T Cells Differential Survival of Naive CD4 And

Differential Survival of Naive CD4 and CD8 T Cells Cristina Ferreira, Thomas Barthlott, Sylvie Garcia, Rose Zamoyska and Brigitta Stockinger This information is current as of September 25, 2021. J Immunol 2000; 165:3689-3694; ; doi: 10.4049/jimmunol.165.7.3689 http://www.jimmunol.org/content/165/7/3689 Downloaded from

References This article cites 44 articles, 19 of which you can access for free at: http://www.jimmunol.org/content/165/7/3689.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

• 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 by guest on September 25, 2021

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. Differential Survival of Naive CD4 and CD8 T Cells1

Cristina Ferreira, Thomas Barthlott, Sylvie Garcia,2 Rose Zamoyska, and Brigitta Stockinger3

In this paper we compare survival characteristics of transgenic and polyclonal CD4 and CD8 T cells. Transgenic CD4 T cells have an intrinsically lower capacity for survival, reﬂected in their gradual disappearance in thymectomized hosts, their increased sensitivity to apoptosis in vitro, and fewer divisions during homeostatic proliferation upon transfer into syngeneic lymphopenic hosts compared with CD8 T cells. Homeostatic proliferation, however, does not generally result in phenotypic conversion of activation markers unless cognate or cross-reactive Ag is present. T cells from the A18 TCR transgenic strain normally selected into the CD4 lineage are fragile as CD4 T cells, yet display the typical robust survival pattern of CD8 T cells when diverted into the CD8 lineage in a CD4-deﬁcient host. Polyclonal CD4 and CD8 T cells also show distinctive patterns of survival, emphasizing that survival signals are relayed differently in the two lymphocyte subpopulations. However, expression levels of Bcl-2 in either transgenic or polyclonal naive CD4 and CD8 T cells are similar, excluding a role for this molecule as a key factor in differential survival of CD4 vs CD8 T cells. The Journal of Immunology, 2000, 165: 3689–3694. Downloaded from

ollowing positive selection and maturation in the thymus, more efﬁciently in lymphopenic hosts. This is presumably due to CD4 and CD8 T cells colonize the periphery and recircu- the absence of cells competing for survival signals, because co- F late through lymphoid organs until they encounter Ag. Ac- transfer of large numbers of T cells can suppress this expansion tivation by Ag results in transient expansion of Ag-speciﬁc clones (15, 16). However, there seem to be substantial differences in the

followed by contraction, so that the total number of peripheral cells expansion capacity of cells with different TCRs. Despite the ab- http://www.jimmunol.org/ remains in homeostatic equilibrium (1). The numbers of different sence of bona fide cognate Ag in the adoptive hosts, some T cells T lymphocyte subpopulations appear to be independently regu- appear to “fill the space” and even up-regulate certain activation lated, suggesting independent niches that control the survival of markers, whereas others show restrained proliferation without naive or activated/memory T cells (2, 3). The mechanisms that changes in such markers (12, 13, 16–19). In this paper we have determine the life span and survival of naive T cells in the periph- analyzed different survival criteria for T cells from TCR transgenic ery are not clearly defined at present. The presence of syngeneic mice and nontransgenic controls. Taken together the data suggest MHC molecules with a diverse repertoire of peptides is essential that CD4 T cells have an intrinsically lower capacity for survival for the survival of naive CD4 and CD8 T cells (4–7), whereas compared with CD8 T cells, reflected in their gradual disappear- memory T cells for both subpopulations can persist for a long time ance in thymectomized hosts, their increased sensitivity to apopto- by guest on September 25, 2021 in the absence of MHC molecules (8, 9). This indicates that TCR- sis in vitro, and their fewer divisions upon transfer into syngeneic mediated signals are essential in the maintenance of the naive T lymphopenic hosts. Interestingly, T cells from the A18 TCR trans- cell pool, but not in the survival of memory cells. genic strain normally selected into the CD4 lineage display the While mice with a polyclonal repertoire maintain their periph- typical survival pattern of CD8 T cells when diverted into the CD8 eral T cell numbers in the absence of thymic output (e.g., after lineage in a CD4-deficient host. Polyclonal CD4 and CD8 T cells thymectomy), this does not appear to be the case for all TCR also show distinctive patterns of survival, emphasizing that sur- transgenic mice, suggesting that individual T cell clones may have vival signals are relayed differently in the two lymphocyte different requirements for their maintenance, probably depending subpopulations. on TCR avidity and the availability of selecting ligands (10, 11). The response to survival signals results in homeostatic prolif- Recent experiments transferring naive T cells labeled with a dye eration, but not in activation, so that naive T cells do not normally that allows monitoring of cell divisions into lymphopenic hosts undergo phenotypic conversion in their expression of characteris- have established the concept of maintenance as a sum of survival and proliferation (reviewed in Ref. 3). Although naive T cells have tic activation markers such as CD44 or CD45RB. However, the been shown to occasionally undergo division in their physiological presence of cognate or potential cross-reactive Ag for transgenic T environment (12–14), both CD4 and CD8 T cells proliferate much cells causes phenotypic conversion, suggesting that under certain circumstances homeostatic proliferation is superseded by Ag driven proliferation. Division of Molecular Immunology, National Institute for Medical Research, London, United Kingdom Received for publication March 13, 2000. Accepted for publication July 17, 2000. Materials and Methods The costs of publication of this article were defrayed in part by the payment of page Animals charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The strains used for these analyses were A18 TCR transgenic (Rag1Ϫ/Ϫ 1 C.F. was supported by a Ph.D. fellowship (Program PRAXIS XXI) from the gov- C5Ϫ/ϪH-2a) (20), A1 TCR transgenic (Rag1Ϫ/Ϫ C5Ϫ/Ϫ H-2a) (21), A18 ernment of Portugal. S.G. was the recipient of an EMBO long-term fellowship. Ϫ/Ϫ Ϫ/Ϫ Ϫ/Ϫ a Ϫ/Ϫ bxk Ϫ/Ϫ b CD4 (Rag1 C5 H-2 ), AND Rag H-2 (22), F5 Rag H-2 Ϫ/Ϫ k Ϫ/Ϫ Ϫ/Ϫ 2 Current address: Unite´ d’Oncologie Virale, Institut Pasteur, 78 rue du Docteur (23), BM3 Rag H-2 (24), as well as syngeneic empty Rag1 C5 Ϫ Ϫ Ϫ Ϫ Roux, 75015 Paris, France. H-2a, Rag1 / H-2k, and Rag1 / H-2b mice. Polyclonal controls were b k a 3 Address correspondence and reprint requests to Dr. Brigitta Stockinger, Division of B10 (H-2 ), CBA (H-2 ), or A/J (H-2 ). All mice were kept in conven- Molecular Immunology, National Institute for Medical Research, The Ridgeway, Mill tional, but pathogen-free, animal facilities at the National Institute for Med- Hill, London, U.K. NW7 1AA. E-mail address: [email protected] ical Research (London, U.K.).

Flow cytometry and mAbs Analytical flow cytometry was conducted using a FACScan (Becton Dick- inson, Mountain View, CA), and the data were processed using CellQuest software (Becton Dickinson). Three-color stainings were performed with FITC-, PE-, and biotin-conjugated mAbs (followed by streptavidin- RED670 (Life Technologies, Paisley, U.K.). Anti-CD4-PE (H129.19) was purchased from PharMingen (San Diego, CA), and anti-CD8 (YTS 169.4), anti-TCR V␤8.3 (7G8.2), and anti-TCR V␤8.2 (F23.2) were conjugated with biotin or FITC using standard procedures. Cells were preincubated with unlabeled mAb to Fc␥RII/III (2.4G2) to minimize unspecific staining. All stainings were performed on ice, and cells were washed with washing buffer (PBS, 2% FCS, and 0.1% azide). For detection of early apoptosis, single-cell suspensions were incubated with PE-conjugated anti-CD4 (PharMingen) and 20 ␮g/ml of 7-aminoactinomycin D (7-AAD)4 (Sigma, St., Louis, MO) for 20 min at 4°C as previously described (25). For analysis of Bcl-2 expression, cells were stained with PE-anti-CD4 or -anti-CD8 mAbs, fixed in PBS/1% paraformaldehyde, and then incubated with FITC- conjugated anti-Bcl-2 mAb or FITC-conjugated isotype control (Phar- Mingen) in saponin buffer for 30 min on ice. Tissue culture FIGURE 1. Four to eight mice per group from five transgenic mouse Downloaded from Single-cell suspensions from spleen and lymph nodes were cultured in strains (all on a RagϪ/Ϫ background) were thymectomized (ATX) or sham- 96-well plates (Falcon, Becton Dickinson) in 200 ␮l of IMDM supple- Ϫ Ϫ thymectomized (control). The figure shows mean Ϯ SEM percentage of mented with 5% heat-inactivated FCS, 5 ϫ 10 5 M 2-ME, 2 ϫ 10 3 M CD4 (A1 and A18 strains) or CD8 T cells (F , A18CD4Ϫ/Ϫ, BM3 strains) L-glutamine, 100 U/ml penicillin, and 100 ␮g/ml streptomycin. Cells were 5 Ⅺ analyzed for apoptosis before and after 1, 2, and 3 days of culture. in blood 1 day before thymectomy (day 0; ) and 60 days after thymectomy (f). The p values, determined by paired t test, show significant dif- Carboxy-fluorescein-diacetate succinimidylester (CFSE) labeling ferences for A1 and A18 before and after thymectomy, whereas all other

and transfer groups are not significantly different. http://www.jimmunol.org/ For CFSE labeling, cells were resuspended in PBS at 107/ml and incubated with CFSE (Molecular Probes, Eugene, OR) at a final concentration of 2.5 ␮M for 10 min at 37°C, followed by two washes in serum-free IMDM cells from different strains show differences in their susceptibility medium. Labeled cells were i.v. injected into syngeneic, untreated RagϪ/Ϫ to apoptosis when removed from their stromal environment, we recipients. cultured spleen cells from the five TCR transgenic strains as well as from nontransgenic controls for 3–4 days in the absence of Ag Results or cytokines and determined apoptosis by staining with 7-AAD. Differential survival of CD4 and CD8 T cells after thymectomy Background levels of apoptosis were between 5 and 10% in T cells

A18 mice that express transgenic TCR specific for an epitope of from all strains at the start of culture. However, there was a strik- by guest on September 25, 2021 C5 presented in the context of H-2Ek were shown to gradually lose ing difference in the apoptosis rate of A18 CD4 T cells compared their peripheral T cells over a period of 6 wk following thymec- with polyclonal CD4 T cells. About 50% of the A18 CD4 T cells tomy, indicating a high turnover of peripheral CD4 T. To inves- showed signs of apoptosis already on day 1 of culture (Fig. 2A). tigate whether this was a general phenomenon in TCR transgenic The percentage of apoptotic cells in A1 CD4 T cells was similar to mice, four additional TCR transgenic strains were thymectomized that of polyclonal cells for the first 2 days of culture, but then their and compared with thymectomized normal mice. All TCR trans- rate of apoptosis exceeded that of polyclonal CD4 cells, similar to genic strains were bred onto a RagϪ/Ϫ background to avoid pitfalls the situation found for A18 T cells (Fig. 2B). In contrast, CD8 T due to the expression of additional unrelated TCR specificities. cells from F5 or BM3 TCR transgenic mice did not differ in their The A1 TCR transgenic strain selects CD4 T cells recognizing a rate of apoptosis from polyclonal CD8 T cells (Fig. 2, C and D). k peptide derived from H-Y presented by H-2E , the F5 TCR trans- Interestingly, CD8 T cells with the A18 receptor showed a similar genic strain selects CD8 T cells reactive to influenza nucleoprotein low increase in the rate of apoptosis over the 4-day culture period epitope 366–374 in the context of H-2Db, and the BM3 TCR trans- (Fig. 2E). Given that even polyclonal CD4 T cells undergo apo- genic strain selects CD8 T cells reactive to Kb. A18 mice when ptosis more readily than polyclonal CD8 T cells (Fig. 2F), it seems bred with mice that lack CD4 (CD4Ϫ/Ϫ) select the A18 TCR ex- that the differential susceptibility to apoptosis is an intrinsic prop- clusively into the CD8 lineage as observed previously in other erty of each lineage, although is not clear whether there is a con- CD4Ϫ/Ϫ TCR transgenic mice (26). Comparison of the percentage nection between apoptosis in vitro and survival in vivo. An obvi- of CD4 or CD8 T cells in the blood before thymectomy and 60 ous candidate molecule that could be involved in the differential days after thymectomy showed a similar trend for the decline in survival of CD4 and CD8 is Bcl-2 (28). However, we could not CD4 T cells in A1 and A18 mice, whereas the percentage of CD8 detect a significant difference in the levels of Bcl-2 expression Ϫ/Ϫ T cells in F5, BM3, and CD4 A18 mice remained largely un- between CD4 and CD8 subsets in transgenic A18 T cells or in affected by thymectomy (Fig. 1). CD44low polyclonal T cells (Fig. 2, G and H).

Susceptibility to apoptosis in vitro Homeostatic proliferation of naive CD4 and CD8 T cells Resting T cells have drastically reduced life spans when removed A number of recent reports described cell division (homeostatic from their stromal environment and cultured in vitro, and survival proliferation) of naive peripheral T cells when these are transferred under these conditions has been shown to be promoted by cyto- into syngeneic, supposedly Ag-free lymphopenic hosts, where they kines such as IL-4 or IL-7 (27). To evaluate whether peripheral T do not encounter competition for survival niches with other lymphocyte populations. Homeostatic proliferation depends on MHC/ 4 Abbreviations used in this paper: 7-AAD, 7-aminoactinomycin D; CFSE, carboxy- peptide recognition, but the nature of the peptides involved and ﬂuorescein-diacetate succinimidylester. their relation to peptides that effect positive selection in the thymus The Journal of Immunology 3691

are still unresolved (16, 29, 30). To compare homeostatic proliferation of different cell populations, spleen cells from the ﬁve TCR transgenic strains as well as from nontransgenic controls were labeled with the tracker dye CFSE (31) and injected into syngeneic RagϪ/Ϫ mice devoid of lymphocytes. Fig. 3 shows the CFSE patterns 9 or 15 days after transfer. While CD8 T cells (Fig. 3, right panels) have undergone several divisions by day 9 after transfer, CD4 T cells (left panels) divide much more slowly. A18 and A1 CD4 T cells show only one peak of division 9 days after transfer and two or three peaks after 15 days. In contrast, homeostatic proliferation of CD8 T cells with the A18 receptor is comparable with that of other CD8 T cells. In polyclonal T cells a fraction of both CD4 and CD8 T cells divided rapidly and lost CFSE label. In all likelihood these were preactivated T cells, as indicated by their high levels of the CD44 activation marker (data not shown; see also Fig. 4). The naive cells, in contrast, showed the same pattern of proliferation for CD4 and CD8 as that seen in transgenic T cells, emphasizing that there must be intrinsic differences in the way Downloaded from homeostatic survival signals are transmitted in CD4 and CD8 T cells.

Homeostatic proliferation does not generally affect the expression of activation markers http://www.jimmunol.org/ Because naive T cells can be distinguished from activated T cells by the expression of activation markers such as CD44 or CD45RB, we wanted to address the question of whether homeostatic proliferation observed after transfer results in phenotypic conversion. T cells from the five TCR transgenic strains used in this analysis did not show any changes in their expression levels of CD44 or FIGURE 2. Spleen and lymph node cells from five transgenic mouse CD45RB during homeostatic proliferation. The majority of naive strains, all on a RagϪ/Ϫ background (Ⅺ), as well as polyclonal T cells (f) polyclonal CD4 and CD8 T cells did not change the expression of were cultured four 3–4 days in medium without Ag (A–E). Polyclonal activation markers, although a small fraction of polyclonal CD8 T by guest on September 25, 2021 controls are from the A/J strain in A, B, E, and F; from CBA mice in D; and cells slightly increased CD44 expression after the fourth or fifth Ϯ Ϫ from B10 mice in C. The percentage of apoptotic cells SEM in the gated division (Fig. 4A). It should be noted that the CFSE , CD44high, CD4 or CD8 population was determined by staining with 7-AAD before CD45 RBlow population visible in recipients of transgenic T cells culture and after each day of culture. G, Expression of Bcl-2 on fresh A18 CD4 T cells (thin line) compared with A18 CD8 T cells (thick line). The is not donor derived and does not express TCR. In contrast, phe- stippled line represents the isotype control curve. H, Mean fluorescence notypic conversion was clearly evident after transfer of CFSE- Ϫ Ϫ values for Bcl-2 expression in fresh polyclonal CD4 or CD8 T cells (from labeled H-Y-specific A1 T cells into syngeneic male Rag / mice, CBA mice). The p values, determined by Student’s t test, were Ͻ0.005 for where they encounter their cognate Ag. This resulted in an increase days 1–3 in A, and Ͻ0.03 for days 3 and 4 in B. The p values in C–E in CD44 expression and a reduction of CD45RB expression al- indicated no statistical differences between cells from transgenic mice and ready evident on day 3 after transfer (Fig. 4B). In contrast, CD4 T control cells. The p values in F were Ͻ0.01 for days 1–3. cells from the AND TCR transgenic strain rapidly proliferated,

FIGURE 3. Homeostatic proliferation of CFSE-labeled lymph node cells from four transgenic mouse strains and polyclonal T cells 9 and 15 days after transfer into untreated syngeneic RagϪ/Ϫ recipients. The histograms show CFSE levels on gated T cell subpopulations. Left panels, CD4 T cell proliferation; right panels, CD8 T cell proliferation. 3692 T CELL SURVIVAL

FIGURE 4. A, The dot plots show staining for CFSE (y-axis) and either CD44 or CD45RB (x-axis) of gated CD4 and CD8 T cell populations 15 days after transfer (9 days after transfer for A18 CD8) into untreated syngeneic RagϪ/Ϫ recipients. B, CD44 and CD45 RB expression on A1 CD4 T cells 3 days after transfer into syngeneic male RagϪ/Ϫ mice expressing cognate Ag. C, CD44 and CD45 RB expression of AND (kxb) CD4 T cells 9 days after transfer into un- Downloaded from treated RagϪ/Ϫ (H-2b) recipients. The CFSE-negative populations (with the exception of those seen in RagϪ/Ϫ recipients of polyclonal T cells) are derived from the host and are not T cells; they are CD44high and negative for TCR expression. http://www.jimmunol.org/ by guest on September 25, 2021

up-regulated CD44 progressively with each division, and down- almost certainly T cells with endogenous or additional TCRs. regulated CD45RB despite the absence of a defined antigenic li- Proof of this assumption requires the generation of H-2Ek/C5 pep- gand in the adoptive host (Fig. 4C). tide tetramers. In this paper we investigated the survival of addi- It is possible that up-regulation of CD44 during proliferation in tional TCR transgenic strains, all on a RagϪ/Ϫ background so that empty hosts signifies T cell activation rather than mere homeo- they express single TCRs exclusively as well as polyclonal T cells. static proliferation. This could be a consequence of recognition of The results obtained highlight three main points: 1) A18 CD4 T bona fide Ag due to expression of additional TCRs in Ragϩ donors cells are not the only ones with survival problems, as we found a or because some T cells are activated under these conditions by similar phenotype in the A1 TCR transgenic strain; 2) CD4 T cells cross-reactive Ags present in the adoptive host (17, 32). This might are per se shorter lived than CD8 T cells whether they are from more easily happen for TCR with high avidity, which can interact transgenic or normal mice, and survival signals are relayed in a with a wide range of selection/survival ligands such as the AND different manner in CD8 T cells compared with CD4 T cells; and TCR (22, 26) 3) activation markers such as CD44 and CD45RB remain un- changed during homeostatic proliferation in lymphopenic hosts, Discussion distinguishing it from Ag-driven proliferation. The survival of naive peripheral T cells is a complicated process involving signals through the TCR and competition with other lymphocytes for environmental niches, ligands, or cytokines. Some T cells are less efficient survivors than others There is a hierarchy of thymic and peripheral selection in which The phenotype of A18 and A1 T cells is that of T cells with a some lymphocytes are more easily replaced by competitors than narrow specificity and/or low affinity TCR recognition. Although it others (33). This is most likely a reflection of the avidity of TCR is still a matter of debate whether ligands involved in thymic se- interactions with a varying choice of selecting and/or survival li- lection are the same as those that procure peripheral survival sig- gands. The A18 TCR transgenic strain on a RagϪ/Ϫ background nals (16, 29, 30), selection in the thymus sets the threshold for the develops few peripheral CD4 T cells, which are relatively short susceptibility to survival signals (34). T cells from the highly se- lived compared with normal CD4 T cells. In contrast, CD4 T cells lecting AND strain show extensive expansion upon transfer into in A18 Ragϩ mice are present in similar numbers as in nontrans- lymphopenic hosts, although susceptibility to apoptosis in vitro is genic controls and show much more robust survival patterns (data similar to that of A18 and A1 T cells (data not shown). The fact not shown). In this situation, however, the better survivors are that proliferation of AND T cells in vivo is accompanied by the The Journal of Immunology 3693 acquisition of activation markers similar to those seen in Ag- similar phenotype seen in recipients of RagϪ/Ϫ transgenic T cells driven expansion indicates that the AND TCR is more promiscu- is not derived from the donor population. It does not express TCR ous, with a lower threshold for survival signals, some of which and is probably identical with a CD4ϩ CD3Ϫ cell population found might represent cross-reactive Ags. A similar phenotype in CD4 T in developing lymph nodes; these were shown to differentiate to cells with hemagglutinin-specific ABII TCR that incorporated bro- APC, NK, and follicular cells, but not to T and B cells (41). Given modeoxyuridine and up-regulated CD44 following thymectomy or that high expression of CD44 is a stable marker that distinguishes transfer into lymphopenic hosts was attributed to recognition of an naive and memory T cells and does not revert (in contrast to unidentified ligand (13). CD45RB expression) (42), it seems unlikely that proliferation of naive T cells in response to mere survival signals generally results CD8 T cells live better than CD4 in phenotypic conversion. Alternative explanations for the discrep- Regardless of the spectrum of survival patterns seen in different ancies described in different systems could be 1) the use of Ragϩ TCR transgenic strains, it was obvious that naive CD8 T cells mice in which bona fide activation of dual TCR-expressing cells or survive better than naive CD4 T cells. This was most clearly il- cells with other endogenous TCR by environmental Ag cannot be lustrated in the different survival characteristics of CD8 vs CD4 T excluded, 2) high affinity TCRs cross-reactive to environmental cells bearing the same A18 TCR. Despite the fact that A18 CD8 T self Ags or hyper-reactive to survival ligands, and possibly 3) the cells are less responsive to cognate Ag than A18 CD4 (our un- use of irradiated hosts. published observation), they do not decay after thymectomy, are Irradiation results in gastrointestinal damage, leading to the re- less fragile in vitro, and undergo more extensive homeostatic pro- lease of LPS (30, 43). This could liberate cross-reactive Ags and liferation. It is unclear at present whether this reflects more effi- result in mobilization of Ag-bearing dendritic cells to lymphoid Downloaded from cient transmission and subsequent signaling upon interaction with tissues, thus promoting T cell activation (44). T cells with high survival ligands or whether CD8 T cells are more responsive to the avidity TCR promiscuously responsive to many ligands such as presence of distinct survival mediators, e.g., cytokines (35, 36). An AND T cells might be genuinely activated by cross-reactive Ags obvious survival factor, such as the anti-apoptotic molecule Bcl-2, when they undergo phenotypic conversion, rather than proliferat- showed similar expression in naive CD4 and CD8 T cells of either ing in response to baseline survival signals. As such they would

transgenic or normal mice, but it is possible that other members of enjoy preferential survival at the expense of T cells with more http://www.jimmunol.org/ the Bcl-2 family, such as Bcl-xL, may be differentially expressed in restricted TCR recognition, for whom survival signals may be lim- the two cell types. iting. The AND and A18 TCRs may represent opposite ends of the CD4 and CD8 T cells appear to occupy overlapping niches in spectrum with respect to their responses to TCR-mediated signals, the periphery, because cotransfer of large numbers of either subset but, in general, homeostatic expansion of naive T cells in condi- can suppress the expansion of both (15, 16). The ratio of CD4 to tions that eliminate competition should be viewed as distinct from CD8 T cells in normal mice is controlled homeostatically, but less Ag-driven expansion followed by phenotypic conversion. is known about its conservation in the absence of thymic output. Recent thymic emigrants contain about 4 times more CD4 than Acknowledgements by guest on September 25, 2021 CD8 T cells, and normalization of the ratio in secondary lymphoid We thank Dimitris Kioussis for critical comments on the manuscript and organs implies that some CD4 T cells are deleted, whereas, in Trisha Norton and Keith Williams for keeping our mice under control. contrast, some CD8 T cells are proliferating (37). The question of how homeostasis is achieved after profound T cell depletion and in References the absence of thymic output has important consequences for ther- 1. Tanchot, C., M. M. Rosado, F. Agenes, A. A. Freitas, and B. Rocha. 1997. apy in humans. It is known that peripheral expansion leads to a Lymphocyte homeostasis. Semin. Immunol. 9:331. 2. Tanchot, C., and B. Rocha. 1995. The peripheral T cell repertoire: independent substantially skewed repertoire dominated by cells with memory/ ϩ homeostatic regulation of virgin and activated CD8 T cell pools. Eur. J. Im- effector phenotypes at the expense of naive cells in addition to low munol. 25:2127. CD4 T cell numbers (38). The observation that the CD4/CD8 ratio 3. Goldrath, A. W., and M. J. Bevan. 1999. Selecting and maintaining a diverse drops in mice as a function of age due to the disappearance of CD4 T-cell repertoire. Nature 402:255. 4. Brocker, T. 1997. Survival of mature CD4 T lymphocytes is dependent on major T cells rather than to the expansion of CD8 T cells (39) is in histocompatibility complex class II-expressing dendritic cells. J. Exp. Med. 186: agreement with their differential survival characteristics, which un- 1223. 5. Kirberg, J., A. Berns, and H. von Boehmer. 1997. Peripheral T cell survival der physiological conditions may only become apparent once thy- requires continual ligation of the T cell receptor to major histocompatibility com- mic output ceases or becomes negligible. plex-encoded molecules. J. Exp. Med. 186:1269. 6. Rooke, R., C. Waltzinger, C. Benoist, and D. Mathis. 1997. Targeted comple- Survival signals do not generally affect activation markers mentation of MHC class II deficiency by intrathymic delivery of recombinant adenoviruses. Immunity 7:123. There are conflicting reports in the literature as to whether the 7. Takeda, S., H. R. Rodewald, H. Arakawa, H. Bluethmann, and T. Shimizu. 1996. ϩ often vigorous proliferation observed after transferring naive T MHC class II molecules are not required for survival of newly generated CD4 T cells, but affect their long-term life span. Immunity 5:217. cells into supposedly Ag-free lymphopenic hosts affects the ex- 8. Swain, S. L., H. Hu, and G. Huston. 1999. Class II-independent generation of pression of activation markers (2, 13, 16, 17, 30, 40). Some studies CD4 memory T cells from effectors. Science 286:1381. even described spontaneous development of effector functions 9. Murali-Krishna, K., L. Lau, S. Sambhara, F. Lemonnier, J. Altman, and R. Ahmed. 1999. Persistence of memory CD8 T cells in MHC class I-deficient upon transfer of transgenic CD8 T cells into irradiated recipients mice. Science 286:1377. (32). In our experiments most polyclonal T cells with a naive phe- 10. Boursalian, T. E., and K. Bottomly. 1999. Survival of naive CD4 T cells: roles of restricting versus selecting MHC class II and cytokine milieu. J. Immunol. 162: notype did not change their expression of CD44 and CD45RB 3795. Ϫ/Ϫ when transferred into syngeneic Rag mice, although they pro- 11. Barthlott, T., R. J. Wright, and B. Stockinger. 1998. Normal thymic selection of liferated vigorously. In polyclonal CD8 T cells there seems to be TCR transgenic CD4 T cells, but impaired survival in the periphery despite the presence of selecting MHC molecules. J. Immunol. 161:3992. a slight increase in CD44 expression after the fourth or fifth round 12. Tough, D. F., and J. Sprent. 1994. Turnover of naive- and memory-phenotype T of division, but it does not affect all cells. The CD44-positive, cells. J. Exp. Med. 179:1127. CFSE-negative population evident in recipients of polyclonal cells 13. Bruno, L., H. von Boehmer, and J. Kirberg. 1996. Cell division in the compart- ment of naive and memory T lymphocytes. Eur. J. Immunol. 26:3179. represents cells with an activated/memory phenotype already 14. Bell, E. B., and S. Sparshott. 1997. The peripheral T cell pool: regulation by present in the initial inoculum. In contrast, the population with a non-antigen induce proliferation? Semin. Immunol. 9:347. 3694 T CELL SURVIVAL

15. Rocha, B., N. Dautigny, and P. Pereira. 1989. Peripheral T lymphocytes: expan- 29. Viret, C., F. S. Wong, and C. S. Janeway. 1999. Designing and maintaining the sion potential and homeostatic regulation of pool sizes and CD4/CD8 ratios in mature TCR repertoire: the continuum of self-peptide:self-MHC recognition. Im- vivo. Eur. J. Immunol. 19:905. munity 10:559. 16. Ernst, B., D. Lee, J. M. Chang, J. Sprent, and C. D. Surh. 1999. The peptide 30. Bender, J., T. Mitchell, J. Kappler, and P. Marrack. 1999. CD4ϩ T cell division ligands mediating positive selection in the thymus control T cell survival and in irradiated mice requires peptides distinct from those responsible for thymic homeostatic proliferation in the periphery. Immunity 11:173. selection. J. Exp. Med. 190:367. 17. Kieper, W. C., and S. C. Jameson. 1999. Homeostatic expansion and phenotypic 31. Lyons, A. B., and C. R. Parish. 1994. Determination of lymphocyte division by conversion of naive T cells in response to self peptide/MHC ligands. Proc. Natl. flow cytometry. J. Immunol. Methods 171:131. Acad. Sci. USA 96:13306. 32. Oehen, S., and K. Brduscha-Riem. 1999. Naive cytotoxic T lymphocytes spon- 18. Rocha, B., and H. von Boehmer. 1991. Peripheral selection of the T cell reper- taneously acquire effector function in lymphcytopenic recipients: a pitfall for T toire. Science 251:1225. cell memory studies? Eur. J. Immunol. 29:608. 19. Sprent, J., M. Schaefer, M. Hurd, C. D. Surh, and Y. Ron. 1991. Mature murine 33. Freitas, A. A., F. Agenes, and G. C. Coutinho. 1996. Cellular competition mod- B and T cells transferred to SCID mice can survive indefinitely and many main- ulates survival and selection of CD8ϩ T cells. Eur. J. Immunol. 26:2640. tain a virgin phenotype. J. Exp. Med. 174:717. 34. Janeway, C. A. J., A. Kupfer, C. Viret, T. Boursalian, J. Goverman, K. Bottomly, 20. Zal, T., A. Volkmann, and B. Stockinger. 1994. Mechanisms of tolerance induc- and D. Sant’Angelo. 1998. T cell development, survival, and signalling: a new tion in major histocompatibility complex class II-restricted T cells specific for a concept of the role of self-peptide:self-MHC complexes. Immunologist 6:5. blood-borne self-antigen. J. Exp. Med. 180:2089. 35. Lodolce, J. P., D. L. Boone, S. Chai, R. E. Swain, T. Dassopoulos, S. Trettin, and 21. Zelenika, D., E. Adams, A. Mellor, E. Simpson, P. Chandler, B. Stockinger, A. Ma. 1998. IL-15 receptor maintains lymphoid homeostasis by supporting lym- H. Waldmann, and S. P. Cobbold. 1998. Rejection of H-Y disparate skin grafts phocyte homing and proliferation. Immunity 9:669. by monospecific CD4ϩ T helper 1(Th1) and T helper 2 (Th2) cells: no require- 36. Nakajima, H., E. W. Shores, M. Noguchi, and W. J. Leonard. 1997. The common ment for CD8ϩ T cells or B cells. J. Immunol. 161:1868. cytokine receptor ␥ chain plays an essential role in regulating lymphoid ho- 22. Kaye, J., M.-L. Hsu, M.-E. Sauron, S. C. Jameson, N. R. J. Gascoigne, and meostasis. J. Exp. Med. 185:189. S. M. Hedrick. 1989. Selective development of CD4ϩ T cells in transgenic mice 37. Berzins, S. P., R. L. Boyd, and J. F. A. P. Miller. 1998. The role of the thymus expressing a class II MHC-restricted antigen receptor. Nature 341:746. and recent thymic migrants in the maintenance of the adult peripheral lymphocyte 23. Mamalaki, C., T. Elliott, T. Norton, N. Yannoutsos, A. R. Townsend,

pool. J. Exp. Med. 187:1839. Downloaded from P. Chandler, E. Simpson, and D. Kioussis. 1993. Positive and negative selection in transgenic mice expressing a T-cell receptor specific for influenza nucleopro- 38. Mackall, C. L., F. T. Hakim, and R. E. Gress. 1997. Restoration of T-cell ho- tein and endogenous superantigen. Dev. Immunol. 3:159. meostasis after T-cell depletion. Semin. Immunol. 9:339. 24. Spoonas, A.-M., P. D. Tomlinson, J. Antoniou, N. Auphan, C. Langlet, 39. Callahan, J. E., J. W. Kappler, and P. Marrack. 1993. Unexpected expansions of B. Malissen, A.-M. Schmitt-Verhulst, and A. L. Mellor. 1994. Induction of tol- CD8-bearing cells in old mice. J. Immunol. 151:6657. 40. Goldrath, A. W., and M. J. Bevan. 1999. Low affinity ligands for the TCR drive erance to self MHC class I molecules expressed under the control of milk protein ϩ or ␤-globin gene promoters. Int. Immunol. 6:277. proliferation of mature CD8 T cells in lymphopenic hosts. Immunity 11:183. 25. Lecoeur, H., H. Ledru, E. Prevost, and M.-C. Gougeon. 1997. Strategies for 41. Mebius, R. E., P. Rennert, and I. L. Weissman. 1997. Developing lymph nodes collect CD4ϩ CD3Ϫ LT␤ϩ cells that can differentiate to APC, NK cells, and

phenotyping apoptotic peripheral human lymphocytes comparing ISNT, an- http://www.jimmunol.org/ nexin-V and 7AAD cytofluorometric stainings methods. J. Immunol. Methods follicular cells but not T or B cells. Immunity 7:493. 209:111. 42. Garcia, S., J. DiSanto, and B. Stockinger. 1999. Following the development of a 26. Matechak, E. O., N. Killeen, S. M. Hedrick, and B. J. Fowlkes. 1996. MHC class CD4 T cell response in vivo: from activation to memory formation. Immunity II-specific T cells can develop in the CD8 lineage when CD4 is absent. Immunity 11:163. 4:337. 43. Hill, G. R., J. M. Crawford, K. R. Cooke, Y. S. Brinson, Y. S. Pan, and 27. Vella, A., T. K. Teague, J. Ihle, J. Kappler, and P. Marrack. 1997. Interleukin 4 J. L. M. Ferrara. 1997. Total body irradiation and acute graft-versus-host disease: (IL-4) or IL-7 prevents the death of resting T cells: stat6 is probably not required the role of gastrointestinal damage and inflammatory cytokines. Blood 90:3204. for the effect of IL-4. J. Exp. Med. 186:325. 44. Reis e Sousa, C., and R. N. Germain. 1999. Analysis of adjuvant function by 28. Veis, D. J., C. M. Sorenson, J. R. Shutter, and S. J. Korsmeyer. 1993. Bcl-2 direct visualization of antigen presentation in vivo: endotoxin promotes accumu- deficient mice demonstrate fulminant lymphoid apoptosis, polycystic kidneys, lation of antigen-bearing dendritic cells in the T cell areas of lymphoid tissue. and hypopigmented hair. Cell 75:229. J. Immunol. 162:6552. by guest on September 25, 2021