Distinct role of CD8 cells and CD4 cells in antitumor immunity triggered by cell apoptosis using a Herpes simplex virus thymidine kinase/ganciclovir system

Abstract

Immune cells can recognize tumor‐associated antigens released from dead tumor cells, which elicit immune responses, potentially resulting in tumor regression. Tumor cell death induced by chemotherapy has also been reported to activate immunity. However, various studies have reported drug‐induced immunosuppression or suppression of inflammation by apoptotic cells. Thus, this study aimed to investigate whether apoptotic tumor cells trigger antitumor immunity independent of anticancer treatment. Local immune responses were evaluated after direct induction of tumor cell apoptosis using a Herpes simplex virus thymidine kinase/ganciclovir (HSV‐tk/GCV) system. The inflammatory response was significantly altered at the tumor site after apoptosis induction. The expression of cytokines and molecules that activate and suppress inflammation simultaneously increased. The HSV‐tk/GCV‐induced tumor cell apoptosis resulted in tumor growth suppression and promoted T lymphocyte infiltration into tumors. Therefore, the role of T cells after inducing tumor cell death was explored. CD8 T cell depletion abrogated the antitumor efficacy of apoptosis induction, indicating that tumor regression was mainly dependent on CD8 T cells. Furthermore, CD4 T cell depletion inhibited tumor growth, suggesting the potential role of CD4 T cells in suppressive tumor immunity. Tumor tissues were evaluated after tumor cell apoptosis and CD4 T cell depletion to elucidate this immunological mechanism. Foxp3 and CTLA4, regulatory T‐cell markers, decreased. Furthermore, arginase 1, an immune‐suppressive mediator induced by myeloid cells, was significantly downregulated. These findings indicate that tumors accelerate CD8 T cell‐dependent antitumor immunity and CD4 T cell‐mediated suppressive immunity. These findings could be a therapeutic target for immunotherapy in combination with cytotoxic chemotherapy.

Tumor cell apoptosis resulted in tumor growth suppression and promoted T lymphocyte infiltration into tumors; CD8 T cells induced antitumor immunity, while CD4 T cells suppressed that immunity. These findings could be a therapeutic target for immunotherapy in combination with cytotoxic chemotherapy.

Abbreviations

APC

antigen‐presenting cell

ARG1

arginase 1

GCV

ganciclovir

HSV‐tk/GCV

Herpes simplex virus thymidine kinase/ganciclovir

IFN

interferon

IL

interleukin

mMDSC

monocytic myeloid suppressor cell

OVA

ovalbumin

PD‐1

programmed cell death‐1

PD‐L1

PD‐1 ligand

PPR

pattern recognition receptor

TAM

tumor‐associated macrophage

TGF‐β1

transforming growth factor beta 1

TLR

Toll‐like receptor

TNFα

tumor necrosis factor alpha

Treg

regulatory T lymphocyte

1. INTRODUCTION

Tumors grow by the rapid proliferation of constituent cell subpopulations, but the growth is limited by ongoing cell death. Although initial studies have reported that dead tumor cells are immunologically inert, recent studies have suggested that they can activate the host immune system. ¹ , ² , ³ Most tissues are maintained by the continuous death of older differentiated cells and renewal by stem cells and progenitors. Apoptosis is a physiologically programmed and controlled cell death pathway that allows cellular remains to be properly processed and phagocytosed by macrophages without inducing inflammation. ⁴ , ⁵ Thus, the control of apoptosis is essential for maintaining immune homeostasis and suppressing the development of autoimmunity. ⁴ , ⁵ , ⁶

Conversely, dead cells not properly degraded by apoptotic pathways can produce danger signals that lead to immune activation, termed immunogenic cell death. ¹ , ² , ³ Endogenous components released from dead cells activate the innate immune system through PPRs expressed primarily by APCs, such as dendritic cells and macrophages, leading to APC maturation. ⁷ , ⁸ However, little is known about the immune responses induced by tumor cell death in vivo. Several studies have examined immune signaling and immune cell activity following the induction of tumor cell death by chemotherapeutic agents or radiotherapy. ¹ , ⁹ , ¹⁰ , ¹¹ However, these stimuli influence the host immune system independently of the tumor response and do not allow for direct examination of the physiological immune response to cancer cell death.

This study aimed to investigate whether apoptotic tumor cells trigger antitumor immunity by inducing cancer cell apoptosis in tumor‐bearing model mice using a HSV‐tk/GCV system to exclude the effects of cytotoxic anticancer therapies on the host immune system. ¹²

2. MATERIALS AND METHODS

2.1. Animals and tumor cell lines

BALB/c and C57BL/6N mice were obtained from Japan SLC at 8 weeks of age, and MyD88 KO mice of a C57BL/6 background were obtained from OrientalBioService and housed under specific pathogen‐free conditions until 10–12 weeks of age, when the experiments started. All experiments were approved by the Institutional Committee for the Use and Care of Laboratory Animals of Tohoku University. Mouse‐derived CT26 colon and 4 T1 breast cancer cell lines were obtained from ATCC. MC38‐OVA^dim cancer cells were derived from MC38 cells as previously described. ¹³

2.2. Reagents

Ganciclovir was obtained from Mitsubishi Tanabe Pharma Corporation and diluted to a working concentration of 5 mg/mL in PBS. Zeocin was obtained from InvivoGen. Anti‐CD8 (2.43), anti‐CD4 (GK1.5), and anti‐PD‐1 Abs (29F.1A12) were purchased from Bio X Cell.

2.3. Stable transfections

CT26 and MC38‐OVA^dim tumor cells were stably transfected with a plasmid DNA vector encoding the HSV‐tk gene (pSELECT‐zeo‐HSV1tkSh; InvivoGen) using Lipofectamine 2000 DNA Transfection Reagent (Invitrogen Life Technologies) according to the manufacturer's protocol. Plasmid‐transfected tumor cells were then selected by continuous passage in 200 μg/mL Zeocin (InvivoGen).

2.4. Flow cytometry

Tumor cells were prepared for flow cytometry as previously described. ¹⁴ Briefly, the cells were washed with PBS, stained with fluorochrome‐conjugated anti‐CD45, anti‐CD3, anti‐CD4, and anti‐CD8 Abs for 30 min, and fixed in fixation buffer for 15 min. All Abs were obtained from BD Pharmingen and BioLegend. Stained cells were then analyzed for surface expression profile using a BD LSRFortessa system (BD Pharmingen). To detect apoptotic cells, cells treated as indicated were stained with annexin V‐FITC and propidium iodide using an Annexin V‐FITC Apoptosis Detection Kit (Nacalai Tesque) according to the manufacturer's instructions and analyzed by flow cytometry.

2.5. In vivo tumor studies

Mice were injected s.c. with tumor cells, and tumor volumes were calculated at the indicated times according to the following formula: (length × width × height)/2. Tumor‐bearing mice were divided into groups adjusted for equal tumor volume and treated as indicated. In most experiments, the experimental group was injected with 500 μg GCV i.p. twice daily for 4 days. Tumor growth curves were generated from three to five mice per treatment group, and all results were derived by combining data from two or three independent experiments.

2.6. Histology

Tumors were removed, fixed in 10% formalin, embedded in paraffin, and sectioned at 3 μm. Sections were then stained for apoptotic cells by TUNEL using the ApopTag Peroxidase In Situ Apoptosis Detection Kit (Merck Millipore) according to the manufacturer's protocol. Histological images were captured using a BZ‐9000 microscope (Keyence). The TUNEL‐positive area was quantified using ImageJ (NIH). For immunostaining, paraffin‐embedded tissues were cut into 3 μm thick sections and stained sequentially with anti‐CD8a (4SM15; eBioscience) and secondary Abs (Histofine Simple Stain MAX PRO; Nichirei Biosciences). Finally, immunolabeling was visualized by incubation in 3,3′‐diaminobenzidine, followed by counterstaining with hematoxylin.

2.7. Preparation of sorted cells

Tumors were digested in Liberase/DNase I (Roche Molecular Biochemicals/Sigma‐Aldrich) solution for 30 min at 37°C. Cells were washed with 2% FCS/PBS, stained with fluorochrome‐conjugated Abs for 30 min at 4°C, and sorted using BD FACSAria to isolate TAMs or mMDSCs, as defined by the following characteristics: CD45⁺, CD11b⁺, Ly6c⁻, and Ly6g⁻ or CD45⁺, CD11b⁺, Ly6c⁺, and Ly6g⁻.

2.8. Quantitative real‐time RT‐PCR

Quantitative real‐time RT‐PCR was carried out as described previously ¹⁵ , ¹⁶ using TaqMan probes and TaqMan Gene Expression Master Mix kit (Applied Biosystems). Relative expression levels of target mRNAs were calculated relative to GAPDH expression.

2.9. Enzyme‐linked immunospot assay

Enzyme‐linked immunospot assays were carried out as described previously. ¹⁵ , ¹⁶ The H‐2Kb‐restricted peptide from OVA (SIINFEKL) was used to stimulate CD8 T cells for 16 h. The number of spots was determined by a single reader blinded to the treatment history. Note that the SIINFEKL peptide given to naïve mice did not differ from the unstimulated control. ¹⁵ No response was observed to the other peptide (AH‐1 peptide) in the preliminary experiment (data not shown).

2.10. Statistical analysis

Tumor growth and cellular responses were compared between treatment and control groups using an independent sample Student's t‐test. All statistical analyses were undertaken using JMP Pro 15 (SAS Institute). A p values <0.05 (two‐tailed) was considered statistically significant for all tests.

3. RESULTS

3.1. HSV‐tk/GCV system induced apoptosis within tumors

Recent studies have reported that the residual components of dead tumor cells can modulate antitumor immunity. ¹ , ⁹ , ¹⁰ However, antitumor treatments targeting tumor cells for destruction influence the host immune system. Therefore, we established tumor cells susceptible to apoptosis in response to the antiherpetic drug GCV by stable transfection of HSV‐tk to examine immune reactions triggered by tumor cell death in the absence of antitumor treatment.

First, we examined the GCV‐induced apoptotic death of stably transfected tumor cell lines in vitro by flow cytometry. The HSV‐tk gene‐transfected CT26 tumor cells (CT26‐HSV‐tk tumor cells) showed a progressive increase in the expression of the apoptosis marker annexin V during GCV treatment (Figure 1A), reaching nearly 100%, which is consistent with previous reports. ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ Then, we examined the GCV responses of tumors derived from HSV‐tk‐expressing cells in mice. Mice were injected s.c. in the right flank with HSV‐tk‐transfected tumor cells. Once these masses reached approximately 4–6 mm in diameter (requiring 5–7 days), the treatment group was injected with GCV (twice daily for 4 days). Consistent with in vitro results, the growth of HSV‐tk‐transfected CT26 cell‐derived tumors was significantly suppressed by GCV treatment for 4 days compared with the untreated control group (Figure 1B). In preliminary experiments, 4 days of GCV treatment made little difference, and 6 days of treatment resulted in the disappearance of many tumors (data not shown). Ganciclovir alone did not suppress tumor growth in cells not transfected with HSV‐tk (Figure 1C). Consistent with this observation, the growth of MC38‐OVA^dim cells transfected with HSV‐tk was significantly delayed (Figure 1D). Apoptosis frequency was evaluated using the TUNEL assay to examine the influence of the HSV‐tk/GCV system on the tumor condition in vivo. At 72 h after repeated GCV treatment, the number of TUNEL‐positive/apoptotic cells was significantly greater (9.4%–23%) than that in the untreated tumors (Figure 1E,F). However, most tumor cells were TUNEL‐negative/alive even after GCV treatment, despite the substantial slowing of the tumor growth rate. This suggests that the HSV‐tk/GCV system induces only limited tumor regression directly and that a downstream process accounts for most of the antitumor effect.

FIGURE 1.

Detection of in vitro and in vivo tumor cell apoptosis induced by the Herpes simplex virus thymidine kinase/ganciclovir (HSV‐tk/GCV) system. (A) CT26 tumor cells stably transfected with HSV‐tk plasmid (CT26‐HSV‐tk tumor cells) were incubated with 20 μM GCV for 48 h. Tumor cells were then stained with annexin V‐FITC and propidium iodide (PI), and apoptotic cell numbers were determined by flow cytometry. Representative cell counting by flow cytometry showing that most untreated cells were negative for annexin V and PI (lower left quadrant), whereas most cells were annexin V‐positive after 48 h of GCV treatment. (B–D) BALB/c or C57BL/6N mice were injected s.c. with (B) CT26‐HSV‐tk, (C) CT26 tumor cells, or (D) MC38‐OVA^dim ‐HSV‐tk (1.0 × 10⁶). Tumor‐bearing mice in the experimental group were then injected i.p. with 500 μg GCV twice daily for 4 days. Data are presented as mean ± SEM of 7–12 mice per treatment group from two or three independent experiments. (E, F) Tumor tissues were removed 1 day after the final GCV treatment and apoptotic cells detected by TUNEL assay. Shown are (E) representative images of TUNEL staining and (F) the TUNEL‐positive area (% of total) in each tumor section (n = 8). ***p < 0.001 compared to the untreated control group.

3.2. Activation of the HSV‐tk/GCV system promoted inflammation within tumors

We investigated the contribution of immunological mechanisms to the regression of HSV‐tk‐expressing tumors following GCV treatment by measuring changes in the expression of immunomodulatory genes within the tumors and the immune cell infiltration rate. The mRNA expression levels of various cytokines and costimulatory molecules were evaluated by real‐time RT‐PCR to examine the immunological signaling within the tumor microenvironment after HSV‐tk/GCV‐mediated tumor cell death. The expression of pro‐inflammatory cytokines such as IL‐1β, TNFα, and IFNβ was markedly upregulated in GCV‐treated tumors compared with the untreated control tumors (Figure 2A). Similarly, the T cell‐polarizing and effector cytokines and molecules encoding IL‐2, IL‐12, IFNγ, IL‐4, CD40, and CD86 were also elevated. Furthermore, the expression levels of immunosuppressive genes or molecules encoding TGF‐β1, ARG1, and PD‐1/PD‐L1 were significantly upregulated in GCV‐treated tumors. These findings indicate that HSV‐tk/GCV‐mediated tumor cell death promoted inflammation within the tumors. Particularly, pro‐inflammatory cytokines are thought to be produced by APC in response to various molecules released from dead cells. Such molecules are reported to be recognized by the TLR‐MyD88 pathway. ² , ³ , ⁹ , ²¹ Therefore, the same analysis was carried out using MyD88 KO mice. Interestingly, apoptosis induction resulted in the same inflammatory response as in WT mice (Figure S1A). Similarly, the inhibition of tumor growth was also observed (Figure S1B).

FIGURE 2.

Expression of mRNA and infiltration of immune cells at the tumor site after apoptosis induction. (A) BALB/c mice were injected s.c. with CT26‐HSV‐tk tumor cells and were then injected i.p. with 500 μg of GCV twice daily for 4 days. Tumors were removed 1, 3, or 5 days after GCV treatment and analyzed for expression of the indicated mRNAs by RT‐PCR. Data presented as the mean ± SEM of 9‐12 mice per treatment group from three independent experiments. Tumors were removed 1, 3, or 5 days after ganciclovir (GCV) treatment and analyzed for expression of the indicated mRNAs by RT‐PCR. Data presented as the mean ± SEM of 9–12 mice per treatment group from three independent experiments. (B, C) Tumors were removed after 5 days and the numbers of tumor‐infiltrating CD45⁺ CD4 cells and CD8 T cells determined by flow cytometry. (B) Representative results from one mouse per group. (C) Results were evaluated independently for each mouse, and the data are presented as mean ± SD of 10 or 11 mice per group from four independent experiments. (D) Tumor sections were removed, immunostained with anti‐CD8 Ab (brown), and counterstained with hematoxylin (blue). *p < 0.05, **p < 0.01, ***p < 0.001 compared to the untreated control group. ARG1, arginase 1; IFN, interferon; IL, interleukin; PD‐1, programmed cell death‐1; PD‐L1, PD‐1 ligand; TGF‐β1, transforming growth factor beta 1; TNF‐α, tumor necrosis factor alpha.

The number of infiltrating CD45⁺ immune cells was analyzed by flow cytometry to identify the immune cells responsible for enhanced inflammatory signaling within the GCV‐treated tumors. The percentage of CD45⁺ immune cells increased within the tumor, and the percentage of CD4⁺ and CD8⁺ T cells increased several fold (6.9‐fold and 4.2‐fold, respectively) following GCV treatment compared with tumors from untreated mice (Figure 2B,C). Moreover, immunohistochemical staining showed that CD8 T cells were more abundantly distributed in the tumors of GCV‐treated mice than in those of untreated mice (Figure 2D). These findings indicate that T cells accumulated within tumors after induction of HSV‐tk/GCV‐mediated tumor cell death.

3.3. Apoptosis‐induced tumor regression was CD8 T cell‐dependent

T cells can recognize tumor cells in an antigen‐specific manner and induce cytotoxicity, thereby contributing to tumor elimination. We compared the induction rates of tumor‐specific CD8 T cells between GCV‐treated and untreated OVA‐expressed MC38 HSV‐tk tumor‐bearing mice to investigate whether apoptosis activated tumor‐specific CD8 T cells. A substantially greater number of splenocytes stimulated ex vivo with MHC class I‐restricted OVA peptide antigen (SIINFEKL) produced IFNγ compared with splenocytes from untreated mice (Figure 3A). These findings suggest that HSV‐tk/GCV‐mediated tumor cell death enhances the number of tumor‐specific CD8 T cells. Furthermore, mice inoculated with CT26‐HSV‐tk tumor cells and showing total tumor regression following GCV treatment were rechallenged with transfected parental CT26 or 4 T1 tumor cells to study potential immunological memory against tumors after induction of tumor cell death. Ninety percent of CT26 tumors were rejected in the rechallenged group (9 of 10 mice cleared), whereas the growth of 4T1 cell‐derived tumors was similar in both groups (Figure 3B,C). These data support tumor‐specific immunity induced by initial activation of the HSV‐tk/GCV system and ensuing tumor cell apoptosis.

FIGURE 3.

Analysis of the role of CD8 T cells in the Herpes simplex virus thymidine kinase/ganciclovir (HSV‐tk/GCV) model. (A) C57BL/6N mice were injected s.c. with MC38‐OVAdim‐HSV‐tk tumor cells and were then injected i.p. with 500 μg GCV twice daily for 4 days. Spleens from tumor‐bearing mice were removed on day 28. The spleen cells (1.0 × 10⁶) were restimulated ex vivo with 1 μg/mL SIINFEKL peptide for 16 h, and monitored for γ‐interferon (IFNγ) secretion using the enzyme‐linked immunospot assay. Data are presented as mean ± SEM from nine mice per group and three independent experiments. (B, C) CT26 or 4T1 tumor cells were injected s.c. into tumor‐naïve mice and mice that had been cleared of previously inoculated tumors using the HSV‐tk/GCV system (rechallenged group). Shown are individual tumor growth curves (CT26: naïve n = 8, all tumor implanted; rechallenged n = 10, 9 mice cleared; 4T1: naïve n = 3, all tumor implanted; rechallenged n = 5, all tumors implanted). (D) CT26‐HSV‐tk tumor cells were injected. On days 0 and 4, mice were injected i.p. with 500 or 250 μg anti‐CD8 Ab to eliminate CD8 T cells. Data are presented as mean ± SEM of six mice per group from two independent experiments. (E) Tumor‐bearing mice were injected i.p. with a suboptimal dose of GCV (500 μg, twice daily for 2 days). Some mice were injected i.p. with 100 μg anti‐programmed cell death‐1 (PD‐1) alone or 1 day before and 4 days after GCV treatment. Data are presented as mean ± SEM of six mice per group from two independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

CT26‐HSV‐tk tumor‐bearing mice received GCV with or without anti‐CD8 Ab treatment to deplete CD8 T cells to further investigate the mechanisms underlying the inhibition of tumor growth. The GCV‐induced inhibition of tumor growth was abrogated by CD8 T cell depletion (Figure 3D), indicating that CD8 T cells contribute substantially to tumor immunity induced by HSV‐tk/GCV‐mediated apoptosis. These results suggest that the inhibition of tumor growth was not caused primarily by direct HSV‐tk/GCV system‐mediated tumor cell death but by subsequent induction of CD8 T cell‐induced antitumor immunity. Additionally, PD‐1 blockade is well known to enhance antitumor immunity by CD8 T cells. ²² , ²³ The upregulation of PD‐1/PD‐L1 was observed in this system. Anti‐PD‐1 alone or half‐dose GCV reduced CT26‐HSV‐tk tumor volume by ~50%, while combination therapy with half‐dose GCV and anti‐PD‐1 Ab induced 100% tumor regression (Figure 3E).

3.4. CD4 T cell depletion augmented apoptosis‐induced antitumor immunity

The fractions of CD4⁺ T cells and CD8 T cells also increased in the tumor. CD4 cells were depleted by anti‐CD4 Abs as well, to investigate the role of these T cells. Interestingly, the apoptosis‐induced inhibition of tumor growth was augmented when CD4⁺ cells were depleted (Figure 4A). All the CD4‐positive cells in the tumors were identified as CD3‐expressing T cells. Furthermore, the administration of anti‐CD4 Abs resulted in the eradication of CD4 T cells. Thus, the expression of mRNA in the tumors was examined 5 days after each treatment to investigate the effect of CD4 removal on tumor immunity. The expression of mRNA related to antitumor immunity, such as IFNγ, IL‐12, and IFN‐induced gene, CXCL10, was upregulated by GCV treatment and further increased by the removal of CD4 cells (Figure 4B). This indicates that CD4 T cells act in a suppressive manner against tumor immunity by apoptosis induction. Suppressive factors for T cells include TGFβ, IL‐10, and ARG1 as mediators and PD‐1 and CTLA4 as costimulatory molecules. ²² , ²³ , ²⁴ , ²⁵ , ²⁶ Therefore, we focused on the inhibitory mediators and molecules of tumor immunity after CD4 T cells were depleted. While the expression of IL‐10, TGFβ, and PD‐1 was unchanged, the expression of ARG1 and CTLA4 was significantly downregulated. Furthermore, the expression of IL‐4 significantly decreased. ²⁵ The suppressor cells of antitumor immunity within the tumor microenvironment are regulatory T cells, MDSCs, and M2 macrophages. ²² , ²³ , ²⁴ , ²⁵ Markers of Tregs, such as Foxp3, and TAMs, such as Chi3l3, and Retnla, were elevated when the inflammatory response was enhanced by apoptosis induction in the tumor. As expected with the CD4 depletion, Foxp3, a marker of Tregs, significantly decreased. However, no change was observed in markers of TAMs, such as Chi3l3 and Retnla (Figure 4B).

FIGURE 4.

Analysis of the role of CD4 T cells in the Herpes simplex virus thymidine kinase/ganciclovir (HSV‐tk/GCV) model. (A) BALB/c mice were injected s.c. with CT26‐HSV‐tk tumor cells and were then injected i.p. with 500 μg of GCV twice daily for 4 days. On days 0 and 6, mice were injected i.p. with 500 μg or 250 μg anti‐CD4 antibody to eliminate CD4 T cells. Data are presented as mean ± SEM of 12‐16 mice per group from four independent experiments. On days 0 and 6, mice were injected i.p. with 500 or 250 μg anti‐CD4 Ab to eliminate CD4 T cells. Data are presented as mean ± SEM of 12–16 mice per group from four independent experiments. (B) Tumors were removed 5 days after GCV treatment and anti‐CD4 Ab and then analyzed for expression of the indicated mRNAs by RT‐PCR. Data presented as the mean ± SEM of six mice per treatment group from two independent experiments. (C, D) Tumors were removed after 5 days and the numbers of tumor‐infiltrating monocytic myeloid suppressor cells (mMDSCs) (CD11b⁺, Ly6c⁺, Ly6g⁻) and tumor‐associated macrophages (TAMs) (CD11b⁺, Ly6c⁻, Ly6g⁻) were determined by flow cytometry. (C) Representative results from one mouse per group. (D) Results were evaluated independently for each mouse, and the data are presented as mean ± SD of four mice per group. (E) Tumor‐infiltrated mMDSCs (CD45⁺, CD11b⁺, Ly6c⁺, Ly6g⁻) and TAMs (CD45⁺, CD11b⁺, Ly6c⁻, Ly6g⁻) were purified by flow cytometry and analyzed for expression of the indicated mRNAs by RT‐PCR. Results represent the mean with SD of results from five independently sorted cell populations. *p < 0.05, **p < 0.01, ***p < 0.001. ARG1, arginase 1; gMDSC, granulocytic myeloid suppressor cells; IFN, interferon; IL, interleukin; PD‐1, programmed cell death‐1; TGF‐β1, transforming growth factor beta 1; TNF‐α, tumor necrosis factor alpha.

3.5. CD4 T cell depletion decreased expression of ARG1 from tumor‐infiltrating myeloid cells

A decrease in ARG1 in tumor tissues due to CD4⁺ cell depletion would activate CD8 T cells, leading to enhanced antitumor immunity. Arginase 1 has been reported to be produced by myeloid cells, and CD4 cell depletion is expected to reduce the number or expression of these cells. The percentage of CD11b⁺ TAMs and mMDSC in the tumor tissues was identified by flow cytometry. There are various reports on markers and classification of TAMs and mMDSCs. We have classified them by the most commonly used markers based on previous reports (Figure 4C). ²⁴ , ²⁵ CD4 T cell depletion did not significantly alter the percentage of TAMs or mMDSCs in the tumor (Figure 4C,D). Next, TAMs and mMDSCs were sorted from the tumor tissues after treatment with GCV and anti‐CD4 Abs and analyzed for mRNA expression. Both cells collected from the CD4‐depleted tumors reduced the expression of ARG1 (Figure 4E). Arginase 1 expressed in TAMs is particularly expressed in the M2 phenotype. Thus, we examined the M1 and M2 phenotypic markers of the sorted macrophages. Based on previous reports, we examined the expression of Chi3l3 and Retnla, which are M2‐type genes, and CXCL10 and TnFα, which are M1‐type genes. As shown in Figure 4E, decreased expression was observed in Chi3l3 and Retnla, as well as in ARG1. In contrast, the markers of M1 phenotypes such as CXCL10 and TnFα remained unchanged.

4. DISCUSSION

Cancer therapies such as chemotherapy and irradiation induce substantial tumor cell death, and immune cells activated by cell component proteins released from dead cells can induce inflammation in the tumor microenvironment, thereby regulating tumor progression. We generated an animal model in which tumor cell death can be induced specifically in target cells using the HSV‐tk/GCV system to examine the immune response triggered by tumor cell death in the absence of immune system dysfunction from cancer chemotherapy. Recent studies have shown that GCV is maintained above the effective concentration for only a limited time (less than 1 h) following injection due to its short biological half‐life and dose‐limiting toxicity. ²⁷ , ²⁸ Thus, direct tumor cell death induced by the system was only a small fraction in vivo. Rather, cell death by the system caused a high inflammatory state within the tumor, thus activating CD8 T cells, which suppress the tumor growth. However, it also enhanced the induction of immune cells, cytokines, and molecules that suppress inflammation after apoptosis induction. Particularly, CD4 T cell depletion and apoptosis induction have been found to not only delete Tregs but also reduce the expression of ARG1 in myeloid cells. These findings could lead to the development of targets that activate antitumor immunity with the induction of tumor cell death.

Generally, the induction of innate immunity is observed before tumor‐specific immunity is induced. Self‐derived intracellular proteins and nucleic acids are released from dead and dying tumor cells. ² , ³ , ⁹ , ²¹ , ²⁹ These substances act as damage‐associated molecular patterns to trigger innate immunity and induce the production of pro‐inflammatory cytokines. ³⁰ , ³¹ , ³² Apetoh et al. reported that chemotherapy‐induced tumor cell death resulted in the release of HMGB1, which acts on APCs through TLR4 to enhance the antitumor immune response. ⁹ Clinical observations also suggest that the release of intracellular proteins by radiotherapy‐induced tumor cell death promotes antitumor immunity, resulting in a potentiated tumor regression effect called the abscopal effect. ³³ Our results using the HSV‐tk/GCV system support these findings. The expression levels of inflammatory cytokines, such as TNFα, IL‐1β, and IL‐12, were markedly enhanced within the tumor as early as 3 days after apoptosis induction (Figure 2). Many damage‐associated molecular patterns have been reported to be recognized and activated by PPRs in the MyD88 molecular pathway, such as TLR2, 4, and 9, to produce these cytokines. ² , ³ , ⁹ , ²¹ Furthermore, a previous paper reported that a similar system eliminates the induction of tumor immunity in MyD88 KO mice. ³⁴ Inconsistent with these reports, current results did not reproduce their results, and the KO mice responded similarly to the WT mice, indicating that they are MyD88‐independent recognition (Figure S1). These results differ significantly from previous findings, which showed that pathogen‐, and damage‐associated molecular patterns use the same molecules for recognition. As a limitation of this conclusion, it is within the scope of the observations made in the HSV‐tk/GCV system.

The activation of immunity by tumor cell death was followed by subsequent activation of tumor‐specific immunity, which is CD8 T cells in the HSV‐tk/GCV system. Consistent with these results, CD8 T cell depletion abolished tumor regression (Figure 3D). This induction of tumor‐specific immunity was acquired as memory and further enhanced by treatment with anti‐PD‐1 Abs (Figure 3B,E). However, a strong increase was observed in suppressive immunity, that is, enhanced expression of suppressive cytokines and molecules after cell apoptosis induction. Foxp3 expressed by Tregs and Chi3l3 and Retnla expressed by M2 macrophages increased in the tumor. In apoptosis‐induced tumors, an increase was observed in suppressive mediators, such as TGFβ and ARG1, and an increase was observed in the expression of the inhibitory molecules, such as PD‐1/PD‐L1 and CTLA4 (Figures 2A and 4B). Two possible mechanisms may account for these observations: (1) the induction of suppressive immunity as a brake on antitumor immune activation, and (2) a system that induces immune tolerance associated with tumor cell death. In particular, such a system in which macrophages phagocytose apoptotic cells and do not induce an immune response against self‐antigens (a system for preventing autoimmunity) has been reported. ⁴ , ⁵ , ⁶ Suppressor cytokines and costimulatory molecules seem to be more potentiated than inflammatory cytokines. Apoptotic cells reprogram macrophages by releasing factors such as sphingosine‐1‐phosphate. ³⁵ The reprogramming of macrophages has been reported to promote a shift to the M2 phenotype in response to factors from the apoptotic tumor and the tumor environment. ³⁶ , ³⁷ These findings are consistent with those of our study.

Furthermore, the study results showed that CD4 T cell depletion potently enhanced the antitumor activity induced by tumor cell death. There are diverse phenotypes of CD4 T cells for each different type of inflammation. ³⁸ Certainly, Tregs play an important role and are activated in the CD4 T cells within the tumor microenvironment. ³⁹ Additionally, the study results showed that CD4 T cell depletion decreased the production of ARG1 in tumors (Figure 4B). It is well established that myeloid cells, such as MDSCs and M2 macrophages, produce ARG1 to inhibit T cells, but T cells do not. ²⁴ , ²⁵ , ⁴⁰ Indeed, MDSCs and M2 macrophages, sorted from tumors, significantly reduced the expression of ARG1. In addition, decreased expression was observed for Chi3l3 and Retnla, which are the markers of M2 macrophages. This implies a relative decrease in the proportion of M2 phenotype. However, markers of the M1 phenotype such as CXCL10 and TNFα remained unchanged, indicating that the percentage of M1 macrophages remained unchanged. These indicate that CD4 T cells are required to acquire the suppressive capacity of myeloid cells within tumors. These results suggest that CD4 T cell depletion could be a mechanism to enhance antitumor immunity and CD8 T cells, which are expected to be therapeutic targets. Consistent with our data, some groups have reported further enhancement of immunotherapy with anti‐CD4 Ab treatment. ⁴¹ , ⁴² , ⁴³ It is well established that IL‐4 promotes the differentiation of M2 macrophages and upregulation of ARG1 activity, which increases the suppressive function of MDSCs and TAMs. Particularly, we previously documented that T follicular helper cells are a major source of IL‐4 in the tumor microenvironment. ⁴⁴

In this study, tumor cell death was associated with dramatic changes in the tumor microenvironment, resulting in the induction of antitumor and suppressive immunity. Thus, qualified analysis of immunity within the tumor induced by tumor cell death by chemotherapy and molecular target drugs could also reveal therapeutic targets. Synergistic effects that were not seen with single agents could be observed by activating or blocking targeted immune molecules with cytotoxic agents. The study results showed that the induction of antitumor immunity by tumor cell death was accompanied by an increase in Tregs and M2 macrophages. CD4 cell depletion also simultaneously blocked these suppressive cells. These could be potential therapeutic targets to enhance antitumor immunity in the administration of cytotoxic anticancer drugs. One group has reported the removal of temporary anti‐CD4 Abs in clinical trials, but in reality, it is challenging to suppress helper T cells, which serve as the control hub of the immune system. ⁴⁵ Blocking these cells simultaneously with chemotherapy while identifying target molecules for Tregs and M2 macrophages within the tumor microenvironment could be an effectively combined immunotherapy.

CONFLICT OF INTEREST STATEMENT

Chikashi Ishioka has received scholarship (incentive) endowments from Takeda, Daiichi‐Sankyo, Ono, Asahi‐Kasei Pharma, Taiho, and Chugai, and research grants from Hitachi and Riken Genesis. Chikashi Ishioka is also an editorial board member of Cancer Science. The other authors have no conflict of interest.

ETHICS STATEMENTS

Approval of the research protocol by an institutional review board: N/A.

Informed consent: N/A.

Registry and registration no. of the study/trial: N/A.

Animal studies: All experiments were approved by the Institutional Committee for the Use and Care of Laboratory Animals of Tohoku University (approved number: 2020 medical recombination‐069‐01, 2020 medical recombination‐070‐02).

Supporting information

Figure S1

Data S1

Umegaki S, Shirota H, Kasahara Y, Iwasaki T, Ishioka C. Distinct role of CD8 cells and CD4 cells in antitumor immunity triggered by cell apoptosis using a Herpes simplex virus thymidine kinase/ganciclovir system. Cancer Sci. 2023;114:3076‐3086. doi: 10.1111/cas.15843