaging of T cell-dendritic cell interactions in lymph nodes. Science 296: 1873 1876. 8. Hugues, S., L. Fetler, L. Bonifaz, J. Helft, F. Amblard, and S. Amigorena. 2004. Distinct T cell dynamics in lymph nodes during the induction of tolerance and immunity. Nat. Immunol. 5: 1235 1242. 9. Haribhai, D., B. Edwards, M. L. Williams, and C. B. Williams. 2004. Functional reprogramming of the primary immune response by T cell receptor antagonism. J. Exp. Med. 200: 1371 1382. 10. Lawrence, C. W., and T. J. Braciale. 2004. Activation, differentiation, and migration of naive virus-specific CD8 T cells during pulmonary influenza virus infection. J. Immunol. 173: 1209 1218. 11. Miller, M. J., O. Safrina, I. Parker, and M. D. Cahalan. 2004. Imaging the single cell dynamics of CD4 T cell activation by dendritic cells in lymph nodes. J. Exp. Med. 200: 847 856. 12. Eichelberger, M., W. Allan, M. Zijlstra, R. Jaenisch, and P. C. Doherty. 1991. Clearance of influenza virus respiratory infection in mice lacking class I major histocompatibility complex-restricted CD8 T cells. J. Exp. Med. 174: 875 880. 13. Graham, M. B., and T. J. Braciale. 1997. Resistance to and recovery from lethal influenza virus infection in B lymphocyte-deficient mice. J. Exp. Med. 186: 2063 2068. 14. Bender, B. S., T. Croghan, L. Zhang, and P. A. Small, Jr. 1992. Transgenic mice lacking class I major histocompatibility complex-restricted T cells have delayed viral clearance and increased mortality after influenza virus challenge. J. Exp. Med. 175: 1143 1145. 15. Lukacher, A. E., V. L. Braciale, and T. J. Braciale. 1984. In vivo effector function of influenza virus-specific cytotoxic T lymphocyte clones is highly specific. J. Exp. Med. 160: 814 826. 16. Topham, D. J., R. A. Tripp, and P. C. Doherty. 1997. CD8 T cells clear influenza virus by perforin or Fas-dependent processes. J. Immunol. 159: 5197 5200. 17. Legge, K. L., and T. J. Braciale. 2003. Accelerated migration of respiratory dendritic cells to the regional lymph nodes is limited to the early phase of pulmonary infection. Immunity 18: 265 277. 18. Drake, D. R., III, R. M. Ream, C. W. Lawrence, and T. J. Braciale. 2005. Transient loss of MHC class I tetramer binding after CD8 T cell activation reflects altered T cell effector function. J. Immunol. 175: 1507 1515. 19. Legge, K. L., and T. J. Braciale. 2005. Lymph node dendritic cells control CD8 T cell responses through regulated FasL expression. Immunity 23: 649 659. 20. Cerwenka, A., T. M. Morgan, and R. W. Dutton. 1999. Naive, effector, and memory CD8 T cells in protection against pulmonary influenza virus infection: homing properties rather than initial frequencies are crucial. J. Immunol. 163: 5535 5543. 21. Chapman, T. J., M. R. Castrucci, R. C. Padrick, L. M. Bradley, and D. J. Topham. 2005. Antigen-specific and non-specific CD4 T cell recruitment and proliferation during influenza infection. Virology 340: 296 306. 22. Redmond, W. L., B. C. Marincek, and L. A. Sherman. 2005. Distinct requirements for deletion versus anergy during CD8 T cell peripheral tolerance in vivo. J. Immunol. 174: 2046 2053. 23. Topham, D. J., M. R. Castrucci, F. S. Wingo, G. T. Belz, and P. C. Doherty. 2001. The role of antigen in the localization of naive, acutely activated, and memory CD8 T cells to the lung during influenza pneumonia. J. Immunol. 167: 6983 6990. 24. Lawrence, C. W., R. M. Ream, and T. J. Braciale. 2005. Frequency, specificity, and sites of expansion of CD8 T cells during primary pulmonary influenza virus infection. J. Immunol. 174: 5332 5340. 25. Hataye, J., J. J. Moon, A. Khoruts, C. Reilly, and M. K. Jenkins. 2006. Naive and memory CD4 T cell survival controlled by clonal abundance. Science 312: 114 116. 26. Zammit, D. J., D. L. Turner, K. D. Klonowski, L. Lefrancois, and L. S. Cauley. 2006. Residual antigen presentation after influenza virus infection affects CD8 T cell activation and migration. Immunity 24: 439 449. 27. Jelley-Gibbs, D. M., D. M. Brown, J. P. Dibble, L. Haynes, S. M. Eaton, and S. L. Swain. 2005. Unexpected prolonged presentation of influenza antigens promotes CD4 T cell memory generation. J. Exp. Med. 202: 697 706. 28. Belz, G. T., C. M. Smith, L. Kleinert, P. Reading, A. Brooks, K. Shortman, F. R. Carbone, and W. R. Heath. 2004. Distinct migrating and nonmigrating dendritic cell populations are involved in MHC class I-restricted antigen presentation after lung infection with virus. Proc. Natl. Acad. Sci. USA 101: 8670 8675. 29. Guermonprez, P., J. Valladeau, L. Zitvogel, C. Thery, and S. Amigorena. 2002. Antigen presentation and T cell stimulation by dendritic cells. Annu. Rev. Immunol. 20: 621 667. 30. Lepault, F., M. C. Gagnerault, C. Faveeuw, and C. Boitard. 1994. Recirculation, phenotype and functions of lymphocytes in mice treated with monoclonal antibody MEL-14. Eur. J. Immunol. 24: 3106 3112. 31. Itano, A. A., and M. K. Jenkins. 2003. Antigen presentation to naive CD4 T cells in the lymph node. Nat. Immunol. 4: 733 739. 32. Eichelberger, M. C., M. L. Wang, W. Allan, R. G. Webster, and P. C. Doherty. 1991. Influenza virus RNA in the lung and lymphoid tissue of immunologically intact and CD4-depleted mice. J. Gen. Virol. 72: 1695 1698. 33. Allan, R. S., C. M. Smith, G. T. Belz, A. L. van Lint, L. M. Wakim, W. R. Heath, and F. R. Carbone. 2003. Epidermal viral immunity induced by CD8 dendritic cells but not by Langerhans cells. Science 301: 1925 1928. 34. den Haan, J. M., S. M. Lehar, and M. J. Bevan. 2000. CD8 but not CD8 dendritic cells cross-prime cytotoxic T cells in vivo. J. Exp. Med. 192: 1685 1696. 35. Moron, G., P. Rueda, I. Casal, and C. Leclerc. 2002. CD8 - CD11b dendritic cells present exogenous virus-like particles to CD8 T cells and subsequently express CD8 and CD205 molecules. J. Exp. Med. 195: 1233 1245. 36. Mellman, I., and R. M. Steinman. 2001. Dendritic cells: specialized and regulated antigen processing machines. Cell 106: 255 258. 37. Alsharifi, M., M. Lobigs, M. Regner, E. Lee, A. Koskinen, and A. Mullbacher. 2005. Type I interferons trigger systemic, partial lymphocyte activation in response to viral infection. J. Immunol. 175: 4635 4640. 38. Shiow, L. R., D. B. Rosen, N. Brdickova, Y. Xu, J. An, L. L. Lanier, J. G. Cyster, and M. Matloubian. 2006. CD69 acts downstream of interferon- / to inhibit S1P1 and lymphocyte egress from lymphoid organs. Nature 440: 540 544. 39. Santis, A. G., M. Lopez-Cabrera, F. Sanchez-Madrid, and N. Proudfoot. 1995. Expression of the early lymphocyte activation antigen CD69, a C-type lectin, is regulated by mrna degradation associated with AU-rich sequence motifs. Eur. J. Immunol. 25: 2142 2146. 40. Kaech, S. M., and R. Ahmed. 2001. Memory CD8 T cell differentiation: initial antigen encounter triggers a developmental program in naive cells. Nat. Immunol. 2: 415 422. 41. Catron, D. M., L. K. Rusch, J. Hataye, A. A. Itano, and M. K. Jenkins. 2006. CD4 T cells that enter the draining lymph nodes after antigen injection participate in the primary response and become central-memory cells. J. Exp. Med. 203: 1045 1054. 42. Soderberg, K. A., G. W. Payne, A. Sato, R. Medzhitov, S. S. Segal, and A. Iwasaki. 2005. Innate control of adaptive immunity via remodeling of lymph node feed arteriole. Proc. Natl. Acad. Sci. USA 102: 16315 16320.