Inhibition of Aurora-A promotes CD8+ T cell infiltration by mediating IL-10 production in cancer cells
Jing Han, Zhen Jiang, Chennan Wang, Xin Chen, Rongqing Li, Na Sun, Xiangye Liu, Hui Wang, Li Hong, Kuiyang Zheng, Jing Yang, Takayuki Ikezoe
1 Jiangsu Province Key Laboratory of Immunity and Metabolism,
2 Department of Pathogenic Biology and Immunology,
3 National Experimental Demonstration Center for Basic Medicine Education, Xuzhou Medical University, Xuzhou, Jiangsu, China;
4 Heilongjiang Province Key Laboratory of Microecology and Immunity,
5 The Department of Pathological Anatomy, Jiamusi University, Jiamusi, Heilongjiang, China; 6The Department of Hematology, Fukushima Medical University, Fukushima, Japan.
Abstract
Intratumoral tumor-specific activated CD8+ T cells with functions in antitumor immune surveillance predict metastasis and clinical outcome in human colorectal cancer (CRC). Intratumoral CD8+ T cells also affect treatment with immune checkpoint inhibitors. Interestingly, inhibition of Aurora kinase A (Aurora-A) by its selective inhibitor alisertib obviously induced infiltration of CD8+ T cells. However, the mechanisms by which inhibition of Aurora-A promotes infiltration of intratumoral CD8+ T cells remain unclear. Our recent results demonstrated that conditional deletion of the AURKA gene or blockade of Aurora-A by alisertib slowed tumor growth in association with an increase in the infiltration of intratumoral CD8+ T cells as well as the mRNA levels of their interleukin-10 receptor (IL-10R). The antitumor effects of targeting Aurora-A were attenuated in the absence of CD8+ T cells. Additionally, antibody-mediated blockade of IL-10R dramatically decreased the percentage of intratumoral CD8+ T cells. In further experiments, we found that the levels of IL-10 were elevated in the serum of Azoxymethane (AOM)/Dextran Sodium Sulfate(DSS)-treated AURKAflox/+;VillinCre+ mice. Unexpectedly, we found that in addition to Aurora-A’s mitotic role, inhibition of Aurora-A elevated IL-10 transcription, which in turn increased the IL-10R mRNA levels in CD8+ T cells. Thus, inhibition of Aurora-A could be a useful treatment strategy for recruiting tumor-specific intratumoral CD8+ T cells.
Implications: Understanding the mechanisms by which inhibition of Aurora-A promotes CD8+ T cell infiltration and activation, as mediated by the IL-10 pathway, could provide a potential strategy for tumor immunotherapy.
Introduction
The extent of intratumoral CD8+ T cells is predictive of the overall survival of individuals with colorectal cancer (1-6), suggesting that the infiltration and activation of intratumoral CD8+ T cells results in an improved prognosis for patients (7-9).
Unfortunately, recent studies have shown that only approximately 10% of intratumoral CD8+ T cells are tumor-specific CD8+ T cells (10). Moreover, multiple studies have demonstrated systemic activation of the T cell repertoire in parallel with dose-limiting autoimmunity, although immunotherapy with multiple immune checkpoint inhibitors improved the clinical responses (11, 12). These findings indicated that elucidating the underlying mechanisms and developing strategies to recruit, activate and expand tumor-specific intratumoral CD8+ T cells are urgently needed.
The capacity of migration and activation of CD8+ T cells is governed in a complex manner by surface expression of chemokine receptors and specific ligands (13). Several cytokines, such as interleukin 10 (IL-10), have been shown to be critical for CD8+ T cell migration and activation. IL-10 is well known for itsanti-inflammatory function mediated by inhibiting the secretion of proinflammatory cytokines and suppressing the expression of major histocompatibility complex (MHC) molecules as well as costimulatory molecules (14, 15). However, mice and humans deficient in IL-10 or the IL-10 receptor develop inflammatory bowel disease and cancer (16, 17). Interestingly, recent studies have found that tumor clearance was enhanced in the presence of IL-10 (18, 19). Additionally, elevation of IL-10 in experimental tumors potentiated the antitumor effects of tumor-specific CD8+ T cells (14, 20-23). Importantly, recent studies have demonstrated that PEGylated IL-10 promoted the expansion of tumor-specific CD8+ T cells and enhanced polyclonal T cell expansion in cancer patients (9, 20, 21). All of these observations suggested that inducing IL-10 secretion could be a useful treatment strategy.
Aurora-A, one of the members of the serine/threonine kinase family, functions asa “mitosis sensor” and is aberrantly expressed in various types of cancers, including colorectal cancer and hematopoietic malignancies (24, 25). During antigen-drivenT-cell activation, Aurora-A was shown to be phosphorylated and recruited to the immunological synapse (IS). Either deletion of AURKA or pharmacological inhibition of Aurora-A function reduced the accumulation of phosphorylated Aurora-A at the IS and impeded the movement of vesicles towards the IS structure but did not result in a global defect in cytoskeleton dynamics (26), indicating that Aurora-A can be involved in controlling the immune response. Treatment of a melanoma xenograft model with alisertib, an oral Aurora-A selective inhibitor, enhanced the infiltration of helper and cytotoxic T-cells into tumors (27), indicating that inhibition of Aurora-A could promote infiltration of intratumoral CD8+ T cells. However, the mechanisms by which blockade of Aurora-A promote tumor-infiltrating lymphocytes may not be related to the critical role of Aurora-A in antigen-driven T-cell activation. A previous study showed that nuclear Aurora-A interacted with heterogeneous nuclear ribonucleoprotein K (hnRNP K) and functioned as a transcription factor (28). To investigate whether inhibition of Aurora-A promoted CD8+ T cell infiltration, we utilized the CT-26 mouse model and the AOM/DSS colitis-associated model of colon cancer. We found that blockade of Aurora-A induced infiltration of CD8+ T cells through elevated IL-10, at least partially.
Materials and Methods Reagents
Alisertib (S1133) was purchased from Selleck Chemicals (Selleckchem, Shanghai, China). The antibody for IFN- (505809), an in vivo anti-CD8 antibody (100701) and an in vivo anti-IL-10R antibody (112711) were purchased from Biolegend (San Diego, CA). Antibodies against CD45 (550994), CD3 (553061), CD8 (552877), CD19 (560375), CD11b (557657), CD206 (565250), and F4/80 (565410) were purchased from BD Biosciences (San Jose, CA). Recombinant mouse IL-10 (abs04067) was purchased from Absin (Shanghai, China).
Isolation of intestinal epithelial cells, cell lines and cell culture
The crypt cells were isolated as previously described (29). Briefly, the intestinal sections were incubated in precooled Hank’s balanced salt solution (HBSS; Gibco, Invitrogen) containing antibiotics at room temperature for 15 min and then cut into 0.5-cm pieces. Crypt epithelial cells were dissociated by repeated vigorous shaking and collected by centrifugation at 200 x g for 15 min at 4°C. Cell pellets were resuspended in 10 mL of HBSS containing 0.4 mg/mL dispase (Life Technologies), 10,000 U/mL penicillin/streptomycin and 5% FBS. After 30 min of incubation at 37°C, the cells were collected by centrifugation at 150 x g for 10 min at 4°C. Murine CT-26 cells were purchased from ATCC (Manassas, VA) and grown in RPMI-1640 containing 10% heat-inactivated fetal bovine serum (FBS). Murine MC-38 cells and human SW480 cells were purchased from FuHeng (FuHeng Cell Center, Shanghai, China) and maintained in DMEM supplemented with 10% FBS.
The identities of the cell lines were verified by short tandem repeat analysis. All cell lines were used in their early passages and routinely tested for mycoplasma contamination by mycoplasma test kit (Clark Bioscience, T102).
Generation of the mice with conditional AURKA knockout
For generation of the mice with conditional knockout of AURKA, an 8.8-kb mouse AURKA gene-containing fragment was subcloned into the pBluescript II SK vector and was used to create an AURKA loxP targeting vector. This vector was constructed by inserting an FRT-PGK-Neo-FRT-loxP cassette into intron 3 and another loxP site into intron 4, followed by an HSV-TK expression cassette at the 3′ end of the construct to be used as a negative selection marker. For excision of the neomycin selection marker, the transgenic mice were mated with Ella-Cre transgenic mice (Jackson Laboratories, stock number 003724), followed by five generations of backcrossing to C57BL/6 mice. The VillinCre+ mice (B6.SJL-Tg(VilCre)997Gum/J) were intercrossed with mice carrying loxP-flanked AURKA alleles (AURKAflox/flox). PCR was utilized to identify the conditional AURKA knockout using primers (p55′-GGTAAGTGGTCTTGGGTGCT-3′; p6 5′- TAGCCAACTCATCTCCTCTG-3′) and VilCre (F 5′- TTTCCCGCAGAACCTGAAGA-3′; R5′-GGTGCTAACCAGCGTTTTCGT-3′) alleles. Mice were strictly bred and maintained under protocols approved by the Institutional Animal Care and Use Committee at Xuzhou Medical University.
Model of colorectal cancer associated with colitis
Colorectal cancer associated with colitis was induced as described previously (29, 30). Briefly, 12- to 14-week-old mice were intraperitoneal (i.p.) injected with 10 mg/kg AOM (Sigma-Aldrich), followed by treatment with 1% DSS (MW 36,000-50,000 Da; MP Biomedicals) in drinking water for 7 consecutive days. Then, the mice were administered normal drinking water for 14 days. This DSS treatment was repeated for three additional cycles. On day 63 of the regime, the mice were euthanized. The intestinal and colon sections were removed, washed with PBS and opened longitudinally for analysis.
Ectopic tumor implantation and weight measurements
The mouse strains were maintained and housed under specific pathogen-free (SPF) conditions. The tumor experiments used 6-week-old female Balb/c mice. The left and right flanks of the mice were shaved, and CT-26 cells (1×106) were injected subcutaneously (s.c.). When palpable tumors formed, the mice were randomized to receive either a drug diluent alone (control) or alisertib (20 mg/kg) orally for 14 consecutive days. The tumor size was measured thrice a week using calipers. The tumor sizes were calculated using the formula a x b x c, where “a” is the length, “b” is the width, and “c” is the height in millimeters. At the end of the experiment, theanimals were killed by CO2 asphyxiation, and the tumor weights were measured after careful resection. The tumor tissues were collected for further analysis.
For CD8+ cell depletion studies, anti-CD8 (clone 53-6.7) antibody at 0.35 mg per mouse (Biolegend) was administered 4 days and 1 day prior to initiation of dosing with alisertib and every 7 days thereafter. The rat IgG2a κ isotype control antibody clone RTK2758 (Biolegend) was used as a negative control. The animals were not dosed with alisertib on the days they received the antibodies to limit anti-antibody responses. Depletion of CD8+ T cells was confirmed by flow cytometry on days 8 and 13.
For the Rag2-/- studies in vivo, wild-type C57BL/6 and Rag2-/- mice were a gift from Hui Wang (Xuzhou Medical University). A total of 1×106 MC-38 cells were injecteds.c. into the left flank of 8-12-week-old wild-type or Rag2-/- mice. When palpable tumors formed, the mice were administered either a drug diluent alone (control) or alisertib (20 mg/kg) orally every day. The tumor size was monitored every other day. After 14 days, the animals were killed by CO2 asphyxiation, and the tumors were isolated to determine the weight following careful dissection.
For the blocking IL-10R studies in vivo, ultralow endotoxin, azide-free purified blocking antibody against the murine IL-10 receptor (IL-10 R; CD210)(α-IL-10R)-clone 1B1.3a or mouse IgG2a, κ isotype control antibody clone MOPC-173 (Biolegend), was administered i.p. six times at a dose of 0.25 mg ofantibody to the mice on days 0, 1, 2, 5, 8, and 11. During the anti-IL-10R antibody treatment, alisertib was administered orally every day.
Tumor digestion and preparation of single cell suspensions
For TIL isolation, the tumors were removed and manually dissociated. The cells were mashed through 70 m filters, washed twice and analyzed by flow cytometry.
Suspensions of spleen cells were obtained by mashing the spleen through a 70-mnylon cell strainer (BD Falcon).
Surface and intracellular staining, flow cytometry and cell sorting
Surface staining was performed with mAbs for 30 min at 4°C in PBS using the indicated antibodies listed. For intranuclear staining, the cells were first stained at thesurface before fixation and permeabilization using a transcription factor staining kit (BD Biosciences) followed by intranuclear staining. Spleens, dLNs and tumors were isolated. Single cell suspensions were used for flow cytometry. Data were analyzed using FlowJo (FlowJo_V10).
For cell sorting, CD8+ cells enriched from the TIL suspensions were stained with anti-CD3e (BD Biosciences, 145-2C11), anti-CD8a (BD Biosciences, 53-6.7) and anti-CD45 (BD Biosciences, 30-F11). CD45+CD3+CD8+ cells were sorted on a FACSAria (BD) flow cytometer. The infiltrating T cells (CD45+CD3+ cells) from the intestinal and colon tissues of Aurora-A conditional knockout mice were stained andsorted. The infiltrating myeloid cells from the intestinal and colon tissues of Aurora-A conditional knockout mice were stained with anti-CD11b (BD Biosciences, M1/70) and anti-CD45 (BD Biosciences, 30-F11). CD45+CD11b+ cells were sorted on a FACSAria (BD) flow cytometer. The purity of the sorted cells was greater than 99%. Real-time reverse transcription-polymerase chain reaction (RT-PCR)
A total of 1×104 cells were lysed in lysis buffer (TaKaRa, 3735A) and subjected to real-time RT-PCR according to the instruments. The primer sets for PCR were shown in Supplementary Table S1.
Histologic analysis
Tissues were fixed in 4% formaldehyde and embedded in paraffin. Sections with a thickness of 4 µm were stained with hematoxylin-eosin (Beyotime Biotechnology, Nantong, Jiangsu, China) as described previously(29, 31).
Immunohistochemistry
Immunohistochemistry staining was performed on formalin-fixed, paraffin-embedded (FFPE) tissue samples purchased from Shanghai Outdo Biotech. The levels of the immunostained tissues were evaluated independently by two pathologists who were blind to the subjects’ clinical information. Between 15 and 20 high-power fields were viewed. Criteria were developed for quantitating the immunoreactivities of Aurora-A and CD8 staining using a score range of 0 to +3, where 0 indicated no positive cell staining, +1 indicated less than 10% positive cell staining, +2 indicated 10-30% positive cell staining, and +3 indicated more than 30% positive cell staining. Similarly, the intensity of staining was also graded as +0, +1, +2, or +3 as previously described (32).
Multiplexed immunofluorescence staining and quantification
Multiplex staining and multispectral imaging were performed to identify the cell subsets expressing CD8a, PD-1, TIM-3, IFN- and Ki-67 in the TME using a PANO 7-plex IHC kit (cat 0004100100, Panovue, Beijing, China). Different primary antibodies were sequentially applied, followed by horseradish peroxidase-conjugated secondary antibody incubation and tyramide signal amplification. The slides were microwave heat-treated after each TSA operation. Nuclei were stained with 4′-6′-diamidino-2-phenylindole (DAPI, Sigma) after all the human antigens had been labelled.
For multispectral images, the stained slides were scanned using the Mantra System (PerkinElmer, Waltham, MA), which captures the fluorescent spectra at 20-nm wavelength intervals from 420 to 720 nm with identical exposure time; the scans were combined to build a single stack image. Quantification of the positively stained cells was verified with manual counting from 10 to 12 random fields at 100× magnification. The mean number from all fields of each tumor sample was taken.
Transwell assay
Transwell assays were performed as previously described(33).
Enzyme-linked immunosorbent assay (ELISA)
The concentrations of IL-10 in the culture medium of the CT-26 cells and the serum of the AOM/DSS-induced mouse model were examined using an ELISA kit (Multi Sciences, Zhejiang, China) according to the manufacturer’s instructions.
Luciferase reporter assays
Luciferase assays were performed as previously described (34). Briefly, either 400 ng pGL4.20 vector or the il10 promoter-driven firefly luciferase reporter plasmid (400 ng) was transfected into cells with 4 µl/well GenXPIII transfection reagent (Probegene, Jiangsu, China). A pRL-TK plasmid (20 ng) was cotransfected as a control for transfection efficiency. Luciferase activity was measured for 20 seconds in a luminometer. The il10 promoter activity of each construct is expressed as fireflyluciferase/Renilla luciferase activity.
Survival analysis and correlation analysis using the GEPIA web tool
The online database Gene Expression Profiling Interactive Analysis (GEPIA2, http://gepia2.cancer-pku.cn/#index) was used to analyze a pairwise gene correlation analysis for any given set of expression data from The Cancer Genome Atlas (TCGA) and/or the Genotype-Tissue Expression (GTEx) using Pearson correlation statistics.
Statistical analysis
Statistical analysis was performed to assess the difference using a paired Student’s t test or one-way ANOVA followed by Bonferroni post-tests by PRISM statistical analysis software (GraphPad Software, San Diego, CA). Data are presented as the mean ± SEM. Significance is indicated as follows: *p < 0.05, **p <0.01 or n.s. for not significant.
Results
Inhibition of Aurora-A by alisertib promoted infiltration of intratumoral CD8+ T cells
We first examined whether inhibition of Aurora-A activity by alisertib could enhance the infiltration of intratumoral CD8+ IFN-γ+ T cells. Balb/c mice were subcutaneously engrafted with murine CT-26 colorectal cancer cells. Tumor growth was monitored with or without oral administration of alisertib. Alisertib dramatically suppressed the tumor growth and the tumor weight without any toxicity (Fig. 1A-C). Consistent with a previous study (27), the therapeutic efficacy was accompanied by an increase in the percentage of intratumoral CD8+ IFN-γ+ T cells (Fig. 1D, E and F).
We next examined the expression of the immune checkpoint molecules and found that in the alisertib-treated tumors, the CD8+ PD-1+ T cells were almost identical to those in the control-treated tumors (Fig. 1D, E and G). However, the CD8+ TIM3+ T cells were significantly increased in response to alisertib treatment (Fig. 1D, E and H). In the absence of alisertib, the CD8+ T cells displayed an exhausted phenotype expressing PD-1 and TIM-3 and had a low proliferative index as measured by the Ki-67+ subset in these cells (Fig. 1E, G and H), consistent with previous studies (9, 35). However, upon alisertib treatment, the population of proliferating CD8+ T cells expressing either PD-1 or TIM-3 was dramatically elevated (Fig. 1E, G and H), indicating that alisertib could induce the activation and expansion of CD8+ T cells expressing the immune checkpoint molecules. Interestingly, CD8+ IFN-γ+ T cells were rarely detected in the spleen and lymph nodes of the mice after administration with alisertib (Fig. 1I, and data not shown), suggesting that activation and expansion of intratumoral CD8+ T cells were in the tumors rather than in the secondary lymphoid organs.
CD8+ T cells were critical for alisertib mediated tumor suppression
To assess whether CD8+ T cells were essential for the alisertib-mediated tumor suppression, we treated the mice with anti-CD8 antibodies, which led to significant depletion of CD8+ T cells in the peripheral blood (Fig. 2A). Importantly, depletion of the CD8+ T cells significantly promoted the tumor growth and the tumor weight inassociation with a decrease in the population of intratumoral CD8+ T cells in the presence of alisertib (Fig. 2B, C and D). To further investigate whether the antitumor effect of alisertib requires immune function, we established syngeneic tumor models by injection of the mouse colon adenocarcinoma cell line MC-38 into age-matched 6- to 8-week-old wild-type C57BL/6 or immunodeficient Rag2-/- mice that lack T cells. We found that alisertib did not suppress tumor growth in the Rag2-/- mice compared with the wild-type mice (Fig. 2E and F), suggesting that the antitumor effect of alisertib was dependent on the extent of infiltration of intratumoral CD8+ T cells, at least partially.
Deletion of Aurora-A decreased the colitis-associated tumor incidence in parallel with an increase in the number of intratumoral CD8+ T cells
Aurora-A is required for early embryonic development, and AURKA knockout (KO) leads to early embryonic lethality (36). To avoid off-target effects and investigate the role of Aurora-A in promoting infiltration of CD8+ T cells, we generated mice with Aurora-A conditional knockout (CKO) by crossing AURKAflox/flox (AURKAf/f) mice (Fig. 3A) with VillinCre+(VilCre+) mice, which induced
Cre-dependent recombination in the intestinal epithelium by embryonic day (E) 12.5. Villin expression was also found in the primitive endoderm, gut, nephron anlagen and developing embryo (37, 38). A lethal phenotype was observed after intercrossing AURKAf/+;VilCre+ with AURKAf/+ (data not shown). We therefore utilized the AURKAf/+;VilCre+ mice. The specific deletion of AURKA alleles in adult AURKAf/+;VilCre+ mice was determined by PCR analysis of genomic DNA (Fig. 3B). As shown in Fig. 3C, the Aurora-A mRNA levels were obviously decreased in crypt cells isolated from the intestines of the AURKAf/+;VilCre+ mice compared to those from the AURKAf/+ mice. Analyses of the intestine and colon sections from the AURKAf/+;VilCre+ and the AURKAf/+ mice at 6 weeks showed that the number of CD8+ T cells was increased in the intestine and colon tissues from the AURKAf/+;VilCre+ mice compared to that of AURKAf/+ mice (Fig. 3D-F). To determine if the deletion had any effects later in development, we established a cohortof male and female mice and monitored them weekly for weight, activity and overt signs of disease. Both the male and female heterozygous deletion mice appeared to be identical to their AURKAf/+ counterparts (data not shown). The AURKAf/+;VilCre+ mice did not demonstrate any notable differences in disease occurrence, spontaneous tumor formation, or mortality compared to the AURKAf/+ mice (data not shown).
To further study whether deletion of AURKA also contributed to recruitment of intratumoral CD8+ T cell infiltration, we treated the mice with a combination of the carcinogen AOM and DSS. The weight loss and AOM/DSS-induced mortality were similar in the AURKAf/+;VilCre+ and AURKAf/+ mice (data not shown).
At the end of the experiments, the mice were euthanized. The tumor incidence was 100% in the AURKAf/+ mice. However, only 6 of 11 AURKAf/+;VilCre+ mice treated with AOM/DSS developed tumors located predominantly in the distal to middle colon (Fig. 3G). Compared with those of the AURKAf/+ mice, the number of tumors and the size of the tumors were substantially decreased in the AURKAf/+;VilCre+ mice (Fig. 3H). Importantly, the population of intratumoral CD8+ T cells was strongly elevated in the tumors of the AURKAf/+;VilCre+ mice (Fig. 3I). To further test whether CD8+ T cells are involved in alisertib-mediated tumor suppression, we treated the AOM/DSS mouse model with alisertib and found that upon alisertib treatment, the tumor sizes decreased along with an increase in the percentage of intratumoral CD8+ T cells in the AOM/DSS-treated AURKAf/+;VilCre+ mice compared with the AOM/DSS-treated AURKAf/+ mice (Supplementary Fig. S1A and B). Consistent with the findings in the CT-26 mouse model, the number of CD8+ T cells in the thymus and lymph nodes of the AURKAf/+;VilCre+ mice was identical to that of AURKAf/+ mice (data not shown), suggesting that inhibition of Aurora-A not only suppressed colon tumor initiation but also suppressed colon tumor development by promoting infiltration of CD8+ T cells.
Inhibition of Aurora-A-mediated infiltration of intratumoral CD8+ T cells was dependent on induction of IL-10R expression in these cells
To determine the mechanisms by which silencing Aurora-A by alisertib or deletion promoted infiltration of intratumoral CD8+ T cells, we performed gene expression profiling of FACS-sorted intratumoral CD8+ T cells. Real-time RT-PCR confirmed that the mRNA levels of interferon- (IFN-), granzyme B (GzmB), and perforin (PFP), an essential player in the cytotoxic activity of cytotoxic CD8+ T cells (39), were significantly increased in the infiltrated CD8+ T cells isolated from the alisertib-treated tumors (Fig. 4A). Notably, the IL-10R mRNA levels were obviously elevated in the CD8+ T cells isolated from the alisertib-treated tumors compared to those of the control-treated tumors (Fig. 4A). IL-10R-deficient children were inclined to develop B cell lymphomas due to a lack of infiltration of cytotoxic T cells (17). We therefore investigated the role of IL-10R in the alisertib-induced recruitment and activation of intratumoral CD8+ T cells utilizing an IL-10R-specific blocking antibody. As shown in Fig. 4B, blockade of IL-10R attenuated the ability of alisertib to suppress tumor growth. Additionally, compared to those of the alisertib alone-treated mice, the tumor weights increased in parallel with a decrease in the population of infiltrated CD8+ T cells in the presence of the anti-IL-10R antibody and alisertib (Fig. 4C and D). All of these observations implicated IL-10R as a key mediator in alisertib-induced infiltration of tumor-specific CD8+ T cells.
In contrast to IL-10 R IL-10R appears to exclusively bind IL-10 (40).
Additionally, upon IL-10 treatment, the cytotoxicity and expansion of tumor-specific CD8+ T cells as well as the expression of IL-10R on these cells were potently induced (9, 20), suggesting that IL-10 could be involved in the alisertib-mediated infiltration of intratumoral CD8+ T cells. To address whether elevated expression of IL-10R in infiltrated CD8+ T cells was a consequence of the induction of IL-10 mediated by Aurora-A inhibition, we exposed freshly sorted CD8+ T cells to IL-10 and found that the IL-10R mRNA levels were elevated in the presence of IL-10 (Supplementary Fig. S2), suggesting that the upregulated expression of IL-10R in CD8+ T cells could be a consequence of the elevated IL-10 production.
IL-10 transcription and production were enhanced
To explore the role of Aurora-A in IL-10 production, we examined the IL-10 levels in the culture medium of the CT-26 cells isolated from either the control or alisertib-treated tumors. We found that the concentration of IL-10 was obviously increased in the culture medium of the CT-26 cells isolated from the alisertib-treated mice compared with that of the CT-26 cells isolated from the control-treated mice (Fig.4E). In addition, the mRNA levels of IL-10 were enhanced in the alisertib-treated CT-26 cells compared with the control-treated CT-26 cells (Fig. 4F). Similarly, the
IL-10 mRNA levels were dramatically elevated in crypt cells but not the infiltrating T cells (CD3+ T cells) or CD11b+ cells enriched in myeloid-derived suppressor cells (MDSCs) that express high levels of IL-10 in colon tissue (41) isolated from the colon of AURKAf/+;VilCre+ mice compared to AURKAf/+ mice (Fig. 4G). The levels of IL-10 in the serum of the AOM/DSS-treated AURKAf/+;VilCre+ mice were also increased compared to those of the AURKAf/+ mice, although IL-10 could not be detected in the serum of both the AURKAf/+;VilCre+ and AURKAf/+ mice (Fig. 4H).
Suppression of Aurora-A induced IL-10 transcription in a mitotic defect-independent manner
Given that IL-10 facilitated the Aurora-A inhibition-induced CD8+ T cell infiltration, we further examined which subset of cells were the major source of IL-10, as well as the mechanisms by which IL-10 was affected. The number of CD19+ B cells was reduced in the alisertib-treated tumors compared to the control-treated tumors (Supplementary Fig. S3A). Similarly, the population of F4/80+CD11b+CD206+ cells was decreased in the tumors isolated from the mice that received alisertib (Supplementary Fig. S3B), indicating that upon alisertib treatment, cancer cells might contribute to elevated IL-10.
To investigate whether cancer cells were involved in mediating IL-10 transcription in the presence of alisertib, we treated CT-26 cells and SW480 cells with various concentrations (0 – 1000 nM) of alisertib for 24 h. We found that a high dosage of alisertib potently increased the IL-10 mRNA levels (Fig. 5A and B).
Additionally, we treated these cells with 1000 nM alisertib for the indicated durations and found that alisertib induced upregulation of the IL-10 mRNA level after exposure of these cells to alisertib for 24 h (Fig. 5C and D). To further determine whether IL-10 was induced as a consequence of mitotic delay and chromosome segregation defects, we treated these two cell lines with nocodazole, which blocked the mitosis process as early as 3 h (42). We found that the IL-10 mRNA levels were decreased after exposure of the CT-26 cells to nocodazole (Fig. 5E). However, the IL-10 mRNAlevels were almost identical to those in the untreated SW480 cells (Fig. 5F). Furthermore, upon alisertib treatment, the transcription levels of IL-10 were enhanced (Fig. 5G), revealing that the expression of IL-10 could be directly suppressed by Aurora-A rather than as a consequence of mitotic defects mediated by Aurora-A inhibition.
Elevated IL-10 mediated by inhibition of Aurora-A promoted CD8+ T cell migration
To examine whether IL-10 secreted by cancer cells indeed recruited tumor-specific CD8+ T cell infiltration, we performed a migration assay. The migrated CD8+ T cells were barely detected when CD8+ T cells were exposed to the medium harvested from the control-treated CT-26 cells. Nevertheless, the number of migrated CD8+ T cells was potently elevated after exposure of these cells to the media from the alisertib-treated CT-26 cells and was decreased in the presence of an anti-IL-10R antibody (Fig. 5H).
Constitutive expression of Aurora-A was correlated with a low density of CD8+ T cells in CRC
To further clarify the clinical relevance of Aurora-A and intratumoral CD8+ T cells in CRC, we examined the expression of Aurora-A as well as the density of CD8+ T cells by immunohistochemistry. We found that 12 of 25 samples highly expressed Aurora-A, whereas the densities of CD8+ T cells in these samples were barely detected (Fig. 6A left panel and Table 1). However, high densities of intratumoral CD8+ T cells were observed in both the tumor tissues and mesenchyme with low expression of Aurora-A (Fig. 6A right panel and Table 1). Moreover, the correlation of Aurora-A and CD8+ T cells was determined using the GEPIA web tool. Our data indicated that the expression of Aurora-A was negatively correlated with the expression of CX3CR1 and FGFBP1 (effector T cells) (Fig. 6B). In addition, the expression of Aurora-A was negatively correlated with the expression of PDCD1 and GZMA (effector memory T cells) (Fig. 6C). Interestingly, the expression of Aurora-A was negatively correlated with the expression of IL-10 (Fig. 6D), indicating that inhibition of Aurora-A could elevate IL-10 transcription and production, stimulating IL-10R expression in intratumoral CD8+ T cells, which could activate its downstream signaling pathway and contribute to enhancements of the functional activity of intratumoral CD8+ T cells (Fig. 6E).
Discussion
The AURKA gene is an oncogene that is amplified or/and overexpressed in many tumors, including leukemia and breast and colorectal cancers. In addition, inhibition of Aurora-A leads to mitotic delays, severe chromosome congression and segregation defects, followed by cell death. Therefore, Aurora-A has become a therapeutic target for the treatment of various malignancies. Alisertib, an oral inhibitor of Aurora-A, has been investigated in clinical trials. Consistent with a previous study showing that alisertib promoted the infiltration of helper and cytotoxic T-cells into melanoma and breast cancer tumors, respectively (27, 43), we found that the antitumor effect of alisertib depended on the recruitment of infiltrated intratumoral CD8+ T cells (Fig. 1 and Fig. 2). Additionally, knockout of AURKA promoted CD8+ T cell infiltration (Fig. 3), indicating that inhibition of Aurora-A could be a new approach for inducing recruitment of intratumoral CD8+ T cells, although a subset of intratumoral CD8+ T cells expressed coinhibitory molecules.
It has been reported that CD8+ T cells expressing coinhibitory receptors, including PD-1, TIM-3, CTLA-4 or LAG-3, are generally considered exhausted T cells with reduced cytotoxic and proliferative capacity (44). However, a recent study demonstrated that CD8+ T cells simultaneously expressing the PD-1, TIM-3 and LAG-3 coinhibitory receptors were also enriched in tumor-specific T cells (45).
Interestingly, it has been shown that pegilodecakin, PEGylated IL-10, induced expansion of and functional activity of exhausted CD8+ T cells in parallel with an increase in the expression of IL-10R of these exhausted CD8+ T cells (9). In addition, the extent of expansion of the tumor-specific CD8+ T cells was correlated with the tumor response (9). Similarly, we found that inhibition of Aurora-A by alisertib dramatically induced recruitment and expansion of infiltrated CD8+ T cells in association with upregulation of the IL-10R levels (Fig. 4A). However, theantitumor effect of alisertib was attenuated in the presence of an anti-IL-10Rantibody (Fig. 4B-D). In further experiments, we found that elevated expression ofIL-10R in the infiltrated CD8+ T cells could be a consequence of the induction of IL-10 mediated by Aurora-A inhibition (Supplementary Fig. S2). All of these dataindicated that alisertib could promote infiltration of tumor-specific CD8+ T cells by controlling the expression of IL-10R. Unlike IL-10 R IL-10R appears to exclusively bind IL-10 (40). Our results demonstrated that the levels of IL-10 were significantly elevated in the culture medium of the alisertib-treated CT-26 cells (Fig. 4E). Additionally, the levels of IL-10 were elevated in the serum of the
AOM/DSS-treated AURKAf/+;VilCre+ mice compared to the AURKAf/+ mice, although IL-10 could not be detected in the serum of both the AURKAf/+;VilCre+ and AURKAf/+ mice (Fig. 4H). Importantly, inhibition of Aurora-A could directly induce an increase in IL-10 transcription, which was not a consequence of mitotic defects mediated by Aurora-A inhibition (Fig. 5A-G). Elevated IL-10 promoted CD8+ T cell migration (Fig. 5H). All of these observations indicated that elevated IL-10 was involved in promoting CD8+ T cell infiltration, although the mechanisms by which Aurora-A inhibited IL-10 transcription need to be investigated in further experiments.
It has been reported that IL-10 plays an important anti-inflammatory role (9).
However, a high dosage of IL-10 could activate the cytotoxicity of intratumoral CD8+ T cells and induce the proliferation of these cells (20-23), contributing to a decrease in tumor burden (9) or tumor clearance (18, 19). A recent study has shown that pegylated IL-10 (pegilodecakin, AM0010) induces tumor-specific CD8+ T cell invigoration and polyclonal T cell expansion in individuals with renal cell cancer (RCC) (9).
Furthermore, one study showed that the combination of AM0010 and 5-FU/LV and oxaliplatin (FOLFOX) was well tolerated in patients with metastatic pancreatic adenocarcinoma (PDAC) (46). An increase in the levels of Th1 and Th2 cytokines was observed in the serum of patients receiving AM0010 in combination with FOLFOX. However, the levels of the immune suppressive molecule TGF- and Th17 cytokines were decreased in the serum of patients treated with AM0010 in combination with FOLFOX, although the median progression-free survival (mPFS) was 3.5 months and the median overall survival (mOS) has not been reached but appears to be beyond 10.2 months (46). This regimen is currently being studied in a phase 3 trial (NCT02923921) (46). All of these observations indicated that inducing
IL-10 secretion could be a useful treatment strategy, although IL-10 hasimmunosuppressive effects.
Overall, the current study demonstrated that inhibition of Aurora-A activity by alisertib or deletion of the AURKA gene induced infiltration and expansion of intratumoral CD8+ T cells by enhancing IL-10.
References
1. Naito Y, Saito K, Shiiba K, Ohuchi A, Saigenji K, Nagura A, et al. CD8(+) T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Research. 1998;58:3491-4.
2. Pages F, Berger A, Camus M, Sanchez-Cabo F, Costes A, Molidor R, et al. Effector memory T cells, early metastasis, and survival in colorectal cancer. The New England journal of medicine. 2005;353:2654-66.
3. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce-Pages C, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313:1960-4.
4. Koelzer VH, Lugli A, Dawson H, Hadrich M, Berger MD, Borner M, et al. CD8/CD45RO T-cell infiltration in endoscopic biopsies of colorectal cancer predicts nodal metastasis and survival. J Transl Med. 2014;12:81.
5. Reissfelder C, Stamova S, Gossmann C, Braun M, Bonertz A, Walliczek U, et al. Corrigendum. Tumor-specific cytotoxic T lymphocyte activity determines colorectal cancer patient prognosis. The Journal of clinical investigation. 2015;125:1364.
6. Siddiqui I, Schaeuble K, Chennupati V, Fuertes Marraco SA, Calderon-Copete S, Pais Ferreira D, et al. Intratumoral Tcf1(+)PD-1(+)CD8(+) T Cells with Stem-like Properties Promote Tumor Control in Response to Vaccination and Checkpoint Blockade Immunotherapy. Immunity. 2019;50:195-211 e10.
7. Rosenberg SA, Dudley ME. Adoptive cell therapy for the treatment of patients with metastatic melanoma. Current opinion in immunology. 2009;21:233-40.
8. Tumeh PC, Harview CL, Yearley JH, Shintaku IP, Taylor EJ, Robert L, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515:568-71.
9. Naing A, Infante JR, Papadopoulos KP, Chan IH, Shen C, Ratti NP, et al. PEGylated IL-10 (Pegilodecakin) Induces Systemic Immune Activation, CD8(+) T Cell Invigoration and Polyclonal T Cell Expansion in Cancer Patients. Cancer cell. 2018;34:775-91 e3.
10. Scheper W, Kelderman S, Fanchi LF, Linnemann C, Bendle G, de Rooij MAJ, et al. Low and variable tumor reactivity of the intratumoral TCR repertoire in human cancers. Nature medicine. 2019;25:89-94.
11. Larkin J, Chiarion-Sileni V, Gonzalez R, Grob JJ, Cowey CL, Lao CD, et al. Combined Nivolumab and Ipilimumab or Monotherapy in Untreated Melanoma. The New England journal of medicine. 2015;373:23-34.
12. Subudhi SK, Aparicio A, Gao J, Zurita AJ, Araujo JC, Logothetis CJ, et al. Clonal expansion of CD8 T cells in the systemic circulation precedes development of ipilimumab-induced toxicities. Proceedings of the National Academy of Sciences of the United States of America. 2016;113:11919-24.
13. Fadel SA, Bromley SK, Medoff BD, Luster AD. CXCR3-deficiency protects influenza-infected CCR5-deficient mice from mortality. European journal of immunology. 2008;38:3376-87.
14. Moore KW, de Waal Malefyt R, Coffman RL, O'Garra A. Interleukin-10 and the interleukin-10 receptor. Annual review of immunology. 2001;19:683-765.
15. Brooks DG, Trifilo MJ, Edelmann KH, Teyton L, McGavern DB, Oldstone MB. Interleukin-10 determines viral clearance or persistence in vivo. Nature medicine. 2006;12:1301-9.
16. Berg DJ, Davidson N, Kuhn R, Muller W, Menon S, Holland G, et al. Enterocolitis and colon cancer in interleukin-10-deficient mice are associated with aberrant cytokine production and CD4(+) TH1-like responses. The Journal of clinical investigation. 1996;98:1010-20.
17. Neven B, Mamessier E, Bruneau J, Kaltenbach S, Kotlarz D, Suarez F, et al. A Mendelian predisposition to B-cell lymphoma caused by IL-10R deficiency. Blood. 2013;122:3713-22.
18. Zheng LM, Ojcius DM, Garaud F, Roth C, Maxwell E, Li Z, et al. Interleukin-10 inhibits tumor metastasis through an NK cell-dependent mechanism. J Exp Med. 1996;184:579-84.
19. Berman RM, Suzuki T, Tahara H, Robbins PD, Narula SK, Lotze MT. Systemic administration of cellular IL-10 induces an effective, specific, and long-lived immune response against established tumors in mice. J Immunol. 1996;157:231-8.
20. Mumm JB, Emmerich J, Zhang X, Chan I, Wu L, Mauze S, et al. IL-10 elicits IFNgamma-dependent tumor immune surveillance. Cancer Cell. 2011;20:781-96.
21. Emmerich J, Mumm JB, Chan IH, LaFace D, Truong H, McClanahan T, et al. IL-10 directly activates and expands tumor-resident CD8(+) T cells without de novo infiltration from secondary lymphoid organs. Cancer research. 2012;72:3570-81.
22. Fujii S, Shimizu K, Shimizu T, Lotze MT. Interleukin-10 promotes the maintenance of antitumor CD8(+) T-cell effector function in situ. Blood. 2001;98:2143-51.
23. MacNeil IA, Suda T, Moore KW, Mosmann TR, Zlotnik A. IL-10, a novel growth cofactor for mature and immature T cells. J Immunol. 1990;145:4167-73.
24. Yang J, Ikezoe T, Nishioka C, Tasaka T, Taniguchi A, Kuwayama Y, et al. AZD1152, a novel and selective aurora B kinase inhibitor, induces growth arrest, apoptosis, and sensitization for tubulin depolymerizing agent or topoisomerase II inhibitor in human acute leukemia cells in vitro and in vivo. Blood. 2007;110:2034-40.
25. Yang J, Ikezoe T, Nishioka C, Nobumoto A, Udaka K, Yokoyama A. CD34(+)/CD38(-) acute myelogenous leukemia cells aberrantly express Aurora kinase A. International journal of cancer. 2013;133:2706-19.
26. Blas-Rus N, Bustos-Moran E, Perez de Castro I, de Carcer G, Borroto A, Camafeita E, et al. Aurora A drives early signalling and vesicle dynamics during T-cell activation. Nature communications. 2016;7:11389.
27. Vilgelm AE, Johnson CA, Prasad N, Yang J, Chen SC, Ayers GD, et al. Connecting the Dots: Therapy-Induced Senescence and a Tumor-Suppressive Immune Microenvironment. Journal of the National Cancer Institute. 2016;108:djv406.
28. Zheng F, Yue C, Li G, He B, Cheng W, Wang X, et al. Nuclear AURKA acquires kinase-independent transactivating function to enhance breast cancer stem cell phenotype. Nature communications. 2016;7:10180.
29. Del Reino P, Alsina-Beauchamp D, Escos A, Cerezo-Guisado MI, Risco A, Aparicio N, et al. Pro-oncogenic role of alternative p38 mitogen-activated protein kinases p38gamma and p38delta, linking inflammation and cancer in colitis-associated colon cancer. Cancer research. 2014;74:6150-60.
30. Neufert C, Becker C, Neurath MF. An inducible mouse model of colon carcinogenesis for the analysis of sporadic and inflammation-driven tumor progression. Nature protocols. 2007;2:1998-2004.
31. Lu J, Wang H, Zhang R, Wan Z, Gao H, Cai J, et al. Effects of Photoperiod on Acetaminophen-Induced Hepatotoxicity in Mice. Digestive diseases and sciences. 2020;65:178-88.
32. Reiter R, Gais P, Jutting U, Steuer-Vogt MK, Pickhard A, Bink K, et al. Aurora kinase A messenger RNA overexpression is correlated with tumor progression and shortened survival in head and neck squamous cell carcinoma. Clinical cancer research : an official journal of the American Association for Cancer Research. 2006;12:5136-41.
33. Nishioka C, Ikezoe T, Furihata M, Yang J, Serada S, Naka T, et al. CD34(+)/CD38(-) acutemyelogenous leukemia cells aberrantly express CD82 which regulates adhesion and survival of leukemia stem cells. International journal of cancer. 2013;132:2006-19.
34. Li Z, Wang C, Jiao X, Lu Y, Fu M, Quong AA, et al. Cyclin D1 regulates cellular migration through the inhibition of thrombospondin 1 and ROCK signaling. Molecular and cellular biology. 2006;26:4240-56.
35. Wherry EJ, Kurachi M. Molecular and cellular insights into T cell exhaustion. Nature reviews Immunology. 2015;15:486-99.
36. Lu LY, Wood JL, Ye L, Minter-Dykhouse K, Saunders TL, Yu X, et al. Aurora A is essential for early embryonic development and tumor suppression. The Journal of biological chemistry. 2008;283:31785-90.
37. Madison BB, Dunbar L, Qiao XT, Braunstein K, Braunstein E, Gumucio DL. Cis elements of the villin gene control expression in restricted domains of the vertical (crypt) and horizontal (duodenum, cecum) axes of the intestine. The Journal of biological chemistry. 2002;277:33275-83.
38. el Marjou F, Janssen KP, Chang BH, Li M, Hindie V, Chan L, et al. Tissue-specific and inducible Cre-mediated recombination in the gut epithelium. Genesis. 2004;39:186-93.
39. Voskoboinik I, Dunstone MA, Baran K, Whisstock JC, Trapani JA. Perforin: structure, function, and role in human immunopathology. Immunological reviews. 2010;235:35-54.
40. Hennig BJ, Frodsham AJ, Hellier S, Knapp S, Yee LJ, Wright M, et al. Influence of IL-10RA and IL-22 polymorphisms on outcome of hepatitis C virus infection. Liver international : official journal of the International Association for the Study of the Liver. 2007;27:1134-43.
41. Ibrahim ML, Klement JD, Lu C, Redd PS, Xiao W, Yang D, et al. Myeloid-Derived Suppressor Cells Produce IL-10 to Elicit DNMT3b-Dependent IRF8 Silencing to Promote Colitis-Associated Colon Tumorigenesis. Cell reports. 2018;25:3036-46 e6.
42. Yang J, Ikezoe T, Nishioka C, Yokoyama A. A novel treatment strategy targeting shugoshin 1 in hematological malignancies. Leukemia research. 2013;37:76-82.
43. Yin T, Zhao ZB, Guo J, Wang T, Yang JB, Wang C, et al. Aurora-A inhibition eliminates myeloid cell-mediated immunosuppression and enhances the efficacy of anti-PD-L1 therapy in breast cancer. Cancer research. 2019.
44. Apetoh L, Smyth MJ, Drake CG, Abastado JP, Apte RN, Ayyoub M, et al. Consensus nomenclature for Alisertib CD8(+) T cell phenotypes in cancer. Oncoimmunology. 2015;4:e998538.
45. Gros A, Parkhurst MR, Tran E, Pasetto A, Robbins PF, Ilyas S, et al. Prospective identification of neoantigen-specific lymphocytes in the peripheral blood of melanoma patients. Nature medicine. 2016;22:433-8.
46. Randolph HJ, Naing A, Falchook G, Patel M, Infante J, Aljumaily R, et al. Immunologic and objective tumor responses to PEGylated human IL-10 (AM0010) with 5-FU/LV and oxaliplatin (FOLFOX) in metastatic pancreatic adenocarcinoma (PDAC). Annals of oncology : official journal of the European Society for Medical Oncology. 2017;28 Suppl 3:iii138.