Control of Syngeneic Tumor Growth by Activation of CD8+ T Cells: Efficacy Is Limited by Migration Away from the Site and Induction of Nonresponsiveness

Protul Shrikant and Matthew F. Mescher
J Immunol March 1, 1999, 162 (5) 2858-2866;

Abstract

Activation of Ag-specific CD8+ T cells in response to syngeneic tumor has been visualized by adoptive transfer of CD8+ T cells from OT-I mice, with a transgenic TCR specific for H-2Kb and an OVA peptide, into Thy-1 congenic recipients. Intraperitoneal challenge with E.G7, the EL-4 thymoma transfected with OVA, results in activation and clonal expansion of the OT-I cells in the peritoneal cavity and transient control of tumor growth. However, within 2 days after becoming activated, the OT-I cells migrate out of the peritoneal cavity into the spleen and lymph nodes, and tumor growth resumes in the peritoneal cavity. The OT-I cells in lymph nodes and spleen have lytic effector activity, but exhibit split anergy in that they cannot proliferate in response to Ag unless exogenous IL-2 is provided. The failure to remain at the tumor site and continue to control tumor growth is not due to selection of Ag loss variants or development of suppression. These results suggest that effective CD8-targeted immunotherapy may depend less on enhancing the initial activation and more on sustaining the response at the appropriate location and/or reactivating cells that have left the site of tumor growth and become nonresponsive.

Numerous peptides that serve as tumor-specific CTL epitopes have been identified 1, 2, 3 , encouraged by a great deal of evidence suggesting that a CTL response can be effective in eliminating tumor. Prior immunization in murine tumor models can establish sufficient immunity so that the host is protected against a subsequent challenge with live tumor 4 , and adoptive transfer of CTL lines or clones can eliminate growing tumors 4, 5, 6, 7 . Tumor-specific CD8+ T cells can clearly be present in the animal, as evidenced by in vitro expansion or protective immunization, but nevertheless fail to protect when the host is challenged with live tumor. This is probably also the case in at least some human cancers where tumor-specific CTL can be isolated from the patient 8, 9, 10 , yet the tumor is progressing. There are numerous reasons why a CTL response may fail to occur or may occur but fail to control tumor growth. Various mechanisms of tolerance or suppression may be operative by which the naive CD8+ T cells may remain ignorant of the presence of tumor in a peripheral site, or a response may occur but be too weak and/or too short-lived to be effective. Therapeutic approaches using cytokines, costimulatory molecules, or peptides to increase response levels can give dramatic results in murine models and are beginning to be evaluated in the clinic.

A major difficulty in attempting to understand why an effective CD8+ CTL response to syngeneic tumor does not normally occur and in developing novel therapeutic approaches that target the CTL response is that the Ag-specific CD8+ T cells are present in very low numbers in the animal and cannot be specifically identified. Thus, their trafficking and response to Ag in vivo cannot be directly tracked. Instead, the location and status of the relevant tumor-specific CD8+ cells can only be assessed by indirect methods that rely on in vitro restimulation and assay of cytolytic activity. These are cumbersome at best and rely upon the ability of the cells to respond in vitro to the stimulus used.

Adoptive transfer of T cells from TCR transgenic mice into normal recipients provides a means of monitoring Ag-specific T cells during a response 11 . The transferred cells can be identified by unique surface markers and serve as an indicator of the endogenous host response. Provided that the cells are transferred in small numbers, to constitute 0.2–0.5% of the spleen and lymph node (LN)3 cells in the recipient, the presence of the transferred cells does not aberrantly skew the normal response, as is the case when attempting to study responses in intact TCR transgenic mice. Adoptive transfer of CD8+ TCR transgenic T cells has been used to study both virus-specific 12, 13 and allogeneic CTL responses 14 . Recent studies of virus-specific responses by adoptive transfer 13 or direct visualization of endogenous Ag-specific cells using tetrameric Ag 15 have shown that the number of specific cells is dramatically underestimated by approaches that rely on in vitro restimulation of the cells, emphasizing the limited information that can be gained by such indirect approaches.

To begin to develop a more detailed understanding of CD8+ T cell responses to syngeneic tumor, we have applied the TCR transgenic adoptive transfer approach to study the response to E.G7, the EL-4 thymoma transfected with the gene for OVA 16 . E.G7 cells grow rapidly in C57BL/6 mice and the host dies within 25–35 days. This cell line, expressing OVA as a pseudo-tumor Ag, has been extensively used in studies of immunotherapy procedures 17, 18, 19, 20, 21, 22 . CD8+ T cells from OT-I mice with the C57BL/6 background express a transgenic TCR that recognizes H-2Kb and OVA257–264 peptide 23 . By adoptively transferring these cells into Thy-1 congenic C57BL/PL mice and challenging them with E.G7, it is possible to identify the Ag-specific cells and determine whether they can respond to the tumor, and if so where and when the response occurs and if it is effective in reducing tumor growth. As described in this report, the OT-I cells do respond at the site of the tumor by clonally expanding and developing cytolytic function, and as a result they transiently control tumor growth. However, at a point where tumor is still present, the OT-I cells migrate from the tumor site into the spleen and LN where they still retain lytic effector function but appear to be anergic to restimulation; the tumor progresses and the mice die. These results clearly define the in vivo induction and progression of a CD8+ T cell immune response to a syngeneic tumor. Furthermore, they identify novel mechanisms involved in the circumvention of the immune response by the tumor and have important implications for devising strategies for effective immunotherapy of cancer.

Materials and Methods

Animals

OT-I TCR transgenic mice 23 , a gift from Dr. Francis Carbone (Monash Medical School, Victoria, Australia), were bred to wild-type C57BL/6 mice to generate mice heterozygous for the OT-I TCR transgene, and these mice were used as the source of transgenic T cells in all experiments. C57BL/PL mice congenic for the Thy-1 marker were purchased from The Jackson Laboratory (Bar Harbor, ME) and were used as recipients in all the experiments described. Animals were housed under specific pathogen-free conditions at all times.

Cell lines

EL-4, a thymoma derived from the C57BL/6 mouse (H-2b), was maintained in vitro in complete RPMI medium; RPMI 1640 (Cellgro, Herndon, VA), 10% FCS (Tissue Culture Biologicals, Tulare, CA), 0.2% l-glutamine, 0.1% penicillin/streptomycin, 0.1% HEPES (BioWhittaker, Walkersville, MD), 0.1% nonessential amino acids, 0.01% sodium pyruvate (Cellgro), and 0.05% 2-ME. E.G7 (OVA-transfected EL-4) 16 was maintained in complete RPMI medium containing 400 μg/ml of G418 (Cellgro). Both cell lines were periodically passaged in vivo. B3, a Kb/OVA257–264-specific CTL cell line was a gift from Drs. S. Jameson and K. Hogquist (University of Minnesota, Minneapolis, MN).

Abs and reagents

Directly conjugated mAbs including anti-CD8α-cychrome, anti-Thy-1.2-PE, anti-CD62L-FITC, anti-CD44-FITC, anti-CD49d-FITC, anti-CD25-FITC, and anti-CD69-FITC were purchased from PharMingen (San Diego, CA). Purified anti-CD16/CD32 (Fc Block) was also purchased from PharMingen.

Adoptive transfer of transgenic cells and tumor challenge

LN cells (axillary, brachial, mesenteric, inguinal, cervical, periaortic, and mediastinal) from heterozygous OT-I transgenic mice were removed, homogenized, and washed three or four times in PBS. CD8+/Thy-1.2+ cells (3–4 × 106) were transferred into sex-matched naive C57BL/PL mice by tail vein injection. Recipient mice were rested for a day and were then challenged by i.p. injection of 4 × 106 EL-4 or E.G7 cells in 0.5 ml of PBS or PBS alone as a control.

Analysis by flow cytometry

Mice were sacrificed at varying times, and the spleen and LN were collected (periaortic, mesenteric, axillary, and brachial), homogenized, and treated with ammonium chloride to remove RBCs. The peritoneal cavity was washed twice with 25 ml of PBS each time, and the resulting peritoneal exudate lymphocytes were adherence-depleted for 90 min in complete medium at 37°C. The total number of cells obtained from each site was determined by counting using a hemocytometer.

Cells (1 × 106) isolated from each site were treated with Fc Block for 30 min on ice and were then stained with anti-CD8α-cychrome, anti-Thy-1.2-PE and a third Ab-FITC specific for phenotypic marker (see below). After a 1-h incubation on ice, the cells were washed twice, resuspended in 0.2 ml of 1% formaldehyde, and analyzed by three-color flow cytometry using the CellQuest software package (Becton Dickinson, San Jose, CA). Transferred OT-I cells were identified as CD8+/Thy-1.2+ cells. Cells from C57BL/PL mice stained in the same way showed no events in the CD8+/Thy-1.2+ gate, and cells from mice that had received OT-I cells by adoptive transfer had <0.05% of events in this gate when the Thy-1.2-PE mAb was replaced with an isotype-matched control Ab labeled with PE. These controls were included in every experiment.

Activation status-associated phenotypic characterization of OT-I cells was performed by staining with a third Ab; these included anti-CD44-FITC, anti-CD25 (IL-2R)-FITC, anti-CD62L (L-selectin)-FITC, anti-CD49d (VLA-4)-FITC, and anti-CD69-FITC. The CD8+/Thy-1.2+ cells were gated, and their numbers and phenotype were determined as a percentage of the total population. In figures examining phenotype, the markers used to determine the percentage of cells expressing high levels of the protein are indicated. The positions of these markers were determined for each Ab by staining with the appropriate FITC-labeled isotype-matched control Ab. The numbers of tumor cells were determined by gating on large granular cells that were Thy-1.2+/CD8−. The total numbers of CD8+/Thy-1.2+ (OT-I) and tumor cells at each location were determined by multiplying the percentage of cells in the population by the total number of cells recovered from each site.

Chromium release assay

The lytic activity of spleen or LN cells obtained from tumor-primed mice was determined using a standard 4-h 51Cr release assay with EL-4 or E.G7 target cells. Briefly, serial dilutions of adherence-depleted spleen and LN cells were plated in 96-well V-bottom plates with 5 × 103 of 51Cr-labeled EL-4 or E.G7 cells in a final volume of 0.2 ml. After 4-h incubation at 37°C, 0.1 ml of supernatant was removed from each well, and radioactivity was determined. The percent specific 51Cr release was determined by subtracting spontaneous release from experimental release and dividing by maximum release minus spontaneous release. Maximum release was determined by treatment of cells with 3% SDS.

Cell proliferation assay

To determine the proliferative ability of spleen or LN cells from tumor-challenged mice, 1 × 105 adherence-depleted spleen or LN cells in complete RPMI medium were cultured in 96-well U-bottom plates with irradiated spleen cells from naive C57BL/6 mice pulsed with OVA257–264 peptide in the presence or the absence of mouse rIL-2 (R&D Systems, Minneapolis, MN). On the indicated days, cultures were pulsed with 1 μCi of [3H]thymidine for 6–8 h and lysed with distilled water, and the amount of [3H]thymidine incorporated into DNA was determined by liquid scintillation counting. The total numbers of CD8+/Thy-1.2+ cells in the cultures was determined before and after the culture period to determine the fold increase in CD8+/Thy-1.2+ cells.

Results

CD8+ OT-I T cells respond in vitro to Ag on E.G7 tumor

E.G7 cells are EL-4 thymoma transfected with the OVA gene 16 and express about 100 Kb/OVA257–264 complexes/cell 24 . Thus, the OVA peptide is present at a low density, as is likely to be the case for physiologically relevant tumor Ags. To insure that T cells from OT-I mice with a TCR-specific for Kb/OVA257–264 23 could respond to Ag on E.G7, proliferation in response to irradiated E.G7 cells was examined (Fig. 1⇓A). OT-I cells made a strong proliferative response to OVA257–264-pulsed spleen cells, and a weaker but still substantial, response to E.G7, while no response occurred to the parental EL-4 cells. E.G7 cells could also be killed by OT-I effector CTL, while lysis of EL-4 was minimal (Fig. 1⇓B).

FIGURE 1.
FIGURE 1.

CD8+ T cells from LNs of OT-I mice recognize E.G7 tumor in vitro. A, Proliferation of OT-I in response to E.G7. CD8+ purified OT-I LN cells (1–5 × 105) were cultured in triplicate in medium alone (Media; ---) or in the presence of 5 × 105 irradiated stimulator cells (EL-4 (○), E.G7 (⊗), or B6 spleen cells pulsed with OVA (×)). After 48 h each well was pulsed with 1 μCi of [3H]thymidine for 12 h, and the incorporation was determined. Results are shown as the mean of triplicate determinations. B, Cytolysis of E.G7 targets by OT-I CTL. Purified CD8+ T cells from LN of OT-I mice were stimulated for 3 days in vitro with B6 spleen cells pulsed with OVA and then assayed for lytic activity in a 4-h 51Cr release assay using 1 × 105 labeled EL-4 or E.G7 as targets. The percent specific lysis was calculated as described in Materials and Methods.

The ability of E.G7 to stimulate Ag-specific responses by OT-I CD8+ T cells confirmed that this system should be suitable for examining the in vivo response to syngeneic tumor by adoptive transfer. Almost all Thy-1.2+ cells from the LN of OT-I mice are also CD8+, and these cells have a naive phenotype: CD25 and CD69 negative, low expression of CD44 and VLA-4, high expression of L-selectin, and forward scatter (FSC) consistent with small resting cells (Fig. 2⇓A). Adoptive transfer was performed by i.v. (tail vein) injection of cells from OT-I mice into Thy-1 congenic C57BL/PL recipients to allow detection of the transferred cells by staining with anti-Thy-1.2 mAb. In C57BL/PL mice that had not received OT-I, virtually no Thy-1.2+CD8+ events could be detected in LN (Fig. 2⇓B) or spleen cells. In mice that received OT-I by i.v. injection (tail vein), the Thy-1.2 and CD8+ OT-I cells could be readily detected in spleen and LN (Fig. 2⇓C). The naive phenotype of the OT-I cells (Fig. 2⇓A) was maintained in the recipients following adoptive transfer (Fig. 2⇓C).

FIGURE 2.
FIGURE 2.

Detection and phenotypic characterization of transferred OT-I cells in C57BL/PL recipients. LN cells were isolated from transgenic OT-I mice (A), from C57BL/PL mice that had not received OT-I cells (B), or from C57BL/PL mice that had received 3–4 × 106 OT-I LN cells 2 days previously (C). The cells were stained with anti-CD8a-cychrome and anti-Thy-1.2-PE mAbs to detect the CD8+ OT-I cells (top panels) and with a third, FITC-conjugated Ab against the indicated activation marker (lower panels). Histograms of activation markers were obtained by gating on CD8+ Thy-1.2+ events.

CD8+ OT-I T cells respond to E.G7 tumor in vivo

EL-4 tumor expresses Ags that can be recognized by C57BL/6 CD8+ T cells 25, 26 , and the E.G7 line expresses in addition the OVA257–264 CD8+ T cell determinant. Nevertheless, i.p. injection of live E.G7 results in progressive tumor growth and death of the host in about 25–35 days. This could be due to failure to generate a CD8+ CTL response or failure of the response to control tumor growth. To examine this, OT-I cells were adoptively transferred by i.v. (tail vein) injection in C57BL/PL recipients and allowed to equilibrate for 24 h so that their distribution in lymphoid organs reflected that of the endogenous CD8+ T cells. One day later the recipients were challenged by i.p. injection of live E.G7, and the number of OT-I cells (CD8+/Thy-1.2+ cells) in the LN, spleen, and peritoneal cavity was determined at subsequent times. By day 4 after challenge, the number of OT-I cells had declined modestly in the spleen and LN (Fig. 3⇓, A and B) and increased substantially in the peritoneal cavity (Fig. 3⇓C), suggesting that they were migrating to the latter site and undergoing clonal expansion. By day 6, OT-I cell numbers had begun to decline in the peritoneal cavity and were increasing in the spleen. Few detectable OT-I cells remained in the peritoneal cavity by day 10, while increased numbers were present in the LN and spleen and remained present at these sites through days 18 (Fig. 3⇓, A and B) and 22 (data not shown). In mice that were not challenged with E.G7 (Fig. 3⇓, OT-I), the number of OT-I cells declined for the first 4 days and then remained constant, with no expansion at any sites. Similarly, no expansion of OT-I cells occurred in adoptively transferred mice that were challenged with EL-4 tumor (data not shown). Thus, OT-I cells exhibit an Ag-specific increases in number at the site of tumor growth in mice challenged with tumor.

FIGURE 3.
FIGURE 3.

Adoptively transferred OT-I CD8+ T cells initially expand in the peritoneal cavity upon Ag challenge and subsequently increase in number in the LN and spleen. OT-I LN cells (3–4 × 106) are transferred i.v. into C57BL/PL recipients. Twenty-four hours later the mice received an i.p. injection of 0.5 ml of PBS (OT-I; ○) or 4 × 106 E.G7 cells in 0.5 ml of PBS (OT-I/E.G7; ⊕). Two animals from each group were sacrificed on the indicated days after tumor challenge, and cells harvested from draining and peripheral LN (A), spleen (B), and peritoneal cavity (C) were analyzed by flow cytometry to determine the total number of OT-I cells (CD8+/Thy-1.2+) at each site. The results shown are the mean ± SD of three separate experiments examining two mice per group at each time.

The increase in OT-I cells in the peritoneal cavity by day 4 suggested that the cells were clonally expanding there, as opposed to simply migrating into the site, and this conclusion was supported by the phenotype of the cells. In contrast to cells before transfer (Fig. 4⇓A), the majority of OT-I in the peritoneal cavities of E.G7-challenged mice had an activated phenotype (Fig. 4⇓C), with increased CD44, CD25, and VLA-4 and decreased L-selectin surface levels, and about half were blasts as judged by FSC. However, the OT-I cells in the spleen and LN on day 4 maintained a naive phenotype, with the exception of increased VLA-4 expression on OT-I cells in the LN (data not shown). When challenge was with EL-4, very few OT-I cells appeared in the peritoneal cavity on day 4 or later, and they did not increase as a percentage of the total CD8+ T cells. The small numbers made precise estimates of phenotypic distribution difficult, but it is clear that the majority of OT-I cells did not become activated in response to the EL-4 tumor (Fig. 4⇓B). Thus, OT-I cells undergo Ag-dependent proliferation in response to the E.G7 tumor, and clonal expansion occurs preferentially at the site of tumor growth rather than in draining LN.

FIGURE 4.
FIGURE 4.

Phenotypic characterization of OT-I cells from the peritoneal cavities of adoptively transferred recipients 4 days after challenge with E.G7, but not EL-4, display an activated phenotype. Cells were stained with CD8a-cychrome and Thy-1.2 PE mAbs, and the double-positive population of cells (OT-I cells) was gated as shown in the top panels and analyzed for expression of CD44, CD25, VLA-4, and L-selectin as described in Materials and Methods. Naive OT-I cells before adoptive transfer are shown in the left panels for comparison (A, pretransfer). The middle and right panels show analysis of peritoneal lavage cells from adoptive transfer recipients challenged by i.p. injection of 4 × 106 EL-4 cells in 0.5 ml of PBS (B) or 4 × 106 E.G7 cells in 0.5 ml of PBS (C).

Activated CTL transiently control tumor growth and then leave the tumor site

While they were present in expanded numbers in the peritoneal cavity (days 4–6), OT-I cells limited the rate of growth of the tumor compared with tumor growth in normal C57BL/PL mice (Fig. 5⇓A). After day 6, when OT-I cells were declining in the peritoneal cavity, tumor growth rate increased (Fig. 5⇓A) and paralleled the growth rate in normal, untransferred C57BL/PL mice. Control of tumor growth by the OT-I cells is Ag specific; transferred OT-I cells have no effect on the growth of EL-4 tumor (Fig. 5⇓B). In five of six experiments the time course of the response, OT-I expansion in the peritoneal cavity, phenotypic changes, and transient control of E.G7 tumor growth were comparable to the representative results shown in Figs. 3–5⇑⇑⇓; tumor failed to grow in the control mice in the sixth experiment. Numerous additional experiments examining more limited time points have further confirmed these findings.

FIGURE 5.
FIGURE 5.

Adoptive transfer of OT-I cells results in transient control of the growth of E.G7, but not EL-4, during the time that CD8+ OT-I cells are present in the peritoneal cavity. A, E.G7 tumor cells (4 × 106) were injected i.p. on day 0 into either normal C57BL/PL mice (E.G7) or C57BL/PL mice that had received OT-I cells by adoptive transfer (OT-I/E.G7). B, EL-4 tumor cells (4 × 106) were injected i.p. on day 0 into either normal C57BL/PL mice (EL-4) or C57BL/PL mice that had received OT-I cells by adoptive transfer (OT-I/EL-4). For both A and B, the number of tumor cells and the number of OT-I cells (A only) in the peritoneal cavities of the mice were determined on the indicated days by flow cytometry as described in Materials and Methods.

The above experiments used OT-I cells from C57BL/6 mice for transfer into C57BL/PL Thy-1 congenic recipients. There was therefore the possibility that a response by the recipient to the Thy-1 difference might influence the results. Experiments were thus performed in which OT-I cells were transferred into normal C57BL/6 recipients and detected by staining for CD8 and Vα2, one of the transgenic TCR chains. This allows detection of OT-I cells with reasonable accuracy over the low background of CD8+Vα2+ host cells, particularly in the peritoneal cavity during the response. The numbers and kinetics of clonal expansion followed by decline in OT-I cells and the numbers and kinetics of tumor cells in the peritoneal cavity were essentially the same as shown in Figs. 3⇑ and 5⇑. Thus, the Thy-1 difference between the OT-I cells and the C57BL/PL recipients did not alter the response, consistent with the low responder status of B6 mice to Thy-1 Ags 27 .

OT-I effector CTL migrate out of the peritoneal cavity

The increasing numbers of OT-I cells in the LN and spleen after day 4 concomitant with the decline in the peritoneal cavity suggested that the cells might be migrating out of the peritoneal cavity. Alternatively, a second wave of clonal expansion might occur in the LN and spleen. Examination of the phenotype of the OT-I cells in spleen and LN cells on day 10 showed the majority to have a phenotype consistent with having been previously activated. The cells had high CD44 and low L-selectin levels, but few were blasts as judged by FSC (Fig. 6⇓), and they did not have increased CD25 levels (data not shown), indicating that they were not actively proliferating. Similar results were obtained when OT-I cells from the LN and spleen were examined on day 8 (data not shown).

FIGURE 6.
FIGURE 6.

OT-I cells in the LN and spleen on day 10 have been previously activated but are not proliferating. LN, spleen, or peritoneal cavity cells were harvested from OT-I-transferred mice 10 days after i.p. injection of 4 × 106 E.G7 tumor cells. Cells were stained for CD8a-cychrome, Thy-1.2-PE, and either CD62L-FITC or CD44-FITC. Gates were set to analyze OT-I cells (CD8+/Thy-1.2+; top panels), and their FSC and levels of CD44 and CD62L were determined as shown in the histograms. Numbers above the markers indicate the percentage of cells with high expression of the indicated phenotype based on the gates shown.

The phenotype of the OT-I cells in the LN and spleen suggested that they had proliferated in the peritoneal cavity and then migrated out. To directly examine this, cells were recovered on day 4 from the peritoneal cavities of mice that had received OT-I cells and been challenged with E.G7. The CD8+ cells (that include OT-I cells) were purified. The purified cells were then transfered into the peritoneal cavities of C57BL/PL mice that had received either PBS or E.G7 tumor cells by i.p. injection 4 days earlier. Three days later, the OT-I cells placed in mice bearing E.G7 had left the peritoneal cavity and appeared in the LN and spleen (Fig. 7⇓, E.G7). In contrast, OT-I cells placed in the peritoneal cavities of mice that received PBS remained in the peritoneal cavities (Fig. 7⇓, PBS). These results are consistent with the suggestion that OT-I cells respond to E.G7 in the peritoneal cavity, control tumor growth transiently, but then migrate out of the site of tumor growth and into the spleen and LN. Furthermore, the retransferred CD8+ T cells did not proliferate in response to tumor, as evidenced by their forward scatter (data not shown).

FIGURE 7.
FIGURE 7.

Activated CD8+ OT-I cells migrate from the peritoneal cavity to the spleen and LN. C57BL/PL mice received 4 × 106 LN cells from OT-I mice and 1 day later were injected i.p. with 4 × 106 E.G7 cells (day 0). Four days later cells were harvested from the peritoneal cavity, adherent cells were depleted, and CD8+ T cells were purified using negative selection columns. CD8+ T cells (3.4 × 106) were injected i.p. into mice that 4 days earlier had received either 0.5 ml of PBS (PBS; right) or 4 × 106 E.G7 cells in 0.5 ml of PBS (E.G7; left). Three days after injection of CD8+ T cells, cells were harvested from LN, spleen, and peritoneal cavities; stained with anti-CD8 and anti-Thy-1.2 mAbs; and analyzed by flow cytometry. Dot plots are of cells gated for lymphocytes. The total number of Thy-1.2+/CD8+ T cells at each site is given. Two mice were analyzed for each condition (PBS vs E.G7) with essentially the same results.

Growing E.G7 tumor remains antigenic following the transient OT-I response

Exit of activated OT-I cells from the site of tumor and resumed tumor growth could occur if cytolysis by the OT-I CTL resulted in selection of an Ag loss variant, E.G7 cells that no longer expressed OVA. However, a Kb/OVA257–264-specific cloned CTL line could kill E.G7 cells isolated on day 18 from the peritoneal cavities of OT-I-transferred, E.G7-challenged mice (Fig. 8⇓A), demonstrating that the majority of E.G7 cells still express sufficient OVA peptide to act as targets.

FIGURE 8.
FIGURE 8.

Persisting E.G7 tumor is antigenic in vitro and in vivo. A, Persisting tumor is lysed by anti-Kb/OVA257–264 CTL. C57BL/PL mice received OT-I LN cells by adoptive transfer and were challenged by i.p. injection of 4 × 106 EL-4 or E.G7 tumor cells. Tumor cells were harvested from the peritoneal cavities of the mice 18 days later, labeled with 51Cr, and used as targets in a standard 4-h 51Cr release assay with B3 effectors, a cloned CTL line specific for H-2b/OVA257–264. B, Naive OT-I cells can respond in vivo following an initial response. C57BL/PL mice received OT-I cells and were challenged on day 0 by i.p. injection of PBS (OT-I; ---). A second group received OT-I cells and was challenged by i.p. injection of 4 × 106 E.G7 tumor on day 0, and then received 3–4 × 106 naive OT-I cells on day 8 as indicated (OT-I/E.G7+ OT-I; ○). A third group did not receive OT-I cells but was challenged with E.G7 tumor on day 0, and then received naive OT-I cells i.v. on day 8 (E.G7+ OT-I; ⊕). The number of OT-I cells present in the peritoneal cavity was determined on the indicated days (averages of two mice per group). C, E.G7 tumor growth is transiently controlled when OT-I cells are present at the tumor site (peritoneal cavity). The number of E.G7 tumor cells recovered from the peritoneal cavities of the groups of mice described above was determined. Results shown are the average of two mice per group.

Antigenicity of the E.G7 tumor following the transient OT-I response was more directly demonstrated by transferring fresh, naive OT-I cells into mice after an initial OT-I response to E.G7. In the initial response, OT-I cells reached high numbers in the peritoneal cavity on day 4 and declined to a low level by day 8 as expected (Fig. 8⇑B). When fresh OT-I cells were then injected i.v., a second wave of OT-I expansion occurred in the peritoneal cavity that peaked 3–4 days after the cells were injected and then declined. Thus, the cells freshly injected on day 8 responded identically to cells present on day 0 at the time of the initial tumor challenge. Tumor growth was transiently controlled during both the first and the second response by OT-I cells (Fig. 8⇑C).

A response also occurred when tumor was allowed to grow uncontrolled in normal C57BL/PL mice (no adoptive transfer) for 8 days before i.v. injection of the OT-I cells (Fig. 8⇑B). Again, clonal expansion of OT-I cells occurred in the peritoneal cavity 4 days after injection, and tumor growth was controlled while OT-I cells were present at the site of the tumor (Fig. 8⇑C). Thus, even when tumor load has reached a high level before introduction of the OT-I cells, they are still able to respond by clonal expansion and to transiently control further tumor growth.

OT-I cells from day 10 spleen have lytic effector function but are anergic

When cells were recovered from the spleens and LN of mice on day 10 following the response to tumor, and the CD8+ cells (which included OT-I cells) were purified, they demonstrated potent cytolytic activity for E.G7 tumor (Fig. 9⇓A, OT-I/E.G7). In contrast, killing of E.G7 was low by cells from mice that had received OT-I by adoptive transfer but had not been challenged with tumor (Fig. 9⇓A, OT-I) or from mice challenged with EL-4 tumor (Fig. 9⇓A, OT-I/EL-4). For all populations there was essentially no lysis of EL-4 tumor cells (Fig. 9⇓B). Thus, the OT-I cells that have migrated out of the peritoneal cavity retain Ag-specific cytolytic activity and could presumably continue to control tumor growth, but they are no longer in the right location to perform this function.

FIGURE 9.
FIGURE 9.

OT-I cells from the LN and spleen on day 10 have Ag-specific cytolytic activity. Cells from LN and spleens were harvested on day 10 from OT-I-transferred mice that had been challenged on day 0 by i.p. injection with PBS (OT-I), EL-4 (OT-I/EL-4), or E.G7 (OT-I/EG7). CD8+ T cells were purified by adherence cell depletion and negative selection columns, and used immediately as effector cells in a standard 4-h 51Cr release assay using E.G7 (A) or EL-4 cells (B) as targets. E:T cell ratios are expressed as the number of OT-I cells per number of target cells.

When the OT-I cells stop expanding and exit the peritoneal cavity, the tumor is clearly still antigenic (Fig. 8⇑). This suggested that although the OT-I cells on day 10 can still lyse target cells (Fig. 9⇑), they may no longer be able to proliferate in response to Ag. We therefore examined whether cells recovered from LN and spleen on day 10 could respond in an in vitro proliferation assay. Cells were taken from mice that had received OT-I cells by adoptive transfer but not been challenged with E.G7 (OT-I), mice that received OT-I cells and were challenged with EL-4 (OT-I/EL-4), or mice that received OT-I cells and were challenged with E.G7 (OT-I/E.G7).

Cells from OT-I or OT-I/EL-4 mice made a vigorous proliferative response to C57BL/6 spleen cells pulsed with OVA257–264 peptide, as measured by [3H]TdR incorporation at 48 h (Fig. 10⇓A). In contrast, the OT-I cells from E.G7-challenged mice did not respond (Fig. 10⇓A, OT-I/E.G7). Addition of exogenous IL-2 to the cultures restored the response of the cells from the OT-I/E.G7 mice (Fig. 10⇓A). IL-2 in the absence of peptide Ag stimulated some response by the cells from the OT-I/E.G7 mice, but the response was much stronger when Ag was also present. To confirm that the [3H]TdR incorporation measured in these experiments reflected the proliferative capacity of the OT-I cells, we also determined the changes in absolute numbers of OT-I cells in the cultures after 3 days (Fig. 10⇓B). This analysis yielded the same conclusion; cells from OT-I/E.G7 mice can respond in the presence of exogenous IL-2, but not in its absence. Thus, the cells from the OT-I/E.G7 mice display an apparent split anergy, able to lyse targets but unable to undergo proliferation unless exogenous IL-2 is provided. Induction of this state depends upon Ag recognition in vivo, since OT-I cells from mice challenged with EL-4 tumor respond without IL-2 addition.

FIGURE 10.
FIGURE 10.

Following expansion in the peritoneal cavity and migration to the LN and spleen, OT-I cells are not responsive to Ag stimulation. CD8+ T cells (1 × 105) derived from LN and spleen of adoptively transferred mice that were challenged with PBS (OT-I), EL-4 (OT-I/EL-4), or E.G7 (OT-I/EG7) were cultured in vitro with 5 × 105 irradiated B6 spleen cells with either no addition (B6 SPL) or pulsed with 1 μM OVA257–264 peptide (OVAp), and with addition of IL-2 (10 U/ml) or anti-mouse CTLA-4 mAb (5 μg/ml) where indicated. A, After 48 h of culture cells were pulsed with [3H]thymidine and harvested to determine the amount of [3H]thymidine incorporation. B, After 72 h of culture, harvested cells were analyzed by flow cytometry for CD8/Thy-1.2 expression (OT-I cells). The mean ± SD are shown. The fold increase was determined based on the input number of OT-I cells. Essentially the same results were obtained in three separate experiments.

Discussion

It has become apparent that most tumors express peptides bound to class I proteins that can be recognized by CD8+ T cells, and this provides considerable encouragement for the possibility of inducing therapeutic immune responses to these epitopes. It has long been recognized, however, that immunogenic tumors grow progressively in the syngeneic host; even in cases where some CD8+ T cell response occurs, the immune system ultimately fails to control the tumor. Numerous suggestions have been made to attempt to explain the inability to mount an effective CTL response, including tolerance due to ignorance or other mechanisms, failure of the antigenic peptide to be presented in a stimulatory context, i.e., with costimulatory ligands and/or cytokines present, selection of Ag loss variants, production by the tumor of Ab-dependent blocking factors that block cytotoxicity, and generation of suppressor cells. The results described here examining the EG.7 tumor with OVA as a pseudo-tumor Ag suggest the additional possibility that activation and development of effector function can be effective, but that the CTL fail to control tumor growth because they do not remain at the site of the tumor.

When adoptively transferred mice are challenged by i.p. injection of EG.7, OT-I CD8+ T cells migrate to the peritoneal cavity within 3 days, undergo substantial clonal expansion, develop lytic activity, and control tumor growth over the next 2 days (Figs. 3⇑ and 5⇑). Thus, the OT-I cells do not remain ignorant and are not tolerant to the Ag; in fact, they make a vigorous response. By day 6, however, the CTL are leaving the peritoneal cavity and migrating to the LN and spleen. They continue to have lytic activity through at least day 10 (Fig. 9⇑), but are ineffective because they are not present in the peritoneal cavity where the tumor has resumed rapid expansion. Failure to remain at the site and control tumor growth is not due to the appearance of Ag loss variants, since tumor cells isolated following the transient CTL response still express Ag and can be killed by the CTL (Fig. 8⇑A).

Failure to sustain a CTL response at the site of tumor also does not appear to be due to development of blocking factors or cells with suppressor activity, since these would be expected to prevent a response when fresh OT-I cells are transferred into the animal, but instead a vigorous response occurs that is essentially identical with the initial response (Fig. 8⇑, B and C). Thus, the environment present in the tumor-bearing host by day 8 and beyond is not inhibitory to the generation of an Ag-specific CD8+ T cell response and does not prevent lytic effector cells from being active in the peritoneal cavity, nor is the tumor resistant to control by the CTL that are generated.

The first, somewhat surprising, observation in this system was that full activation of the OT-I cells, as assessed by CD25 up-regulation and blast transformation, was first detected in the peritoneal cavity at the site of tumor rather than in draining LN (Fig. 4⇑ and data not shown). In fact, most of the clonal expansion that was observed occurred in the peritoneal cavity, with the cells subsequently migrating to the LN and spleen (Fig. 7⇑) where they expressed a high level of CD44, indicating that they had already responded to Ag, but where few were blasts. Failure to detect full activation and expansion in the draining LN was unexpected. However, although blast transformation was not detected in LN at early times, the level of VLA-4 expression on the OT-I cells did increase in an Ag-dependent manner (data not show). This occurred at the same time that OT-I cells were decreasing in number in the spleen and LN and appearing in the peritoneal cavity. Similar changes were seen when examining 2C TCR transgenic CD8+ T cells responding to allogeneic tumor challenge in the peritoneal cavity 14 . These observations suggest the hypothesis that recognition of Ag by naive CD8+ T cells results initially in altered expression of receptors involved in trafficking, including VLA-4, and that this promotes selective migration of the Ag-specific cells to the peripheral site of tumor growth where they become fully activated. This could potentially explain the somewhat paradoxical observation that expression of B7-1 on a tumor cell enhances the CD8+ T cell response to the tumor, but that the responding T cells are restricted only by class I expressed by host APC; T cells specific for class I expressed just by the tumor make only a small contribution 28 . Ag-specific CD8+ T cells may need to recognize Ag presented by host APC in the draining LN, i.e., by cross-priming, so that they then migrate to the tumor site and undergo potent activation and expansion in response to the tumor expressing the B7-1 costimulatory ligand. Most T cells specific for class I on the tumor that is not expressed by the host APC may never gain access to the tumor.

Although the EG.7 tumor used in these experiments does not express B7 ligands, vigorous clonal expansion of OT-I cells occurs. This may involve cross-priming 28, 29 . Alternatively, ICAM expressed by the tumor may provide sufficient LFA-1-dependent costimulation to support proliferation 30, 31 . Host CD4+ Th cells do not appear to be involved, since the OT-I response through at least day 10 is essentially identical in recipients that have been depleted of CD4+ T cells by in vivo anti-CD4 mAb administration (data not shown).

Following the peak of clonal expansion on day 4, the number of OT-I cells declines in the peritoneal cavity (Fig. 3⇑). This is probably due in part to cell death; by using mAb staining for annexin V 32 , about 30% of OT-I cells in the peritoneal cavity were found to be positive on days 4–6 (data not shown). It is also clear, however, that much of the decline in number of effector cells at the tumor site is due to migration out of the peritoneal cavity into LN and spleen (Fig. 7⇑). Additional experiments have shown that migration out of the peritoneal cavity is not Ag specific; activated OT-I cells placed in the peritoneal cavities of mice that have been injected i.p. with either EL-4 tumor or LPS also leave and appear in the LN and spleen (P. Shrikant and M. F. Mescher, manuscript in preparation). Thus, it appears that exit from the peritoneal cavity requires inflammation, but not recognition of specific Ag by the OT-I cells. OT-I cells placed in the peritoneal cavities of E.G7-bearing mice can be found in LN and spleen as early as 2 days after injection (data not shown).

Why do the active lytic effector OT-I cells leave the peritoneal cavity while tumor cells are still present in large numbers? This is not because the tumor no longer expresses Ag (Fig. 8⇑). It also does not appear to be due to the tumor creating an environment that either prevents T cells from entering the peritoneal cavity or causes them to leave rapidly, since naive OT-I transferred late can migrate there, respond, and again transiently control tumor growth (Fig. 8⇑, B and C). The response of 2C CD8+ T cells to allogeneic tumor in the peritoneal cavity shows almost identical kinetics of clonal expansion followed by exit from the peritoneal cavity 14 as does the response of OT-I cells to EG.7, the difference being that allogeneic tumor has been eliminated by the time the CTL leave, while the syngeneic tumor is still present in large numbers.

Taken together, these observations raise the possibility that the CTL developmental program may dictate this behavior and that it is largely Ag independent once the initial activation has occurred. What regulates the location of the CTL is unclear. One possibility is that naive or recently activated CD8+ T cells home very effectively to an inflammatory site, resulting in their accumulation in the tumor-bearing peritoneal cavity, but that within a short time of gaining effector function the cells lose this ability. Thus, once they exit the peritoneal cavity they would be unable to efficiently reenter the site. Another possibility is that Ag recognition during the initial activation results in the cells being held at the site of the tumor, perhaps through constitutive up-regulation of adhesion receptors. Subsequent down-regulation would then allow the cells to release and exit the site; the effector cells might continue to migrate through the peritoneal cavity, but fail to accumulate there after the initial activation period. Consistent with this suggestion, we have found that CD8+ T cells display constitutively active binding to ICAM-1 and fibronectin within about 24 h of in vitro stimulation with Ag, and this disappears by about 72 h when the cells develop effector function (J. M. Curtsinger, D. Lins, and M. F. Mescher, manuscript in preparation).

Whatever the mechanism responsible, it is clear that failure of the cytolytic effector cells to remain at the site of the tumor is the major limitation in the ability of the CD8+ T cell response to control tumor growth. A similar phenomenon may occur for CD8+ T cell responses to human tumors. In recently reported trials 33 , administration of melanoma-specific peptides resulted in increased Ag-specific reactivity of cells from the peripheral blood of the majority of patients, but few tumor responses were seen. In contrast, administering peptide along with IL-2 had little effect on the reactivity of peripheral blood cells, but induced tumor responses in a significant fraction of the patients. The investigators suggested that IL-2 administration might cause the Ag-specific effector CTL to traffic to the tumor site, while in its absence increased numbers of CTL are still generated but are ineffective because they leave the tumor site. We have, in fact, found that IL-2 administration at the appropriate time can cause OT-I cells to remain in the peritoneal cavity in large numbers and control tumor growth for longer times than in the absence of IL-2 (P. Shrikant and M. F. Mescher, unpublished observations).

The OT-I cells that have responded to tumor and then exited the peritoneal cavity are anergic, in that they cannot respond to Ag by proliferating unless exogenous IL-2 is provided (Fig. 10⇑). This somewhat resembles the split anergy described for cloned CD8+ T cell lines 34 , where a signal 1 stimulus is sufficient to mediate cytolysis but renders the cells unresponsive to subsequent costimulation. It also resembles the anergy described for CD4+ T cells 35, 36, 37 . In both of those cases, however, the cell are rendered anergic when they recognize Ag in the absence of costimulation and fail to make an initial response. In contrast, the results described here demonstrate that OT-I cells become nonresponsive following a vigorous initial response to the EG.7 tumor. This induction of anergy may not be unique to tumor Ag, since CD8+ T cell nonresponsiveness has also been observed following in vivo stimulation with peptides and superantigen 38, 39, 40, 41, 42 . Whether activation-induced nonresponsiveness limits the ability of CD8+ T cells to control tumor growth is unclear, given that the cells are not at the site of Ag in any case.

Being able to visualize the response of Ag-specific CD8+ T cells to a tumor has provided some novel insights into the factors that lead to the immune system failing to control tumor growth. In the system studied here, this is clearly not a result of a failure of the CD8+ T cells to become activated. Rather, it is a failure of the activated cells to remain at the site of tumor growth, where they could continue to control or eliminate it. Thus, successful immunotherapy in this case would not require manipulations to initiate a response as this occurs anyway, but, rather, manipulations to sustain the response and, critically, to sustain it at the appropriate location.

Acknowledgments

We thank Drs. Bruce Blazar, E. N. Benveniste, and Stephen Jameson for critical reading of this manuscript, and Dr. Marc Jenkins for helpful advice and discussion.

Footnotes

  • ↵1 This work was supported by Grants RO1AI31524 and RO1AI34824 (to M.F.M.) from the National Institutes of Health, and a National Multiple Sclerosis Society postdoctoral fellowship (to P.S.).

  • ↵2 Address correspondence and reprint requests to Dr. Matthew F. Mescher, Center for Immunology, Box 334 Mayo, 420 Delaware St. S.E., Minneapolis, MN 55455. E-mail address: mesch001{at}maroon.tc.umn.edu

  • ↵3 Abbreviations used in this paper: LN, lymph nodes; PE, phycoerythrin; VLA-4, very late antigen-4; FSC, forward scatter.

  • Received September 1, 1998.
  • Accepted November 12, 1998.

References

  1. ↵
    Boon, T., J.-C. Cerottini, B. Van den Eynde, P. van der Bruggen, A. Van Pel. 1994. Tumor antigens recognized by T lymphocytes. Annu. Rev. Immunol. 12: 337
    OpenUrlCrossRefPubMed
  2. ↵
    Pardoll, D.M.. 1994. Tumor antigens for the 1990’s. Nature 369: 357
    OpenUrlCrossRefPubMed
  3. ↵
    Slingluff, C., D. Hunt, V. Engelhard. 1994. Direct analysis of tumor-associated peptides. Curr. Opin. Immunol. 6: 733
    OpenUrlCrossRefPubMed
  4. ↵
    Dye, E., R. North. 1981. T cell-mediated immunosuppression as an obstacle to adoptive immunotherapy of the P815 mastocytoma and its metastases. J. Exp. Med. 154: 1033
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Cheever, M., P. Greenberg, A. Fefer. 1981. Specific adoptive therapy of established leukemia with syngeneic lymphocytes sequentially sensitized in vivo and in vitro and nonspecifically expanded by culture with interleukin 2. J. Immunol. 126: 1318
    OpenUrlPubMed
  6. ↵
    Dailey, M., E. Pillemer, I. Weissman. 1982. Protection against syngeneic lymphoma by a long-lived cytotoxic T-cell clone. Proc. Natl. Acad. Sci. USA 79: 5384
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Rosenstein, M., T. Eberlein, S. Rosenberg. 1984. Adoptive immunotherapy of established syngeneic solid tumors: role of T lymphoid subpopulations. J. Immunol. 132: 2115
    OpenUrl
  8. ↵
    Hom, S., S. Topalian, T. Simonis, M. Mancini, S. Rosenberg. 1991. Common expression of melanoma tumor-associated antigens recognized by human tumor infiltrating lymphocytes: analysis by human lymphocyte antigen restriction. J. Immunother. 10: 153
  9. ↵
    Chen, Q., P. Hersey. 1992. MHC-restricted responses of CD8+ and CD4+ T-cell clones from regional lymph nodes of melanoma patients. Int. J. Cancer 51: 218
    OpenUrlPubMed
  10. ↵
    Schendel, D. J., R. Oberneder, C. S. Falk, P. Jantzer, S. Kressenstein, B. Maget, A. Hofstetter, G. Riethmuller, E. Nossner. 1997. Cellular and molecular analyses of major histocompatibility complex (MHC) restricted and non-MHC-restricted effector cells recognizing renal cell carcinomas: problems and perspectives for immunotherapy. J. Mol. Med. 75: 400
    OpenUrlCrossRefPubMed
  11. ↵
    Kearney, E., K. Pape, D. Loh, M. Jenkins. 1994. Visualization of peptide-specific T cell immunity and peripheral tolerance induction in vivo. Immunity 1: 327
    OpenUrlCrossRefPubMed
  12. ↵
    Zimmerman, C., K. Brduscha-Riem, C. Blaser, R. M. Zinkernagel, H. Pircher. 1996. Visualization, characterization and turnover of CD8+ memory T cells in virus-infected hosts. J. Exp. Med. 183: 1367
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Butz, E. A., M. J. Bevan. 1998. Massive expansion of antigen-specific CD8+ T cells during an acute virus infection. Immunity 8: 167
    OpenUrlCrossRefPubMed
  14. ↵
    Kedl, R. M., M. F. Mescher. 1997. Migration and activation of antigen-specific CD8+ T cells upon in vivo stimulation with allogeneic tumor. J. Immunol. 159: 650
    OpenUrlAbstract
  15. ↵
    Murali-Krishna, K., J. D. Altman, M. Suresh, D. J. D. Sourdive, A. J. Zajac, J. D. Miller, J. Slansky, R. Ahmed. 1998. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 8: 177
    OpenUrlCrossRefPubMed
  16. ↵
    Moore, M., F. Carbone, M. Bevan. 1988. Introduction of soluble protein into the class I pathway of antigen processing and presentation. Cell 54: 777
    OpenUrlCrossRefPubMed
  17. ↵
    Newman, M. J., J.-Y. Wu, B. H. Gardner, K. J. Munroe, D. Leombruno, J. Recchia, C. R. Kensil, R. T. Coughlin. 1992. Saponin adjuvant induction of ovalbumin-specific CD8+ cytotoxic T lymphocyte responses. J. Immunol. 148: 2357
    OpenUrlAbstract
  18. ↵
    Zhou, F., B. T. Rouse, L. Huang. 1992. Prolonged survival of thymoma-bearing mice after vaccination with a soluble protein antigen entrapped in liposomes: a model study. Cancer Res. 52: 6287
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Fenton, R. G., D. D. Taub, L. W. Kwak, M. R. Smith, D. L. Longo. 1993. Cytotoxic T-cell response and in vivo protection against tumor cells harboring activated ras proto-oncogenes. J. Natl. Cancer Inst. 85: 1294
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Minev, B. R., B. J. McFarland, P. J. Spiess, S. A. Rosenberg, N. P. Restifo. 1994. Insertion signal sequence fused to minimal peptides elicits specific CD8+ T-cell responses and prolongs survival of thymoma-bearing mice. Cancer Res. 54: 4155
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Porgador, A., D. Snyder, E. Gilboa. 1996. Induction of antitumor immunity using bone marrow-generated dendritic cells. J. Immunol. 156: 2918
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Brossart, P., A. W. Goldrath, E. A. Butz, S. Martin, M. J. Bevan. 1997. Virus mediated delivery of antigenic epitopes into dendritic cells as a means to induce CTL activity. J. Immunol. 158: 3270
    OpenUrlAbstract
  23. ↵
    Hogquist, K., S. Jameson, W. Heath, J. Howard, M. Bevan, F. Carbone. 1994. T cell receptor antagonist peptides induce positive selection. Cell 76: 17
    OpenUrlCrossRefPubMed
  24. ↵
    Rotzschke, O., K. Falk, S. Stevanovic, G. Jung, P. Walden, H.-G. Rammensee. 1991. Exact prediction of a natural T cell epitope. Eur. J. Immunol. 21: 2891
    OpenUrlCrossRefPubMed
  25. ↵
    Vallera, D. A., P. A. Taylor, S. L. Aukerman, B. R. Blazar. 1993. Antitumor protection from the murine T-cell leukemia/lymphoma EL4 by the continuous subcutaneous coadministration of recombinant macrophage-colony stimulating factor and interleukin-2. Cancer Res. 53: 4273
    OpenUrlAbstract/FREE Full Text
  26. ↵
    Chen, L., P. McGowan, S. Ashe, J. Johnston, Y. Li, I. Hellstrom, K. E. Hellstrom. 1994. Tumor immunogenicity determines the effect of B7 costimulation on T cell-mediated tumor immunity. J. Exp. Med. 179: 523
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Zaleski, M., J. Klein. 1978. Genetic control of the immune response to Thy-1 antigens. Immunol. Rev. 38: 120
    OpenUrlCrossRefPubMed
  28. ↵
    Huang, A. Y. C., A. T. Bruce, D. M. Pardoll, H. I. Levitsky. 1996. Does B7–1 expression confer antigen-presenting cell capacity to tumors in vivo. J. Exp. Med. 183: 769
    OpenUrlAbstract/FREE Full Text
  29. ↵
    Bevan, M. J.. 1976. Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. J. Exp. Med. 143: 1283
    OpenUrlAbstract/FREE Full Text
  30. ↵
    van Seventer, G., Y. Shimizu, K. Horgan, S. Shaw. 1990. The LFA-1 ligand ICAM-1 provides an important costimulatory signal for T cell receptor-mediated activation of resting T cells. J. Immunol. 144: 4579
    OpenUrlAbstract
  31. ↵
    Kuhlman, P., V. Moy, B. Lollo, A. Brian. 1991. The accessory function of murine ICAM-1 in T lymphocyte activation: contributions of adhesion and co-activation. J. Immunol. 146: 1773
    OpenUrlAbstract
  32. ↵
    Vermes, I., C. Haanen, H. Steffens-Nakken, C. Reutelingsperger. 1995. A novel assay for apoptosis: flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled annexin V. J. Immunol. Methods 184: 39
    OpenUrlCrossRefPubMed
  33. ↵
    Rosenberg, S. A., J. C. Yang, D. J. Schwartzentruber, P. Hwu, F. M. Marincola, S. L. Topalian, N. P. Restifo, M. E. Dudley, S. L. Schwarz, P. J. Spiess, et al 1998. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nat. Med. 4: 321
    OpenUrlCrossRefPubMed
  34. ↵
    Otten, G., R. Germain. 1991. Split anergy in a CD8+ T cell: receptor-dependent cytolysis in the absence of interleukin-2 production. Science 251: 1228
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Jenkins, M., R. Schwartz. 1987. Antigen presentation by chemically modified splenocytes induces antigen-specific T cell unresponsiveness in vitro and in vivo. J. Exp. Med. 165: 302
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Quill, H., R. Schwartz. 1987. Stimulation of normal inducer T cell clones with antigen presented by purified Ia molecules in planar lipid membranes: specific induction of a long-lived state of proliferative non-responsiveness. J. Immunol. 138: 3704
    OpenUrlAbstract
  37. ↵
    Mueller, D., M. Jenkins, R. Schwartz. 1989. Clonal expansion vs functional clonal inactivation. Annu. Rev. Immunol. 7: 445
    OpenUrlCrossRefPubMed
  38. ↵
    Rocha, B., H. von Boehmer. 1991. Peripheral selection of the T cell repertoire. Science 251: 1225
    OpenUrlAbstract/FREE Full Text
  39. ↵
    Bhandoola, A., E. Cho, K. Yui, H. Saragovi, M. Greene, H. Quill. 1993. Reduced CD3-mediated protein tyrosine phosphorylation in anergic CD4+ and CD8+ T cells. J. Immunol. 151: 2355
    OpenUrlAbstract
  40. ↵
    Kyburz, D., P. Aichele, D. Speiser, H. Hengartner, R. Zinkernagel, H. Pircher. 1993. T cell immunity after a viral infection versus T cell tolerance induced by soluble viral peptides. Eur. J. Immunol. 23: 1956
    OpenUrlCrossRefPubMed
  41. ↵
    Gaus, H., T. Miethke, H. Wagner, K. Heeg. 1994. Superantigen-induced anergy of Vβ8+ CD4+ T cells induces functional but non-proliferative T cells in vivo. Immunology 83: 333
    OpenUrlPubMed
  42. ↵
    Sundstedt, A., I. Hoiden, J. Hansson, G. Hedlund, T. Kalland, M. Dohlsten. 1995. Superantigen-induced anergy in cytotoxic CD8+ T cells. J. Immunol. 154: 6306
    OpenUrlAbstract
View Abstract
Back to top

In this issue

The Journal of Immunology: 162 (5)
The Journal of Immunology
Vol. 162, Issue 5
1 Mar 1999
Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section