Apoptotic Death of CD8⁺ T Lymphocytes After Immunization: Induction of a Suppressive Population of Mac-1⁺/Gr-1⁺ Cells

Cell lines

CT26.WT, the β-gal-expressing CT26.CL25, and EL4 thymoma and its β-gal-expressing (LacZ transfected) subclone, E22, have been described (18, 19). The first two cell lines are H-2^d, while EL4 and E22 are H-2^b. BSC-1 cells (CCL26, American Type Culture Collection, Manassas, VA) and HeLa S3 (CCL2.2, American Type Culture Collection) were used to prepare all the VV stocks. Cell lines were maintained in culture media consisting of RPMI 1640, 10% heat-inactivated FBS (Biofluids, Rockville, MD), 0.03% l-glutamine, 100 mg/ml streptomycin, 100 mg/ml penicillin, and 50 mg/ml gentamicin sulfate (National Institutes of Health Media Center, Bethesda, MD). CT26.CL25 and E22 were maintained in culture media containing 400 μg/ml G418 (Life Technologies, Grand Island, NY). BSC-1 and HeLa S3 were maintained in DMEM.

Recombinant vaccinia viruses

All rVV used in this study were generated by insertion of the foreign genes into the VV thymidine kinase gene by homologous recombination (20). Virus preparations were propagated from plaque-purified crude virus stocks, as described previously (21). Briefly, 175-cm² flasks of HeLa S3 or BSC-1 cells were infected with 5 plaque-forming units (PFU)/cell and incubated at 37°C for about 72 h. Infected cells were harvested and centrifuged at 1000 × g for 10 min. Cells were resuspended in 10 mM Tris (pH 9.0) and lysed by 30 strokes of a dounce homogenizer. Nuclei and cell debris were partially removed by centrifugation for 5 min at 1000 × g, viral particles were collected after purification by centrifugation over a sucrose cushion, and stocks were aliquoted and stored at −80°C.

Viral concentrations were determined by plaque titration on BSC-1 cells. rVV used in a single experiment were titered concurrently to maximize accuracy. Preparation of rVV expressing the influenza A/PR/8/34 nucleoprotein (NP), NP-rVV, was previously described (22). Murine IL-2 cDNA was amplified by PCR from pBMGNeomIL2 and ligated into the SmaI-BamHI site of vaccinia expression vector, pMJ601, which contains the lacZ gene under the control of the natural p_7.5 early promoter (18). IFN-γ was inserted into the VV genome using a similar procedure, as reported (23). In the VJS6 construct, the Escherichia coli lacZ gene was under the control of the natural p_7.5 early/late promoter element from plasmid pSC65 (24); this construct was named β-gal-rVV for simplicity. Wild-type (WT) VV strain WR was kindly provided by J. Yewdell and J. Bennink (National Institute of Allergy and Infectious Diseases, Bethesda, MD).

Peptides

The following synthetic peptides were synthesized by Peptide Technologies (Washington, D.C.) to a purity of greater than 99% as determined by HPLC and amino acid analysis: TPHPARIGL (amino acids 876–884 of β-gal, H-2L^d-restricted (25)), DAPIYTNV (amino acids 96–103 of β-gal, H-2K^b-restricted, (26)).

Antibodies

FITC- or phycoerythrin-labeled mAb recognizing mouse CD8, CD4, CD11b (Mac-1), Lyt-6G (Gr-1), and the isotype-matched controls were purchased from PharMingen (San Diego, CA). Concentrations used for cytofluorometry ranged between 1 and 12 μg/10⁶ cells depending on the Ab. The mAb 24G.2 (CD16/CD23, PharMingen), which reacts with a common epitope of the extracellular domain of the mouse FcγRII/FcγRIII, was used to block the nonspecific binding of mAb during staining. For the in vitro and in vivo depletion studies, mAb were extensively dialyzed against PBS to remove the sodium azide.

Evaluation of CTL responses

Eight- to 12-wk-old female BALB/c, (Animal Production Colonies, Frederick Cancer Research Facility, National Institutes of Health, Frederick, MD), C57BL/6J, or MRL-lpr/lpr mice (The Jackson Laboratory, Bar Harbor, ME) were immunized with various doses (5 × 10⁶ to 2 × 10⁷ PFU/mouse) of different rVV. The spleens were collected on day 6 after immunization, separated into a single-cell suspension, and tested for their ability to lyse β-gal-positive targets in a 6-h ⁵¹Cr release assay as previously described (18). Briefly, 2 × 10⁶ target cells were incubated with 200 μCi Na⁵¹CrO₄ (⁵¹Cr) for 90 min (together with 1 μg/ml of peptide or 100 μl of crude VV-WT preparation, where designated). After labeling, the targets were washed and diluted to 10⁵ viable cells/ml. Targets were then plated at 0.1 ml/well in 96-well plates (10⁴ cells/well) and effectors were added at the indicated ratio. Plates were incubated for 6 h before harvesting. The amount of ⁵¹Cr released was determined by γ-counting and the percentage of specific lysis was calculated from triplicate samples using the formula: [(experimental cpm − spontaneous cpm)/(maximal cpm − spontaneous cpm)] × 100. In some experiments, splenocytes, homogenized to a single cell suspension, were cultured at 5 × 10⁶ cells/ml in 75-cm² flasks (Costar, Cambridge, MA) with 30 ml of RPMI 1640 containing 10% FCS (Biofluids), 0.1 mM nonessential amino acids, 1 mM sodium pyruvate (Biofluids), and 5 × 10⁻⁵ M 2-ME (Life Technologies, Rockville, MD). For the in vitro stimulation, the peptide (1 μg/ml) or irradiated tumor cells at a responder-to-stimulator ratio of 40:1 were added to the cultures. After 6 days, effectors were harvested and tested in a 6-h ⁵¹Cr release assay, as indicated above. In cell separation experiments, splenocytes were cultured at the same cell concentration in 24-well plates (Costar) containing a culture chamber insert with 0.4-μM pores (Millipore, Bedford, MA).

In vivo studies

In protection studies, BALB/c mice (5/group) were immunized with 5 × 10⁶ PFU of rVV, boosted on day 6 with various Ags, and inoculated i.v. 21 days after the boosting with 5 × 10⁵ tumor cells. Negative controls were always included and consisted of mice inoculated with only the vehicle used to resuspend the Ag. Mice were sacrificed on day 12 following tumor inoculation and lung metastases were enumerated in a coded, blind fashion.

Detection of apoptosis in lymphocyte subpopulations

A modification of the method described by Sherwood and Schimke (27) was employed. Briefly, 10⁶ cells were stained with a FITC-conjugated anti-CD8 or anti-CD4 mAb (PharMingen) for 30 min at 4°C in FACS buffer (Ca²⁺- and Mg²⁺-free HBSS containing 0.5% BSA and 0.02% sodium azide). Cells were washed three times in cold FACS buffer and fixed with the addition of 70% ice-cold ethanol. After 1 h incubation at 4°C, cells were washed twice with PBS and resuspended in propidium iodide (PI) staining solution (PBS containing 100 μg/ml RNase A and 50 μg/ml PI, both from Sigma, St. Louis, MO). Flow cytometry was performed on a Becton Dickinson (San Jose, CA) FACScan using a 488 Argon Laser. Data analysis was performed on either the Cellfit (Becton Dickinson) or Modfit (Verity Software House, Topsham, ME) software packages. In some experiments, PI staining was compared with staining with the TUNEL method used by the in situ cell death detection kit (Boehringer Mannheim, Indianapolis, IN), obtaining comparable results (not shown).

Isolation of splenic populations

A panning technique employing flasks coated with mouse anti-rat Abs (T-25 AIS MicroCell, AIS, Santa Clara, CA) was used to deplete specific populations from spleens. Spleens were depleted of red cells with ACK lysis buffer (Biofluids) and resuspended in HBSS containing 1 mM EDTA and 10% mouse serum (HBSS-EDTA-MS). The remaining cells were placed on ice, incubated for 30 min with the primary Ab at a concentration of 10 μg/10⁷ cells, and washed three times with cold HBSS-EDTA. Next, the cells were resuspended in ice-cold HBSS-EDTA-MS at a concentration of 4 × 10⁷/ml, transferred to flasks coated with the secondary Ab (anti-rat), and incubated for 1 h at 4°C. Finally, the nonadherent cells were dislodged and collected as a negative fraction. Fluorescence labeling confirmed >95% depletion. The cells that remained attached to the flasks were provided with RPMI 1640 complete medium and incubated for further studies. Adherent cells were recovered by gentle scraping in Versene solution (1:5000, Biofluids). In some experiments, 2 h adherence to plastic or ingestion of carbonyl iron (Myloclear, Cedarlane Laboratories, Accurate Chemical and Scientific Corp., Westbury, NY) followed by magnetic sorting were used to deplete spleens of mature monocytes/macrophages. CD8⁺ lymphocytes were separated through affinity columns (R & D Systems, Minneapolis, MN) according to the manufacturer’s instructions. The percentage of CD8⁺ cells after enrichment was usually 80–90%.

Statistical analysis

The Wilcoxon-Mann-Whitney U test was used to examine the null hypothesis of identity of ranks between two sets of data. All the p values were reported as two-sided.

The “mirror image” result: the most powerful immunogens elicit the weakest secondary immune responses

In an effort to increase the immunogenicity of recombinant anti-cancer vaccines, we previously inserted a variety of cytokine genes into the genome of a β-gal-expressing rVV (18). Insertion of the gene-encoding IL-2 into a construct-encoding β-gal significantly augmented primary cytolytic T lymphocyte responses specific for VV. Conversely, insertion of the gene-encoding IFN-γ had the opposite effect of decreased CTL activity (Fig. 1⇓A). However, when an aliquot of the same splenic preparation was evaluated for the generation of β-gal-specific cytotoxicity after a 6-day incubation in vitro with the β-gal-immunodominant peptide (TPHPARIGL), a nearly opposite result was obtained in which IL-2-rVV-treated mice showed greatly diminished CTL responses, while IFN-γ-rVV-treated mice exhibited potent cytotoxicity (Fig. 1⇓C). Thus, while splenocyte cultures derived from mice that had been primed in vivo with rVV containing the genes for IL-2 and β-gal showed no cytotoxicity against β-gal-positive tumor cells upon restimulation with the β-gal peptide, cultures from mice infected with virus-encoding IFN-γ with β-gal displayed excellent killing capabilities. Mice immunized with β-gal-rVV, which did not encode any cytokine, showed only a weak cytolytic response.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Immunization with powerful immunogens induces a profound loss of cytolytic capacity upon restimulation with peptide. IL-2-rVV immunization augments primary cytolytic responses. Two BALB/c mice were immunized i.v. with 5 × 10⁶ PFU/mouse of different rVV. After 6 days, the spleens were removed, pooled, and tested in a 6-h ⁵¹Cr release assay against CT26.WT tumor cells infected with VV-WT (A) or pulsed with the β-gal peptide (B). Secondary responses are lost in mice immunized with IL-2-rVV. The splenocytes used for the previous experiment were further incubated in vitro with 1 μg/ml of the synthetic β-gal peptide for 6 days and then assayed in a 6-h ⁵¹Cr release assay against the CT26.CL25 β-gal-positive clone (C) or the CT26.WT cells pulsed with the β-gal peptide (D). Purification of CD8⁺ cells restores specific reactivity in mice immunized with IL-2-rVV. CD8⁺ cells were enriched through affinity columns from splenocytes used in Fig. 1⇓C and D. This enriched population was admixed with naive, CD8-depleted splenocytes from syngeneic mice to constitute about 10% of total cells in culture. After a 6-day in vitro restimulation with β-gal peptide, the mixture was tested in a 6-h ⁵¹Cr release assay against CT26.WT cells pulsed with the β-gal peptide (E). Cytotoxicity toward the CT26.WT cells or the irrelevant E22 target cells was always <5% even at the highest E:T ratio (not shown). The E:T ratio was 33:1, then diluted threefold (11:1, 4:1, 1:1). The experiment was repeated four times with similar results.

Similar effects were observed using an L^d-restricted β-gal peptide pulsed onto target cells. While the addition of the gene-encoding IL-2 greatly enhanced peptide-specific cytolytic responses of fresh splenocytes (Fig. 1⇑B), restimulation of these same splenocytes with peptide for 6 days in vitro resulted in the complete disappearance of cytotoxic activity (Fig. 1⇑D). On the other hand, no primary β-gal-specific cytolysis was observed in mice primed with the IFN-γ-containing virus, but splenocytes from these mice were specifically cytolytic after identical stimulation with peptide (Fig. 1⇑, B vs D). The phenomenon of apparent suppression of cytolytic activity was observed only upon “early” restimulation: when splenocytes were harvested 14 days after immunization with the panel of rVV-containing cytokine genes and stimulated with peptide in vitro, all groups mounted efficient CTL responses (data not shown).

A trivial explanation for the apparent lack of CD8⁺ T cell function was “fratricide,” i.e., the killing of one lymphocyte by another after the addition of soluble peptide. However, the cytolytic capacity of IL-2rVV-primed splenocytes was also eliminated by restimulation with tumor cells that had been transfected with the β-gal gene (CT26.CL25) rather than pulsed with peptide (not shown).

Lack of a cytolytic response is not associated with irreversible damage of CD8⁺ T cells

Induction of an unresponsive state by hyperstimulation of the immune response has been explained in other models by several mechanisms such as anergy, propriocidal apoptosis, and clonal exhaustion, which all focus on properties inherent to T lymphocytes. We thus set out to separate and characterize the CD8⁺ T lymphocytes using negative separation methods that eliminated other cells from the cultures, without ligation of differentiation Ags on the CD8⁺ cells. However, when purified populations of CD8⁺ lymphocytes derived from day 6 spleens of IL-2rVV-inoculated mice were cultured together with naïve splenocytes depleted of CD8⁺ cells and pulsed with β-gal peptide, excellent CTL responses were observed (Fig. 1⇑E). Indeed, comparable responses were generated irrespective of the rVV inoculated 6 days before the cultures were established with the expected exception of the β-gal-negative virus, NP-rVV. Thus, the suppression observed was not a characteristic of the T cells, but was dependent on some other element(s) in the splenic population.

Suppression of CD8⁺ T lymphocyte activity occurs in vivo and is long-lived

To evaluate the effects of “early” boosting on the suppression of immune responses in vivo, we employed a tumor challenge model. We have previously established the importance of CD8⁺ T cells in the effective immune response to CT26.CL25 (19, 28). Mice were immunized with either carrier (HBSS) or with rVV-encoding β-gal in combination with IL-2 or IFN-γ, boosted 6 days later with β-gal protein, then challenged 3 weeks after the boost with a syngeneic murine carcinoma-expressing β-gal. The protection from tumor challenge was nearly complete in mice receiving an initial inoculum of IFN-γ-rVV (Fig. 2⇓A). However, while the IL-2-rVV was also protective when boosted with an irrelevant protein, OVA, tumor protection was abrogated in mice that were initially primed with the IL-2-rVV and boosted with β-gal protein. Similar results were obtained in IL-2-rVV-immunized mice when a recombinant fowlpox virus (FPV)-expressing β-gal was used as a booster (Fig. 2⇓B). FPV-β-gal and, to a lesser extent, the β-gal protein were immunogenic because they induced protection from tumor challenge when used alone (p = 0.001 and 0.005, respectively, Fig. 2⇓, A and B). Thus, the suppression could be observed in vivo, was Ag-specific, and was relatively long-lived.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Impairment of immune response after boosting occurs in vivo and is long lasting. Five different BALB/c mice in each treatment group were injected i.v. with 5 × 10⁶ PFU of the rVV indicated at the bottom of the boxes. HBSS alone was used as vehicle control. Six days later, mice were boosted with two i.v. injections of PBS alone (not shown) or PBS containing 400 μg of either OVA or β-gal protein (top). Alternatively, mice were boosted with a single i.v. inoculation of 10⁷ PFU/mouse of wild-type (FPV-WT) or β-gal-expressing FPV (bottom). On day 21 following the in vivo restimulation, mice were challenged i.v. with 5 × 10⁵ CT26.CL25 cells to establish pulmonary metastases. Twelve days later, mice were sacrificed, and pulmonary nodules were counted in a coded and blinded fashion. Two separate, independent experiments were performed under identical experimental conditions with overall similar results. An arbitrary value of 250 was assigned when metastases were too numerous to count. Data represent the mean ± SD of pulmonary metastases.

CD8⁺ T cells from unresponsive cultures die an apoptotic death

We hypothesized that the unresponsive state observed above could be due to anergy or apoptosis. To explore the death rate under various conditions of restimulation, we employed a double-staining protocol designed to evaluate the percentage of hypodiploid (apoptotic) cells among lymphocytes that were positive for CD8 (Fig. 3⇓). In splenocyte cultures from mice primed with V69 rVV, the mean percentage of apoptosis upon in vitro stimulation was 22.4%. This baseline value of apoptosis was not Ag specific, because V69 encoded the influenza NP gene in lieu of LacZ (NP-rVV) and did not prime CTL responses (Fig. 1⇑). The number of apoptotic cells after stimulation with antigenic peptide rose to 46.0% in cultures derived from mice that had been immunized with VJS6, which encodes β-gal without the heterologous addition of cytokines. Levels of apoptosis comparable to negative controls were detected in cultures derived from IFN-γ-rVV immunized mice. Apoptosis of CD8⁺ T cells was significantly increased in mice immunized with the IL-2-rVV. Similar patterns of apoptotic death was observed in CD8⁺ cultures that were restimulated with tumor cells expressing the β-gal Ag; however, CD4⁺ cells present in the same experiment did not follow the same pattern of apoptosis (not shown). Thus, apoptotic death appeared to be limited to the CD8⁺ compartment and was not a generalized death of all lymphocytes present in the culture.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

CD8⁺ T lymphocytes die an apoptotic death after immunization with IL-2-rVV in vivo and peptide restimulation in vitro. The cells used in this experiment are from the splenocytes used in the experiment shown in Fig. 1⇑, C and D. Two spleens from BALB/c mice immunized 6 days earlier with different rVV were stimulated in vitro for 6 days with 1 μg/ml of β-gal peptide. One million cells were stained with FITC-conjugated mAb against the CD8 Ag. After staining, cells were fixed in ethanol and exposed to propidium iodide solution to allow detection of hypodiploid nuclei. Figures indicate the percentage of hypodiploid cells among CD8⁺ lymphocytes. The apoptosis percentage in cultures of spleens from mice immunized with NP-rVV, V69, was used as the background line. The experiment was repeated two additional times with similar results.

Suppression of CD8⁺ T cell function depends on membrane contact and is not Fas (CD95)-mediated

We next investigated the molecular mechanism causing apoptosis of CTL in acutely infected mice. The production of soluble factors secreted by the suppressive splenocytes was ruled out by mixing splenocytes from mice infected with different rVV in diffusion chambers. The experiments clearly showed that cell-cell contact was required for the immunosuppressive effect, because suppression of CTL generation only occurred when the suppressor-containing population was not separated by a membrane from the population responsive to peptide stimulation (Fig. 4⇓).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Suppression of CD8⁺ T cell function requires cell-cell contact. A suspension of three spleens from BALB/c mice infected 6 days earlier with 5 × 10⁶ PFU/mouse of IFN-γ-rVV was incubated with a β-gal peptide either alone (control) or together with splenocytes of mice similarly infected with IL-2-rVV (1:1). The latter were either admixed in the same well (cell contact) or separated by a semipermeable membrane (no cell contact). After 6 days, CTL activity in the cultures was assayed against the same panel of target cells used in previous experiments. For simplicity, only the cytolytic activity against CT26.CL25 at an E:T ratio of 10:1 is shown. Control wells containing 1:1 mixture of splenocytes from IFN-γ-rVV-infected mice and normal mice did not show any difference from control culture whether they were admixed or separated by the insert membrane (not shown).

We tested whether Fas-Fas ligand (FasL) interactions were responsible for the observed cell membrane-associated suppression as this interaction is a primary inducer of apoptosis in activated lymphocytes. To elucidate this, we immunized MRL-lpr/lpr mice, which do not express a functional FasR (CD95) (Fig. 5⇓, top), with IL-2-rVV or IFN-γ-rVV. As before, primary and secondary responses against the tumor cell line, EL-4, VV-WT infected EL-4, or the β-gal-expressing line, E22, were tested in ⁵¹Cr release assays (Fig. 5⇓, bottom). An identical pattern of suppression was observed in the Fas-deficient mice as in normal C57BL/6J mice. This generation of the same IL-2-rVV-induced suppression in MRL-lpr/lpr mice indicated that the cell membrane-associated mechanism did not involve Fas-FasL interaction.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Inhibition of CD8⁺ T cell killing is not Fas-dependent. Lymphoblasts obtained from ConA-stimulated splenocytes from C57BL/6J and MRL-lpr/lpr mice were evaluated for CD95 (Fas/Apo1) expression by cytofluorometric analysis (A, left and right, respectively). Mice were immunized i.v. with 5 × 10⁶ PFU/mouse of IL-2-rVV or IFN-γ-rVV (B). Six days later, the spleens from three different mice were removed, pooled, and tested in a 6-h ⁵¹Cr release assay against EL-4, EL-4 tumor cell line infected with VV-WT, or the β-gal-expressing line, E22 (Primary, top). The splenocytes used for the previous experiment where further incubated in vitro with 1 μg/ml of the H-2K^b-restricted, synthetic β-gal peptide for 6 days and then assayed in a 6-h ⁵¹Cr release assay against CT26.CL25, EL-4, or E22 (Secondary, bottom). The experiment was repeated with similar results.

An increase in Mac-1 and Gr-1 double-positive cells correlates with the induction of primary cytolytic T cells

The lack of response in the whole spleens taken from IL-2-rVV-infected mice could be explained with either the absence of competent APC or with the presence of regulatory suppressor elements distinct from CD8⁺ lymphocytes. The first possibility was addressed by mixing experiments in which we found that splenocytes from IL-2-rVV-infected mice could suppress the response of IFN-γ-rVV mice even when mixed at a 1:2 ratio before culture with β-gal peptide (not shown). Additionally, Fig. 4⇑ shows a similar mixing experiment of splenocytes from IL-2-rVV-infected mice and IFN-γ-rVV-infected mice, which demonstrates the same pattern of suppression. These findings disfavored the hypothesis of insufficient APC function and supported the presence of suppressive elements.

To identify candidate cells with suppressive activity, we cytofluorometrically evaluated surface markers on the spleen cells of naive mice or mice immunized with IL-2-rVV or IFN-γ-rVV. As can be seen in the upper panels of Fig. 6⇓, both Mac-1 and Gr-1 were present on the surfaces of a population of cells that was increased in the spleens of mice that had been immunized with IL-2-rVV. Gr-1 is a marker that is normally expressed by granulocytes, monocytes, and immature myeloid precursors in the bone marrow but is expressed at a very low level in the spleens of normal mice (29). This population of cells expressing both Ags was significantly increased (p < 0.01) in the spleens of IL-2-rVV-immunized mice but not in mice immunized with IFN-γ-rVV (means ± SD of three experiments were: 3.03 ± 0.87, 2.92 ± 0.22, and 7.93 ± 0.68 for naïve, IFN-γ-rVV- and IL-2-rVV-inoculated mice, respectively). Moreover, in vivo depletion with anti-Gr-1 mAb but not with the control mAb reduced the double-positive cells in the spleens of mice inoculated with IL-2-rVV to levels comparable to those detected in naïve or in IFN-γ-rVV-immunized mice. The depletion was even more effective in vitro (Fig. 6⇓, bottom).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Immunization with IL-2-rVV results in an increase in the percentage of Mac-1⁺ and Gr-1⁺-double-positive splenocytes. Splenocyte pools used in the experiment shown in Fig. 7⇓ were stained with FITC-anti-Gr-1 and phycoerythrin-anti-Mac-1 mAbs in the presence of an Ab blocking the Fcγ-RII/Fcγ-RIII (2.4G2). Upper panels show unseparated splenocytes from unimmunized mice (naïve) or mice immunized 6 days earlier with either IFN-γ-rVV or IL-2-rVV. Lower panels show the IL-2-rVV-immunized splenocytes after depletion with the indicated mAbs. Not shown are the results of the in vitro depletion with IgG2b mAb used as negative control. As for the in vivo data, depletion with this mAb did not change the percentages of double-positive cells seen in untreated controls. A secondary anti-rat IgG Ab revealed the number of positive cells to be <1% after depletion, confirming that the negativity was not due to Ab competition. Data are representative of four separate, independently performed experiments.

Mac-1⁺/Gr-1⁺ cells mediate the elimination of CD8⁺ T cell function in vivo

To ascertain whether these cells were suppressing the generation of CTL in our cultures, different mAbs were used to deplete, in vitro or in vivo, specific populations before restimulation of the spleens of mice infected 6 days earlier with rVV immunogens. As shown in Fig. 7⇓A, depletion of cells positive for Gr-1 or Mac-1 completely restored the capacity of cells to mount cytolytic responses. Cytotoxicity recovered in Ab-depleted cultures was similar to that obtained with a population of CD8⁺ lymphocytes enriched from the same spleens or with the other positive controls included in the assay (IFN-γ-rVV-immunized mice or mice immunized 14 days earlier with IL-2-rVV). An isotype-matched Ab (rat IgG2b) did not produce the same effect, ruling out an in vitro artifact related to the experimental protocol. Moreover, a second Ab directed against a mouse macrophage Ag, Mac-3, only restored 20% of the CTL activity seen in the cultures of CD8⁺ enriched populations (not shown). The data shown in Fig. 7⇓A further supports the hypothesis that the suppression of CD8⁺ cell responses is not a characteristic of the T cells, but rather is a quality of some other splenocyte component. In vivo depletion of Gr-1⁺ cells by repeated i.p. inoculations of the Ab during the first days of infection with IL-2-rVV resulted in a complete recovery of the deficient CTL response, while no effect was observed with the control IgG2b (Fig. 7⇓B). Thus, Mac-1⁺/Gr-1⁺ cells appear to mediate the suppression of CD8⁺ T cells in vitro and in vivo.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

In vitro and in vivo depletion of either Mac-1⁺ or Gr-1⁺ cells restores CTL response in splenocyte cultures from mice immunized with IL-2-rVV. A, In vitro depletion of Mac-1⁺/Gr-1⁺ cells restores the immune response. Two to five BALB/c mice/group were inoculated with HBSS (naïve), 5 × 10⁶ PFU of IL-2 rVV, or 5 × 10⁶ PFU of IFN-γ-rVV. After 6 days, spleens were pooled together and depleted in vitro of Mac-1⁺ or Gr-1⁺ cells. IgG2b isotype-matched mAb was used as negative control of the depletion protocol. Splenocytes collected after depletion were cultured with β-gal peptide as described previously. The same panel of target cells was used to evaluate the cytolytic activity (A). As positive controls, the following experimental groups were also included: CD8⁺ splenocytes from mice inoculated with IL-2-rVV and cultured together with CD8-depleted naïve spleens (∗); mice inoculated 6 days earlier with IFN-γ-rVV; and mice inoculated with IL-2-rVV 14 days before the in vitro restimulation. Data with E22 tumor cells were negative and are not shown for simplicity. The E:T ratio is 100:1 followed by threefold dilutions (33:1, 11:1, 4:1, 1:1, 0.3:1). The experiment was repeated four more times with similar results. B, In vivo depletion of Mac-1⁺/Gr-1⁺ cells restores the immune response. The experiment was performed as in A, with the exception that 200 μg Gr-1 mAb and the IgG2b controls were injected i.p. into mice on day 2 and 4 following immunization with the rVV (B).