APOBEC3B Improves Response to PD-1 Blockade Through DSBs–ATR–CHK1 Pathway

doi:10.21203/rs.3.rs-1606737/v1

Download PDF

Research Article

APOBEC3B Improves Response to PD-1 Blockade Through DSBs–ATR–CHK1 Pathway

https://doi.org/10.21203/rs.3.rs-1606737/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like (APOBEC) family enzymes are considered to induce base substitutions and strand breaks, which contribute significantly to mutation and genomic instability. In the APOBEC family, APOBEC3B (A3B) is a major mutational source in multiple cancers, which fuels tumor diversity and subclonal evolution. Although the APOBEC mutational signature is enriched in immune checkpoint inhibitor responders, it remains unclear whether and how APOBEC influences the immunotherapy.

Methods

The association between A3B and anti-PD-1 treatment sensitivity was evaluated with mouse bearing subcutaneous Hepa1-6 and 4T1 tumors which was trasfected with mouse APOBEC3, immunohistochemical staining and flow cytometry were utilized to estimated the tumor immune microenvironment. The effects of A3B on programmed cell death-ligand 1 (PD-L1) expression and mechanistic detection were investigated with bioinformatics analysis, the results were validated in vitro and in vivo.

Results

High A3B expression increased sensitivity to anti-PD-1 antibody, accompanied by PD-L1 upregulation, and improved tumor-infiltrating T cells and NK cells. We further clarified that A3B induced double-stranded DNA breaks, thereby activating the ATR–Chk1–STAT–IRF1 pathway and resulting in PD-L1 overexpression in tumor. Moreover, Chk1 inhibitor, MK8776, notably reversed the PD-L1 upregulation, aggravated the tumor microenvironment, and eventually abrogated the enhanced sensitivity to anti-PD-1 antibody.

Conclusions

There findings support the notion that A3B had increased sensitivity to anti-PD-1 antibody, providing direct evidence to illustrate the potential of A3B as a marker of response to immunotherapy.

APOBEC3B

PD-1/PD-L1

immunotherapy

CD8+ T cells

DNA damage response

Immune checkpoint inhibitor (ICI)–based immunotherapy, especially anti-PD-1/PD-L1 antibodies, has achieved great success in cancer therapy. However, the response rate is not satisfactory, and toxicities can be severe in some patients [1]. These facts highlight the importance of a biomarker for identifying patients most likely to benefit from anti-PD-1/PD-L1 immunotherapy. Currently, predictive candidates under development include PD-L1 expression, tumor-infiltrating lymphocytes, T-cell receptor clonality, tumor mutational burden (TMB), neoantigen burden, and immune gene signatures [2, 3]. It has also been reported that a subset of patients with positive expression of PD-L1 experienced deteriorated disease after receiving anti-PD-1/PD-L1 therapy [4]. High TMB has been shown to correlate with immunotherapy response given that tumor-related antigens can be affected by TMB [5, 6]; however, there are some nonresponders even among patients with high TMB, and different mutations show variable immunogenicity [7]. Gut microbiota and biomarkers in peripheral blood have not been established steadily in effect prediction [8–10]. Taken together, there has not been an individual predictor with sufficient reliability. Accordingly, to reflect tumor immune microenvironment and select sensitive patients more accurately, it emphasized an urgent requirement for exploring new predictive markers.

Apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like (APOBEC) is a family member of Zn²⁺-dependent polynucleotide cytosine deaminases; it plays a role in broad retrotransposon and retrovirus restriction activities by eliciting DNA cytosine deaminase activity [11, 12]. By converting cytosines in DNA strands to uracils, APOBEC family enzymes are considered to induce base substitutions and strand breaks, which contribute significantly to mutation and genomic instability [13]. Therefore, APOBEC is the most frequent source of mutations related to tumor development and progression [14]. In addition, the APOBEC mutational signature is abundantly enriched in patients with better response to PD-l blockade [15, 16]. However, it remains unclear whether and how APOBEC affects the efficacy of immunotherapy. Although APOBEC activity has been shown to increase TMB in cancers [17], the correlation between APOBEC-driven mutagenesis and immunotherapy seems to be independent of TMB [18, 19]. In the APOBEC family, APOBEC3B (A3B) is the major mutational source in a variety of cancers, including breast, head/neck, lung, bladder, cervical, and ovarian cancers [17]. A3B has also been shown to own powerful features that fuel tumor diversity and promote subclonal evolution [20]. Therefore, we propose that A3B may be a potential marker of response to immunotherapy.

Herein, we investigated the influence of A3B expression on the efficacy of PD-1/PD-L1 blockade in multiple types of cancer. We also examined the mechanism of A3B-induced PD-L1 expression and tumor immune infiltration. Our findings support the notion that A3B increases sensitivity to anti-PD-1 antibody, providing evidence that A3B may be a marker of response to immunotherapy.

Cells and reagents

Breast carcinoma cell line 4T1 of BALB/c mice, liver carcinoma cell line Hepa1-6 and Lewis lung carcinoma cell line LLC of C57/L mice, and H2228 human lung carcinoma cell line were purchased from the National Collection of Authenticated Cell Cultures (Shanghai, China). MC38 colon adenocarcinoma cells of C57/L mice and human renal-derived 293FT cell lines were bought from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). Human lung adenocarcinoma carcinoma cell line was purchased from the American Type Culture Collection (ATCC, Manassas, Virginia, USA). 4T1, MC38, and H2228 cell lines were cultivated with RPMI 1640 supplemented medium, while Hepa1-6, LLC, HCC4006, and 293FT cell lines were maintained in Dulbecco’s modified Eagle’s medium (DMEM). The cells were incubated with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin in the medium, at an environment of 37°C and 5% CO₂.

MK8776, a Chk1 inhibitor, was obtained from Adooq Bioscience (Irvine, California, US), and dissolved as 5 mmol/L stocks in dimethyl sulfoxide (DMSO, MP Biomedicals, Santa Ana, California, USA) for in vitro studies. For in vivo utility, 4 mg/mL MK8776 was dissolved in 4% DMSO and 30% propylene glycol (Sigma-Aldrich, St. Louis, Missouri, USA) and administered through intraperitoneal (IP) injection. Anti-PD-1 monoclonal antibody (catalog number: BE0146; clone RMP1-14) was purchased from Bio X Cell (West Lebanon, New Hampshire, USA).

Lentiviral infection

The mouse APOBEC3 (mA3)–expressing cell lines were established with retroviral transduction using a pEZ-Lv201 vector (GeneCopoeia). To generate lentivirus, 293FT cells were incubated in a lentiviral transduction system that consisted of psPAX2, pMD2.G plasmids, and Lipofectamine 3000 (Invitrogen Corp., Carlsbad, California, US). After 48 hours of transfection, the cell supernatant was collected and filtered. We cultured the mouse cell lines in virus supernatant for 24 hours and then selected mA3-expressing cells with 2 µg/mL puromycin (Sigma-Aldrich) for another 48 hours. A3B-expressing human cell lines were incubated with tet-on lentivirus (Shanghai Genechem Co., Ltd.) for 24 hours, and then the cells were selected by puromycin at a concentration of 3.00 μg/mL for another 72 hours after transfection.

Cell proliferation assays and growth inhibition assays

Both cell growth assays and growth inhibition assays were conducted using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium (MTT) assay. Tumor cells were transplanted into 96-well plates with a culture medium, and the cell number was kept at around 2000 cells/well. In cell proliferation assays, MTT solution was added into the culture medium and incubated for two hours every day from day 0 to day 4. In growth inhibition assays, multiple concentrations of the indicated drug were added to each well at 24 hours after seeding and further cultured for 72 hours. Then, MTT solution was added to each well and incubated for another 2 hours. DMSO was alternatively added into 96-well plates to dissolve the dark blue crystals after removing the culture medium containing MTT solution. The growth status of the cells was reflected in absorbance, which was read by a microplate reader with reference wavelengths at 490 and 550 nm.

Quantitative real-time PCR (qRT-PCR) analysis

Total RNA of the cell lines and tumor tissues was extracted in accordance with the manufacturer’s standard protocol of RNAiso Plus (TaKaRa, Dalian, China) and then reverse-transcribed to complementary DNA (cDNA) using PrimeScript RT Master Mix (TaKaRa). Real-time PCR was conducted using SYBR Premix Ex Taq (Tli RNase HPlus; TaKaRa), and the signal was detected and recorded using Roche LightCycler 480 instrument (Roche Diagnostics). We used the following primers: mA3, forward: CAGAGCAGGTACTAAGGTTCCT; reverse: TTCTGGGTCCCGTATGTTGTA; ATM, forward: GATCTGCTCATTTGCTGCCG; reverse: GTGTGGTGGCTGATACATTTGAT; ATR, forward: TTGGCCTCGAAAAGGGTCTC; reverse: TCCGTCAACAACGACCAAAGG; Irf1, forward: ATGCCAATCACTCGAATGCG; reverse: TTGTATCGGCCTGTGTGAATG. Relative expression was calculated using the −2^ΔΔCt method.

Western blotting

The cell lines and tumor tissues were lysed in RIPA lysis buffer (Beyotime), which contained proteinase and phosphatase inhibitor cocktails (Sigma). The BCA Protein Assay Kit (Thermo Fisher Scientific) was used to estimate the protein concentrations, and then denature the proteins at 98°C for 5 min. We added 40 µg of protein per well into 10% SDS-polyacrylamide gel and transferred the protein blot to polyvinylidene difluoride (PVDF) membrane (Millipore). The membrane was soaked into a quick blocking solution (Beyotime) for 30 minutes and then incubated with a specific primary antibody at 4°C overnight. The following primary antibodies were used in this study: PD-L1 (#13684, Cell Signaling Technology); PD-L1 (66248-1-Ig, Proteintech); β-Actin (#4970, Cell Signaling Technology); GAPDH (#2118, Cell Signaling Technology); Flag Tag (#2368, Cell Signaling Technology); phospho-Chk1 (#2348, Cell Signaling Technology); Chk1 (#2360, Cell Signaling Technology); γ-H2AX (#2577, Cell Signaling Technology); p-STAT1 (#9167, Cell Signaling Technology); STAT1 (#14994, Cell Signaling Technology); Phospho-Stat3 (#9145, Cell Signaling Technology); STAT3 (#12640, Cell Signaling Technology); Irf1 (#8478, Cell Signaling Technology). We visualized the protein blotting after incubating the PVDF membrane with horseradish peroxide (HRP)–conjugated secondary antibody (#7074, Cell Signaling Technology) for one hour.

In vivo experiment

Female wild-type C57BL/6, BALB/c mice, 3–4-week-old, were obtained from the Experimental Animal Center of Southern Medical University (Guangzhou, China), their care was in accordance with institution guidelines. The mice were subcutaneously injected with Hepa1-6 (1 × 10⁶, C57BL/6) or 4T1 (0.5 × 10⁶, BALB/c) in 100 μL phosphate-buffered saline (PBS) to establish the tumor models. When the comparable average tumor size reached 80–90 mm³, tumor-bearing mice were randomly assigned into afferent treatment groups. The anti-PD-1 antibody matched control IgG isotype antibody, and Chk1 inhibitor (MK8776) were intraperitoneally injected every 3 days, with 200 µg anti-PD-1 or IgG antibody and 50 mg/kg MK8776 per dose.

Immunohistochemistry and immunofluorescence

For immunohistochemistry (IHC) staining, tissue sections were incubated in hydrogen peroxide for 30 minutes to reduce endogenous peroxidase activity. Next, the sections were incubated overnight at 4°C with primary antibodies CD8 alpha (ab209775, ab237709; Abcam) and A3B (ab191695; Abcam). After reacting with a 3,30-diaminobenzidine (DAB) kit, a horseradish peroxidase–streptavidin detection system was used to detect immunoreactivity, and hematoxylin was used for counterstaining. Immunofluorescent staining was conducted following a standard protocol. Namely, after incubating the sections with a primary antibody (PD-L1, 66248-1-Ig, Proteintech) at 4°C overnight, a secondary antibody conjugated with fluorescence was added to the sections and incubated for another hour at room temperature.

Flow cytometric analysis and FACS

Subcutaneous tumor tissue samples were minced using scalpels, incubated with 1 mg/mL collagenase IV (Sigma Aldrich), 0.1 mg/mL DNase I (Solarbio, Beijing, China), RPMI 1640 for 30 minutes at 37 °C. ACK lysis buffer (Leagene, Beijing, China) was added to remove red blood cells after mechanical disruption. Subsequently, the cells were incubated for 30 minutes at 4°C with the following antibodies: FITC-conjugated anti-CD45 (clone 30-F11, eBioscience), PerCP-Cy5.5-conjugated anti-TCRβ (clone H57-597, eBioscience), APC-Cy7-conjugated anti-CD4 (clone GK1.5, eBioscience), APC-conjugated anti-CD8 (clone 53-6.7, eBioscience), PE-eFluor 610-conjugated anti-PD-1 (cloneJ43, eBioscience), PE-conjugated anti-TIM-3 (cloneRMT3-23, eBioscience), PerCP-Cy5.5-conjugated anti-CD3e (BD clone 145-2C11, Biosciences), and APC-Cy7-conjugated anti-NK1.1 (clonePK136, BD Biosciences). Stained cells were analyzed using BD FACSCANTO II and data were analyzed using FlowJo software.

ELISA

Peripheral blood samples collected from the mice after different treatments were centrifuged to obtain the serum. We detected the concentration of IFN-gamma following the standard protocol of the mouse IFN-gamma ELISA kit (70-EK280/3-96, MultiSciences).

Bioinformatics analyses

The TCGA Pan-Cancer cohort, which contained 10,535 samples from 33 types of cancer, was downloaded from the UCSC database (https://xena.ucsc.edu/). Liver Hepatocellular Carcinoma (TCGA, both PanCancer Atlas) and Breast Invasive Carcinoma (TCGA, Cell 2015) datasets were acquired from cBioPortal database (http://www.cbioportal.org/). Patients with A3B mRNA expression above the third quartile were defined as high A3B group, while those below the first quartile were defined as low A3B group. Levels of tumor-related total proteins were obtained from TCGA protein microarray sequencing data. R package "estimate" was used to calculate the Estimation of Stromal and Immune Cells in Malignant Tumor Tissues Using Expression Data (ESTIMATE) immune scores of each sample [21], which represented the abundance of immune components in different A3B expression groups, respectively. Survival analysis was done using the R package "survminer," and the R package "survival" was utilized to draw the survival curves. Correlations of A3B mRNA expression with PD-L1 level and immune infiltration were analyzed in the TIMER database (http://timer.cistrome.org/). Differential gene analysis was based on R package "limma"; the results were presented with heat maps and volcano plots by R package "ggplot2." The version of R was 4.1.0. The difference between the two groups was analyzed with an unpaired t test.

Statistical analysis

SPSS 22.0 software was used for statistical analysis. The experimental data were expressed as mean and standard deviation. The homogeneity of variance was tested by Levene's test. The means of two samples were compared by t test for two independent samples (independent-sample t test), and the means of more than two groups of samples were compared by analysis of variance (two-way ANOVA). Kaplan–Meier survival curves and log-rank test were used to analyze the relationship between the expression of A3B and overall survival (OS), and Pearson coefficient was used to analyze the correlation. P < 0.05 was considered statistically significant.

A3B increases sensitivity to anti-PD-1 antibody

We first analyzed the differences in A3B gene expression between cancer tissues and normal tissues in the TCGA pan-cancer sample (TCGA Pan-Cancer, 10535 cases). As shown in Fig. 1A, A3B was highly expressed in many types of cancers, including breast cancer (BRCA), liver hepatocellular carcinoma (LIHC), cervical squamous cell carcinoma and adenocarcinoma (CESC), head and neck squamous cell carcinoma (HNSC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), and others. However, there was no significant association between the expression of A3B and OS in these cancers (Supplemental Fig. 1).

We next explored the relationship of A3B level with immune infiltration proportion and TMB, which may predict the response to immunotherapy. The results showed that in LUAD, BRCA, LIHC, skin cutaneous melanoma (SKCM), diffuse large B-cell lymphoma (DLBC), kidney renal clear cell carcinoma (KIRC), kidney chromophobe (KICH), HNSC, and other tumors, ESTIMATE immune score was higher in tumors with a high expression of A3B (Fig. 1B). Additionally, A3B expression displayed a significant and positive correlation with TMB (Supplemental Fig. 2) in DLBC, BRCA, LIHC, LUAD, SKCM, and stomach adenocarcinoma (STAD). To verify our conjecture, we transfected mA3 gene into two mouse cell lines, Hepa1-6 (mouse liver cancer) and 4T1 (mouse breast cancer), and established subcutaneous tumors in C57BL/6 mice. Transfection of mA3 did not affect the cell viability in vitro (Supplemental Fig. 3), although the mA3 mRNA was increased markedly in both cell lines (Fig. 1C). In mouse tumors derived from Hepa1-6/vector and 4T1/vector, 200 µg of anti-PD-1 antibody inhibited tumor growth, whereas a low dose of anti-PD-1 antibody (100 μg) did not (Fig. 1D). In contrast, tumor growth of Hepa1-6/mA3 and 4T1/mA3 cells was significantly suppressed by 100 μg of anti-PD-1 antibody (Fig. 1E). These results suggested that the high expression of A3B augmented the sensitivity to anti-PD-1 antibody immunotherapy.

A3B upregulates PD-L1 expression in pan-cancers.

Next, we investigated the mechanism by which A3B improves sensitivity to PD-1 blockade. Since PD-L1 expression was closely correlated with sensitivity to anti-PD-1 antibody, we analyzed the relationship between the expression of A3B and PD-L1 (CD274) in the pan-cancer dataset of the TIMER database. We found that in most cancers, including BRAC, LIHC, LUAD, LUSC, DLBC, and SKCM, the expression levels of A3B and PD-L1 were positively correlated (Fig. 2A). We further evaluated the correlation of A3B with PD-1, PD-L1, and PD-L2 in LIHC and BRAC using the cBioPortal database. Consistently, A3B expression was positively associated with these immune checkpoint molecules (Fig. 2B). In immunofluorescence staining, we found that Hepa1-6/mA3 and 4T1/mA3 cells expressed much higher levels of PD-L1 than Hepa1-6/vector and 4T1/vector cells (Fig. 2C). PD-L1 overexpression in mA3-transfected Hepa1-6 and 4T1 cells was also detected by western blotting (Fig. 2D). Moreover, PD-L1 expression levels were elevated markedly in mouse xenografts derived from Hepa1-6/mA3 and 4T1/mA3 cells, compared with xenografts derived from Hepa1-6/vector and 4T1/vector cells (Fig. 2E).

We further confirmed the effect of A3B on PD-L1 expression in multiple other cancer cells, including MC38 (mouse colon cancer cells), LLC (mouse lung cancer cells), H2228 (human lung cancer cells), and HCC4006 (human lung adenocarcinoma cells). By western blot and immunofluorescence staining, we detected much higher levels of PD-L1 expression in all of the abovementioned cells after A3B gene transfection (Fig. 2F–G). These results demonstrated that A3B upregulated PD-L1 expression in pan-cancers, suggesting that A3B-induced PD-L1 overexpression may be involved in improved sensitivity to PD-1 blockade.

A3B improves tumor immune microenvironment

The tumor immune microenvironment is a key factor in response to immunotherapy. To clarify the change of tumor immune microenvironment by A3B, we explored tumor-infiltrating lymphocytes (TILs) by flow cytometry using xenografts from Hepa1-6/vector and Hepa1-6/mA3 cells. The results clearly showed that A3B significantly enhanced tumor-infiltrating TCRβ⁺ T cells, including CD8⁺ T cells and CD4⁺ T cells, and the levels of these infiltrating T cells were further augmented by anti-PD-1 antibody (Fig. 3A). In the Hepa1-6/vector tumors, anti-PD-1 antibody increased the percentage of CD8⁺ T cells and decreased the proportion of CD4⁺ T cells, while the percentage of CD8⁺ T cells was high in Hepa1-6/mA3 tumors even without anti-PD-1 antibody (Fig. 3B). Similarly, the proportion of NK cells was higher in Hepa1-6/mA3 tumors than in Hepa1-6/vector tumors. Anti-PD-1 antibody further enhanced the proportion of NK cells in Hepa1-6/mA3 tumors (Fig. 3C). Exhausted TIM3⁺PD-1⁺CD8⁺T cells were more abundant in Hepa1-6/mA3 tumors than in Hepa1-6/vector tumors (Fig. 3D, Supplemental Fig. 4A), which might be due to the high levels of PD-L1 expression. Anti-PD-1 antibody successfully decreased the abundance of exhausted CD8⁺T and CD4⁺T cells in Hepa1-6/mA3 tumors, whereas this phenomenon was not observed in Hepa1-6/vector tumors (Fig. 3D, Supplemental Fig. 4A and B). Anti-PD-1 antibody dramatically increased the levels of INF-γ in serum of mice with Hepa1-6/mA3 tumors or 4T1/mA3 tumors (Fig. 3E, Supplemental Fig. 4C). By IHC, we further confirmed that infiltrating CD8+ T cells were more abundant in Hepa1-6/mA3 and 4T1/mA3 tumors (Fig. 3F). These results demonstrated that A3B improved tumor immune microenvironment by upregulating tumor-infiltrating T cells and NK cells, improving the response to anti-PD-1 antibody.

To further verify the correlation between A3B and tumor-infiltrating T cells in pan-cancers, we stained A3B and CD8 in tumor specimens of 92 breast cancer patients from Hunan Provincial People's Hospital. Our results showed that the levels of infiltrating CD8⁺ T cells were much higher in high-A3B tumors (P = 0.008, Fig. 3G). Then, we explored the relationship between A3B and CD8⁺T cells in pan-cancerous tumors in the TIMER database. The results showed a correlation between A3B and infiltration by CD8⁺T cells not only in BRCA and LIHC (Supplemental Fig. 4D) but also in many other tumors. In some tumors, such as HNSC, A3B also positively correlated with CD4⁺T cells and NK cells (Fig. 3H). These results indicate that A3B boosted infiltrating T cells and NK cells to improve tumor immune microenvironment in pan-cancers.

A3B activates the ATR–CHK1–STAT–IRF1 pathway

To clarify the mechanism by which A3B promotes PD-L1 expression, we compared the gene variation between A3B low and high cancers using LIHC pan-cancer dataset (Fig. 4A). Then, we selected the abovementioned significantly upregulated genes in the LIHC data set in the DAVID database for Gene Ontology (GO) analysis. The results showed that the upregulated genes were generally enriched in double-strand break repair, DNA damage checkpoints, and other pathways (Fig. 4B). DNA break repair genes were upregulated in tumors with high levels of A3B (Supplemental Fig. 5A). Consistently, the protein levels of CHK1 and CHK2 in A3B-high tumors were higher than those in A3B-low tumors in the BRCA dataset (Supplemental Fig. 5B), confirming the findings in the LIHC dataset.

The motivation of DNA damage response depends on two checkpoint protein kinases, ataxia-telangiectasia mutated (ATM) and ATR [22]. To validate the results, we then examined ATM and ATR in Hepal1-6 and 4T1 cells with and without mA3 transfection. The results showed that the mRNA expression levels of ATR increased after mA3 transfection, whereas the expression of ATM did not change significantly (Fig. 4C). To further verify the activation of ATR-mediated checkpoint pathways, we examined the protein expression levels of downstream transducer CHK1. As shown by western blot, elevated p-CHK1 and γ-H2AX, the marker of DNA damage, were detected in Hepa1-6/mA3 and 4T1/mA3 cells (Fig. 4D), indicating that A3B induced DNA damage, thereby activating the ATR–CHK1 pathway. Meanwhile, the downstream p-STAT1, p-STAT3, and IRF1 were enhanced when mA3 was upregulated in Hepa1-6/mA3 and 4T1/mA3 cells (Fig. 4E–F). These results suggest that A3B activates the ATR–CHK1–STAT–IRF1 pathway.

A3B-induced PD-L1 expression can be reversed by CHK1 inhibition

Then, we evaluated whether the ATR–CHK1–STAT–IRF1 pathway was responsible for A3B-induced PD-L1 expression. We blocked this pathway by CHK1 inhibitor MK8776 and detected the related molecules. MK8776 did not influence cell viability of Hepa1-6 and 4T1 cells before and after mA3 transfection (Supplemental Fig. 6), but MK8776 prevented A3B-induced PD-L1 expression at both mRNA and protein levels (Fig. 5A, B). As expected, P-STAT1, P-STAT3, and IRF1 in Hepal1-6/mA3 and 4T1/mA3 cell lines were significantly downregulated by MK8776 (Fig. 5C).

We also examined the status of PD-L1 when the anti-PD-1 antibody was combined with MK8776. We found that MK8776 decreased PD-L1 expression with or without anti-PD-1 antibody, indicating that A3B-induced PD-L1 expression was mediated mainly by the ATR–CHK1 pathway (Fig. 5D). Similarly, MK8776 decreased P-STAT1, P-STAT3, and IRF1 regardless of the anti-PD-1 antibody (Fig. 5E). These results clearly showed that A3B enhanced PD-L1 expression through the ATR–CHK1–STAT–IRF1 pathway.

A3B improves T cell infiltration and sensitivity to the anti-PD-1 antibody through the ATR–CHK1 pathway

To test the role of the ATR–CHK1 pathway on A3B-mediated response to immunotherapy in vivo, we combined a low dose of anti-PD-1 Ab (100 μg) with MK8776 to treat the mouse xenografts of Hepal1-6/mA3 and 4T1/mA3 cells. The xenografts were sensitive to a low dose of anti-PD-1 Ab, while MK8776 alone did not affect tumor growth. However, MK8776 completely circumvented the antitumor effect of anti-PD-1 Ab in these tumor models (Fig. 6A, B). Although PD-L1 was highly expressed in Hepa1-6/mA3 and 4T1/mA3 tumors, MK8776 significantly downregulated PD-L1 expression in these tumors (Fig. 6C). Consistent with flow cytometry analysis of tumor-infiltrating T cells (Fig. 3A), we found that the anti-PD-1 antibody was able to enhance infiltrating CD8⁺ T cells in both Hepa1-6/mA3 and 4T1/mA3 tumors (Fig. 6D). Interestingly, MK8776 markedly reduced the infiltrating CD8⁺ T cells in both tumors (Fig. 6D). These results showed that CHK1 inhibition completely prevented A3B-improved response to anti-PD-1 Ab through decreasing PD-L1 expression and T cell infiltration.

Taken together, our results demonstrated that A3B activated the ATR–CHK1 signaling to augment T cell infiltration, and further promoted PD-L1 expression through the DSB–ATR–CHK1–STAT–IRF1 pathway, thereby increasing sensitivity to anti-PD-1 antibody in pan-cancers.

The successes and failures of immunotherapy highlight the importance of searching for biomarkers and developing strategies to improve response to immunotherapy. APOBEC mutation signature and A3B expression have been reported to be related to clinical benefits of immunotherapy, but the direct evidence is lacking and the related mechanism remains unknown. A3B is the most popular member of the APOBEC family and a gene-editing enzyme with cytidine deaminase activity. In this study, we showed that A3B was highly expressed in multiple tumor tissues, but it was not related to prognosis, which is consistent with previous observations [17, 23, 24]. Interestingly, we found that A3B positively correlated with PD-L1 expression, immune cell infiltration, and response to anti-PD-1 antibody in multiple cancers. Our data suggest that A3B may be a potential biomarker for immunotherapy.

A3B mutation signal and A3B (del) polymorphism have been reported to correlate with better response to immunotherapy in patients with non-small cell lung cancer and breast cancer [15, 25]. In this study, we confirmed the association between A3B and immunotherapy in pan-cancers, and further clarified the underlying mechanisms. We showed that A3B improved the efficacy of the anti-PD-1 antibody by increasing PD-L1 expression and T cell infiltration. Despite the fact that high PD-L1 expression is regarded as a sign of immune escape, a large number of clinical trials showed that patients with PD-L1 positive tumors had increased response rate and clinical benefit while receiving anti-PD-1/anti-PD-L1 therapy [26–28]. Hence, PD-L1 as a single indicator maybe not enough to predict the efficacy of therapy, a comprehensive prediction framework involving such as PD-L1 expression, proportion of tumor infiltrating lymphocyte and neogenic antigen would be more accurate to select sensitive patients. A3B upregulation promoted tumor infiltrating T cells aggregation, moreover, elevated PD-L1 expression led to a more dependent and vital value for PD-L1 blockade to activate cytotoxic T cells to kill tumor cells. On the other hand, improved PD-L1 expression caused by DNA damage, for example utilization of PARP inhibitors and ionizing irradiation [29, 30], significantly enhanced the therapeutic effect of PD-L1 blockade. Based on our discovery, we illustrated the relationship of DNA stranded damage induced by A3B and immunotherapy effect, which provided new evidence that genomic instability determines sensitivity to immune checkpoint blockade. Although both PD-L1 overexpression and elevated immune cell infiltration could be beneficial for the response to anti-PD-1 antibody, A3B itself did not influence prognosis in patients who did not receive anti-PD-1 antibody therapy. This may be because the high PD-L1 level suppressed the infiltrating T cells, as verified by the high level of exhausted CD8 + T cells in A3B-transfected tumors (Fig. 3D). Thus, although A3B did not influence the prognosis, it made the tumor a better candidate for anti-PD-1 immunotherapy.

As the only member of the deaminase family with constitutive nuclear localization [31, 32], A3B is highly expressed in various cancers [17, 33–35]. A3B-catalyzed genomic uracil damage is the cause of a large number of dispersed and aggregated mutations in many different tumors [17]. DNA damage repair pathway has been reported to denpend on checkpoint kinases ATM, ATR and Chk1, and consequently activate STAT signalling to upregulate PD-L1 expression [36], and we further confirmed that the A3B-related nucleotide damage regulates the expression of PD-L1 mainly through the activation of ATR-CHK1. In addition, our bioinformatics-based research further confirmed the positive association of TMB and A3B in various cancers. However, previous studies have shown that APOBEC mutagenesis and immunotherapy may be independent of TMB [18, 19]. Moreover, inhibition of PD-L1 expression by blocking the DSBs–ATR–CHK1 pathway results in complete loss of sensitivity to anti-PD-1 immunotherapy. Therefore, the upregulation of TMB induced by A3B cannot be said to play a major role in enhanced immunotherapy. However, uracil damage is processed into gaps by the synergistic action of uracil DNA glycosylase and alkaline endonuclease, which can lead to single-stranded or double-stranded DNA breaks [37–39]; other members of the APOBEC3s family do not have the same effect as A3B [40–45].

Consistently, in the present study, by unbiased gene variation scan, pathway enrichment analysis, and pathway inhibition, we demonstrated that the DSBs–ATR–CHK1 pathway is essential in A3B-induced PD-L1 expression and immune cell infiltration. Thus, PD-1 blockade sensitization by A3B is mainly through the activation of the DSB–ATR–CHK1 pathway and the subsequent PD-L1 expression and T cell infiltration, but not through hypermutation. Moreover, we illustrated that A3B-induced PD-L1 upregulation depended on the activated STAT–IRF1 signaling, which is located downstream of the ATR–CHK1 pathway. However, how the ATR–CHK1 pathway influences immune infiltration needs to be explored in subsequent studies.

In summary, our results showed that A3B significantly increased the sensitivity of cancers to the anti-PD-1 antibody by enhancing PD-L1 expression and immune cells infiltration through the DSB–ATR–CHK1 pathway. These findings suggested that A3B may be a promising biomarker of immunotherapy.

APOBEC Apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like

A3B APOBEC3B

PD-L1 Programmed cell death-ligand 1

ICI Immune checkpoint inhibitor

TMB Tumor mutational burden

BRCA Breast cancer

LIHC Liver hepatocellular carcinoma

CESC Cervical squamous cell carcinoma and adenocarcinoma

HNSC Head and neck squamous cell carcinoma

LUAD Lung adenocarcinoma

LUSC Lung squamous cell carcinoma

SKCM Skin cutaneous melanoma

DLBC Diffuse large B-cell lymphoma

KIRC Kidney renal clear cell carcinoma

KICH Kidney chromophobe

STAD Stomach adenocarcinoma

TILs Tumor-infiltrating lymphocytes

GO Gene Ontology

Ethics approval and consent to participate：We obtained patient material from the Hunan People’s Hospital, First Affiliated Hospital of Hunan Normal University (Hunan, China). All patients provided written informed consent for detailed pathological and clinical data, and for participation to the study. The study was approved by the Medical Ethics Committee of Hunan People’s Hospital of Hunan Normal University.

Consent for publication：Not applicable.

Availability of data and material：Data are available on reasonable request.

Competing interests：All authors declare that they have no competing interests.

Funding：This work was supported by the National Natural Science Foundation of China (81572966).

Authors’contributions: Hong Yang: Project conceptualization, experimental design, investigation, validation. Shiyu Chen: Project conceptualization, experimental design, investigation, validation, methodology, bioinformatics analysis. Yueyun Ma: Writing-original draft and editing. Rong Wang: Experimental design, investigation, methodology. Yanxia Wang: Experimental design, investigation, methodology. Qin Zeng: Experimental design, investigation, methodology. Wei Guo: Experimental design, investigation, methodology. Bin Xie: Experimental design, investigation, methodology. Xiaomei Zhou: Scientific input, supervision. Tongxin Liu: Scientific input, supervision. Ziyi Luo: Scientific input, supervision. Wei Wang and Shuai Chu: Project conceptualization, supervision, funding acquisition, experimental design, investigation, project administration, writing-original draft, writing-review and editing.

Acknowledgements: We thank LetPub (www.letpub.com) for its linguistic assistance during the preparation of this manuscript.

Bagchi, S., R. Yuan and E.G. Engleman, Immune Checkpoint Inhibitors for the Treatment of Cancer: Clinical Impact and Mechanisms of Response and Resistance. Annu Rev Pathol, 2021. 16: p. 223–249.
Gibney, G.T., L.M. Weiner and M.B. Atkins, Predictive biomarkers for checkpoint inhibitor-based immunotherapy. Lancet Oncol, 2016. 17(12): p. e542-e551.
Yi, M., et al., Biomarkers for predicting efficacy of PD-1/PD-L1 inhibitors. Mol Cancer, 2018. 17(1): p. 129.
Davis, A.A. and V.G. Patel, The role of PD-L1 expression as a predictive biomarker: an analysis of all US Food and Drug Administration (FDA) approvals of immune checkpoint inhibitors. J Immunother Cancer, 2019. 7(1): p. 278.
Samstein, R.M., et al., Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat Genet, 2019. 51(2): p. 202–206.
Chan, T.A., et al., Development of tumor mutation burden as an immunotherapy biomarker: utility for the oncology clinic. Ann Oncol, 2019. 30(1): p. 44–56.
Jardim, D.L., et al., The Challenges of Tumor Mutational Burden as an Immunotherapy Biomarker. Cancer Cell, 2021. 39(2): p. 154–173.
Gopalakrishnan, V., et al., The Influence of the Gut Microbiome on Cancer, Immunity, and Cancer Immunotherapy. Cancer Cell, 2018. 33(4): p. 570–580.
Nabet, B.Y., et al., Noninvasive Early Identification of Therapeutic Benefit from Immune Checkpoint Inhibition. Cell, 2020. 183(2): p. 363–376.e13.
Jin, Y., et al., The predicting role of circulating tumor DNA landscape in gastric cancer patients treated with immune checkpoint inhibitors. Mol Cancer, 2020. 19(1): p. 154.
Henderson, S. and T. Fenton, APOBEC3 genes: retroviral restriction factors to cancer drivers. Trends Mol Med, 2015. 21(5): p. 274–84.
Desimmie, B.A., et al., APOBEC3 proteins can copackage and comutate HIV-1 genomes. Nucleic Acids Res, 2016. 44(16): p. 7848–65.
Siriwardena, S.U., K. Chen and A.S. Bhagwat, Functions and Malfunctions of Mammalian DNA-Cytosine Deaminases. Chem Rev, 2016. 116(20): p. 12688–12710.
Ladanyi, A. and J. Timar, Immunologic and immunogenomic aspects of tumor progression. Semin Cancer Biol, 2020. 60: p. 249–261.
Wang, S., et al., APOBEC3B and APOBEC mutational signature as potential predictive markers for immunotherapy response in non-small cell lung cancer. Oncogene, 2018. 37(29): p. 3924–3936.
Barroso-Sousa, R., et al., Prevalence and mutational determinants of high tumor mutation burden in breast cancer. Ann Oncol, 2020. 31(3): p. 387–394.
Burns, M.B., N.A. Temiz and R.S. Harris, Evidence for APOBEC3B mutagenesis in multiple human cancers. Nat Genet, 2013. 45(9): p. 977–83.
Boichard, A., I.F. Tsigelny and R. Kurzrock, High expression of PD-1 ligands is associated with kataegis mutational signature and APOBEC3 alterations. Oncoimmunology, 2017. 6(3): p. e1284719.
Boichard, A., et al., APOBEC-related mutagenesis and neo-peptide hydrophobicity: implications for response to immunotherapy. Oncoimmunology, 2019. 8(3): p. 1550341.
Swanton, C., et al., APOBEC Enzymes: Mutagenic Fuel for Cancer Evolution and Heterogeneity. Cancer Discov, 2015. 5(7): p. 704–12.
Yoshihara, K., et al., Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun, 2013. 4: p. 2612.
Jazayeri, A., et al., ATM- and cell cycle-dependent regulation of ATR in response to DNA double-strand breaks. Nat Cell Biol, 2006. 8(1): p. 37–45.
Harris, R.S., Molecular mechanism and clinical impact of APOBEC3B-catalyzed mutagenesis in breast cancer. Breast Cancer Res, 2015. 17: p. 8.
Fujiki, Y., et al., APOBEC3B gene expression as a novel predictive factor for pathological complete response to neoadjuvant chemotherapy in breast cancer. Oncotarget, 2018. 9(55): p. 30513–30526.
Kim, S.H., et al., Identifying germline APOBEC3B deletion and immune phenotype in Korean patients with operable breast cancer. Breast Cancer Res Treat, 2020. 183(3): p. 697–704.
Reck, M., et al., Pembrolizumab versus Chemotherapy for PD-L1-Positive Non-Small-Cell Lung Cancer. N Engl J Med, 2016. 375(19): p. 1823–1833.
Ding, W., et al., Pembrolizumab in patients with CLL and Richter transformation or with relapsed CLL. Blood, 2017. 129(26): p. 3419–3427.
Balar, A.V., et al., First-line pembrolizumab in cisplatin-ineligible patients with locally advanced and unresectable or metastatic urothelial cancer (KEYNOTE-052): a multicentre, single-arm, phase 2 study. Lancet Oncol, 2017. 18(11): p. 1483–1492.
Jiao, S., et al., PARP Inhibitor Upregulates PD-L1 Expression and Enhances Cancer-Associated Immunosuppression. Clin Cancer Res, 2017. 23(14): p. 3711–3720.
Deng, L., et al., Irradiation and anti-PD-L1 treatment synergistically promote antitumor immunity in mice. J Clin Invest, 2014. 124(2): p. 687–95.
Lackey, L., et al., APOBEC3B and AID have similar nuclear import mechanisms. J Mol Biol, 2012. 419(5): p. 301–14.
Burns, M.B., et al., APOBEC3B is an enzymatic source of mutation in breast cancer. Nature, 2013. 494(7437): p. 366–70.
Burns, M.B., et al., APOBEC3B is an enzymatic source of mutation in breast cancer. Nature, 2013. 494(7437): p. 366–70.
Ng, J., et al., Pan-cancer transcriptomic analysis dissects immune and proliferative functions of APOBEC3 cytidine deaminases. Nucleic Acids Res, 2019. 47(3): p. 1178–1194.
Roberts, S.A., et al., An APOBEC cytidine deaminase mutagenesis pattern is widespread in human cancers. Nat Genet, 2013. 45(9): p. 970–6.
Sato, H., et al., DNA double-strand break repair pathway regulates PD-L1 expression in cancer cells. Nat Commun, 2017. 8(1): p. 1751.
Yamazaki, H., et al., Endogenous APOBEC3B Overexpression Constitutively Generates DNA Substitutions and Deletions in Myeloma Cells. Sci Rep, 2019. 9(1): p. 7122.
Nowarski, R. and M. Kotler, APOBEC3 cytidine deaminases in double-strand DNA break repair and cancer promotion. Cancer Res, 2013. 73(12): p. 3494–8.
Taylor, B.J., et al., DNA deaminases induce break-associated mutation showers with implication of APOBEC3B and 3A in breast cancer kataegis. Elife, 2013. 2: p. e00534.
Refsland, E.W., et al., Quantitative profiling of the full APOBEC3 mRNA repertoire in lymphocytes and tissues: implications for HIV-1 restriction. Nucleic Acids Res, 2010. 38(13): p. 4274–84.
Koning, F.A., et al., Defining APOBEC3 expression patterns in human tissues and hematopoietic cell subsets. J Virol, 2009. 83(18): p. 9474–85.
Fujino, T., N. Navaratnam and J. Scott, Human apolipoprotein B RNA editing deaminase gene (APOBEC1). Genomics, 1998. 47(2): p. 266–75.
Muramatsu, M., et al., Specific expression of activation-induced cytidine deaminase (AID), a novel member of the RNA-editing deaminase family in germinal center B cells. J Biol Chem, 1999. 274(26): p. 18470–6.
Hultquist, J.F., et al., Human and rhesus APOBEC3D, APOBEC3F, APOBEC3G, and APOBEC3H demonstrate a conserved capacity to restrict Vif-deficient HIV-1. J Virol, 2011. 85(21): p. 11220–34.
Land, A.M., et al., Endogenous APOBEC3A DNA cytosine deaminase is cytoplasmic and nongenotoxic. J Biol Chem, 2013. 288(24): p. 17253–60.

No competing interests reported.

SupplementalFigs.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

APOBEC3B Improves Response to PD-1 Blockade Through DSBs–ATR–CHK1 Pathway

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Materials And Methods

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1