Anti–4-1BB and cisplatin showed synergistic anticancer effects in the CT-26 colon carcinoma model, producing complete regression in >60% of mice with either preventive or therapeutic treatment. The tumor-free mice formed long-lasting CD8+ T cell–dependent tumor-specific memory. Anti–4-1BB induced rapid repopulation of T and B cells from cisplatin-mediated lymphopenia and differentiation and expansion of IFN-γ+CD11c+CD8+ T cells. Cisplatin facilitated expansion of naïve, effector, and memory CD8+ T cells; combination therapy produced almost twice as many lymphoid cells as anti–4-1BB alone. Cisplatin increased 4-1BB on antigen-primed T cells and induced 4-1BB de novo on kidney tubular epithelium. Cross-linking of 4-1BB protected the T cells and kidney epithelium from cisplatin-mediated apoptosis by increasing expression of antiapoptotic molecules. Thus, cisplatin-induced 4-1BB provided a mechanism for amelioration of the lymphopenia and nephrotoxicity inherent in cisplatin treatment. We concluded that chemoimmunotherapy with anti–4-1BB and cisplatin is synergistic in tumor killing and prevention of organ-specific toxicity. [Cancer Res 2008;68(18):7264–9]
Immunotherapy can complement anticancer chemotherapy ( 1– 6). Agonistic anti–4-1BB (CD137), an immunotherapeutics, displays a strong anticancer effect ( 7– 9) and is undergoing clinical trials against various tumors. Genotoxic agents, γ-irradiation, and chemotherapeutics increased 4-1BB expression on T cells, and cross-linking of the 4-1BB enhanced T-cell survival ( 10). We, therefore, hypothesized that the chemotherapeutics that enhanced 4-1BB expression could potentiate the anticancer activities of an agonistic anti–4-1BB.
Cisplatin is widely used as a chemotherapeutic agent to treat a variety of tumors ( 11). The efficacy of cisplatin is dose dependent, although it induces thrombocytopenia and leukocytopenia ( 12). Nephrotoxicity frequently hinders the use of higher doses that would maximize its antineoplastic effects ( 13, 14).
Here, we investigated whether anti–4-1BB augmented the anticancer effect of cisplatin and ameliorated cisplatin-induced loss of leukocytes and renal dysfunction. The results revealed that combination therapy with agonistic anti–4-1BB and cisplatin enhanced primary and memory antitumor immune responses, which were mediated by a preferential enhancement of antigen-specific CD8+ T cells, especially CD11c+CD8+ T cells. Moreover, anti–4-1BB treatment attenuated cisplatin-induced nephrotoxicity by protecting renal tubular epithelial cells from cisplatin-induced apoptosis.
Materials and Methods
Mice. Six- to eight-week-old BALB/c, C57BL/6, and OT-1 mice (Jackson Laboratory), and Rag2γc−/− (CIEA) were housed in a specific pathogen-free facility at the Immunomodulation Research Center (IRC) at the University of Ulsan. All animal experiments were performed in accordance with the approved protocols and guidelines of the IRC's Institutional Animal Care and Use Committee.
Antibodies and peptides. Anti–m4-1BB hybridoma cells (3E1) were a gift from Dr. Robert Mittler (Emory University, Atlanta, GA). Anti-mCD4 (GK1.5) and anti-mCD8 (2.43) hybridomas were obtained from American Type Culture Collection. Purified rat IgG was obtained from Sigma-Aldrich. Flow cytometric monoclonal antibodies (mAb) were purchased from BD PharMingen. H-2Kb/OVA257-264- phycoerythrin (PE) pentamers were purchased from ProImmune Ltd. OVA257-264 (SIINFEKL) was synthesized by Peptron.
Tumor challenge and treatment. Mice were injected s.c. with 3 × 105 CT26 cells on the dorsal surface [posttumor injection (PI) day 0]. As a standard treatment, tumor-challenged mice were injected i.p. with 100 μg of rat IgG or anti–4-1BB mAb (3E1) every 5 d for 5 times, and a single injection of 10 mg/kg cisplatin on PI day 0. To examine therapeutic effects of combined therapy, the mice received the standard therapy beginning 6 d PI. For tumor rechallenge, tumor-free mice were injected s.c. in the opposite flank 100 d after the primary tumor inoculation with CT26 or RENCA cells (3, 6, or 12 × 105 cells per mouse). Mice were injected i.p. with 400 μg of anti-CD4 or anti-CD8, or 100 μg of anti–asialo-GM1 antibody every 5 d for 5 cycles. The efficacy of immune subset depletion (>99%) was confirmed by flow cytometry.
Flow cytometry. Lymphocytes were incubated with the Fc blocker 2.4G2 for 10 min at 4°C and stained with specific antibodies to surface markers. To identify IFN-γ–producing cells, tumor-draining lymph node (TDLN) cells were first stimulated with 50 ng/mL phorbol 12-myristate 13-acetate and 500 ng/mL ionomycin for 6 h in a medium containing Brefeldin A. Cells were then stained with FITC-anti-CD8 or anti-CD4, and intracellularly stained PE-anti–IFN-γ by using Cytofix/cytoperm kit. All samples were analyzed using a FACScalibur (BD Biosciences).
Assessment of renal function. Blood samples were obtained from before (0 h) and at 24, 48, and 72 h after treatment with cisplatin and/or anti–4-1BB mAb. Serum creatinine was measured as a marker of renal dysfunction using a Roche Cobas Rara automated system (Roche) with a Creatinine kit (R&D Systems).
Histologic examination. The kidney paraffin block was cut into 4-μm sections and stained with H&E. Renal tubular damage was assessed using a semiquantitative scale. A pathologist who was blinded to the experiments scored the degree of tubular damage. The magnitude of tubular epithelial cell loss and necrosis was scored on the basis of the percentage of affected tubules in a high-power field under a light microscope, as follows: 0, normal; 0.5, <10%; 1, 10% to 25%; 2, 25% to 50%; 3, 50% to 75%; and 4, >75%. For anti–4-1BB immunostaining of the kidney, thin cryosections from kidneys were stained with FITC-anti–m4-1BB (3E1) mAb, and visualized with an Olympus FV500 confocal laser scanning microscope.
Reverse transcription-PCR. Total RNA was extracted from cells and tissues using TRIzol (Invitrogen Life Technologies) and reverse transcribed into cDNA using a SuperScriptII (Invitrogen Life Technologies). Gene-specific primers used were as follows: mouse 4-1BB, forward 5′-tgacatttcgccatgggaaacaactgt-3′, reverse 5′-tcccggtcttaagcacagaccttccgt-3′; human 4-1BB, forward 5′-cgcgtggatccacaagatcattgcaggatcc-3′, reverse 5′-ttgagctcgaggatctgcggagagtgtcctg-3′; glyceraldehyde-3-phosphate dehydrogenase (GAPDH), forward 5′-gaacgggaagcttgtcatcaa-3′, reverse 5′-ctaagcagttggtggtgcag-3′. PCR conditions were as follows: for 4-1BB, 94°C for 5 min, followed by 35 cycles of 30 s at 94°C, 30 s at 55°C, and 1 min at 72°C with a final extension at 72°C for 5 min; for the GAPDH reverse transcription-PCR (RT-PCR), conditions were identical except that 30 cycles were used instead of 35.
Real-time PCR. SYBR Green I–based real-time quantitative PCR was carried out on a continuous fluorescence detection system (Opticon DNA Engine; MJ Research, Inc.) with first-strand cDNA of kidney. Gene-specific primers used were as follows: mouse Bcl-2, forward 5′-tgggatgcctttgtggaactat-3′, reverse 5′-agagacagccaggagaaatcaaac-3′; mouse Bcl-XL, forward 5′-ctgggacacttttgtggatctct-3′, reverse 5′-gaagcgctcctggccttt-3′; mouse Bfl-1, forward 5′-gcagattgccctggatgtat-3′, reverse 5′-agaaaagtcagccagccacta-3′; mouse Bax, forward 5′-gagatgaactggacagcaat-3′, reverse 5′-atcccaggaaaatgtcataa-3′; GAPDH, forward 5′-gaacgggaagcttgtcatcaa-3′, reverse 5′-ctaagcagttggtggtgcag-3′. The relative gene expression was determined using the comparative computed tomography (CT) method and normalized to the housekeeping gene GAPDH. To simplify the representation of Bcl-2, Bcl-XL, Bfl-1, and Bax expression, the gene expression data normalized for GAPDH are shown as fold increases compared with levels in the rat IgG group and are means ± SD of triplicate experiments.
Kidney epithelial cells. Kidneys were removed, sliced, and digested with collagenase type IV in HBSS for 30 min at 37°C. The digest was centrifuged, and the cell pellet was resuspended in a cold culture medium with 10% fetal bovine serum. The cell suspension was plated onto tissue culture dishes. Cells were allowed to adhere for 2 h. Unbound cells were removed by washing thrice with PBS. Only adherent cells were maintained as primary cultures at 37°C for 7 d; these cells were then used for experiments.
Western blot. Samples of 20-μg kidney protein were diluted with 4× SDS sample buffer, applied to SDS-PAGE gels, fractionated, and transferred to nitrocellulose membranes (Millipore). The membranes were incubated with anti–4-1BB mAb (3E1) and anti-rat IgG–horseradish peroxidase, and bound Ab was detected by enhanced chemiluminescence (Amersham Pharmacia Biotech).
Small interfering RNA transfection and treatment. The cultured kidney epithelial cells from Rag2γc- or 4-1BB–deficient mice were transfected with either gene-specific small interfering RNA (siRNA) against Bcl-XL, Bfl-1, and Bax or nontargeting control siRNA according to the manufacturer's directions (Santa Cruz Biotechology). Transfection efficiencies were assessed by RT-PCR analysis 36 h after gene transfer. Two days after the transfection, cells were treated with 2 μg/mL cisplatin and/or 10 μg/mL rat IgG or anti–4-1BB. The absolute numbers of viable cells were measured in primary cultures of mouse kidney epithelial cells at the indicated time points.
Statistics. Student's t test and one-way ANOVA were performed using the Prism software program (GraphPad Software).
Results and Discussion
We screened a variety of anticancer chemotherapeutics to select one that enhanced anticancer activity of the agonistic anti–4-1BB. A synergistic anticancer effect was observed when cisplatin was used in combination with anti–4-1BB in B16 melanoma and CT26 carcinoma models. Mice receiving combination therapy seldom showed the hunched posture, ruffled fur, and diarrhea often seen with cisplatin alone.
We examined mechanisms involved in this anticancer synergy and reduced nephrotoxicity in the CT26 model. Mice were treated with anti–4-1BB and/or cisplatin beginning on PI day 0 or 6: >50% of 4-1BB–treated mice showed delayed tumor growth and 10% to 30% were tumor free, whereas rat IgG- or cisplatin-alone–treated mice were dead or sacrificed by PI day 35 due to the large tumor burden ( Fig. 1A and B ). Anti–4-1BB plus cisplatin resulted in 60% to 70% tumor-free mice in both preventive and therapeutic models. Mice tumor free for >100 days after combination treatment were rechallenged with CT26 or RENCA cells. CT26 did not grow, whereas RENCA did ( Fig. 1C). Depletion of CD8+ T cells from the tumor-free mice followed by rechallenge with CT26 produced tumor growth; depletion of CD4 or natural killer (NK) cells did not ( Fig. 1C). These results suggest that cisplatin enhances 4-1BB–mediated anticancer effects and that combination therapy leads to CD8+ T cell–mediated long lasting memory. Anti–4-1BB increased the numbers of IFN-γ+CD4+ and CD8+ T cells including AH-1–specific CD8+ T cells, and cisplatin provided further enhancement ( Fig. 1D; Supplementary Fig. S1). CD11c+CD8+ T cells ( 9) were the major source of IFN-γ–producing CD8+ T cells ( Fig. 1D; Supplementary Fig. S2).
Combination therapy showed enhanced expansion of all three phenotypes of CD8+ ( Fig. 2A ) and CD4+ T cells (Supplementary Fig. S3)—naïve (CD62LhighCD44low), effector (CD62LlowCD44high), and memory (CD62LlowCD44high)—compared with other monotherapies. Anti–4-1BB alone expanded effector and memory CD8+ T cells significantly compared with IgG or cisplatin alone. 4-1BB effectively expanded T cells after cisplatin-induced lymphopenia and expansion after combination therapy was more vigorous than with 4-1BB alone ( Fig. 2A). Reduction of Foxp3+ Treg cells in TDLN and spleen was observed at 2 to 7 days after cisplatin treatment (Supplementary Fig. S4). This may contribute to the cisplatin-mediated anticancer effect.
Absolute numbers of viable cells in the TDLNs on PI day 5 showed that cisplatin reduced total lymphoid cells by >78%, T cells by >70%, and B cells by >90% ( Fig. 2B). Combination therapy enhanced lymphoid cells by PI day 5 compared with cisplatin alone: total numbers of TDLN cells increased >11-fold, T cells by >7-fold (CD4, >5-fold; CD8, >11-fold), and B cells by >54-fold. Combination therapy was also more effective in repopulating lymphoid cells than anti–4-1BB alone (total cells, 1.7-fold; B cells, 2.5-fold; Fig. 2B). Analysis of TDLN cells on PI day 17 showed a 4-fold increase in CD8+ T cells and a 2-fold increase in CD4+ T cells with combination therapy, compared with rat IgG treatment (Supplementary Fig. S5).
We hypothesized that cisplatin enhanced the expression of 4-1BB on the T-cell surface, cross-linking of which produced more antiapoptotic molecules, resulting in increased lymphocytes after combination therapy. To create a system that measure 4-1BB expression in both antigen-primed and naïve CD8+ T cells simultaneously, OT-1 CD8+ T cells were adoptively transferred into congenic C57BL/6 mice, which were treated with OVA and/or cisplatin. Cisplatin increased expression of 4-1BB on OVA/H2Kb-positive CD8+ T cells from day 1 of antigen injection; the higher levels persisted until day 6 to 8, in which 4-1BB in cisplatin plus OVA-treated mice ∼2-fold higher than in OVA alone group ( Fig. 2C). In separate experiments, we showed that enhanced release of tumor antigen by cisplatin did not contribute to the enhanced expression of 4-1BB and expansion of CD8+ T cells (Supplementary Fig. S6).
Interestingly, cross-linking of 4-1BB in the presence of cisplatin increased expression of Bcl-2 and Bfl-1 almost 2-fold compared with anti–4-1BB alone. Expression of Bax, a molecule known to induce cisplatin-mediated apoptosis of kidney epithelium ( 15), did not change ( Fig. 2D). Cisplatin-mediated enhancement of 4-1BB expression on the T-cell surface and concomitant enhancement of antiapoptotic molecules with anti–4-1BB treatment may contribute to the rapid expansion of T cells and enhanced tumor killing activity.
Cisplatin increased renal tubular damage dose dependently; damage was limited and localized at 200 μg but widespread in all parts of the cortex at 400 or 800 μg. However, high-dose tubular damage was ameliorated with anti–4-1BB (Supplementary Fig. S7). Combination treatment reduced the semiquantitative scores for renal tubular epithelial cell necrosis (Supplementary Fig. S8) and loss by 2-fold compared with cisplatin alone ( Fig. 3A ). The concentration of serum creatinine increased dose dependently with cisplatin and decreased with addition of anti–4-1BB ( Fig. 3B).
To explain 4-1BB–mediated amelioration of cisplatin-induced nephrotoxicity, expression of 4-1BB ( 16) was examined in renal tissue after cisplatin treatment in ex vivo. To minimize infiltration of 4-1BB–expressing immune cells into the kidney, cisplatin was injected into Rag2γc DKO mice at doses indicated. 4-1BB was detected only in the kidneys from mice receiving 400 μg or more cisplatin but not in those receiving lower doses ( Fig. 3CI and II). 4-1BB–positive cells were located in the kidney cortex where renal tubular damage was increased by cisplatin ( Fig. 3D).
4-1BB expression was detected in cisplatin-treated in vitro cultured kidney epithelium but not in PBS-treated cells ( Fig. 4A ). Human embryonic kidney cells (HEK 293) constitutively expressed 4-1BB, which was marginally increased by cisplatin (Supplementary Fig. S9). We tested whether cross-linking of 4-1BB on cultured kidney epithelium enhanced the expression of antiapoptotic molecules as with T cells ( 17, 18). Cisplatin decreased mRNA of antiapoptotic molecules such as Bcl-2, Bcl-XL, and Bfl-1; only the decrease of Bfl-1 mRNA was statistically significant. Conversely, combination with anti–4-1BB significantly increased the expression of Bfl-1. The proapoptotic gene Bax increased with cisplatin treatment and decreased with anti–4-1BB in vivo ( Fig. 4B).
We tested whether the 4-1BB–mediated survival of kidney epithelium was related to the changes in Bfl-1 and Bax expression levels using gene-specific siRNA. Transfection of siRNA specific to Bfl-1 or Bax decreased mRNA levels of Bfl-1 or Bax by 40% to 50% of control siRNA in primary cultures of kidney epithelium ( Fig. 4C). Bfl-1 siRNA reduced anti–4-1BB–mediated survival of kidney epithelium by 32.5% compared with the control, whereas Bax siRNA enhanced the survival of kidney epithelium by 2.14-fold compared with cisplatin alone and by 1.46-fold compared with cisplatin plus anti–4-1BB ( Fig. 4D). We concluded that cisplatin directly enhances 4-1BB expression in renal tubular epithelial cells, which is involved in the anti–4-1BB–mediated protection of kidney epithelial cells by regulating the expression of antiapoptotic and proapoptotic molecules.
It seems that ciplatin does not produce any tumor-killing activities against CT26 ( Fig. 1); the therapeutic effects of anti–4-1BB plus cisplatin were completely dependent on both primary and secondary antitumor activities of CD8+ T cells. This suggests that cisplatin rather preconditions the host, possibly dampening the immunosuppressive network, and providing the necessary lymphoid space and cytokine milieu for augmented 4-1BB–mediated expansion and differentiation of CD8+ T cells.
It is noteworthy that B cells were the most sensitive to cisplatin toxicity and anti–4-1BB was very effective in repopulating B cells ( Fig. 2). Anti–4-1BB may cross-link 4-1BB on the B-cell surface ( 19) or alternatively repopulated CD4+ T cells may provide growth factors for B cells. Cisplatin enhanced 4-1BB expression on antigen-primed CD8+ T cells and anti–4-1BB markedly enhanced the repopulation of lymphocytes ( Fig. 2). Because 4-1BB is induced in an antigen-specific manner, the majority of repopulated T cells are likely against tumor antigens. Although we could not directly quantify the tumor-specific CD8+ T cells in current studies because of the unavailability of CT26-specific tetramers, we found that, in another tumor model (B16 melanoma), the major portion of the CD8+ T cells repopulated by anti–4-1BB were tumor-antigen specific. 4
It is a novel finding that cisplatin induced 4-1BB on the kidney epithelium ( Fig. 3). Provision of the antiapoptotic signals to renal tubular epithelium by anti–4-1BB seemed to contribute to the improvement in cisplatin-induced nephrotoxicity. Both enhancement of Bfl-1 and suppression of Bax ( Fig. 4) seem important factors involved in anti–4-1BB–mediated amelioration of nephrotoxicity.
Current studies suggest that one single type of chemotherapy or immunotherapy may not be able to deal with all facets of tumor progression, and tumor- and immune system–derived regulatory barriers. Therefore, a rational combination of synergistic chemotherapy and immunotherapy could provide a successful clinical approach to human cancers.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Grant support: National Cancer Center, Korea (NCC-0890830-1 and NCC-0810720-1), Korean Research Foundation (KRF-2005-201-E00008, KRF-2005-084-E00001), Arthritis Foundation (Innovative Research Award), Stem Cell Project (MEST-2006-04212), and NIH grant R01EY013325.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).
↵4 Unpublished observation.
- Received April 11, 2008.
- Revision received July 2, 2008.
- Accepted July 22, 2008.
- ©2008 American Association for Cancer Research.