Clin Epigenetics. 2023;15(1):75.

Epigenetic priming improves salvage chemotherapy in diffuse large B-cell lymphoma via endogenous retrovirus-induced cGAS-STING activation.

, , , , , , , , , ,

제1저자 리우준 · 교신저자 홍준식

PMID 37138342 PMC10155448 DOI 10.1186/s13148-023-01493-x IF 5.7 혈액종양내과

소속 (11)

College of Medicine, Zhejiang University, Hangzhou, China.
Center for Medical Innovation, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Republic of Korea.
Liangzhu Laboratory, Zhejiang University Medical Center, Hangzhou, China.
Cancer Research Institute, Seoul National University College of Medicine, Seoul, Republic of Korea.
Department of Internal Medicine, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul, 03080, Republic of Korea.
Biomedical Research Institute, Seoul National University College of Medicine, Seoul, Republic of Korea.
Division of Hematology/Oncology, Department of Internal Medicine, Vanderbilt University Medical Center, Nashville, TN, USA.
Center for Medical Innovation, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Republic of Korea. hongjblood@snu.ac.kr.
Cancer Research Institute, Seoul National University College of Medicine, Seoul, Republic of Korea. hongjblood@snu.ac.kr.
Department of Internal Medicine, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul, 03080, Republic of Korea. hongjblood@snu.ac.kr.
Biomedical Research Institute, Seoul National University College of Medicine, Seoul, Republic of Korea. hongjblood@snu.ac.kr.

Abstract

BACKGROUND Although most patients with diffuse large B-cell lymphoma (DLBCL) achieve complete remission after first-line rituximab-containing immunochemotherapy, up to 40% of patients relapse and require salvage therapy. Among those patients, a substantial proportion remain refractory to salvage therapy due to insufficient efficacy or intolerance of toxicities. A hypomethylating agent, 5-azacytidine, showed a chemosensitizing effect when primed before chemotherapy in lymphoma cell lines and newly diagnosed DLBCL patients. However, its potential to improve outcomes of salvage chemotherapy in DLBCL has not been investigated.

RESULTS In this study, we demonstrated the mechanism of 5-azacytidine priming as a chemosensitizer in a platinum-based salvage regimen. This chemosensitizing effect was associated with endogenous retrovirus (ERV)-induced viral mimicry responses via the cGAS-STING axis. We found deficiency of cGAS impaired the chemosensitizing effect of 5-azacytidine. Furthermore, combining vitamin C and 5-azacytidine to synergistically activate STING could be a potential remedy for insufficient priming induced by 5-azacytidine alone.

CONCLUSIONS Taken together, the chemosensitizing effect of 5-azacytidine could be exploited to overcome the limitations of the current platinum-containing salvage chemotherapy in DLBCL and the status of cGAS-STING has the potential to predict the efficacy of 5-azacytidine priming.

Keywords

5-AzacytidineCisplatinDLBCLERVsSTING

Figures

**Fig. 1** 5-Azacytidine pretreatment resensitizes cisplatin-resistant DLBCL cell lines to cisplatin to varying degrees. A Plasma levels of cisplatin in human patients following intravenous injection of an 80 mg/m² dose. Red points represent the determined value for the patient at that time point reported by a widely cited paper. Black points represent the mPBPK model-predicted value using typical physiological and biochemical parameters. The dashed line indicates the top clinically achievable dose. B The IC50 value was calculated using a cell viability assay. The dashed line indicates the top clinically achievable dose. C Bar chat of IC50 values for cisplatin in six DLBCL cell lines; *p < 0.05, **p < 0.01 and ***p < 0.001 (two-way ANOVA). D Bar chat of log2-transformed IC50 fold change values. E–J. Subcutaneous tumor growth, xenograft tumor images and tumor weights of xenografts from OCI-LY1 and SU-DHL2 tumor-bearing athymic nude mice. E and H *p < 0.05, **p < 0.01 and ***p < 0.001 (two-way ANOVA). G and I *p < 0.05, **p < 0.01 and ***p < 0.001 (t test). AZA: 5-azacytidine

**Fig. 2** Low-dose 5-azacytidine induces DNA demethylation without DNA damage. A Phospho-H2AX expression in the indicated DLBCL cell lines was detected using Western blotting; UV-treated cell was serving as positive control. B 5mC DNA dot blots showing the effect of low-dose 5-azacytidine on the global DNA methylation level for each cell line, and methylene blue served as the DNA loading control. C DNMT1 protein levels were detected using Western blotting in PBS or 5-azacytidine-treated cells

**Fig. 3** Low-dose 5-azacytidine treatment altered the expression of cisplatin sensitivity-related genes but did not have a dominant effect on the transcriptome. A Hierarchal clustering of the top variable gene expression profiles (interquartile range of probe sets between cell lines > 1). B Overlapping genes between the cisplatin sensitivity-related gene set and the 5-azacytidine treatment-regulated gene set. C The gene expression profiles of the 122 overlapping genes are shown using a heatmap

**Fig. 4** 5-Azacytidine activates ERV expression in DLBCL cell lines. A ERV expression profiles are shown using a heatmap. B Histogram of the number of reads attributed to each of the 3220 ERV loci sorted in order of highest to lowest expressed ERVs for each cell line, x-axis: 3220 ERV loci in percentage scale, y-axis: scaled expression levels of ERVs in each cell line. C The total ERV locus number is shown using a bar chat. D The total ERV normalized expression is shown using a bar chat. E Venn diagram showing the number of baseline ERV, post-treatment ERV, and 5-azacytidine-induced ERV conditions in each cell line. F The number of 5-azacytidine-induced ERV loci in each cell line is shown using a bar chat. G The 5-azacytidine-induced ERV normalized expression level in each cell line is shown using a bar chat. H Correlation between the 5-azacytidine-induced ERV locus number and ERV expression. I–L The expression levels of several ERVs were detected using qPCR. I–L *p < 0.05, **p < 0.01 and ***p < 0.001 (two-way ANOVA). AZA: 5-azacytidine

**Fig. 5** 5-Azacytidine induces chemosensitization mediated by the cGAS-STING axis. A–H The expression levels of target proteins were measured using Western blotting and RNA-sequencing data. MDA5, RIG-I, cGAS, and STING are encoded by the IFIH1, DDX58, MB21D1, and TMEM173 genes, respectively. I Pearson correlation between the dsDNA sensor activation level and cisplatin sensitivity. J Pearson correlation between cGAS-STING activation and the chemosensitization effect. K, L The expression and phosphorylation levels of IRF3 and IκBα were measured using western blotting

**Fig. 6** cGAS-STING-mediated viral mimicry was dependent on reverse transcriptase activity. A–C Reverse transcriptase activity was measured after PBS, AZA, RTi, and AZA + RTi treatment on OCI-LY1, SU-DHL2, and SU-DHL8, respectively. D–F IC50 values were calculated using cell viability measured on day 5 after pretreatment with PBS, AZA, RTi, and the AZA + RTi combination on OCI-LY1, SU-DHL2, and SU-DHL8, respectively. G–I The expression of cGAS and STING and phosphorylation levels of IRF3 and IκBα were measured using western blotting after PBS, AZA, RTi, and AZA + RTi treatment on OCI-LY1, SU-DHL2, and SU-DHL8, respectively. J–L The expression of cGAS and STING and phosphorylation levels of IRF3 and IκBα were measured in the negative control and knockdown cell lines with and without 5-azacytidine treatment on SU-DHL8 cells. A and B *p < 0.05, **p < 0.01 and ***p < 0.001 (two-way ANOVA). AZA: 5-azacytidine, RTi: reverse transcriptase inhibitor (delavirdine mesylate)

**Fig. 7** A combination of vitamin C and 5-azacytidine could be a potential remedy for insufficient epigenetic priming. A IC50 values were calculated using cell viability measured on day 5. *p < 0.05, **p < 0.01 and ***p < 0.001 (two-way ANOVA). B, STING expression levels were measured using western blotting. C, D GLUT3 expression levels were measured using RNA-sequencing data and western blotting, GLUT3 is encoded by the SLC2A3 gene. E GLUT3 expression levels were measured using western blotting of empty vector-expressing and SLC2A3-overexpressing OCI-LY1 cell lines, OE: overexpression. F IC50 values were calculated using cell viability measured on day 5. *p < 0.05, **p < 0.01 and ***p < 0.001 (two-way ANOVA). G STING expression levels were measured using western blotting. H–J Subcutaneous tumor sizes, xenograft tumor images, and tumor weights of xenografts from empty vector-expressing or SLC2A3-overexpressing OCI-LY1 tumor-bearing models are shown. *p < 0.05, **p < 0.01 and ***p < 0.001 (t test). AZA: 5-azacytidine, VC: vitamin C, OE: overexpression

**Fig. 8** Epigenetic priming-induced chemosensitization is a potential solution for the unmet needs of salvage chemotherapy. 5-azacytidine epigenetic priming improved platinum-based salvage chemotherapy in DLBCL via ERV-induced cGAS-STING activation rather than via dsRNA sensor activation, cGAS-STING activation is likely involved over dsRNA detection due to the putatively high levels of reverse transcriptase in DLBCL cells, and the cGAS/STING deficiency could impair the chemosensitizing effect

그림 출처: PubMed Central 공개 논문 (PMC10155448)

Conflict of interest statement

The authors declare no competing interests.

그 밖의 정보