Downloaded From

Home , Naive T cell

Two Distinct Stages in the Transition from Naive CD4 T Cells to Effectors, Early Antigen-Dependent and Late Cytokine-Driven Expansion and This information is current as Differentiation of October 2, 2021. Dawn M. Jelley-Gibbs, Nancy M. Lepak, Michael Yen and Susan L. Swain J Immunol 2000; 165:5017-5026; ;

doi: 10.4049/jimmunol.165.9.5017 Downloaded from http://www.jimmunol.org/content/165/9/5017

References This article cites 52 articles, 31 of which you can access for free at: http://www.jimmunol.org/content/165/9/5017.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

by guest on October 2, 2021 • 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 © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Two Distinct Stages in the Transition from Naive CD4 T Cells to Effectors, Early Antigen-Dependent and Late Cytokine-Driven Expansion and Differentiation1

Dawn M. Jelley-Gibbs, Nancy M. Lepak, Michael Yen, and Susan L. Swain2

Efﬁcient peptide presentation by professional APC to naive and effector CD4 T cells in vitro is limited to the ﬁrst 1–2 days of culture, but is nonetheless optimum for effector expansion and cytokine production. In fact, prolonging Ag presentation leads to high levels of T cell death, decreased effector expansion, and decreased cytokine production by recovered effectors. Despite the absence of Ag presentation beyond day 2, T cell division continues at a constant rate throughout the 4-day culture. The Ag- independent later stage depends on the presence of IL-2, and we conclude optimum effector generation depends on an initial 2 days of TCR stimulation followed by an additional 2 days of Ag-independent, cytokine driven T cell expansion and differentiation. The

Journal of Immunology, 2000, 165: 5017–5026. Downloaded from

primary CD4 T cell response can be conceptually di- drive polarization are most effective when present at the initiation vided into three distinct phases involving activation, of T cell activation before the ﬁrst round of cell division (15, 16). A proliferation, and differentiation of responding T cells Once cell proliferation commences, highly activated T cells divide leading to the generation of large numbers of highly potent effec- rapidly and can continue through multiple rounds of division, result-

tors with the capacity to rapidly produce high levels of multiple ing in an impressive expansion of the initial population (17, 18). Little http://www.jimmunol.org/ cytokines after reencounter with Ag (1, 2). However, each of these is known about the critical sequence of external signals required for phases may in fact occur sequentially and/or simultaneously. driving cell division over the later phase of the 4-day period necessary Which stimulatory signals are required to trigger each phase, and to generate large numbers of highly activated effectors. the sequence in which they must be delivered to the T cell, are also Differentiation, leading to changes which effect the future func- unclear. For the purpose of these studies, we have defined effectors tional and survival potential of the progeny cells, also occurs during as activated T cells with the capacity to rapidly produce high levels effector generation. The differentiation of naive CD4 T cells into po- of multiple cytokines after restimulation (1, 2). We have found that larized effectors in vitro is dependent on high doses of high avidity full development of effector function requires 4 days of in vitro Ag, the interaction of multiple costimulation receptor/coreceptor culture, (3), and we have also seen a similar requirement for 4 days pairs, growth-promoting cytokines, like IL-2, and polarizing cyto- by guest on October 2, 2021 or more to generate fully activated effector T cells in vivo (4). kines (1–3, 16, 19, 20). IL-12 or IFN-␥, and the lack of IL-4, generate We consider the activation phase to be the first step that occurs Th1-polarized effectors (capable of IL-2, IFN-␥, TNF-␣, and TGF-␤ before division, which commits the cell to enter the cell cycle, production), whereas the presence of IL-4 and the lack of IFN-␥ gen- requiring a number of signals to promote progression through the erate Th2-polarized effectors (capable of IL-4, -5, -10, and -13 pro- cell cycle. There is at least a 24-h lag while a resting (Go) T cell duction) (16). Although many of the signals required for effector dif- becomes activated and thus competent to progress through the cell ferentiation are known, it is not clear whether these and other cycle (5, 6). During this phase, many genes and their products are differentiation signals are delivered early or late during a primary T expressed in preparation for DNA replication and cell division cell response to drive the development of functional and phenotypic (7–12). Epigenetic remodeling changes resulting in “on” configu- characteristics critical for optimal effector function. It is likely that rations of polarized cytokine genes account for some of the very various differentiation signals and differentiation-associated changes earliest changes that occur during T cell activation as the cell un- are available throughout in vivo effector generation at different times. dergoes its first round of DNA replication before the first cell di- For example, for polarization, the presence of polarizing cytokines are vision (10, 13, 14). Therefore, it is notable that cytokines which required only during the early phase of effector generation, supporting that the signals required for polarized T cell differentiation begin im- mediately after stimulation of naive CD4 T cells (21–23). However, Trudeau Institute, Saranac Lake, NY 12983 the ability of effectors to become competent to produce large amounts Received for publication June 19, 2000. Accepted for publication August 10, 2000. of polarized cytokines (i.e., IFN-␥ and IL-4) rapidly upon restimula- The costs of publication of this article were defrayed in part by the payment of page tion takes 2–4 days (3, 15, 23, 24). Additionally, the growth-promot- charges. This article must therefore be hereby marked advertisement in accordance ing cytokine IL-2 is only produced at maximal levels by the respond- with 18 U.S.C. Section 1734 solely to indicate this fact. ing CD4 T cells during the later phase of effector generation (1, 2, 21, 1 This work was supported by National Institutes of Health Grant AI26887. 24, 25). Another important change that appears to occur during the 2 Address correspondence and reprint requests to Dr. Susan L. Swain, Trudeau In- stitute, P.O. Box 59, Saranac Lake, NY 12983. E-mail address: [email protected] late phase of effector differentiation includes the induction of susceptibility to activation-induced cell death (AICD),3 which requires early 3 Abbreviations used in this paper: AICD, activation-induced cell death; Tg, transgene; HA, hemagglutinin; HNT T cells, TCR Tgϩ HNT (influenza HA-specific) exposure to IL-2 and at least 2 days of Ag/APC (26, 27). CD4ϩ T cells; CFSE, 5-carboxyfluorescein diacetate; PCCF, pigeon cytochrome c Although effects of the duration of Ag stimulation on the ability fragment; AND T cells, TCR Tgϩ AND (PCCF-specific) CD4ϩ T cells; PI, propidium iodide; DC, dendritic cell; MitC, mitomycin c; ICCS, intracellular cytokine of naive T cells to proliferate has been studied previously, equally staining; FSC, forward scatter; SSC, side scatter; MFI, mean fluorescent intensity. important functional characteristics including cytokine production,

Copyright © 2000 by The American Association of Immunologists 0022-1767/00/$02.00 5018 Ag-DEPENDENT AND IL-2-DRIVEN STAGES FROM NAIVE TO EFFECTOR activation status, and survival of the generated effectors are com- Cell isolations monly overlooked. An additional weakness of these earlier studies The isolation of pooled spleen and lymph node cells enriched for naive is that they involved the use of non-physiological plate-bound anti- CD4 T cells has been described previously (34). In brief, the pooled spleen TCR/CD3 mAb plus anti-CD28 mAb as an Ag/APC surrogate, and lymph node cells were passed through a nylon wool column, and the which may not lead to an accurate picture of Ag-induced responses nonadherent cells treated with a panel of CD8, heat-stable Ag, and class II MHC-depleting Ab and complement followed by Percoll gradient separa- in situ (16, 20, 28). tion. The purified cell populations were routinely Ͼ95% CD4ϩ cells, 90– Here, we determine the impact of duration of TCR stimulation 95% of which had a naive phenotype (CD45RBhigh, CD62Lhigh, CD44low, on the quantity and quality of effectors generated. To directly as- and CD25low) and expressed the TCR Tg. sess the requirements of naive CD4 T cells for, and response to, Immunofluorescent staining defined durations of Ag presentation, we used a TCR-transgenic #mouse model system in which we isolate a homogeneous popu- All staining was done at 4°C in PBS (Life Technologies, Grand Island, NY) with 1% BSA (Sigma, St. Louis, MO) and 0.1% NaN3 (Sigma). FITC- lation of naive Ag-specific CD4 T cells. We evaluate several TCR labeled Ab are produced in our laboratory, and PE-biotin- and APC-labeled agonists, including plate-bound anti-TCR mAb and two different Ab were purchased where indicated. The following Ab and fluorescent physiologically relevant Ag/APC model systems. Because Rogers reagents were used: PE-labeled, CyChrome-labeled, and APC-labeled anti- and Croft (29) have previously reported on the effects of Ag con- CD4 (clone RM4–5; Caltag, Burlingame, CA), FITC- and PE-labeled anti- CD44 (clone IM7; PharMingen, San Diego, CA), FITC- and PE-labeled centration on T cell stimulation, we restricted these studies to op- anti-CD62L (clone Mel 14; Caltag), FITC- and PE-labeled anti-CD45RB timal concentrations of Ag as previously determined for each Ag (clone 23G2; PharMingen), FITC- and PE-labeled anti-CD25 (IL-2R␣ to be used in these studies. We know that suboptimal or variable chain, clone PC61; PharMingen), FITC-labeled anti-V␣11 (clone RR8-1), PE- and biotin-labeled anti-V␤3 (clone KJ25), PE-labeled anti-V␤8.3

levels of Ag, costimulation, and growth factors (i.e., IL-2) can Downloaded from have major negative impacts on the generation of CD4 effectors; (clone 1B3.3; PharMingen), streptavidin-APC (PharMingen), PE-labeled antimouse IL-2 (clone S4B6; PharMingen), PE-labeled antimouse IFN-␥ therefore, we chose to limit the scope of these studies to deter- (clone XMG.1; PharMingen), and PE-labeled antimouse IL-4 (clone BVD- mining the effects of Ag duration under conditions of controlled 24G2; Caltag). All isotype control Abs were purchased from PharMingen: and optimal Ag dose, costimulation, and IL-2 (1, 19, 29, 30, 31). FITC- and PE-labeled hamster Ig, FITC- and PE-labeled rat IgG1, FITC- With the use of vital dyes and FACS analysis, we have examined and PE-labeled rat IgG2a, and FITC- and PE-labeled rat IgG2b. Flow cytometry was conducted using FACScan or FACScalibur flow cytometers, the impact of varied durations of TCR stimulation on both T cell and the data were analyzed with CellQuest software (all from Becton Dick- http://www.jimmunol.org/ division and effector expansion, and confirm that a short duration, inson, San Jose, CA). 1–2 days, of TCR stimulation promotes optimal proliferation and Labeling of cells with fluorescent dyes effector expansion during culture. We also measured the functional capacity of the resulting effectors as the ability to produce high T cell populations were stained with the vital fluorescent dye 5 (and 6-)- levels of polarized cytokines upon restimulation, and we find that carboxyfluorescein diacetate succinimidyl ester (CFSE) or BODIPY Red (Molecular Probes, Eugene, OR) as previously described (25, 35). In brief, a short duration, 1–2 days, of TCR stimulation in the presence of cells were resuspended in serum-free RPMI 1640 (Life Technologies) at growth-promoting and polarizing cytokines is both sufficient and 10 ϫ 106/ml. CFSE or BODIPY Red was added to the cell suspension at optimal to promote stable polarized cytokine production. In fact, 1 ␮M final concentration and incubated in a 37°C shaking water bath for increasing the duration of TCR stimulation beyond 2 days leads to 13 min. Cells were then washed twice, recounted, and cultured as described by guest on October 2, 2021 below. high levels of CD4 T cell death and a significant decrease in the ability of the remaining viable effectors to produce polarized cy- Cell culture and effector generation tokines upon restimulation. We observed that a TCR stimulation- All in vitro cultured cells were cultured in RPMI 1640 supplemented with independent expansion phase exists during the late phase, days penicillin (200 ␮g/ml; Sigma), streptomycin (200 ␮g/ml; Sigma), glu- 2–4, of effector generation and is dependent on IL-2. These studies tamine (4 mM; Sigma), 2-ME (50 ␮M; Sigma), HEPES (10 mM; Sigma), suggest that optimal CD4 effector generation occurs only when and 8% FBS (Intergen, Purchase, NY). DCEK-ICAM, fibroblast cell line that expresses B7.1 constitutively and there is an initial 2 days of Ag presentation followed by an Ag- is engineered by stable transfection with ICAM-1 and class II MHC (I-Ek) independent and IL-2-driven cell expansion phase lasting an ad- molecules, was used as APC at 2:1 T cell:APC. These cells do not express ditional 2 days. These observations predict that sequential com- other costimulatory molecules such as LFA-1, CD48, or heat-stable Ag and partmentalization and redistribution of T cells and Ag-bearing were originally generated by R. Germain (National Institutes of Health, Bethesda, MD). T-depleted spleen APC blasts were prepared as described APC during in vivo cognate interactions are necessary to achieve previously (22). Briefly, splenic cell suspensions were depleted of T cells this precise sequence of T cell stimulatory conditions, linking ob- using anti-Thy1.2 (clones HO13.14 and F7D5), anti-CD4 (clone RL172.4), anti-CD8 (clone 3.155), and complement. The resulting cells were small servations of compartmentalization of cognate T cell-APC inter- ϩ actions to the regulation of TCR stimulation and signals provided and 90–95% B220 . The T-depleted spleen APC were then stimulated at 5 ϫ 105/ml for 3 days with LPS (25 ␮g/ml; Sigma) and dextran (25 ␮g/ml; by growth-promoting cytokines on the generation of optimally Sigma). All APC were treated with 100 ␮g/ml of mitomycin c (MitC; functional effectors (18, 32, 33). These results have important im- Sigma) for 45 min at 37°C before use. plications for in vitro model systems in which TCR stimulation is CD4 effectors were generated by culturing Tgϩ CD4 cells as previously ϩ ϫ 6 provided through plate-bound anti-TCR Ab. Our results would described (22). Briefly, naive Tg CD4 AND T cells (3 10 cells/ml, Fig. 1, and 3 ϫ 105 cells/ml, Figs. 2–6) were cultured with syngeneic predict that prolonged TCR stimulation may select for a subpopu- T-depleted splenic APC blasts (3 ϫ 105 cells/ml, Fig. 2) or DCEK-ICAM lation of less functional effectors and preclude recovery of the most APC (1.5 ϫ 106 cells/ml, Fig. 1, and 1.5 ϫ 105 cells/ml, Figs. 3–6) plus active populations of effectors. 5 ␮M KAERADLIAYLKQATAK pigeon cytochrome c fragment (PCCF peptide) (Figs. 1–6). We have compared stimulation of T cells by culture with syngeneic T-depleted splenic APC blasts to that of DCEK-ICAM APC and seen no differences (data not shown). In Fig. 2, 5 ␮M HNT- Materials and Methods NGVTAACSHE influenza hemagglutinin (HA peptide; New England Pep- Animals tide, Fitchburg, MA) was added in addition to the PCCF peptide to stimulate the CD4 HNT T cells. Alternatively, Tgϩ CD4 cells (3 ϫ 105 cells/ H-2k/k (B10.Br) and H-2b/b (C57BL/6) V␣11/V␤3 AND TCR transgene ml, Fig. 5) were stimulated with plate-bound anti-V␤3 (10 ␮g/ml) and k ␤ ␣ d (Tg; I-E restricted), B10.Br.HNT.B10.D2/F1 (V 8.3/V 2) TCR Tg (I-A soluble anti-CD28 (10 ␮g/ml). Recombinant murine cytokines IL-2 and k/d restricted), and H-2 B10.Br.B10.D2/F1 mice were used at 2–6 mo of age IL-4 were obtained from culture supernatant of X63.Ag8-653 cells trans- and were bred in the animal facilities at the Trudeau Institute. fected with cDNA for the respective cytokines (36). Recombinant murine The Journal of Immunology 5019

IL-12 was a gift from Stanley Wolf (Genetics Institute, Cambridge, MA). IL-2-polarized effectors (Th0) were generated with IL-2 (80 U/ml). Th1 effectors were generated with IL-2, IL-12 (2 ng/ml), and anti-IL-4 (clone 11B11; 10 ␮g/ml). Th2 effectors were generated with IL-2, IL-4 (200 U/ml), and anti-IFN-␥ (clone XMG1.2; 10 ␮g/ml). Culture durations varied as indicated in each individual experiment. At the end of culture, live T cells were enumerated by direct cell counts of trypan blue-excluding cells.

Cytokine detection Culture supernatants collected after 24 h, from cultures of effectors (5 ϫ 105 cells/ml) restimulated with Ag/APC (2.5 ϫ 105 cells/ml) or with plate- bound anti-V␤3 (10 ␮g/ml), were assayed for the presence of IL-2 in a bioassay with NK-3 cells and for IL-4 by ELISA as previously described (37). IL-2 concentrations are expressed in U/ml (1 U of IL-2 is equal to 1.2 ng). IL-4 concentrations are expressed in nanograms per milliliter. Alternatively, intracellular cytokine staining (ICCS) was detected in restimulated effectors as previously described (38, 39). In brief, effectors (5 ϫ 105 cells/ml) were restimulated for 24 h with DCEK-ICAM APC (2.5 ϫ 105 cells/ml) with PCCF (5 ␮M) or restimulated for 4 h with PMA (10 ng/ml; Sigma) and ionomycin (50 ng/ml; Sigma). Negative controls (indicated by dotted lines) indicate ICCS of T cells that are not restimulated ␮

and thus do not express the indicated cytokine. Brefeldin A (10 g/ml final Downloaded from concentration; Sigma) was added 2 h after culture initiation and maintained throughout the ICCS protocol. At the end of the specified restimulation period, cells were collected and surface stained for CD4 and Tg expression as described above. The cells were then divided into separate tubes, washed, fixed in 75 ␮l of fresh 4% paraformaldehyde (Sigma) plus 25 ␮l of PBS containing 10 ␮g/ml brefeldin A, and incubated for 20 min at room temperature. The tubes were then washed once with PBS and resuspended

␮ http://www.jimmunol.org/ in 50 l saponin buffer (PBS containing 1% FBS, 0.1% NaN3, and 0.1% saponin (Sigma), pH 7.4–7.6). PE-labeled antimouse IL-2, PE-labeled antimouse IFN-␥, or PE-labeled antimouse IL-4 was added and incubated with the T cells for 30 min at room temperature. All samples were then washed with PBS/BSA ϩ 0.2% azide and analyzed on a FACScan or FACScalibur cytometer.

Results and Discussion Kinetics of CD4 effector expansion during the proliferative phase by guest on October 2, 2021

In order for naive CD4 T cells to progress from a resting (G0) state to an activated effector state and initiate division, they must transit

from G0 into the cell cycle. Because the transition from a resting G0 state into the cell cycle involves expression of multiple genes FIGURE 1. Kinetics of naive CD4 cell division. Naive CD4 AND T and their products in a coordinated fashion, this often results in a cells were labeled with CFSE and cultured at 3 ϫ 106 T cells/culture with delay between entry into the cell cycle and the first round of cell 1.5 ϫ 106 DCEK-ICAM, PCCF, and IL-2. a, Analysis of CFSE dye loss proliferation (5, 6). To analyze the rate of division of naive CD4 T was measured at days 1–4 during effector generation by FACS analysis as cells following Ag stimulation, we stained naive CD4 T cells with indicated. b, Number of cell divisions per 24-h time period was determined by counting the number of cell divisions, as measured by CFSE dye loss, the vital dye CFSE, which binds to intracellular proteins and is that occurred between each successive 24-h interval. Duration per cell di- partitioned to daughter cells with each division (35, 40). Analysis vision was determined by dividing the number of cell divisions per 24-h of the CFSE profiles on each of the 4 days following culture ini- interval by 24 h. c, Actual effector recovery was measured by direct cell tiation indicates that naive CD4 T cells have a delay of ϳ24 h counts. Predicted effector recovery was estimated on the basis of starting between initial TCR stimulation and the first round of cell division cell number and average number of cell divisions. Data are representative (Fig. 1, a and b), but that after this initial lag period the cells of three independent experiments. progress through multiple cell cycles at a constant rate of 7–10 h/cell division throughout the remaining 3 days of culture (Fig. 1b). After day 2, only cells that have divided are detected, indi- erative phase has the potential to generate ϳ250-fold cell expan- cating either that all of the cells have been stimulated to undergo sion, conditions in vitro are such that the viable effector expansion division or that cells which have not divided fail to survive (21). is limited to ϳ10- to 20-fold (Fig. 1c). The nature of this cell death Moreover, when we compare the predicted to the actual effector is unclear, but we and others also observe that naive CD4 T cell recovery for each day during effector generation, with the calcu- expansions in vivo in response to peptide Ag in adjuvant are also lations based on the starting cell number of 3 ϫ 105 cells and the less than would be predicted by the CFSE profiles (E. Roman, G. average number of cell divisions (Fig. 1, a and b) for each day, we E. Huston, and S. L. Swain, unpublished observations; Refs. find that the actual effector recovery is much lower than the pre- 41–43). dicted effector recovery (Fig. 1c). This suggests that there are sig- One possible explanation for limited CD4 expansion, despite the nificant levels of cell death which occur throughout effector gen- high rate of cell division, is that Ag presentation is of limited eration, leading to lower effector recoveries than would be duration in vitro and cells need repeated or ongoing TCR trigger- predicted if all of the cells generated survived the 4-day culture ing to persist after division. Several groups have reported that period. These results suggest that although the CD4 T cell prolif- DNA synthesis, measured via [3H]thymidine incorporation, or cell 5020 Ag-DEPENDENT AND IL-2-DRIVEN STAGES FROM NAIVE TO EFFECTOR proliferation, measured via 5-bromo-2Ј-deoxyuridine incorporation, of naive T cells is largely determined by the strength and/or duration of Ag (2, 16, 20, 44, 45). Although division of responding lymphocytes is certainly an important step in the development of an immune response, we would argue that the development of increased numbers of highly potent effectors capable of rapidly producing multiple cytokines upon restimulation with Ag/APC is the most important criteria for measuring the generation of an effective immune response. We particularly wanted to address the impact of prolonging TCR stimulation. To better understand what happens normally in our in vitro cultures, we started by evaluating how long Ag presentation was actually occurring.

Efficient Ag presentation is restricted to the first 1–2 days of culture We know from previous studies that MitC-treated APC disappear from culture by 48 h, as measured by surface staining for ICAM-1 and class II (17, 46). This phenomenon only occurs during culture conditions where Ag presentation occurs. MitC-treated APC cul- Downloaded from tured with T cells and without Ag or MitC-treated APC cultured with Ag and without T cells survive for 4 days of culture, whereas MitC-treated APC cultured with T cells and Ag are rapidly deleted from culture (data not shown). However, since it is possible that FIGURE 2. Efficient Ag presentation only occurs for the first 1–2 days membrane fragments of no longer intact APC might still present of culture. Naive CD4 HNT T cells were BODIPY Red-labeled and cul- ϫ 5 ϫ 5 ␮

tured at 3 10 T cells/culture with 3 10 splenic APC blasts, 5 MHA http://www.jimmunol.org/ Ag, we sought to determine the time at which cultures of MitC- peptide, 5 ␮M PCCF peptide, and under IL-2-polarizing (Th0), Th1-, or treated APC are no longer capable of effectively stimulating naive Th2-polarizing cytokine environments. Fresh naive CD4 AND T cells were CD4 T cell proliferation or cytokine production by effectors in CFSE labeled and added at 3 ϫ 105 T cells/culture to the ongoing HNT T standard in vitro cultures. We developed a cell mixing system cell:Ag/APC cultures at days 0–3 as indicated. The AND T cells were where we initiate cultures of BODIPY Red-labeled naive HNT T harvested 3 days after their addition to the cultures. a, Cell division was cells, specific for influenza HA peptide, with an activated APC measured by CFSE dye loss of live V␤3ϩ (AND)/CD4ϩ cells, and the ϩ population containing not only B cell blasts, but also a small num- percent undivided cells are indicated by FACS gating. b, Live V␤3 ϩ ber of other APC populations including macrophages and dendritic (AND)/CD4 cell recoveries were determined by performing direct cell cells. We then loaded the splenic APC blasts with HA and PCCF counts of total cells and extrapolating them from the percentages of CFSE- ␤ ϩ ϩ by guest on October 2, 2021 peptides such that the APC are capable of efficient presentation of labeled live V 3 (AND)/CD4 cells present in the whole culture. Start- ing AND T cell numbers were 3 ϫ 105. Data are representative of three both the influenza HA peptide to HNT T cells and the PCCF pep- independent experiments. tide to AND T cells within the same culture. At various time points after initiation of the HNT T cell:Ag/APC culture, we isolated fresh naive AND T cells from spleens and lymph nodes from AND vision (Fig. 2a) and expansion (Fig. 2b) of naive AND T cells TCR Tg mice, labeled these cells with CFSE, and introduced them declined during the first day of culture and was negligible by 2 to the ongoing HNT T cell:Ag/APC cultures. Three days after the days after the original Ag/APC cultures were initiated. introduction of the AND T cells to the ongoing HNT T cell:Ag/ To determine whether cytokines present during effector polar- APC cultures, the T cells were harvested and analyzed for cell ization might prolong the ability of the splenic APC blasts to division and effector expansion. With these experiments, we were present Ag to naive CD4 cells, we designed this experiment under able to easily identify the two populations of T cells with anti-TCR Th0 (IL-2 only)-, Th1-, or Th2-polarizing conditions. The ability ␤-chain-specific Ab, and based on their BODIPY Red vs CFSE of the Ag/APC cultures to stimulate naive AND T cell prolifera- staining, measure the capacity of the newly added AND T cells to tion was similar under all three polarization conditions (Fig. 2a). respond to whatever APC were still actively presenting Ag at var- Therefore, the cytokines present during in vitro effector polariza- ious time points after initiation of HNT T cell:Ag/APC cultures, tion did not enhance the ability of the splenic APC blasts to present and verify that the proliferative capacity of the HNT T cells added Ag to naive CD4 T cells. We repeated this experiment using the at the initiation of the T cell:Ag/APC cultures was not perturbed by DCEK-ICAM fibroblast APC line and have seen no difference in the addition of the fresh naive AND T cells. the ability of the two APC populations to stimulate naive CD4 When naive AND T cells were added to HNT T cell:Ag/APC AND T cell expansion and differentiation (data not shown; Refs. 1, cultures at day 0, Ͻ10% of the AND T cells could be detected as 2, 30). nondivided cells under all three polarizing conditions (Fig. 2a), The recovery of the AND T cells after 3 days in culture mirrored indicating that these culture conditions are fully capable of stim- their division. Naive AND T cells added at the initiation of HNT ulating optimal effector generation. However, when the naive T cell:Ag/APC culture (day 0) expanded markedly by 3 days, AND T cells were added after 1 day of Ag/APC culture, ϳ40% of whereas AND T cells added after 1 or 2 days of Ag/APC culture the AND T cells recovered after 3 days had not divided, and add- underwent little expansion or even declined over the 3 days (Fig. ing naive AND T cells to Ag/APC cultures at day 2 resulted in 2b). Th2-polarizing conditions supported slightly greater effector little AND T cell division (Fig. 2a). Our results indicate that al- recoveries as compared with IL-2-polarizing (Th0) and Th1-po- though the original HNT T cells proliferate normally, regardless of larizing conditions, suggesting that the presence of IL-4 in the the time of naive AND T cell introduction to the cultures (data not cultures may enhance T cell survival. However, all AND T cell shown), the ability of the splenic APC blasts to stimulate cell di- effector recoveries had decreased to numbers below starting cell The Journal of Immunology 5021 numbers when AND T cells were added on either day 2 or 3 of they exhibit significantly reduced intracellular cytokine production Ag/APC culture. Overall, these results suggest that MitC-treated (Fig. 3, solid lines). These results indicate that the ability of Ag/ splenic APC blasts are no longer capable of supporting naive T cell APC to stimulate cytokine production by even highly activated proliferation and effector generation when cultured for more than effectors is lost as early as 1 day after in vitro culture of Ag/APC 2 days with Ag-specific T cells, regardless of the T cell-polarizing with T cells. Together, these results reveal that efficient Ag pre- cytokine milieu. sentation to both naive and effector T cells by professional APC is Since naive CD4 T cells require a significantly greater amount limited to the first 1–2 days in standard in vitro culture systems of TCR stimulation and costimulation to enter into the cell cycle (Figs. 2 and 3), and any potential residual APC are incapable of and rapidly produce high levels of cytokines, as compared for supporting naive T cell proliferation and cytokine production or 4-day effectors, we examined the ability of freshly prepared (day even the response of highly activated effectors. 0) and 1-day Ag/APC cultures to stimulate 4-day effectors to produce polarized cytokines (1, 30). We generated 4-day Th0-or Th2- Prolonged Ag stimulation leads to detrimental effects on effector polarized effectors and restimulated them with freshly prepared expansion and function DCEK-ICAM APC loaded with PCCF peptide (Fig. 3, positive In all cases discussed thus far in these studies, the APC used were control, filled histograms), freshly prepared DCEK-ICAM APC MitC treated; therefore, their short life span was expected, but it is without Ag (negative control, dotted lines), or 1-day cultures of important to note that under these standard in vitro culture condi- DCEK-ICAM APC loaded with PCCF peptide (solid lines). Twen- tions used by many groups of investigators, Ag presentation is of ty-four hours after the 4-day Th0 and Th2 effectors were added to short duration but nonetheless results in highly efficient T cell ex- the various APC cultures, we harvested the T cells and measured pansion and effector generation. However, in vivo there is the po- intracellular cytokine production by effectors. Because we needed tential for APC to interact with T cells throughout the process of Downloaded from to culture APC with Ag and T cells in order for the APC to be effector generation and to potentially provide T cells with pro- deleted from the cultures, and these T cells could potentially con- longed or repeated TCR stimulation. Since our in vitro Ag/APC tribute secreted cytokines to culture supernatants used in ELISA culture system results in the disappearance of APC between 1 and cytokine detection methods, we used ICCS to ascertain the cyto- 2 days, to determine whether increased durations of exposure to kine production of the 4-day Th0 or Th2 effectors while gating out Ag/APC results in more efficient effector generation, we examined any cytokine contribution of the naive T cells used to generate the the effects of increasing the duration of Ag presentation on effector http://www.jimmunol.org/ 1-day Ag/APC cultures. When 4-day effectors were stimulated for generation by supplementing cultures with additional Ag/APC 24 h with freshly prepared Ag/APC, they were induced to secrete throughout the 4-day culture period. We incubated naive CD4 T readily detectable levels of appropriate cytokines (positive con- cells under IL-2-polarizing (Th0) and Th2-polarizing conditions trol), whereas 4-day effectors stimulated for 24 h with freshly pre- and supplemented the cultures with fresh peptide-pulsed APC once pared APC without Ag do not produce detectable levels of cyto- at day 0 (normal culture conditions), twice at days 0 and 1, three kines (negative control). Th0 effectors typically produce high times at days 0–2, and four times at days 0–3. The effectors levels of IL-2 and slightly lower production of IFN-␥, whereas Th2 present after 4 days were then analyzed for effector recovery and effectors produce measurable levels of IL-4 as determined by ICCS their ability to secrete cytokines upon restimulation. by guest on October 2, 2021 (Fig. 3, shaded histograms, positive controls). These results can be The results indicate that effector recovery in both the IL-2-po- mimicked by the recovery of cytokines in culture supernatants larized (Th0) (Fig. 4a) and Th2-polarized (Fig. 4b) subsets is sig- from purified 4-day effector populations (data not shown). In con- nificantly decreased when standard in vitro T cell cultures are sup- trast, when 4-day Th0 or Th2 effectors are added to Ag/APC cul- plemented with extended durations of Ag/APC exposure. tures which have interacted for as little as 1 day with naive T cells, Although the Th2-polarized subset benefited slightly from one additional Ag/APC supplement, extended durations of Ag/APC exposure resulted in markedly decreased effector recoveries (Fig. 4b, shaded bars). Interestingly, the viable cells recovered after 4 days all maintained similar proliferation profiles, as measured by CFSE dye loss (data not shown), suggesting that the decrease in effector recovery which results from extended durations of Ag/APC exposure is not due to an impairment in proliferation potential of surviving cells. Therefore, we hypothesized that the decreased effector recovery (Fig. 4) indicates that effectors stimulated with Ͼ2 days of Ag presentation undergo high-level death as they respond. Using forward scatter and side scatter (FSC ϫ SSC) FACS profiles of the CD4 Tg cells, we were able to show that, in fact, the percentages of cell death increased from 11 to 55% for the IL-2 (Th0)- polarized effectors and from 18 to 34% for the Th2 effectors as FIGURE 3. Loss of Ag presentation to CD4 effectors by Ag/APC cul- Ag/APC exposure was extended. These numbers may seem too tures. Naive CD4 AND T cells were stimulated for 4 days under IL-2- low to account for the low cell recoveries, but the cell death mea- polarizing (Th0) or Th2-polarizing conditions. The resulting Th0 and Th2 surements taken at day 4 do not reflect the integrated effect of cell 4-day effectors were CFSE labeled and then reintroduced to culture with death during the culture. Dead cells typically disappear from cul- freshly prepared DCEK-ICAM APC without PCCF peptide (A, negative ture within 24–36 h of their death, and therefore a cell that dies at control: dotted lines), 24-h cultured PCCF peptide-pulsed DCEK-ICAM days 2–3 of culture would not be detectable on day 4. Additionally, (B, solid lines), or freshly prepared PCCF peptide-pulsed DCEK-ICAM (C, positive control: shaded histograms). ICCS for IL-2 and IFN-␥ was mea- if a cell dies on day 2 its potential capacity to expand had it lived sured on CFSEϩ-, V␤3ϩ-, CD4ϩ-gated Th0 effectors after 24 h of restimu- must be taken into account when adjusting for cell recovery at day lation. IL-4 was measured on CFSEϩ-, V␤3ϩ-, CD4ϩ-gated Th2 effectors 4. Thus, the cell recovery at day 4 has the potential to be much after 24 h of restimulation. Data are representative of between three and lower than what would be expected based on calculations of per- four independent experiments. centages of cell death. 5022 Ag-DEPENDENT AND IL-2-DRIVEN STAGES FROM NAIVE TO EFFECTOR

prolonged Ag presentation when the time of exogenous IL-2 addition was delayed from day 0 to day 1 or 2 (data not shown). The ability of effectors to rapidly produce high levels of multiple cytokines after restimulation is an important functional attribute which is critical to regulate the immune response to an invading pathogen. Therefore, we also tested the ability of the surviving viable 4-day effectors to produce polarized cytokines. Because T cell recovery was much lower in the cultures with prolonged Ag/ APC exposure, we readjusted the 4-day effectors recovered to equivalent cell concentrations before restimulation with Ag/APC for 24 h. We measured IL-2 levels produced by IL-2-polarized (Th0) effectors (Fig. 4a) and IL-4 levels produced by Th2-polarized effectors (Fig. 4b). Equivalent numbers of effectors obtained from cultures with prolonged Ag/APC stimulation produced much lower levels of cytokines compared with effectors resulting from the standard (shorter) Ag/APC exposure (Fig. 4). Together, these studies show that exposure to Ag/APC beyond 2 days during effector generation not only leads to decreased effector recovery,

which is most likely due to increased cell death, but also decreases Downloaded from the ability of surviving T cells to function properly and produce polarized cytokines upon restimulation. Thus, the susceptibility of restimulated developing effectors to cell death may limit the cytokine accumulation upon restimulation or select for less differentiated T cells. Supplementing T cell cultures with Ag/APC beyond 2 days was http://www.jimmunol.org/ detrimental to the recovery and function of resulting effectors; however, it is possible that this result might be due to the effects of culture crowding by the added APC or some other negative impact FIGURE 4. Negative effect of prolonged Ag stimulation with Ag/APC. of adding fresh APC to the culture. Therefore, we used plate- Naive CD4 AND T cells were CFSE labeled and cultured at 3 ϫ 105 T bound anti-TCR mAb stimulation plus soluble anti-CD28 mAb cells/culture with 1.5 ϫ 105 DCEK-ICAM pulsed with PCCF peptide under IL-2-polarizing (Th0) or Th2-polarizing cytokine environments. A total costimulation conditions to independently define the optimal du- of 1.5 ϫ 105 freshly prepared PCCF peptide-pulsed DCEK-ICAM was ration of Ag presentation. These studies were designed to provide added once at day 0 only, twice at days 0 and 1, three times at days 0–2, optimal stimulatory conditions throughout the culture period with ϩ or four times at days 0–3. a, Th0 effector recoveries of live V␤3 (AND)/ the only variable being the duration of TCR stimulation. Naive by guest on October 2, 2021 ϩ CD4 T cells were determined at day 4 of culture by direct cell counts. Cell AND T cells were incubated in anti-V␤3 mAb-coated wells with death was measured by FSC ϫ SSC FACS profiling and determined to be soluble anti-CD28 mAb and IL-2 alone (Th0) or under Th2-po- 11, 36, 52, and 55%, respectively. IL-2 production was measured by bio- larizing conditions. The T cell cultures were then transferred to assay of supernatants recovered after 24 h of restimulation of 4-day Th0 noncoated tissue culture wells, without washing, over a broad effectors with PCCF peptide-pulsed DCEK-ICAM. b, Th2 effector recoveries of live V␤3ϩ(AND)/CD4ϩ T cells were determined at day 4 of cul- range of time points. Effector recovery, effector death, and polar- ture by direct cell counts. Cell death was measured by FSC ϫ SSC FACS ized cytokine-producing capacity of the resulting effectors were profiling and determined to be 18, 13, 23, and 34%, respectively. IL-4 measured at the end of a total 4-day culture period. production was measured by ELISA of supernatants recovered after 24 h of Our results indicate that the highest level of effector recovery restimulation of 4-day Th2 effectors with PCCF peptide-pulsed DCEK- results from 2 days of TCR stimulation in both the IL-2-polarized ICAM. Data are representative of four independent experiments. (Th0) (Fig. 5a) and Th2-polarized (Fig. 5c) populations, with shorter and longer durations of TCR stimulation resulting in low numbers of effectors (Fig. 5, a and c). Stimulation with anti-TCR Ͻ In these studies, we provide exogenous IL-2 at the beginning of mAb for 2 days resulted in effectors that appeared poorly acti- T cell culture (day 0) to promote optimal T cell expansion. How- vated as indicated by changes in activation marker expression and ever, previous studies have shown that production and accumula- small cell size, whereas exposure to anti-TCR mAb beyond 2 days tion of autocrine IL-2 by naive T cells takes 20–30 h following resulted in effectors that did display a highly activated phenotype stimulation with optimum levels of TCR stimulation and costimu- (data not shown). We hypothesized that the decrease in effector lation (Ref. 11; L. Haynes and S. L. Swain; unpublished data). IL-2 recovery resulting from increased durations of Ag/APC exposure is known to regulate AICD such that T cells previously exposed to (Figs. 1c and 4) is due to AICD as effectors are reexposed to IL-2 undergo apoptosis rapidly following Ag receptor stimulation Ag/APC or anti-TCR Ab, rather than due to a decrease in induction (26, 47). Moreover, the kinetics of IL-2 addition to culture has of cell proliferation. Therefore, we examined effector death, as been shown to have profound effects on the susceptibility and res- measured by propidium iodide (PI) uptake or FSC ϫ SSC FACS cue of T cells from Fas-mediated AICD (17). Therefore, we ex- profiles of the cultures. Indeed, longer durations of TCR stimula- amined the effects of increasing the duration of Ag presentation on tion resulted in high levels of effector death (Fig. 5, a and c). Taken effector generation by supplementing cultures with additional Ag/ together, the decreased effector recovery and increased cell death APC throughout the 4-day culture period, as described in Fig. 4, (Fig. 5, a and c) by cultures stimulated with Ͼ2 days of TCR and adding exogenous IL-2 on day 0 (normal culture conditions), stimulation support the hypothesis that prolonging Ag presentation 1, or 2. We found there to be no difference in the kinetics of results in high levels of T cell death, which directly results in de- decreased effector recovery and increased cell death caused by creased effector recovery rather than increased effector responses. The Journal of Immunology 5023

FIGURE 5. Negative effect of prolonged Ag stimulation with anti-V␤3 (anti-TCR). Naive CD4 AND T cells were CFSE labeled and cultured at 3 ϫ 105 T cells/culture in anti- V␤3-coated wells with soluble anti-CD28 under IL-2-polarizing (Th0) or Th2-polarizing cytokine environments. Total T cell cultures and supernatants were transferred to noncoated tissue culture wells without washing at the times indicated. a, Th0 effector recoveries of live V␤3ϩ(AND)/CD4ϩ T cells were determined at day 4 of culture by direct cell counts. Percent effector mortality was determined by PI uptake (percent PIϩ)ofV␤3ϩ(AND)/CD4ϩ as measured by FACS analysis. b, IL-2 production was measured by bioassay of supernatants recovered after 24 h of restimulation of 4-day Th0 effectors with PCCF peptide-pulsed DCEK-ICAM. ND, Not determined because there were not enough cells to restimulate. c, Th2 effector recoveries of live V␤3ϩ(AND)/CD4ϩ T cells were deter- Downloaded from mined at day 4 of culture by direct cell counts. Percent effector mortality was determined by FSC ϫ SSC FACS proﬁling of V␤3ϩ(AND)/ CD4ϩ T cells. d, IL-4 production was measured by ELISA of supernatants recovered after 24 h of restimulation of 4-day Th2 effectors with http://www.jimmunol.org/ PCCF peptide-pulsed DCEK-ICAM. Data are representative of ﬁve independent experiments.

We also analyzed the ability of effectors generated in the anti- beled the T cells at day 2 of culture before dividing the cells into TCR mAb cultures to be restimulated to produce polarized cyto- the three respective groups. Because all of the T cells are already kines. We found that a slightly longer Ag presentation of 21⁄2 days cycling, we observed a relatively synchronized cell proliferation by guest on October 2, 2021 resulted in IL-2-polarized (Th0) effectors best capable of produc- profile unlike that seen when naive or resting T cells are CFSE ing high levels of IL-2 upon restimulation, with lesser and longer labeled and then cultured to proliferate. Naive or resting T cells durations of Ag presentation resulting in lower levels of cytokine enter the cell cycle in an asynchronous manner which has a delay production (Figs. 4a and 5b). Whereas 2 days of Ag presentation of 24–32 h, and then they divide at a relatively constant rate of resulted in Th2 effectors best capable of producing high levels of every 6–8 h (Fig. 1). Alternatively, actively cycling cells (those IL-4 upon restimulation (Figs. 4b and 5d). The need for a shorter seen on days 2–4 of culture) appear to be relatively synchronized TCR stimulation in the Th2-polarizing cultures could be due to the in their cell cycles. This synchronized cell cycling results in CFSE presence of IL-4, which promotes a more rapid activation as com- profiles which are less defined for specific division peaks. How- pared with IL-2 alone that is present in Th0 cultures (D. M. Jelley- ever, the numbers of divisions each cell has undergone can be Gibbs and S. L. Swain, unpublished observations). easily calculated by measuring the mean fluorescent intensity (MFI) of the CFSE as dye loss in each daughter cell being equal to Cytokine dependence during Ag-independent late phase of exactly 50% of that of the parent cell. For example, if the MFI of effector generation CFSE labeling on day 2 is 1000, the MFI of a cell that had divided Our next set of experiments was aimed at examining the mecha- one time would be 500, twice would be 250, and three times would nism responsible for driving the constant rate of cell division seen be 125. Using these mathematical calculations, we were able to in the late phase (days 2–4) of our standard in vitro cultures (Fig. determine the average number of divisions occurring between days 1), despite the clear lack of Ag presentation beyond the early phase 2 and 4 of effector generation. At the end of the 4-day culture of effector generation (days 0–2). We hypothesized that the Ag- period, the effector recovery and CFSE profiles were examined, independent constant rate of cell division observed in the late and the ability of the resulting viable effectors to produce polarized phase of cell expansion is either driven by growth-promoting cy- cytokines upon restimulation was measured by ICCS. tokines, like IL-2, or is a consequence of early and irreversible Under conditions where effectors received no ongoing IL-2 programming events within the T cell, or a combination of both. stimulation (blocking with anti-IL-2R␣ Ab for the last 2 days of To test whether IL-2 is required to drive the late phase of effector culture), there was a greatly reduced effector recovery (Fig. 6a), generation, naive CD4 cells were cultured with Ag/APC under which was paralleled by a decreased rate of cell division between standard Th2-polarizing conditions for 2 days. The T cells were days 2 and 4 as determined by CFSE dye loss (Fig. 6b). The de- then washed to remove any exogenous and secreted IL-2, CFSE creased effector recovery of ϳ50% of that found when IL-2 was labeled, and recultured for an additional 2 days with either 1) cul- present was accounted for by a reduction of an average number of ture supernatants added back, 2) fresh T cell media plus exogenous cell divisions of 4.25 divisions for the cultures containing IL-2, to IL-2, or 3) fresh T cell media plus anti-IL-2R␣-blocking Ab. To 2 divisions for the cultures without IL-2. We quantitated cell death obtain high levels of synchronized CFSE labeling, we CFSE la- of these cultures by FSC ϫ SSC FACS profiling and found no 5024 Ag-DEPENDENT AND IL-2-DRIVEN STAGES FROM NAIVE TO EFFECTOR

by the lack of IL-2 stimulation, but the resulting effectors were incapable of optimum effector function, as measured by cytokine production. This supports the hypothesis that there are two distinct phases of cell activation and expansion such that effectors ﬁrst undergo a 1–2-day Ag-dependent phase of cell activation and expansion, which is then followed by an Ag-independent and IL-2- driven late phase of effector differentiation and expansion into fully functional effectors (Figs. 1 and 6) (25). This pattern suggests that mechanisms may well exist in situ to ensure that once fully activated, responding CD4 T cells must be segregated from Ag- bearing APC. In these studies, which use highly puriﬁed naive CD4 T cells and APC populations, autocrine IL-2 and exogenous IL-4 are the only IL-2R␥ chain-binding, growth-promoting cytokines available. It is likely that other IL-2R␥ chain-binding cytokines, or other growth-promoting cytokines, may drive in vivo responses (48, 49). Thus, it will be important to determine whether other cytokines can promote this Ag-independent phase in vivo and to determine what regulates their availability.

By supplementing T cell cultures with more prolonged Ag pre- Downloaded from sentation, either in the form of Ag/APC or plate-bound anti-TCR Ab, we found that Ag presentation beyond day 2 leads to the disappearance and death of responding CD4 T cells, thus resulting in recovery of fewer effectors at day 4. Moreover cytokine production by surviving effectors is decreased by the prolonged Ag exposure (Figs. 4 and 5). Many responding CD4 T cells, which continue to http://www.jimmunol.org/ encounter Ag/APC, apparently undergo AICD, which results in the selection of a population of cells which produce less cytokine, perhaps reflecting a selection for the least differentiated cohorts of cells. This may be a means to prevent uncontrolled effector expansion, which could lead to detrimental inflammatory responses. These results augment many previous in vitro studies whereby restimulated effector T cells undergo cell deletion via AICD (17, 50). We have recently confirmed that CD4 effectors transferred to

adoptive hosts also disappear when hosts are immunized with Ag by guest on October 2, 2021 (E. Roman, G. E. Huston, and S. L. Swain, unpublished observations). As discussed below, we suggest that these findings may indicate that productive in vivo responses must involve a process which ensures responding CD4 T cells are not reexposed to Ag FIGURE 6. Cytokine dependence during Ag-independent phase of ef- after 2 days. fector generation. Naive CD4 AND T cells were cultured at 3 ϫ 105 T The results presented in these studies also have important im- cells/culture with 1.5 ϫ 105 PCCF peptide-pulsed DCEK-ICAM under plications for the popular in vitro model system in which TCR Th2-polarizing conditions for 2 days. To synchronize the CFSE labeling, stimulation is provided through plate-bound anti-TCR Ab. These the cultured cells were then washed to remove exogenous and secreted studies are commonly performed by incubating T cells in tissue IL-2, CFSE labeled, and cultured an additional 2 days with either their culture plates coated with anti-CD3 Ab, anti-TCR Ab, or tetramers original culture supernatants added back, T cell media plus only exogenous for the entire duration of the experiment, which typically extends ␣ ␤ ϩ IL-2, or T cell media plus anti-IL-2R -blocking Ab. a, Live V 3 (AND)/ to 4–7 days of culture. These results predict that prolonged TCR CD4ϩ effector recoveries were determined at day 4 of culture by direct cell stimulation, beyond the first 2 days of culture for naive T cells, counts. Cell death was determined by FSC ϫ SSC FACS profiling and was 17, 14, and 25%, respectively. b, FACS analysis of CFSE dye loss of live may select for a subpopulation of less functional effectors and V␤3ϩ(AND)/CD4ϩ effectors was used to measure cell division during the preclude recovery of the most active populations of effectors. last 2 days of culture. c, ICCS for IL-2 and IL-4 was measured on live These results also predict that extended TCR stimulation would CFSEϩV␤3ϩ(AND)/CD4ϩ effectors restimulated for 4 h with PMA ϩ promote the selection of a subpopulation of effectors resistant to ionomycin. Data are representative of three independent experiments. cell death. These results present a need to evaluate the functional and phenotypic qualities of effectors generated using this particular significant increases in cell death as a result of blocking IL-2. in vitro model system for T cell stimulation. Therefore, instead of proliferating every 8 h, effectors cultured Since optimal CD4 effector generation and function is depen- during the late phase of effector generation with anti- dent on an initial 1–2 days of Ag presentation followed by the IL-2R␣-blocking Ab only underwent one round of proliferation absence of Ag presentation during an IL-2-driven cell expansion approximately every 16 h (Fig. 6b). Moreover, effectors cultured phase lasting an additional 2 days (Fig. 6), it is important to con- under conditions where IL-2 stimulation was blocked were capable sider the possibility that in vivo CD4 T cell responses may be of producing little IL-2 and no IL-4, whereas cultures reconstituted compartmentalized to achieve this set of conditions. Several pub- with their own supernatants or with exogenous IL-2 produced nor- lished reports support the concept of chronologically and anatom- mal levels of IL-2 and IL-4 (Fig. 6c). Therefore, not only was the ically separate phases in CD4 T cell response. Ingulli et al. (18) rate of cell proliferation during the final 2 days of culture reduced have shown that Ag (OVA)-specific CD4 T cells and Ag-pulsed The Journal of Immunology 5025 dendritic cells (DC) accumulate to a maximal level in the T cell- 12. Powell, J. D., Ragheb, J. A., Kitagawa-Sakakida, S., and Schwartz, R. H. 1998. rich paracortical regions of spleen and lymph nodes within 24 h of Molecular regulation of interleukin-2 expression by CD28 co-stimulation and anergy. Immunol. Rev. 165:287. i.v. injection of the Ag-pulsed DC and that the accumulation of 13. Fitzpatrick, D. R., K. M. Shirley, and A. Kelso. 1999. Cutting edge: stable epi- DC:T cell clusters then declined by 48 h (18). This dissociation genetic inheritance of regional IFN-␥ promoter demethylation in CD44highCD8ϩ could have been due to limitations of Ag or instead an active T lymphocytes. J. Immunol. 162:5053. 14. Richter, A., M. Lohning, and A. Radbruch. 1999. Instruction for cytokine ex- process associated with the CD4 T cell response. By 48 h, Ag- pression in T helper lymphocytes in relation to proliferation and cell cycle pro- pulsed DC rapidly disappeared from the lymph nodes, either be- gression. J. Exp. Med. 190:1439. cause they migrate out of the lymph node or are killed by the 15. Swain, S. L., A. D. Weinberg, M. English, and G. Huston. 1990. IL-4 directs the responding T cells. Two days after immunization, both Ag-specific development of Th2-like helper effectors. J. Immunol. 145:3796. 16. Lanzavecchia, A., G. Iezzi, and A. Viola. 1999. From TCR engagement to T cell T cells and B cells moved toward each other from their separate activation: a kinetic view of T cell behavior. Cell 96:1. starting locations and cognate T cell-B cell interactions could be 17. Zhang, X., . Giangreco, H. E. Broome, C. M. Dargan, and S. L. Swain. 1995. visualized at the edge of the B cell-rich follicles at the border Control of CD4 effector fate: TGF␤1 and IL-2 synergize to prevent apoptosis and promote effector expansion. J. Exp. Med. 182:699. between the follicles and the T cell areas in the spleen (33). By 18. Ingulli, E., A. Mondino, A. Khoruts, and M. K. Jenkins. 1997. In vivo detection days 3 and 4, the Ag-specific T cells are no longer found near the of dendritic cell antigen presentation to CD4ϩ T cells. J. Exp. Med. 185:2133. Ag-presenting B cells (18, 33). Perhaps those T cells destined to 19. Rogers, P. R., G. Huston, and S. L. Swain. 1998. High antigen density and IL-2 expand and produce large populations of effectors relocate as well are required for generation of CD4 effectors secreting Th1 rather than Th0 cytokines. J. Immunol. 161:3844. to a separate location rich in growth and differentiation-promoting 20. Lanzavecchia, A. 1998. From antigen presentation to T-cell activation. Res. Im- factors. Indeed, a recent study by Vasseur et al. (32) suggests that munol. 149:626. Ag-specific T cell accumulation in the lymph nodes is very tran- 21. Carter, L. L., X. Zhang, C. Dubey, P. Rogers, L. Tsui, and S. L. Swain. 1998. sient (1–2 days), and proliferating cells recirculate in the blood Regulation of T cell subsets from naive to memory. J. Immunother. 21:181. Downloaded from 22. Swain, S. L. 1994. Generation and in vivo persistence of polarized Th1 and Th2 rapidly after Ag presentation and T cell activation (32). Thus, the memory cells. Immunity 1:543. concept that the period where Ag is presented to CD4 T cells is of 23. Swain, S. L., L. M. Bradley, M. Croft, S. Tonkonogy, G. Atkins, A. D. Weinberg, limited duration seems to be mirrored by the in vivo migratory D. D. Duncan, S. M. Hedrick, R. W. Dutton, and G. Huston. 1991. Helper T-cell subsets: phenotype, function and the role of lymphokines in regulating their de- patterns of responding CD4 T cells and APC. Together, these stud- velopment. Immunol. Rev. 123:115. ies support a model in which CD4 T cells are activated in the first 24. Bird, J. J., D. R. Brown, A. C. Mullen, N. H. Moskowitz, M. A. Mahowald, 1–2 days of an immune response by Ag-presenting DC and begin J. R. Sider, T. F. Gajewski, C. R. Wang, and S. L. Reiner. 1998. Helper T cell http://www.jimmunol.org/ division. Then the T cells migrate to either an area in the lymphoid differentiation is controlled by the cell cycle. Immunity 9:229. 25. Haynes, L., P.-J. Linton, S. M. Eaton, S. L. Tonkonogy, and S. L. Swain. 1999. compartment where they can interact with Ag-presenting B cells or IL-2, but not other common ␥ chain (␥c)-binding cytokines, can reverse the they may migrate away from the Ag/APC environments altogether. defect in generation of CD4 effector T cells from naive T cells of aged mice. Whether there is a defined site of additional CD4 T cell expansion J. Exp. Med. 190:1013. 26. Lenardo, M. J. 1991. Interleukin-2 programs mouse ␣␤ T lymphocytes for apo- and differentiation is unclear (18, 32, 33, 51, 52). However, the ptosis. Nature 353:858. need for responding CD4 T cells to encounter IL-2 or an in vivo 27. Rafaeli, Y., L. V. Parijs, C. A. London, J. Tschopp, and A. K. Abbas. 1998. surrogate for full expansion and effector development would sug- Biochemical mechanisms of IL-2-regulated Fas-mediated T cell apoptosis. Im- gest that they may do so in a defined niche. munity 8:615. 28. Iezzi, G., K. Karjalainen, and A. Lanzavecchia. 1998. The duration of antigenic by guest on October 2, 2021 stimulation determines the fate of naive and effector T cells. Immunity 8:89. 29. Rogers, P. R., and M. Croft. 1999. Peptide dose, affinity, and time of differen- References tiation can contribute to the Th1/Th2 cytokine balance. J. Immunol. 163:1205. 30. Croft, M.,L. M. Bradley, and S. L. Swain. 1994. Naive versus memory CD4 T 1. Dubey, C., M. Croft, and S. L. Swain. 1995. Costimulatory requirements of naive ϩ cell response to antigen: memory cells are less dependent on accessory cell co- CD4 T cells: ICAM-1 or B7-1 can costimulate naive CD4 T cell activation but stimulation and can respond to may antigen-presenting cell types including rest- both are required for optimum response. J. Immunol. 155:45. ing B cells. J. Immunol. 152:2675. 2. Dubey, C., M. Croft, and S. L. Swain. 1996. Naive and effector CD4 T cells differ 31. Seder, R. A. 1996. High-dose IL-2 and IL-15 enhance the in vitro priming of in their requirements for T cell receptor versus costimulatory signals. J. Immunol. ϩ naive CD4 T cells for IFN-␥ but have differential effects on priming for IL-4. 157:3280. J. Immunol. 156:2413. 3. Swain, S. L., M. Croft, C. Dubey, L. Haynes, P. Rogers, X. Zhang, and 32. Vasseur, F., A. Le Campion, J. H. Pavlovitch, and C. Penit. 1999. Distribution of L. M. Bradley. 1996. From naive to memory T cells. Immunol. Rev. 150:143. cycling T lymphocytes in blood and lymphoid organs during immune responses. 4. Bradley, L. M., D. D. Duncan, S. Tonkonogy, and S. L. Swain. 1991. Charac- J. Immunol. 162:5164. terization of antigen-specific CD4ϩ effector T cells in vivo: immunization results Ϫ Ϫ 33. Garside, P., E. Ingulli, R. R. Merica, J. G. Johnson, R. J. Noelle, and in a transient population of MEL-14 , CD45RB helper cells that secretes in- M. K. Jenkins. 1998. Visualization of specific B and T lymphocyte interactions ␥ terleukin 2 (IL-2), IL-3, IL-4, and interferon . J. Exp. Med. 174:547. in the lymph node. Science 281:96. 5. Gramaglia, I., D. Cooper, K. T. Miner, B. S. Kwon, and M. Croft. 2000. Co- 34. Jaiswal, A. I., and M. Croft. 1997. CD40 ligand induction on T cell subsets by stimulation of antigen-specific CD4 T cells by 4-1BB ligand. Eur. J. Immunol. peptide-presenting B cells: implications for development of the primary T and B 30:392. cell response. J. Immunol. 159:2282. 6. Ullman, K. S., J. P. Northrop, C. L. Verweij, and G. R. Crabtree. 1990. Trans- 35. Wells, A. D., H. Gudmundsdottir, and L. A. Turka. 1997. Following the fate of mission of signals from the T lymphocyte antigen receptor to the genes respon- individual T cells throughout activation and clonal expansion: signals from T cell sible for cell proliferation and immune function: the missing link. Annu. Rev. receptor and CD28 differentially regulate the induction and duration of a prolif- Immunol. 8:421. erative response. J. Clin. Invest. 100:3173. 7. Boehme, S. A., and Lenardo, M. J. 1993. Propriocidal apoptosis of mature T 36. Karasuyama, H., and F. Melchers. 1988. Establishment of mouse cell lines which lymphocytes occurs at S phase of the cell cycle. Eur. J. Immunol. 23:1552. constitutively secrete large quantities of interleukin 2, 3, 4 or 5, using modified 8. Radvanyi, L. G., Mills, G. B., and Miller, R. G. 1993. Regulation of the T cell cDNA expression vectors. Eur. J. Immunol. 18:97. receptor after primary activation of mature T cell inhibits proliferation and in- 37. Linton, P. J., L. Haynes, N. R. Klinman, and S. L. Swain. 1996. Antigen-inde- duces apoptotic cell death. J. Immunol. 150:5704. pendent changes in naive CD4 T cells with aging. J. Exp. Med. 184:1891. 9. Fitzpatrick, D. R., K. M. Shirley, L. E. McDonald, H. Bielefeldt-Ohmann, 38. Openshaw, P., E. E. Murphy, N. A. Hosken, V. Maino, K. Davis, K. Murphy, and G. F. Kay, and A. Kelso. 1998. Distinct methylation of the interferon ␥ (IFN-␥) A. O’Garra. 1995. Heterogeneity of intracellular cytokine synthesis at the single- ϩ and interleukin 3 (IL-3) genes in newly activated primary CD8 T lymphocytes: cell level in polarized T helper 1 and T helper 2 populations. J. Exp. Med. 182: regional IFN-␥ promoter demethylation and mRNA expression are heritable in 1357. high ϩ CD44 CD8 T cells. J. Exp. Med. 188:103. 39. Assenmacher, M., J. Schmitz, and A. Radbruch. 1994. Flow cytometric determi- 10. Bix, M., and Locksley, R. M. 1998. Independent and epigenetic regulation of the nation of cytokines in activated murine T helper lymphocytes: expression of interleukin-4 alleles in CD4ϩ T cells. Science 281:1352. interleukin-10 in interferon-␥ and in interleukin-4-expressing cells. Eur. J. Im- 11. Zuckerman, L. A., Pullen, L., and Miller, J. 1998. Functional consequences of munol. 24:1097. costimulation by ICAM-1 on IL-2 gene expression and T cell activation. J. Im- 40. Lyons, A. B., and C. R. Parish. 1994. Determination of lymphocyte division by munol. 160:3259. flow cytometry. J. Immunol. Methods 171:131. 5026 Ag-DEPENDENT AND IL-2-DRIVEN STAGES FROM NAIVE TO EFFECTOR

41. Pape, K. A., E. R. Kearney, A. Khoruts, A. Mondino, R. Merica, Z. M. Chen, 47. Fournel, S., L. Genestier, E. Robinet, M. Flacher, and J.-P. Revillard. 1996. E. Ingulli, J. White, J. G. Johnson, and M. K. Jenkins. 1997. Use of adoptive Human T cells require IL-2 but not G1/S transition to acquire susceptibility to transfer of T-cell-antigen-receptor-transgenic T cell for the study of T-cell acti- Fas-mediated apoptosis. J. Immunol. 157:4309. vation in vivo. Immunol. Rev. 156:67. 48. Anderson, D. M., S. Kumaki, M. Ahdieh, J. Bertles, M. Tometsko, A. Loomis, 42. McHeyzer-Williams, M. G., J. D. Altman, and M. M. Davis. 1996. Tracking J. Giri, N. G. Copeland, D. J. Gilbert, and N. A. Jenkins. 1995. Functional char- antigen-speciﬁc helper T cell responses. Curr. Opin. Immunol. 8:278. acterization of the human interleukin-15 receptor ␣ chain and close linkage of 43. Mikszta, J. A., L. J. McHeyzer-Williams, and M. G. McHeyzer-Williams. 1999. IL15RA and IL2RA genes. J. Biol. Chem. 270:29862. Antigen-driven selection of TCR in vivo: related TCR ␣-chains pair with diverse 49. DiSanto, J. P., D. Guy-Grand, A. Fisher, and A. Tarakhovsky. 1996. Critical role TCR ␤-chains. J. Immunol. 163:5978. for the common cytokine receptor ␥ chain in intrathymic and peripheral T cell 44. Rogers, P. R., H. M. Grey, and M. Croft. 1998. Modulation of naive CD4 T cell selection. J. Exp. Med. 183:1111. activation with altered peptide ligands: the nature of the peptide and presentation 50. Lenardo, M., F. K.-M. Chan, F. Hornung, H. McFarland, R. Siegel, J. Wang, and in the context of costimulation are critical for a sustained response. J. Immunol. L. Zheng. 1999. Mature T lymphocyte apoptosis: immune regulation in a dy- 160:3698. namic and unpredictable antigenic environment. Annu. Rev. Immunol. 17:221. 45. Lanzavecchia, A. 1997. Understanding the mechanisms of sustained signaling 51. Fuchs, E. J., and P. Matzinger. 1992. B cells turn off virgin but not memory T and T cell activation. J. Exp. Med. 185:1717. cells. Science 258:1156. 46. Swain, S. L., H. Hu, and G. Huston. 1999. Class II-independent generation of 52. Eynon, E. E., and D. C. Parker. 1992. Small B cells as antigen-presenting cells CD4 memory T cells from effectors. Science 286:1381. in the induction of tolerance to soluble protein antigens. J. Exp. Med. 175:131. Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021