SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

We have reported for the first time the significance of effector T-cell multifunctionality in antitumor immunity, suggesting that the appearance of multifunctional/polyfunctional tumor-specific CD8+ T cells in vivo is a critical determinant of the success of antitumor immunotherapy, and a strategy to induce multifunctionality in effector cells is required for the successful immunotherapy of hosts with progressing tumor. Glucocorticoid-induced tumor necrosis factor receptor (GITR) stimulation has been shown to enhance antitumor immune response. However, its functional impact on adoptively transferred T cells remains unclear. Here, we analyzed the impact of GITR stimulation in vivo on the functional profiles of adoptively transferred CD8+ T cells specific for murine fibrosarcoma CMS5. GITR stimulation was found to enhance multifunctionality (interferon (IFN)-γ and tumor necrosis factor (TNF)-α production and CD107a mobilization as a degranulation marker) in transferred cells at the single-cell level. These cells exhibited upregulated expression of CD25 in draining lymph nodes and increased infiltration in tumor. Mice that received T-cell therapy with GITR stimulation showed reduced Foxp3+CD4+ T cells among tumor infiltrating lymphocytes and increased in vivo cytotoxic T lymphocytes (CTL) activity even with progressing tumor, resulting in enhanced tumor regression. These data strengthen the idea that effector T-cell multifunctionality is a sensitive immune correlate for successful immunotherapy against malignancy and provide an immunological rationale for effective T-cell therapy combined with GITR stimulation. (Cancer Sci 2009; 100: 1317–1325)

Abbreviations:
ACT

adoptive cell transfer

APC

allophycocyanin

CFSE

5-(and-6)-carboxyfluorescein succinimidyl ester (CFSE)

CTL

cytotoxic T lymphocytes

DLN

draining lymph node

FcR

Fc receptor

GITR

glucocorticoid-induced tumor necrosis factor receptor

GITRL

glucocorticoid-induced tumor necrosis factor receptor ligand

IFN

interferon

mERK

mutated mitogen-activated protein kinase

NDLN

non-draining lymph node

NK

natural killer

PE

phycoerythrin

TCR

T cell receptor

TIL

tumor infiltrating lymphocytes

TNF

tumor necrosis factor

Treg

regulatory T cells

Multifunctionality is the ability of T cells to exhibit multiple effector functions such as the production of multiple cytokines, chemokines, or molecules to mediate cytotoxicity at the single-cell level. Multifunctional T cells are beginning to receive increasing appreciation as an important and sensitive immune correlate for the immunological control of infectious disease. The importance of T-cell multifunctionality has been reported in preclinical infection models,(1,2) as well as in humans reactive to human immunodeficiency virus (HIV), cytomegalovirus, hepatitis B virus, or tuberculosis.(2–9) These analyses were made possible by recent advances in the parameters that can be detected simultaneously with polychromatic flow cytometry.(10) T-cell multifunctionality in tumor immunity, however, is only starting to be considered.

We have demonstrated for the first time the significance of T-cell multifunctionality in tumor immunity, showing that tumor progression inhibited the induction of multifunctionality in adoptively transferred tumor-specific CD8+ T cells via induction of Treg, resulting in unsuccessful tumor eradication.(11) We used an adoptive T-cell therapy model utilizing a TCR transgenic mouse in which CD8+ T cells express an H-2Kd-restricted transgenic TCR specific for the syngeneic CMS5 fibrosarcoma-derived endogenous tumor-rejection antigen, mutated extracellular signal-regulated kinase (mERK2).(12,13) In a previous study, we found that adoptively transferred tumor-specific T cells failed to acquire a multifunctional phenotype (as assessed by the production of cytokines such as IFN-γ and TNF-α and mobilization of CD107a as a cytotoxic degranulation marker) when transferred into hosts with progressing tumor resulting in unsuccessful tumor eradication. Depletion of Treg prior to tumor challenge facilitated the induction of high multifunctionality of transferred cells even in hosts with progressing tumor, leading to enhanced tumor regression. T cells with high multifunctionality harvested from hosts with successful therapy induced tumor regression when retransferred into the tumor-bearing hosts, indicating that multifunctional tumor-specific CD8+ T cells were responsible for tumor regression. These data suggest that a strategy to induce multifunctionality in transferred cells via activation of effecter cells and/or inhibition of Treg is required for an effective adoptive T-cell therapy in hosts with progressing tumor.

GITR is a type I transmembrane protein with homology to TNF family members.(14) GITR is constitutively expressed at high levels on Foxp3+CD4+ Treg and at low levels on resting conventional CD4+ and CD8+ T cells, and is upregulated following their activation.(15–17) GITR is a receptor that provides a costimulatory signal for conventional CD4+ and CD8+ T cells to enhance their proliferation and effector function with promotion of their resistance to Treg suppression.(17–22) Additionally, signaling through GITR was reported to directly abrogate the suppressive function of Treg.(15,16) Engagement of GITR in vivo by agonistic mAb (DTA-1) or its physiological ligand, GITRL, has been reported to enhance antitumor immunity in a number of models including our report.(22–28) However, the functional impact of GITR stimulation on adoptively transferred tumor-specific T cells remains unclear.

Here, we questioned whether effector T-cell multifunctionality could predict the control of a tumor when we intensified the immunotherapy. We examined the impact of GITR stimulation in vivo on the functional profiles of adoptively transferred tumor antigen-specific CD8+ T cells. We combined GITR stimulation in vivo with adoptive T-cell therapy, and addressed the relationship of the multifunctionality in tumor-specific effector T cells to their activation phenotype, tumor infiltration, Treg ratio among TIL, in vivo CTL activity, and tumor rejection.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Mice.  Studies were conducted on 7–10-week-old female BALB/c mice (CLEA Japan, Osaka, Japan). DUC18 mice, transgenic for αβ-TCR reactive with Kd-restricted 136–144 peptide of mERK2 were established as described previously,(13) and used at 7–10 weeks of age. Mice were maintained at the Animal Center in Mie University Graduate School of Medicine. The experimental protocol was approved by the Ethics Review Committee for Animal Experimentation of Mie University Graduate School of Medicine.

Cell lines.  CMS5 is a methylcholanthrene-induced sarcoma cell line of BALB/c origin expressing mERK2.(12) P1.HTR is a subline of mastocytoma P815 of DBA/2 origin.(29) CMS5 and P1.HTR do not express GITRL (data not shown).

Antibodies and reagents.  Anti-GITR mAb (DTA-1; rat IgG2a) was produced from a hybridoma and purified with protein G columns for in vivo administration. R-PE-conjugated anti-TCRVβ8.3 (1B3.3), PerCP-Cy5.5-conjugated anti-CD4 (RM4-5), PerCP-Cy5.5-conjugated anti-CD8 (53-6.7), APC-Cy7 conjugated anti-CD25 (PC61), APC-conjugated anti-IFN-γ (XMG1.2), and PE-Cy7 conjugated anti-TNF-α (MP6-XT22) mAb were purchased from BD Bioscience (San Diego, CA, USA). PE-conjugated anti-CD107a (1D4B), PE-conjugated anti-Foxp3 (FJK-16s), APC-conjugated anti-CD25 (PC61) and anti-GITR (DTA-1) mAb were purchased from eBioscience (San Diego, CA, USA). APC-anti-CD69 mAb (H1.2F3) was purchased from BioLegend (San Diego, CA, USA). 5-(and-6)-carboxyfluorescein succinimidyl ester (CFSE) was purchased from Molecular Probes (Eugene, OR, USA). Synthetic mERK2 9m peptide QYIHSANVL(12) and HER2 oncoprotein-derived HER2p63–71 (T) peptide TYLPTNASL(30) have been described and were purchased from Qiagen (Hilden, Germany).

Tumor challenge.  CMS5 tumor cells (1 × 106 in 0.2 mL PBS) were inoculated subcutaneously into the right flank of mice. Tumor size represented the average of two orthogonal diameters measured with calipers every other day, until the mean diameter of tumors reached 20 mm according to institutional guidelines.

Adoptive transfer.  The CD8+ fraction of cells was obtained from the spleens of DUC18 mice using a MACS mouse CD8 isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer's instructions, obtaining a purity of >95% (data not shown). After three washes in PBS, CD8+ cells (1 × 106) were suspended in 0.2 mL PBS and injected intravenously into a lateral tail vein. When indicated, cells labeled with 2.5 µM CFSE were used to monitor T-cell division in vivo. To evaluate the effect of GITR stimulation, 0.25 mg anti-GITR (DTA-1) mAb in 0.2 mL PBS was administrated intravenously.

In vitro stimulation and staining of cells.  Cells harvested from DLN were incubated at an effector/stimulator ratio of two for 2 h at 37°C with irradiated (45 Gy) P1.HTR cells, which had been pulsed with 1 µM 9m peptide, in the presence of 1 µg/mL PE-conjugated anti-CD107a. We then incubated these samples for an additional 5 h after the addition of 1 µL/mL GolgiPlug (BD Bioscience). After stimulation, samples were washed with RPMI-1640 and live cells were enriched with NycoPrep 1.077A (Axis-Shield, Rodeløkka, Norway). After this procedure, less than 0.5% of all cells were propidium iodide-positive (data not shown). Cells were incubated for 15 min at 4°C with anti-CD16/CD32 FcR blockade (93; eBioscience), and then stained with PerCP-Cy5.5-conjugated anti-CD8 and APC-Cy7-conjugated anti-CD25 mAb. After permeabilization and fixation using a Cytofix/Cytoperm Kit (BD Bioscience), cells were stained intracellularly with APC-conjugated anti-IFN-γ and PE-Cy7-conjugated anti-TNF-α mAb.

Flow cytometric analysis.  For polychromatic analysis, between 100 000 and 200 000 CD8+ events were collected for each sample as described previously.(11) For phenotypic analysis, we collected 20 000 CD8+Vβ8.3+ events from each sample. TIL were obtained as described previously.(11) Lymph node cells or TIL were incubated for 15 min at 4°C with anti-CD16/CD32 FcR blockade and stained with cell-surface mAb.

Stained cells were acquired using a FACSCanto flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA), and data were analyzed using FACSDiva (Becton Dickinson) and FlowJo (Tree Star, Ashland, OR, USA) software.

In vivo CTL assay.  BALB/c splenocytes were separated into two populations. One was pulsed with 20 µM 9m peptide, incubated at 37°C for 90 min, and labeled with a high concentration (2.5 µM) of CFSE (CFSEhigh cells). The second population served as a control; these cells were left untreated and labeled with a low concentration (0.25 µM) of CFSE (CFSElow cells). Recipient mice were injected intravenously with 1 × 107 of each target cell population. Twenty hours later, spleen cells were analyzed by flow cytometry for CFSE-positive target cells. The percentage of specific lysis was normalized to the results seen in naive BALB/c mice:

  • Percentage specific lysis = (1 – [{CFSEhighexp/CFSElowexp}/{CFSEhighBALB/c/CFSElowBALB/c}]) × 100.

Statistical analysis.  Data were expressed as the mean + SD. Differences between groups were examined for statistical significance using Student's t-test. A P-value less than 0.05 denoted a statistically significant difference.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Anti-GITR agonistic mAb enhances the antitumor effect of adoptively transferred tumor-specific CD8+ T cells in hosts with progressing tumor.  We reported previously that CMS5 tumor progression limits the efficacy of immunotherapy with DUC18 TCR transgenic mouse-derived CD8+ T cells specific for the CMS5 tumor antigen mERK2.(11) As shown in Table 1, transfer of 1 × 106 DUC18-derived CD8+ T cells 2 days after tumor challenge induced tumor regression in 70% (7/10) of the mice as a cumulative result. However, no effect on tumor rejection (0/13) was observed when T cells were transferred on day 7.

Table 1. Tumor rejection in mice that received adoptive T-cell therapy combined with or without glucocorticoid-induced tumor necrosis factor receptor stimulation
Adoptive cell transferTreatmentIncidence of tumor rejection
Exp. 1Exp. 2Exp. 3Total
Day 2PBS2/55/5NT 7/10 (70%)
Day 7PBS0/50/50/3 0/13 (0%)
Day 7DTA-13/53/56/612/16 (75%)

We examined the antitumor effect of the combination of GITR stimulation and ACT in mice with day 7-established tumors. As shown in Table 1, administration of the anti-GITR agonistic mAb DTA-1 enhanced the effect of DUC18 T-cell transfer on day 7. We observed that 75% (12/16) of total mice from three independent experiments rejected tumors. As shown in Figure 1, a representative experiment, administration of DTA-1 on day 7 without ACT had only a marginal effect on tumor growth compared to the control group treated with PBS. ACT alone had some effect on tumor growth in this experiment, but did not induce tumor regression in any of the mice. In contrast, the combination of DTA-1 administration with ACT inhibited tumor growth in all mice, and induced tumor regression in 60% (3/5) of mice.

image

Figure 1. Glucocorticoid-induced tumor necrosis factor receptor (GITR) stimulation combined with adoptive T-cell therapy induces tumor regression in hosts with progressing tumor. BALB/c mice (n = 5 per group) were inoculated subcutaneously with 1 × 106 CMS5 cells on day 0. On day 7, mice were administrated with the anti-GITR mAb DTA-1 or PBS, and a total of 1 × 106 CD8+ cells purified from splenocytes isolated from DUC18 transgenic mice were adoptively transferred intravenously in the groups indicated as adoptive cell transfer; tumor growth was monitored over time. The mean tumor diameter of each group is represented as the average + SD of five mice. Results are representative of three independent experiments.

Download figure to PowerPoint

GITR stimulation increases the multifunctionality of adoptively transferred CD8+ T cells.  Our previous report indicated that appearance of multifunctional CD8+ effector T cells in vivo is a critical determinant of the success of antitumor immunity and transferred DUC18-derived T cells failed to acquire high multifunctionality when transferred into mice with day 7-established CMS5 tumor.(11) To understand the impact of GITR stimulation on the functional signature of transferred tumor-specific CD8+ T cells, we undertook a comprehensive functional analysis including the examination of multifunctionality in day 7-tumor-bearing mice that received ACT with or without DTA-1 administration. DUC18 mouse-derived CD8+ T cells, which had been labeled with CFSE before transfer into tumor-bearing hosts, were recovered from DLN 3 days after transfer to assess their IFN-γ and TNF-α production and CD107a mobilization using multiparameter flow cytometric analysis.

We first compared the quantities of cytokine production and CD107a mobilization assessed by the mean fluorescence intensity (MFI) of each parameter on the cells (Fig. 2a,b). With administration of DTA-1, IFN-γ and TNF-α production and CD107a expression were higher than those without DTA-1. All three parameters increased with the number of cell divisions determined by CSFE intensity when the mice received ACT with DTA-1, whereas neither TNF-α nor CD107a expression changed significantly according to cell division when treated without DTA-1.

image

Figure 2. Glucocorticoid-induced tumor necrosis factor receptor (GITR) stimulation enhances the multifunctionality of adoptively transferred tumor-specific CD8+ T cells. BALB/c mice (n = 5 per group) were injected subcutaneously with 1 × 106 CMS5 cells. Adoptive transfer of 1 × 106 CFSE-labeled DUC18 CD8+ cells was performed with administration of DTA-1 or PBS on day 7 of tumor challenge. Three days after adoptive cell transfer (ACT), we used polychromatic flow cytometry to determine antigenic peptide 9 m-specific cytokine production and CD107a mobilization by transferred CD8+ T cells harvested from draining lymph nodes. (a) Representative histograms for CFSE and dot plots for CFSE versus IFN-γ, TNF-α, or CD107a are shown. (b) MFI of IFN-γ, TNF-α, and CD107a staining at each cell division for cells derived from mice that received ACT with DTA-1 or PBS are represented as the means + SD of five mice. (c) Responses are grouped and color-coded according to the number of acquired functions (IFN-γ, TNF-α, CD107a). Data are summarized in the pie chart, where each wedge represents the frequency of CFSE+CD8+ cells expressing all three functions as IFN-γ+TNF-α+CD107a+ (3), any two functions including IFN-γ+TNF-α+CD107a, IFN-γ+TNF-αCD107a+, and IFN-γTNF-α+CD107a+ (2), a single function including IFN-γ+TNF-αCD107a, IFN-γTNF-α+CD107a, and IFN-γTNF-αCD107a+ (1), or no function (0). Results are representative of three independent experiments. (d) Functional composition of the CD8+ T-cell response. Every possible combination of responses is shown in the x-axis. Boxes represent interquartile ranges; the line in the box represents the median, and minimum/maximum lines are shown. (e) Analysis of the frequency of each population with different numbers of acquired functions in cells at each cell division, as assessed by CFSE intensity. Results are representative of three independent experiments.

Download figure to PowerPoint

Next, we addressed the multifunctionality of the transferred cells by calculating the relative frequency of several distinct populations: cells with all three functional characteristics, the populations with varying combinations of two functions, the subsets displaying only one function, and those without any of these functions (Fig. 2c). When ACT alone was carried out 7 days after tumor challenge, 14.1% of the transferred T cells acquired all three functions, 33.3% of cells gained two functions, and 45.6% of all cells were monofunctional. In contrast, 38.9% of the transferred T cells acquired all three functions, whereas 20.6% of the cells were monofunctional when ACT was carried out in combination with DTA-1. Therefore, tumor-specific CD8+ T cells gained significantly augmented multifunctionality when transferred with GITR stimulation in vivo.

Among the cells with two functions, IFN-γ+CD107a+ T cells were the major population that increased with GITR stimulation (Fig. 2d), suggesting the importance of these two functions in the antitumor immune response. Among the monofunctional T cells, the major population was IFN-γ+ cells regardless of DTA-1 treatment.

We calculated the relative frequency of each of these multifunctional populations at each cell division subgroup, as determined by CSFE intensity (Fig. 2e). When ACT was carried out with DTA-1 administration, a large number of the transferred T cells gained all three functions from the early phases of cell division; the frequency increased with increasing number of cell divisions. In contrast, the acquisition of multifunctionality was low and did not increase significantly according to cell division in mice that received only ACT on day 7. Therefore, the transferred cells proliferated without acquisition of sufficient functionality in tumor-bearing mice without GITR stimulation.

GITR stimulation-induced multifunctional tumor-specific CD8+ T cells exhibit superior cytokine secretion and cytolytic degranulation.  Several groups including our laboratory have reported that multifunctional CD8+ T cells produce greater amounts of IFN-γ than cells that make IFN-γ alone.(2,6,11) In the present study, we harvested the transferred cells from mice treated with GITR stimulation, and compared the production of IFN-γ, TNF-α, and expression of CD107a among the cells with different functionality (Fig. 3). Multifunctional tumor-specific T cells produced more IFN-γ than cells with bifunctionality or monofunctionality. TNF-α production in cells with all three functions was higher than that seen in cells lacking CD107a expression. The majority of the cells demonstrating TNF-α production also produced IFN-γ; therefore, neither TNF-α+IFN-γCD107a+ nor TNF-α+IFN-γCD107a populations were observed. Similarly, CD107a expression in cells with all three activities was higher than that in cells lacking TNF-α production. Most of the cells exhibiting CD107a expression also produced IFN-γ. These data imply that the superior cytokine secretion and degranulation by multifunctional tumor-specific CD8+ T cells that were induced with GITR stimulation generated a better antitumor response in vivo.

image

Figure 3. Multifunctional T cells exhibit more cytokine production and degranulation than cells with less functionality. MFI of IFN-γ, TNF-α, and CD107a staining of antigen specific tri-, bi-, or mono-functional cells from mice subjected to adoptive cell transfer combined with DTA-1 administration 7 days after tumor challenge are shown. Data represent the mean + SD of five mice. Differences between groups were examined for statistical significance using Student's t-test. *P < 0.01, **P < 0.05.

Download figure to PowerPoint

Transferred T cells showed enhanced proliferation, activated phenotype, and augmented tumor infiltration when combined with GITR stimulation in vivo.  We addressed the relationship between multifunctionality of effector cells and their proliferation, cell surface phenotype, and localization. As shown in Figure 4(a), the transferred T cells entered the cell division in DLN as detected by CFSE dilution, whereas they appeared to stay without division in NDLN. Therefore, the transferred tumor antigen-specific CD8+ T cells selectively proliferated in the tumor DLN. Administration of DTA-1 significantly accelerated the cell division of transferred DUC18 T cells.

image

Figure 4. Glucocorticoid-induced tumor necrosis factor receptor (GITR)-induced effector T-cell multifunctionality is associated with their enhanced proliferation, increased expression of CD25, and augmented infiltration into tumor. (a,b) CFSE-labeled DUC18 CD8+ T cells were adoptively transferred and administered DTA-1 or PBS on day 7 after tumor challenge (five mice per group). (a) Lymphocytes were harvested from draining lymph nodes (DLN), non-draining lymph nodes (NDLN), or tumor, and analyzed for their proliferation and cell surface expression of CD25 or CD69 by flow cytometry. (b) We analyzed 9m peptide-specific cytokine production, CD107a mobilization, and CD25 surface expression by transferred CD8+ T cells harvested from DLN in mice treated with DTA-1 or PBS. MFI of CD25 staining for tri-functional, bi-functional, mono-functional, or non-functional cells are indicated as the mean + SD of five mice. Data show the CD8+Vβ8.3+ population. Results are representative of three independent experiments.

Download figure to PowerPoint

Next, we analyzed the cell surface markers related to cellular activation on the transferred T cells. Upon GITR stimulation, the transferred T cells in DLN upregulated CD25 expression particularly when the cells entered high numbers of cell division (Fig. 4a). Such upregulation was not observed on the cells harvested from mice that were not administrated with DTA-1. DTA-1 did not have a clear influence on the expression of the early phase activation marker CD69.

When we analyzed the transferred cells that infiltrated into tumors, we could detect a small number of these cells when transferred into mice with day 7-established tumor, consistent with our previous report.(11) GITR stimulation markedly enhanced the infiltration of transferred cells into tumor even when ACT was carried out on the mice with progressing tumor (Fig. 4a). The transferred cells in tumor showed well-proliferated and CD25high phenotypes regardless of GITR stimulation, suggesting that these phenotypes correlate with the ability of transferred cells to migrate into tumor.

Cells with high multifunctionality show CD25high phenotype.  As we found that the transferred T cells significantly upregulated the expression of CD25 when combined with in vivo GITR stimulation, we compared the expression of CD25 between the cells with different multifunctionality. As shown in Figure 4(b), highly multifunctional T cells expressed more CD25 than cells with less multifunctionality. Furthermore, administration of DTA-1 upregulated CD25 expression of cells in every group of different multifunctionality. These data suggest that the acquisition of activation-related phenotype is directly associated with the functional enhancement of adoptively transferred cells with GITR stimulation.

GITR stimulation combined with ACT reduces regulatory T cells in tumor.  We have shown that Treg plays an important role in the impaired induction of effector multifunctionality in hosts with progressing tumor.(11) Another report has shown that DTA-1 administration could reduce the ratio of Foxp3+ cells among tumor-infiltrating CD4+ T cells.(24) Therefore, we analyzed the CD4+Foxp3+ Treg in DLN, NDLN, spleen, and tumor in mice with ACT 7 days after tumor challenge with or without DTA-1 administration. As shown in Figure 5, the proportion of Foxp3+ cells among CD4+ cells was higher in tumors than in DLN, NDLN, or spleen in mice that received ACT without GITR stimulation. Combination of DTA-1 with ACT significantly reduced the proportion of Treg in tumor. The proportion of Treg in DLN, NDLN, and spleen did not change with DTA-1 administration in contrast to the marked reduction among tumor-infiltrating CD4+ T cells.

image

Figure 5. Glucocorticoid-induced tumor necrosis factor receptor (GITR) stimulation combined with adoptive cell transfer (ACT) reduces the ratio of Foxp3+ cells among tumor infiltrating CD4+ T cells. (a,b) BALB/c mice (n = 5 per group) were inoculated subcutaneously with 1 × 106 CMS5 cells on day 0, and received 1 × 106 DUC18 mouse-derived CD8+ T cells combined with administration of DTA-1 or PBS on day 7. Three days after ACT, lymphocytes in draining lymph nodes (DLN), non-draining lymph nodes (NDLN), spleen, or tumor were harvested and stained with anti-Foxp3 and anti-CD4 mAb. (a) Representative histogram for Foxp3 staining among CD4+ cells in each organ is indicated. Similar results were obtained in three independent experiments. (b) The ratio of Foxp3+ cells among CD4+ cells in each organ is indicated. Data are the mean + SE of three independent experiments. Differences between groups were examined for statistical significance using Student's t-test. *P < 0.01.

Download figure to PowerPoint

Increased GITR expression on effector CD8+ T cells in DLN and on Treg in tumor.  We harvested DLN, NDLN, spleen, and tumor from mice that were subjected to ACT on days 2 or 7 of tumor inoculation to examine GITR expression on effector CD8+ T cells, non-specific CD8+ T cells, Foxp3+CD4+ Treg, and Foxp3CD4+ T cells in each organ (Fig. 6). The GITR expression of effector CD8+ T cells was higher than that of non-specific CD8+ T cells in DLN, but not in NDLN, spleen, or tumor. Conversely, GITR expression of Treg cells was significantly higher in tumor than in other organs, although Foxp3+CD4+ Treg always expressed higher GITR than Foxp3CD4+ T cells. These results suggest that effector CD8+ T cells in DLN are more sensitive to DTA-1 stimulation than those in other organs and Treg cells in tumor are highly sensitive to DTA-1 stimulation, implying the major site of DTA-1 stimulation for each target cell population.

image

Figure 6. Glucocorticoid-induced tumor necrosis factor receptor (GITR) expression is prominent on effector cells in draining lymph nodes (DLN) and on Treg in tumors. BALB/c mice were inoculated subcutaneously with 1 × 106 CMS5 cells on day 0, and CFSE-labeled DUC18 CD8+ T cells were adoptively transferred on day 2 or day 7. Three days after adoptive cell transfer, lymphocytes in DLN, non-draining lymph nodes (NDLN), spleen, or tumor were harvested and stained with anti-CD8, anti-CD4, anti-Foxp3, and anti-GITR mAb. For CD8+ T cells, the solid line indicates the staining of CFSE-positive effector T cells, and the dotted line represents CFSE-negative non-specific CD8+ T cells. For CD4+ T cells, the solid line indicates Foxp3+CD4+ T cells, and the dotted line represents Foxp3CD4+ T cells. The filled histogram represents control staining.

Download figure to PowerPoint

GITR stimulation induces increased cytotoxic activity in vivo.  Finally, we evaluated the antigen-specific in vivo cytolytic activity in mice subjected to ACT using an in vivo killing assay with CSFE-labeled splenocytes pulsed with the CTL epitope-peptide 9m, as a measurement of overall activity of immunotherapy (Fig. 7). The in vivo CTL activity in mice that received ACT on day 7 of tumor inoculation decreased significantly compared to that seen in mice subjected to ACT on day 2. Combinations of DTA-1 administration with ACT markedly enhanced in vivo CTL activity, even in mice that received ACT on day 7 of tumor challenge.

image

Figure 7. Glucocorticoid-induced tumor necrosis factor receptor (GITR) stimulation enhances the in vivo CTL activity in mice that received adoptive cell transfer on day 7 of tumor challenge. 3 × 105 DUC18 CD8+ cells were transferred into day 2-CMS5 tumor-bearing mice or day 7-CMS5 tumor-bearing mice in the absence or presence of DTA-1 treatment. CFSE-labeled target cells were prepared and transferred on day 10. Twenty hours later, target cells were harvested from spleens and analyzed by flow cytometry. The percentage of specific lysis of target cells is indicated with representative histograms, in which 9m-pulsed and -unpulsed target cells were labeled with high and low concentrations of CFSE respectively. The control indicates the result of naive mice in the absence of therapy. Data are the mean + SD of three mice. *P < 0.01, **P < 0.05. Results are representative of three independent experiments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Recent technological advances in flow cytometry now allow for the simultaneous detection of a number of functional, phenotypic, and lineage-associated markers on T cells at the single-cell level, dramatically enhancing the information on the qualitative evaluation of T-cell response.(1–9) We demonstrated for the first time the significance of effector T-cell multifunctionality in a systematic analysis of antitumor immunity utilizing a murine tumor model.(11) Simultaneously, the NY-ESO-1-specific T-cell response in patients treated with the anticytotoxic T lymphocyte-associated antigen 4 antibody (ipilimumab) was reported as a multifunctional response,(31) suggesting an important role for multifunctional T cells also in the human immune response against tumor. The present study here, demonstration of the impact of GITR stimulation on multifunctionality of adoptively transferred effector T cells and its correlation with the enhanced efficacy of ACT (Figs 1,2) is in line with the above reports. This is the first report to systemically evaluate the impact of immunological modulation on effector T-cell multifunctionality in a therapeutic condition, and strengthens the viewpoint that effector T-cell multifunctionality is a sensitive immune correlate for successful control of tumor. As we discussed previously, multifunctional analysis does not aim at pointing out the single function that plays a major role in disease control. Rather, superior expression of each function in multifunctional T cells (Fig. 3) ensures the sensitivity of this analysis for successful disease control. A function that has not yet been identified, such as the ability to migrate to tumor sites, may play an important role in tumor rejection and closely correlate with the result of our multifunctional analysis.

Tumor and its stroma mediate molecules and cellular populations that inhibit the effector cells in their generation and expansion, activation and acquisition of effector functions, or migration and homing to the tumor site.(32–43) The disruption of Treg-mediated suppression on immune system has become an important issue to be considered in immunotherapy of cancer patients.(44–48) Because the signal through GITR has been reported to act as a costimulator for effector T cells and to induce suppression in Treg,(15–18,49,50) it is a promising strategy to be combined with cancer immunotherapy in hosts with progressing tumor. As a refinement of this understanding, in the present report GITR stimulation was shown to increase effector multifunctionality with reduced Foxp3+CD4+ T cells among TIL and enhanced effector T-cell migration in tumor (Figs 2,4,7). Therefore, the study here supports our concept that it is important to induce high effector T-cell multifunctionality in association with Treg suppression for successful eradication of tumor by immunity. The development of GITRL–Fc, an extracellular fragment of GITRL fused to an immunoglobulin Fc portion, as a new soluble reagent to induce GITR stimulation is in progress in our laboratory.(51)

Whether the in vivo activity of GITR stimulation is dependent on direct activation of effector T cells,(17,18) accompanied by increased resistance of effector T cells to Treg suppression,(21,22) loss of the suppressive effect of Treg,(15,16,49) or both has been a matter of debate.(50) Our results here suggest that the reduction of Treg ratio among TIL (Fig. 5) is involved in the enhanced multifunctionality in effector cells and the successful tumor rejection in hosts treated with the combination of ACT and GITR stimulation. Additionally, our observations clearly showed that GITR was highly expressed on effector CD8+ T cells in DLN and on Treg in tumor (Fig. 6). These results suggest that anti-GITR antibody directly stimulates effector T cells predominantly in DLN, and acts on Treg mainly in tumor. It is possible that CD4+ helper T cells and antigen-presenting cells in DLN also play important roles in the enhanced multifunctionality of effector CD8+ T cells in DLN. From this viewpoint, we believe that the decrease in Treg in tumor affects the quality of antigen-presenting cells and CD4+ helper T cells in DLN. However, a definitive identification of the initial target cells of DTA-1 with spaciotemporal analysis of downstream effects in vivo remains a future challenge, similar to the limitation in previous reports.(24,28,52) An adoptive transfer experiment utilizing immunodeficient mice reconstituted with several different lymphocyte subpopulations derived from GITR-deficient or -sufficient mice will be useful to address this issue. Alternatively, it will be interesting to use conditional GITR gene-disrupted mice that lack GITR in specific lymphocyte subpopulations. It was reported that GITR is expressed on human natural killer (NK) cells and its signal might impair NK cell effector function.(53,54) However, GITR is consistently costimulatory for effector T cells in both mice and humans,(50) and the overall impact of GITR stimulation on immunosurveillance of human tumors needs further investigation. Nevertheless, stimulation through GITR will be beneficial in the T-cell mediated immunotherapy.

Systemic administration of an agonistic antibody for GITR in adult mice has been reported without histologically evident autoimmune diseases.(24) However, a careful attention on the break of self-tolerance and development of autoimmunity will be required when we plan a systemic GITR stimulation in vivo in a clinical setting.(15,26) Therefore, it will be ideal if future applications of GITR stimulation in patients is achieved with technological advances to provide GITR stimulation mainly at local sites. The multifunctional analysis as presented here will help evaluation of the effectiveness of such future immunotherapy with technological improvement of GITR stimulation. Moreover, the combination of ACT with GITR stimulation will help to reduce the required amount of GITR stimulants.

In summary, the present study suggests that GITR stimulation is a promising strategy to be combined with adoptive T-cell therapy as it enhances the multifunctionality of tumor-specific effector T cells even in hosts with progressing tumor, resulting in enhanced tumor regression. The data here add strong evidence for the significance of effector T-cell multifunctionality in the antitumor immune response and provide an immunological rationale for GITR stimulation to overcome the immunosuppressive mechanism on effector T-cell functionality in tumor-bearing hosts.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

We thank Dr Shimon Sakaguchi (Kyoto University) for generously providing DTA-1 mAb-producing hybridoma cells, and Kazuko Mori and Makiko Yamane for their expert technical assistance. This work was supported by Grants-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References