Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Immunol. Author manuscript; available in PMC 2011 May 15.
Published in final edited form as:
PMCID: PMC3016941

Self-specific CD8+ T cells maintain a semi-naïve state following lymphopenia-induced proliferation


Upon transfer into T-cell deficient hosts, naïve CD8+ T cells typically undergo “lymphopenia-induced proliferation” (LIP, also called “homeostatic proliferation”), and develop the phenotypic and functional characteristics of memory CD8+ T cells. However, the capacity of T cells with self-peptide/MHC specificity to respond in this way has not been intensively studied. We examined pmel-1 TCR transgenic CD8+ T cells which are specific for an epitope from gp100, a protein expressed by melanoma cells and normal melanocytes. Despite their self-specificity, naïve pmel-1 cells were inefficient at LIP in typical lymphopenic hosts. In CD132 (common-γ chain) deficient hosts, pmel-1 CD8+ T cells underwent extensive proliferation but, surprisingly, the majority of these cells retained certain naïve phenotypic traits (CD44low, CD122low), rather than acquiring the expected central-memory phenotype. Following LIP, pmel-1 T cells acquired the capacity to control B16F10 tumor growth, but only in common-γ chain deficient host mice. Together, these data suggest that LIP does not always favor expansion of self-specific CD8 T cells, and that sustained extensive lymphopenia is required for such cells to exhibit tumor control.


In response to T-cell lymphopenia, most naïve T cells undergo a slow proliferative response termed homeostatic proliferation or lymphopenia-induced proliferation (LIP). In addition to expansion, LIP induces naïve T cells to differentiate into memory-like cells, which display surface markers and functional properties similar to antigen primed central memory (TCM), despite the fact that LIP memory cells have not encountered foreign antigen (1, 2). However, not all T cells undergo LIP, and it has been proposed that TCR affinity for self peptide/MHC ligands dictates whether, and to what extent, a T cell will respond to lymphopenic “space” (1, 35). For example, use of TCR transgenic CD8 T cells reveals that cells bearing the OT-I, 2C and P14 TCR are efficient at LIP, while cells bearing the H-Y TCR are unable to undergo LIP at all in typical lymphopenic hosts (3, 6), but can be driven into LIP in the extreme lymphopenia of γC (CD132) deficient hosts (7). Based on the fact that the H-Y TCR induces inefficient thymic positive selection and naïve T cell survival, it has been proposed that the failure of these cells to undergo LIP reflects scarcity or poor engagement with suitable self peptide/MHC ligands. However, this has not been directly tested using T cells with known self-specificity.

CD8 T cells expressing the pmel-1 TCR transgene have been extensively studied for their capacity to control syngeneic melanoma in B6 mice (810). This TCR is specific for an epitope in gp100, a protein produced by melanocytes but overexpressed in melanoma (8). Adoptive cell immunotherapy using activated pmel-1 CD8 T cells allows control of established melanoma, but this requires induction of extensive lymphopenia in the host animal (11). Efficient activation of pmel-1 also leads to overt self-reactivity, as revealed by induction of normal melanocyte death (vitiligo). As pmel-1 has a self-specific TCR, it was possible that these cells would exhibit enhanced or altered homeostatic characteristics in lymphopenic hosts. However, the capacity of naïve pmel-1 cells to undergo LIP in lymphopenic hosts has not been carefully studied.

We found that, despite their intrinsic self-reactivity, naïve pmel-1 CD8 T cells undergo very inefficient LIP in conventional lymphopenic models. Interestingly, this correlates with the phenotype of pmel-1 cells, which resembles H-Y TCR transgenic cells. Extensive pmel-1 LIP could be induced in highly lymphopenic CD132-deficient hosts but, surprisingly, these cells did not acquire a typical memory phenotype and upregulate CD44 and CD122. Furthermore, although LIP promoted the capacity of pmel-1 to control B16F10 melanoma, this response required sustained extreme lymphopenia in the host. These unusual responses of pmel-1 leads to the surprising conclusion that at least some self-specific CD8 T cells are unfavored for expansion in lymphopenic conditions. We also conclude that differentiation toward a memory phenotype is not an automatic consequence of extensive naïve T cell proliferation, nor is it indicative of a functional response.

Materials and Methods


C57BL/6 were purchase from the NCI. Pmel-1 TCR Tg were purchased from Jackson Laboratory (Bar Harbor, Maine) and crossed onto a Rag 1−/− Thy1.1 background. Common-γ chain deficient, Rag−/− deficient mice were purchased from Taconic (Germantown, NJ) and were maintained or crossed onto a wild-type background. HY.Rag−/− were also purchased from Taconic. OT-I Rag−/− mice are maintained in our colony (12). Mice were bred and maintained under specific pathogen-free conditions at the University of Minnesota (Minneapolis, MN). Experiments were conducted with approval by the University of Minnesota Institutional Animal Care and Usage Committee.


The melanoma B16F10 were maintained at 37°C with 5.5% CO2 in DMEM supplemented with 10% FCS, non-essential amino acids, penicillin, streptomycin, gentamicin sulfate, 2-mercaptoethanol, L-glutamine, and HEPES. Aliquots were thawed and grown for one week prior to injecting into mice. Cells were passed one day prior to injection. Tumors were trypsinized and washed once in HBSS and once in PBS to remove FBS. Cells were suspended at 2×106 cells/ml. Mice were anesthetized prior to receiving 100μl of B16F10 suspension subcutaneously in the flank. Tumor size was assessed by measuring the greatest length and width to determine area. Animals were euthanized when the tumor reached > 400 mm2, the tumor was necrotic, or the animals appeared moribund.

Isolation of Tumor Infiltrating Lymphoctyes

Subcutaneous tumors were removed and digested with 2mg/ml Collagenase D (Roche, Mannheim, Germany) for 45 minutes at 37°C. Digested tissue was then passed through 35μM filters. TIL were isolated by density centrifugation at room temperature using 80% and 40% Percoll (GE Healthcare, Piscataway, NJ). The interface was isolated and washed twice with PBS prior to staining and in vitro stimulation.

Adoptive Transfer of T cells

CD44low OT-I or pmel-1 were purified by negative selection using MACS columns (Miltenyi Biotec, Bergisch Gladbach, Germany) as previously described (13). Cells were labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE) as described (14). 1×106 cells injected either via tail vein or retro-orbital injection. For adoptive transfer of cells that had already undergone HP, spleens were harvested and the percentages of donor cells were determined by congenic marker (pmel-1) or Kb-OVA tetramer staining (OT-I). Cells were then CFSE labelled and a sufficient number was transferred so that 1×106 donor cells from the original host were injected.

Recombinant cytokine Treatment

Murine IL-15 or IL-15 complexed to its receptor was given one day after adoptive transfer of pmel-1. Recombinant mouse IL-15Rα-Fc chimeric molecule (R&D Systems, Minneapolis, MN) was complexed with IL-15 by suspending both in PBS and incubating for 30 minutes at 37°C. Each mouse was injected with 2.5μg of IL-15, or the same amount of IL-15 complexed with 15μg of IL-15Rα-Fc. In other experiments, mice were treated with recombinant IL-12 (1μg in a total volume of 200μl PBS, administered via tail vein injection) (R&D Systems) on days 1, 2 and 3 after adoptive transfer of pmel-1 into sub-lethally irradiated hosts.

Viral Infection

Vaccinia virus encoding the human epitope of gp100 (VV-hgp100) was a kind gift of Dr. Restifo (NCI, Bethesda, MD). Mice were infected with 2×106 PFU i.p one day after transfer of 4×104 naïve pmel-1 into wild-type hosts. Mice were bled at day 5 to check for expansion of pmel-1. Spleens were harvested >2 months after infection and CD90.1 was used to determine VV-hgp100 memory CD8+ T cells.

Vaccination of Tumor Bearing Mice

Mice that had received subcutaneous B16F10 were given low dose irradiation (420 cGy) one day prior to receiving 1×106 pmel-1 that had been isolated from CD132° host spleens. Twenty-four hours after adoptive transfer, some mice received 1μg hgp100 (KVPRNQDWL) and 53.8μg of αCD40 (FGK4.5, BioXcell, West Lebanon, NH) in a total volume of 200μl PBS via retro-orbital vein.

Flow Cytometric Analysis

FACS analysis was performed on a LSRII with BD FACS Diva software (BD Biosciences, San Jose, CA). Data analysis was performed using FlowJo software (Treestar, Ashland, OR). For in vitro stimulation assays, cells were stimulated in cDMEM/10%FBS with or without peptide for 6 hours in the presence of GolgiStop (BD Biosciences) in 96-well round bottom plates. For pmel-1 stimulation 1μg/ml of mgp100 peptide (EGSRNQDWL) was used. OVAp (SIINFEKL) was used at a final concentration of 250nM for OT-I. Cells were then washed and surface stained prior to fixation and permeablilization and staining for IFN-γ and TNF-α. For surface detection of LAMP, CD107a and CD107b conjugated to FITC (BD Biosciences) were included during peptide stimulation.


When applicable, the data is expressed as the mean ± SD. Unpaired, two-tailed Student's t- test was performed to determine significance using Prism Version 4.0a software (GraphPad Software, La Jolla, CA). Differences were considered statistically significant when p-values were <0.05. Statistical significance is indicated as follows: * p< 0.05, ** p<0.01, *** p<0.001.


Pmel-1 undergo modest LIP in irradiated wild-type hosts

Given that pmel-1 CD8+ T cells express a self-specific TCR, it might be expected to promote expansion in a lymphopenic host (a process driven by recognition of self peptide/MHC (15). On the other hand, published studies suggested pmel-1 T cells failed to proliferate when transferred into Rag−/− recipients (16), yet this was measured at an early time point (4 days) and assessed in tumor bearing hosts, leaving the question of conventional LIP unresolved. In order to directly test the ability of naïve pmel-1 to undergo LIP, we adoptively transferred CD44low Rag−/− pmel-1 CD8 T cells into sublethally irradiated C57BL/6 hosts, and allowed them to homeostatically proliferate for either10 days for 6 weeks (Fig. 1a, top panels). Limited proliferation, as measured by CFSE dye dilution, was observed at 10 days and no further cell division was observed by 6 weeks. This was unexpected given the rapid LIP exhibited by most TCR transgenic and polyclonal CD8 T cells (17). Furthermore, we observed minimal upregulation of CD44 by pmel-1 cells exposed to this lymphopenic environment (Fig 1a), suggesting ineffective conversion of naïve to “homeostatic” memory cells. We and others have reported that cytokines and cytokine complexes, such as IL-12, IL-15 and IL-15/IL-15Rα can augment LIP of naïve CD8 T cells (13, 18, 19). However, we observed minimal enhancement of pmel-1 cell proliferation or CD44 upregulation by these treatments (Fig 1a,b). Similar poor proliferation of naïve Rag−/− pmel-1 CD8 T cells was observed in Rag−/− deficient hosts (data not shown), suggesting that these findings were not exclusive to the sub-lethal irradiation model of lymphopenia. These studies used Rag−/− pmel-1 T cells: analysis of CD8 T cells from Rag+ pmel-1 mice revealed slightly enhanced LIP (Supplementary Fig 1a), but the naïve population from these animals includes T cells which fail to bind mgp100/Db multimers (Supplementary Fig 1b) and hence may express TCRs derived from endogenous rearrangements. For this reason, further experiments utilized Rag−/− pmel-1 CD8 T cells.

Figure 1
pmel-1 undergo limited homeostatic proliferation in irradiated wild-type hosts

Pmel-1 Express High Levels of CD8α and Low Levels of CD5 correlating with poor LIP Potential

Previous studies have suggested that the capacity of TCR transgenic CD8 T cells to undergo LIP may correlate with their expression levels of CD5, CD8 and/or CD127 (IL-7Rα) (5, 20). These parameters were analyzed for naïve pmel-1 CD8 T cells, in comparison with OT-I and female H-Y TCR transgenic T cells, which exhibit strong and weak LIP respectively (3, 21, 22). Polyclonal B6 naïve CD8 T cells were included as a control. We found an inverse correlation between expression levels of CD5 and CD8α (Fig 2a) and expression levels of CD8α and CD127 (Fig 2b) in these populations. Interestingly, these patterns of cell surface expression (especially high CD5 and low CD8α expression) correlate with the ability of the naïve T cell populations to undergo LIP (Fig 2a), as determined from our studies (Fig 1) and published reports (3, 21, 22). Whether and how these expression characteristics dictate the LIP potential of naïve CD8 T cells is not yet clear, but these data suggested the poor proliferative capacity of pmel-1 could have been predicted by its clustering with H-Y female TCR transgenic T cells using these markers. It will be interesting to see whether CD5, CD8α and IL-7Ra expression levels correlate with LIP in additional TCR transgenic systems.

Figure 2
Similar expression of CD5, CD8α and CD127 on naïve pmel-1 T cells and female H-Y TCR transgenic T cells

Pmel-1 acquire a unique semi-naïve phenotype after rapid proliferation in Common-γ chain deficient Hosts

Although female H-Y TCR transgenic CD8 T cells also show poor LIP in typical lymphopenic hosts (3, 21), a recent report showed that, when transferred into CD132 (“Common-γ chain”)-deficient hosts, naïve H-Y TCR transgenic CD8 T cells undergo extensive LIP, including upregulation of memory markers (7). CD132° animals are massively lymphopenic, lacking B, NK, and CD8 cells, although they retain a population of host CD4 T cells (23). Given the similarities we observed in the LIP potential and phenotype of naïve H-Y and pmel-1 CD8 T cells, we next tested whether pmel-1 LIP was enhanced in CD132° hosts. Indeed, in CD132° recipients, naïve pmel-1 underwent extensive cell division, as evidenced by CFSE dye dilution (Fig 3a). Furthermore, the proliferation and accumulation of pmel-1 in these hosts was similar to that of OT-I T cells (Fig 3a,b). Interestingly, while OT-I cells underwent LIP in both CD132° and CD132°/Rag° hosts, pmel-1 proliferation was minimal in the CD132°/Rag° recipients (Supplementary figure 2), suggesting a role for residual host lymphocytes in this model.

Figure 3
Pmel-1 rapidly expand in Common-γ chain Deficient (CD132°) Hosts

Surprisingly, while LIP of OT-I T cells in CD132° hosts was accompanied by the expected upregulation of the memory marker CD44 and retention of CD62L, the majority of proliferating pmel-1 retained a CD44low phenotype (Fig 3a,c), and a significant proportion of these cells were low for CD62L (Fig 3c). This phenotype was not correlated simply with the extent of cell division, since CFSE dye dilution was extensive regardless of pmel-1 phenotype (Fig 3d).

The unusual CD44low phenotype of the LIP expanded pmel-1 pool could simply reflect an inability of pmel-1 cells to upregulate CD44 after any stimulation. However, previous studies have shown CD44 upregulation on antigen activated pmel-1 (24), and we confirmed this result by priming pmel-1 in vivo with vaccinia virus expressing the stimulatory hgp100 antigen (Fig 4a). It was also possible that the CFSElow CD44low pmel-1 pool was derived from co-transferred stem cells, which would have developed in the thymus and been exported as naïve phenotype, dye-diluted cells. This is unlikely, since we assessed LIP at 7-days, before a cohort of precursors could have been exported, and we failed to observe CD8 SP in the thymii of the host mice (data not shown).

Figure 4
Pmel-1 bear some markers of memory-phenotype cells after LIP in Common-γ chain Deficient Hosts

Two previous studies have reported a population of mature CD8 T cells that, despite extensive proliferation, failed to upregulate CD44 (25, 26). In both cases, the stem-cell antigen-1 (sca-1) was used to differentiate proliferated CD44low T cells from naïve (25, 26). Sca-1 has been used as a useful marker of virus-specific memory cells, although there is no attributed function on CD8 T cells (27). Intriguingly, we found that the pmel-1 population which had expanded in CD132° hosts had indeed upregulated Sca-1, to similar levels as antigen primed memory pmel-1 (Fig 4a).

It has been previously demonstrated that LIP induces cells which are able to function as memory cells, with the characteristic ability to produce IFN-γ upon in vitro stimulation with antigen (2). We confirmed this finding with OT-I T cells which had undergone LIP in CD132° hosts (Fig 4b). Interestingly, despite their lack of CD44 upregulation, pmel-1 cells which had expanded in CD132° hosts attained a similar capacity to make IFN-γ. Naïve pmel-1 and OT-I produced TNF-α but not IFN-γ upon cognate antigen stimulation (Fig 4c), as expected. Taken together, these data suggest that, despite their CD44low phenotype, pmel-1 cells which undergo LIP in CD132° hosts express memory-like functions.

Memory phenotype CD8 T cells (whether generated by antigen-driven responses or LIP) typically upregulate IL-2Rβ (CD122) and show heightened responsiveness and dependence on IL-15 for survival (1, 2, 28). Such upregulation was clearly seen on OT-I T cells undergoing LIP in CD132° hosts, but in contrast, LIP of pmel-1 lead to modest elevation of CD122 expression (Fig. 5a). Indeed, the CD122 levels expressed by LIP pmel-1 cells were lower than that of polyclonal CD8 memory-phenotype cells, and more closely resembled those on memory-phenotype CD4 T cells (Fig 5b). This is potentially significant for understanding the poor LIP of pmel-1 cells in conventional lymphopenic hosts since it has been shown that IL-15 driven LIP of CD4 memory cells is inefficient when CD8+ and NK cells are present, likely to due to competition for IL-15 (29). These low levels of CD122 on both naïve and LIP pmel-1 cells may also explain the poor response of pmel-1 cells to IL-15/Rα complex treatment (Fig 2b).

Figure 5
Pmel-1 do not upregulate CD122 (IL-2Rβ) extensively in CD132° Hosts

Pmel-1 Require Sustained Lymphopenia for Homeostatic Proliferation

Typical memory phenotype CD8 cells are able to undergo rapid proliferation in lymphopenic hosts (1, 2, 30). However, LIP of pmel-1 cells induced some but not all memory phenotype features, hence it was unclear whether expansion of these cells in a CD132° host imbued them with the capacity to expand in more conventional lymphopenic environments. To test this we recovered pmel-1 or OT-I cells which had undergone LIP for 7 days in a CD132° host, labelled the cells with CFSE and the adoptively transferred the populations into CD132° or sub-lethally irradiated B6 hosts. After one week, CFSE dilution was evaluated. While LIP OT-I cells underwent extensive expansion in both adoptive hosts (Fig 6a,b), LIP pmel-1 cells proliferated only in the secondary CD132° hosts, failing to proliferate (Fig. 6a) or accumulate (Fig 6b) in the irradiated B6 host. Furthermore, the continued proliferation of pmel-1 LIP cells in secondary CD132° hosts failed to change their CD44lo phenotype (Fig 6a, right panel). These results indicate that even after LIP in an CD132° environment, pmel-1 cells behave like their naïve counterparts in their inability to expand in a conventional lymphopenic environment, further suggesting that these cells fail to compete efficiently for homeostatic cues.

Figure 6
Pmel-1 require the unique environment of common- γ chain deficient hosts for sustained homeostatic proliferation

Sustained LIP of pmel-1 allows for control of Established B16F10 Tumors

Our data suggested that pmel-1 cells could undergo LIP in certain lymphopenic hosts, but that these cells retained some features of naïve CD8 T cells. This raised the question of whether LIP pmel-1 cells were capable of controlling tumor growth. Initial experiments tested control of B16F10 tumor cells injected into animals which had already received pmel-1 cells. Transfer of naïve pmel-1 cells into CD132° hosts prevented B16F10 growth (Supplementary Fig. 3a), but irradiated B6 mice given pmel-1 cells (either naïve or LIP pmel-1 recovered from primary CD132° hosts) failed to control tumor growth at all (Supplementary Fig. 3b). These data suggested LIP of naïve pmel-1 was sufficient to eliminate tumors, but these experiments test the capacity of pmel-1 to control melanoma in a prophylactic rather than in a therapeutic assay. To assess whether LIP pmel-1 cells were capable of controlling established tumors, we injected B16F10 cells into B6 or CD132° animals and allowed the tumor to grow for 6 days. At this point, LIP pmel-1 were harvested after expansion in primary CD132° hosts, and transferred into the tumor bearing hosts. To be consistent between the groups, all recipient mice were given a sublethal dose of irradiation prior to pmel-1 transfer. In addition, some mice received hgp100 peptide and αCD40 one day after T cell transfer, to yield antigen primed pmel-1 cells. LIP pmel-1 were ineffective at controlling tumor growth when transferred into irradiated B6 mice, and this situation was only modestly enhanced by peptide vaccination (Fig 7a). These data parallel reports by others using in vitro activated pmel-1 CTL rather than LIP pmel-1 cells (810). However, LIP pmel-1 transferred into CD132° provided effective tumor control, whether or not the hosts were vaccinated (Fig 7a). Interestingly, localized vitiligo was observed around the tumor injection site of both vaccinated and unvaccinated CD132° hosts (data not shown). These data suggested that continued LIP of pmel-1 was sufficient to promote tumor clearance, and prompted the question of what impact peptide priming had on pmel-1 phenotype, numbers and function. After tumor control in CD132° hosts, pmel-1 isolated from the spleen of hgp100 peptide vaccinated hosts were almost exclusively CD44hi, while those from the unvaccinated group retained a substantial fraction of CD44lowCD62Lhi naïve phenotype cells, suggesting not all these cells had encountered tumor antigen (Fig 7b). However, total numbers of pmel-1 in the spleen of vaccinated mice were only slightly higher than unvaccinated animals, indicating that the vaccination had played a relatively small role in expanding the pmel-1 pool (Fig 7c). Recovery of pmel-1 from the spleen indicated that the transferred cells were able to make IFN-γ, TNF-α, and express LAMP at the cell surface upon in vitro stimulation with cognate peptide mgp100, although unvaccinated mice were slightly better at producing cytokine IFN-γ and TNF-α (Fig 7d). However, when IFN-γ, TNF-α, and LAMP triple expression was assessed, the vaccinated and LIP pmel-1 were equivalent. Thus, given a sustained lymphopenic environment, LIP was sufficient to promote melanoma control by pmel-1 cells, and deliberate antigen priming only modestly enhanced this response.

Figure 7
Pmel-1 are able to control established B16F10 melanoma in the absence of vaccination


Central tolerance mechanisms delete highly self-reactive clones, limiting the potential for autoimmunity. However, some self-reactive clones do persist and these cells may be kept in check by antigen ignorance and/or peripheral tolerance mechanisms. Indeed, the pmel-1 TCR was originally identified from melanoma infiltrating lymphocytes. It was found to bear a receptor specific for a self peptide (derived from mgp100, recognized in the context of Db), and suitably activated pmel-1 T cells have been shown to be capable of responding to normal self tissue, manifest as vitiligo (8). Given that the reduced constraints in a lymphopenic environment might be expected to favor expansion of self-reactive T cells, we were initially surprised to observe very limited proliferation of pmel-1 cells in lymphopenic hosts (Fig 1). Furthermore, we found that naïve pmel-1 cells were CD5low, CD127low, CD8high, a phenotype which was previously observed for female H-Y TCR transgenic T cells and was associated with inefficient LIP (5, 20). It is interesting to contrast this with the CD8low, CD127high phenotype of male H-Y TCR transgenic T cells (20), which are known to be exposed to stimulatory self- antigen. Earlier studies proposed that the phenotypic traits and poor LIP of H-Y TCR T cells was due to a poor affinity for self peptide/MHC ligands expressed in the host (5, 20), but this seems unlikely to account for the impaired LIP of pmel-1 which can be overtly activated by self mgp100/Db (8). Instead, our data might suggest naïve pmel-1 TCR transgenic T cells are ignorant of stimulatory self peptide/MHC ligands, potentially due to the fact that gp100 expression is restricted to the skin. Immunologic ignorance is known to be a key mechanism in preventing some forms of autoimmunity, as is well documented by studies on LCMV-specific TCR transgenic animals which also express the target antigen in the pancreas (31). Alternatively, pmel-1 may have undergone some form of active tolerance following exposure to gp100 during development or homeostasis (as has previously been proposed (8)). Future studies in which gp100 expression is manipulated (e.g. by gene disruption or transgenic expression for broader tissue distribution) would allow for better dissection of these possibilities. In any case, our data reinforce the concept that the capacity of CD8 T cells to undergo LIP in conventional lymphopenic hosts is predictable by the CD5, CD8 and CD127 phenotype of the naïve cells.

Further supporting similarities with female H-Y TCR transgenic T cells (7), we found that pmel-1 undergo vigorous LIP in common-γ chain deficient hosts. However, despite extensive proliferation, the pmel-1 showed the unusual feature of being CD44low, in contrast with the substantial upregulation of CD44 observed in most situations of LIP (1), including OT-I cells which expand in CD132° hosts (Fig 3) (7). Pmel-1 cells, which expand in a CD132° host, were also CD122int, more similar to the expression levels seen on memory CD4 T cells rather than memory CD8 T cells. On the other hand, LIP of pmel-1 did result in other hallmarks of memory CD8 T cells, including upregulation of sca-1 and the capacity to rapidly express IFN-γ following activation, similar to OT-I cells analyzed in parallel. The basis for the atypical phenotype of pmel-1 cells following LIP is unclear. Two previous reports described CD8+ T cells that are phenotypically naïve, as characterized by CD44 expression, but are antigen experienced by virtue of CFSE dilution and high levels of sca-1 (25, 26). Restifo and colleagues have shown that mimicking wnt signalling during in vitro stimulation results in limited expression of CD44low, but that these cells are sca-1hi and go on to protect against established B16F10 (26). A recent study also suggests that constitutive wnt signalling promotes establishment of a larger memory pool, although the CD44 phenotype of these cells was not reported (32). Our studies focused on LIP rather than antigenic stimulation in production of pmel-1 proliferation, making it difficult to directly compare these systems. However, it is interesting to speculate that LIP in the CD132° environment may more efficiently stimulate wnt signalling, leading to the CD44low phenotype of pmel-1 cells. Nevertheless, such explanations would have to reconcile the fact that CD44 is upregulated by OT-I T cells in CD132° hosts, indicating that the host environment is not the only determinant of the CD44low LIP phenotype. Furthermore, there appears to be an important contribution of residual host lymphocytes to the expansion of pmel-1 CD8 T cells in CD132° hosts, since pmel-1 T cells failed to proliferate in CD132°/Rag° hosts. The nature of this contribution by host T and/or B cells is unclear, but it has previously been noted that a population of CD4 T cells with activated phenotype appear in such mice (33). Further studies will be needed to identify whether these host cells are relevant for establishing essential lymphoid architecture, or generation of factors which support pmel-1 expansion.

Ramsey et al. suggest that increased levels of IL-15 in the CD132° host play an important role in driving homeostatic proliferation (7). The finding that pmel-1 undergoing LIP do not upregulate CD122 to the same extent as OT-I and polyclonal CD8 memory T cells may account for the inability of pmel-1 to expand in a conventional lymphopenic host, where residual endogenous CD8 and NK cells would compete for IL-15. In CD132° hosts, the lack of common-γ chain signalling resulting in severe lymphopenia and increased availability of IL-15, provides an ideal environment to facilitate homeostatic proliferation of pmel-1. Even following LIP in CD132° hosts, we found that pmel-1 cells were inefficient at proliferating in an irradiated host (Fig 6), yet continued to proliferate extensively in a secondary CD132° environment. This may further indicate that the cells are unable to compete with other lymphocyte populations for suitable resources. This correlated further with the inability of post-LIP pmel-1 cells to protect against B16 melanoma in irradiated recipients.

The use of T cells specific for non-mutated tumor associated antigens is at its most basic targeted autoimmunity. Hence, it made sense to use a tumor model to determine if homeostatically proliferated self/tumor-specific CD8 could be useful. Brown et al. had suggested that homeostatic proliferation alone was sufficient to promote tumor rejection using the 2C TCR Tg and P815 mastocytoma (34). However, the 2C TCR is capable of strong LIP (5), potentially allowing for efficient generation of tumor controlling cells. In addition, Brown et al. tested rejection of an allogeneic tumor, while we have focused on a syngeneic model. Our results indicate that sustained lymphopenia (specifically, the extreme lymphopenia of a CD132° host) is needed for naïve pmel-1 to gain the ability to control established B16F10 melanomas. While this does confirm that LIP alone can promote generation of CD8 T cells capable of tumor elimination, the data also highlight that the threshold for efficient induction of this response may be very high for T cells bearing certain self-reactive TCR.

Supplementary Material

Supplementary Figs.


The authors would like to thank Sara Hamilton, Kris Hoqquist, and Colleen Winstead for critical comments on the manuscript and the members of the Jamequist lab for helpful discussions.

This work was supported by an NIH award (R37 AI38903) to SCJ.


1. Surh CD, Sprent J. Homeostasis of Naive and Memory T Cells. Immunity. 2008;29:848–862. [PubMed]
2. Hamilton SE, Wolkers MC, Schoenberger SP, Jameson SC. The generation of protective memory-like CD8+ T cells during homeostatic proliferation requires CD4+ T cells. Nature immunology. 2006;7:475–481. [PubMed]
3. Ernst B, Lee DS, Chang JM, Sprent J, Surh CD. The peptide ligands mediating positive selection in the thymus control T cell survival and homeostatic proliferation in the periphery. Immunity. 1999;11:173–181. [PubMed]
4. Goldrath AW, Bevan MJ. Low-affinity ligands for the TCR drive proliferation of mature CD8+ T cells in lymphopenic hosts. Immunity. 1999;11:183–190. [PMC free article] [PubMed]
5. Kieper WC, Burghardt JT, Surh CD. A Role for TCR Affinity in Regulating Naive T Cell Homeostasis. J Immunol. 2004;172:40–44. [PubMed]
6. Baldwin TA, Sandau MM, Jameson SC, Hogquist KA. The timing of TCR alpha expression critically influences T cell development and selection. The Journal of experimental medicine. 2005;202:111–121. [PMC free article] [PubMed]
7. Ramsey C, Rubinstein MP, Kim DM, Cho J-H, Sprent J, Surh CD. The Lymphopenic Environment of CD132 (Common {gamma}-Chain)-Deficient Hosts Elicits Rapid Homeostatic Proliferation of Naive T Cells via IL-15. J Immunol. 2008;180:5320–5326. [PubMed]
8. Overwijk WW, Tsung A, Irvine KR, Parkhurst MR, Goletz TJ, Tsung K, Carroll MW, Liu C, Moss B, Rosenberg SA, Restifo NP. gp100/pmel 17 is a murine tumor rejection antigen: induction of “self”-reactive, tumoricidal T cells using high-affinity, altered peptide ligand. The Journal of experimental medicine. 1998;188:277–286. [PMC free article] [PubMed]
9. Gattinoni L, Powell DJ, Jr., Rosenberg SA, Restifo NP. Adoptive immunotherapy for cancer: building on success. Nature reviews. 2006;6:383–393. [PMC free article] [PubMed]
10. Muranski P, Boni A, Wrzesinski C, Citrin DE, Rosenberg SA, Childs R, Restifo NP. Increased intensity lymphodepletion and adoptive immunotherapy--how far can we go? Nature clinical practice. 2006;3:668–681. [PMC free article] [PubMed]
11. Wrzesinski C, Paulos CM, Gattinoni L, Palmer DC, Kaiser A, Yu Z, Rosenberg SA, Restifo NP. Hematopoietic stem cells promote the expansion and function of adoptively transferred antitumor CD8 T cells. The Journal of clinical investigation. 2007;117:492–501. [PMC free article] [PubMed]
12. Hogquist KA, Jameson SC, Heath WR, Howard JL, Bevan MJ, Carbone FR. T cell receptor antagonist peptides induce positive selection. Cell. 1994;76:17–27. [PubMed]
13. Kieper WC, Prlic M, Schmidt CS, Mescher MF, Jameson SC. IL-12 Enhances CD8 T Cell Homeostatic Expansion. J Immunol. 2001;166:5515–5521. [PubMed]
14. Quah BJC, Warren HS, Parish CR. Monitoring lymphocyte proliferation in vitro and in vivo with the intracellular fluorescent dye carboxyfluorescein diacetate succinimidyl ester. Nat. Protocols. 2007;2:2049–2056. [PubMed]
15. Boyman O, Purton JF, Surh CD, Sprent J. Cytokines and T-cell homeostasis. Current opinion in immunology. 2007;19:320–326. [PubMed]
16. Antony PA, Piccirillo CA, Akpinarli A, Finkelstein SE, Speiss PJ, Surman DR, Palmer DC, Chan CC, Klebanoff CA, Overwijk WW, Rosenberg SA, Restifo NP. CD8+ T cell immunity against a tumor/self-antigen is augmented by CD4+ T helper cells and hindered by naturally occurring T regulatory cells. J Immunol. 2005;174:2591–2601. [PMC free article] [PubMed]
17. Goldrath AW, Sivakumar PV, Glaccum M, Kennedy MK, Bevan MJ, Benoist C, Mathis D, Butz EA. Cytokine requirements for acute and Basal homeostatic proliferation of naive and memory CD8+ T cells. The Journal of experimental medicine. 2002;195:1515–1522. [PMC free article] [PubMed]
18. Rubinstein MP, Kovar M, Purton JF, Cho JH, Boyman O, Surh CD, Sprent J. Converting IL-15 to a superagonist by binding to soluble IL-15R{alpha} Proceedings of the National Academy of Sciences of the United States of America. 2006;103:9166–9171. [PubMed]
19. Stoklasek TA, Schluns KS, Lefrancois L. Combined IL-15/IL-15Ralpha immunotherapy maximizes IL-15 activity in vivo. J Immunol. 2006;177:6072–6080. [PMC free article] [PubMed]
20. Park J-H, Adoro S, Lucas PJ, Sarafova SD, Alag AS, Doan LL, Erman B, Liu X, Ellmeier W, Bosselut R, Feigenbaum L, Singer A. `Coreceptor tuning': cytokine signals transcriptionally tailor CD8 coreceptor expression to the self-specificity of the TCR. Nature immunology. 2007;8:1049–1059. [PubMed]
21. Kieper WC, Jameson SC. Homeostatic expansion and phenotypic conversion of naive T cells in response to self peptide/MHC ligands. Proceedings of the National Academy of Sciences of the United States of America. 1999;96:13306–13311. [PubMed]
22. Rocha B, von Boehmer H. Science. Vol. 251. New York, N.Y: 1991. Peripheral selection of the T cell repertoire; pp. 1225–1228. [PubMed]
23. Cao X, Shores EW, Hu-Li J, Anver MR, Kelsail BL, Russell SM, Drago J, Noguchi M, Grinberg A, Bloom ET, Paul WE, Katz SI, Love PE, Leonard WJ. Defective lymphoid development in mice lacking expression of the common cytokine receptor [gamma] chain. Immunity. 1995;2:223–238. [PubMed]
24. Gattinoni L, Klebanoff CA, Palmer DC, Wrzesinski C, Kerstann K, Yu Z, Finkelstein SE, Theoret MR, Rosenberg SA, Restifo NP. Acquisition of full effector function in vitro paradoxically impairs the in vivo antitumor efficacy of adoptively transferred CD8+ T cells. The Journal of clinical investigation. 2005;115:1616–1626. [PMC free article] [PubMed]
25. Zhang Y, Joe G, Hexner E, Zhu J, Emerson SG. Host-reactive CD8+ memory stem cells in graft-versus-host disease. Nature medicine. 2005;11:1299–1305. [PubMed]
26. Gattinoni L, Zhong XS, Palmer DC, Ji Y, Hinrichs CS, Yu Z, Wrzesinski C, Boni A, Cassard L, Garvin LM, Paulos CM, Muranski P, Restifo NP. Wnt signaling arrests effector T cell differentiation and generates CD8+ memory stem cells. Nature medicine. 2009;15:808–813. [PMC free article] [PubMed]
27. Whitmire JK, Eam B, Whitton JL. Mice deficient in stem cell antigen-1 (Sca1, Ly-6A/E) develop normal primary and memory CD4+ and CD8+ T-cell responses to virus infection. European journal of immunology. 2009;39:1494–1504. [PMC free article] [PubMed]
28. Sandau MM, Winstead CJ, Jameson SC. IL-15 is required for sustained lymphopenia-driven proliferation and accumulation of CD8 T cells. J Immunol. 2007;179:120–125. [PubMed]
29. Purton JF, Tan JT, Rubinstein MP, Kim DM, Sprent J, Surh CD. Antiviral CD4+ memory T cells are IL-15 dependent. J. Exp. Med. 2007;204:951–961. [PMC free article] [PubMed]
30. Tan JT, Ernst B, Kieper WC, LeRoy E, Sprent J, Surh CD. Interleukin (IL)-15 and IL-7 jointly regulate homeostatic proliferation of memory phenotype CD8+ cells but are not required for memory phenotype CD4+ cells. The Journal of experimental medicine. 2002;195:1523–1532. [PMC free article] [PubMed]
31. Ohashi PS, Oehen S, Buerki K, Pircher H, Ohashi CT, Odermatt B, Malissen B, Zinkernagel RM, Hengartner H. Ablation of “tolerance” and induction of diabetes by virus infection in viral antigen transgenic mice. Cell. 1991;65:305–317. [PubMed]
32. Zhao DM, Yu S, Zhou X, Haring JS, Held W, Badovinac VP, Harty JT, Xue HH. Constitutive Activation of Wnt Signaling Favors Generation of Memory CD8 T Cells. J Immunol. 2009 [PMC free article] [PubMed]
33. Nakajima H, Shores EW, Noguchi M, Leonard WJ. The common cytokine receptor gamma chain plays an essential role in regulating lymphoid homeostasis. The Journal of experimental medicine. 1997;185:189–195. [PMC free article] [PubMed]
34. Brown IE, Blank C, Kline J, Kacha AK, Gajewski TF. Homeostatic proliferation as an isolated variable reverses CD8+ T cell anergy and promotes tumor rejection. J Immunol. 2006;177:4521–4529. [PubMed]