ahmed_mohamed

INTRODUCTION

Squamous cell carcinoma is the most globally occurring neoplasm in the oral cavity. The genetic and epigenetic regulations of oral squamous cell carcinoma (OSCC) have been widely studied in the last few years (1). Epigenetics is the term describing the reversible changes in gene expression without any difference in DNA sequences, a process achieved through DNA methylation, histone modification, chromatin remodeling, and post-transcriptional modifications (2, 3).

Transferring a methyl group to the C5 position of the cytosine during DNA methylation regulates gene expression either by activating proteins engaged in repressing gene expression or by weakening the binding of the transcription factors to DNA. This process occurs, in normal development and cancer, under the control of a group of enzymes called DNA methyltransferases (2, 4).

DNA methyl transferase 3A (DNMT3A) is an enzyme responsible for de novo methylation which means the addition of methyl group to unmodified DNA. DNMT3A increased expression was proved to be related to progression, poor prognosis, aggressiveness, and recurrence in hepatocellular carcinoma, breast cancer, vulvar cancer, and OSCC (5-8). In their study, Daniel et al. (7) declared that DNMT3A has an early significant role in the carcinogenesis of OSCC when compared to the other studied methyltransferases.

The effect of inflammatory mediators on carcinogenesis has been widely investigated (9, 10). Cytokines like Interleukin- 1, Interleukin-6 (IL-6), transforming growth factor beta (TGF β), Interferon-γ, and many others were reported to have high expression in skin SCC and OSCC (11-13).

The chemokine (C-X-C motif) ligand 1 (CXCL1) is one of the CXC family (14). This agent is either secreted by the tumor cells (15, 16) or by cells of the tumor microenvironment (17,18). It is considered a promoter of cancer cell proliferation, angiogenesis, and metastasis in numerous cancer types including OSCC (15, 18-21). Previous studies on different epithelial malignancies attempted to clarify the interaction between different chemokines and epigenetic regulation of cancer cells whether on the expression of methyltransferase enzymes or specific gene silencing (22-24). Therefore, we aimed to detect the immunohistochemical correlation of the potent epigenetic marker DNMT3A with epithelial and mesenchymal CXCL1, as well as detect their association with different histological grades and clinical parameters in OSCC.

MATERIALS AND METHODS

SPECIMEN COLLECTION

Hundred and twenty specimens of formalin-fixed paraffin-embedded (FFPE) blocks were included in this study. The blocks were obtained from the archives of the Oral Pathology Department, Faculty of Dentistry, Ain Shams University, and General Pathology Department Faculty of Medicine Cairo University, Cairo, Egypt. The blocks used were present for diagnostic purposes from years (2015-2020). The work was done according to the Declaration of Helsinki. The ethics committee of the Faculty of Dentistry, Ain Shams University considered the study with exemption from consent (FDASU-Rec ER072201), as all the blocks were coded, the patients were unidentified and their information was anonymous.

The cases were divided into 30 normal control samples from blocks of gingival tissue obtained during implant surgeries. The other 90 were OSCC cases of primary neoplasms. The cases were divided into 30 each well differentiated, moderately differentiated, and poorly differentiated according to the WHO grading system (25). Two pathologists not familiar with clinical data of patients re-diagnosed the cases.

IMMUNOHISTOCHEMISTRY

FFPE tissues of 4-μm thick sections were mounted on positively charged glass slides. Sections were deparaffinized, rehydrated, and antigen retrieved using Trilogy (Cell Marque, Millipore, Sigma-Aldrich, Germany). The slides were stained automatically and counterstained by Hematoxyllin according to the manufacturer’s protocol (Ventana-Benchmark GX, Tucson, AZ, USA). The primary rabbit polyclonal antibodies; anti-DNMT3A (HPA026588) and anti-CXCL1 (SAB4301833) were obtained from (Sigma-Aldrich, Germany) and diluted to (1:50) and (1:100) respectively.

At least 4 fields from each slide at magnification 400X were captured using (Canon EOS 650D) digital camera mounted on a light microscope (BX60, Olympus, Japan). We assessed DNMT3A expression in the nucleus of epithelial cells only (7), while CXCL1 immunoreactivity was assessed in either the cytoplasm or the nucleus of epithelial and mesenchymal cells.

The Immunohistochemical (IHC) scoring was the used semiquantitative system for immunohistochemical expression detection. The percentage and intensity of positively stained epithelial cells to the total number of epithelial cells were determined in both DNMT3A and CXCL1 sections, while those of the mesenchymal CXCL1 stained cells, were detected in relation to the total number of mesenchymal cells. Then, the IHC scoring formula was applied as follows: 3 × percentage of strongly stained cells (brownish black) + 2 × percentage of moderately stained cells (brown) + 1 × percentage of weakly stained cells (light brown) + 0 × percentage of negatively stained cells (26). Two independent observers counted the cells and evaluated the intensity.

The IHC cutoff value was chosen based on the mean IHC score for each marker (27). The IHC score was defined as high when the DNMT3A score was higher than 174 while the high IHC score for CXCL1 was defined as higher than 125.

STATISTICAL ANALYSIS

Data were coded and entered using the statistical package for the Social Sciences (SPSS) version 28 (IBM Corp., Armonk, NY, USA). Data was summarized using mean and standard deviation for quantitative variables and frequencies (number of cases) and relative frequencies (percentages) for categorical variables. Data was double-checked for normality using normality plots and Shapiro Wilk test and proved not to deviate from normal distribution. Comparisons of IHC scores between normal mucosa and different tumor grades were done using analysis of variance (ANOVA) with multiple comparisons (Tukey) post hoc test. Correlations between the pathologic variables and DNMT3A, epithelial, and mesenchymal CXCL1 were performed using the Chi-square test. Correlations between DNMT3A and epithelial CXCL1 besides DNMT3A and mesenchymal CXCL1 were done using Pearson’s correlation coefficient. P-values less than 0.05 were considered statistically significant.

RESULTS

Eighty-three male and 37 female FFPE tissue samples with an age range from (31-92) years were used in this study. Of them, 65 males and 25 females were OSCC patients with an age range of (40-92).

DNMT3A AND CXCL1 IMMUNOHISTOCHEMICAL EXPRESSION IN DIFFERENT HISTOLOGICAL GRADES

The epithelial tissues of all groups showed positive nuclear immunoreactivity of DNMT3A (Figure 1.a-d). A statistically significant increase in the IHC score of DNMT3A protein was observed in each OSCC grade in comparison to the normal mucosa. Besides a significant increase between well differentiated when compared to moderately and poorly differentiated and between moderately differentiated in comparison to poorly differentiated groups (p<0.001) (Table 1).

CXCL1 immunostaining in epithelial cells revealed cytoplasmic positive expression in the normal group and all OSCC groups, while nuclear immunostaining appeared in some tumor cells of the poorly differentiated group (Figure 1.e-h). The peripheral cells in cell nests showed higher staining intensity than central cells (Figure 1.g). Mesenchymal cells showed immunoreactivity of CXCL1 in all study groups (Figure 1.e-h). A significant rise in the mean IHC score of both epithelial and mesenchymal CXCL1 was observed in each grade when compared to normal mucosa and between one grade and the other (p<0.001) (Table 1).

CORRELATION of DNMT3A AND CXCL1

WITH CLINICAL PARAMETERS

Next, we examined the correlation of DNMT3A, epithelial, and mesenchymal CXCL1 with different clinical parameters. Correlation with age was adjusted to cut off value of 60 years which was the mean age of the OSCC cases used in the study.

Chi-square test analysis of the relation of clinical data with the IHC expressions of DNMT3A, epithelial, and mesenchymal CXCL1 did not reveal any association of any of the markers with age, sex, or tumor site. However, all three markers showed a correlation with lymph node metastasis (p=0.010, 0.018, and 0.019 respectively). Moreover, all DNMT3A, epithelial, and mesenchymal CXCL1were seen to be related to the change in tumor size from T1/T2 to T3/T4 (p=0.020, 0.040, and 0.040 respectively) and tumor stage from I/ II to III/ IV (p=0.008, 0.009 and 0.010 respectively) (Table 2).

CORRELATION OF DNMT3A EXPRESSION WITH EPITHELIAL AND MESENCHYMAL CXCL1 EXPRESSIONS

Pearson’s correlation test revealed a statistically significant strong correlation between DNMT3A IHC score and that of epithelial and mesenchymal CXCL1 (r=0.937, p<0.001 and r=0.917, p<0.001 respectively) (Table 3).

DISCUSSION

DNMT3A is an epigenetic regulator, acting by silencing tumor suppressor genes favoring proliferation and carcinogenesis (2, 28). Understanding the effect of cytokines on epigenetic regulation paves the way for more treatment strategies for OSCC. CXCL1 whether produced from tumor or stromal cells was reported as a predictive factor for cancer progression (15, 18, 29). Although some studies shed light on the cross-talk between cytokines and epigenetic management (22, 23), the complete picture in OSCC remains unclear.

Our study revealed an increase in the IHC score of DNMT3A in different OSCC grades from normal mucosa and between tumor grades. Previously, the expression of the methyltransferase in OSCC was detected to be statistically significant with the non-neoplastic tissue (7, 8, 30). It was declared that the high expression of de novo DNMT3A renders it responsible at least partly for the silencing of tumor suppressor genes during oral carcinogenesis (7). In addition, DNMT3A immunoreactivity was found to increase from nonneoplastic tissues through high-grade nodules to advanced carcinoma in liver cancer (31). It should be noted here that only the nuclear expression was regarded in scoring, as the enzyme is considered active just when transported to the nucleus (32).

The association of DNMT3A with tumor size, stage, and lymph node metastasis was found to be statistically significant, a finding indicating a correlation between the expression and poor prognosis. Yang J. et al. (33) proposed similar results in gastric cancer. They detected an association between DNMT3A immunoscoring, III /IV TNM stage, and lymph node involvement. Moreover, DNMT3A overexpression was found to be associated with high recurrence in vulvar carcinoma (6). Similar to our results, Daniel et al., did not detect any association between DNMT3A expression and age or sex in OSCC (7).

The expression of CXCL1 revealed a statistically significant increase in epithelial and mesenchymal immunoscoring between normal mucosa and OSCC and by the transition from one grade to another. Following our results, a study on bladder cancer detected higher IHC scores in high-grade than low-grade tumors (34). The mechanisms of CXCL1-induced tumorigenesis have been well-established (18, 21, 30,35). Lee et al. (21) indicated that CXCL1 was involved in the interleukin 1 β (IL-1 β)-mediated phosphorylation of epidermal growth factor receptors promoting proliferation of oral premalignant cells. It was shown that lung cancer growth was induced by tumor cell-secreted CXCL1 via recruitment of neutrophils into the tumor (35). Also, gastric cancer cells induced apoptosis, and decreased proliferation and migration were reached by CXCL1 silencing (36).

The nature of CXCL1-positive mesenchymal cells in different cancer types was verified by dual immunostaining and cell line studies to be cancer-associated fibroblasts (CAF) and tumor-associated macrophages (TAM) (18,29,37). Wei et al. (19) reported that besides the high expression of CXCL1 in OSCC cells, IL-1β expressed in tumor cells significantly increased CXCL1 production by CAF, which in turn promoted oral malignant cell invasion and migration. CXCL1 produced by TAMs stimulated breast cancer metastasis via Nuclear factor-κB/SOX4 induction (37). TAMs/ CAFs-secreted CXCL1 enhanced tumor growth of subcutaneous bladder cancer tumors in nude mice and was responsible for repeated recurrence and disease progression through its ability to enhance invasion (29).

Comparing the CXCL1 IHC score in tumor and mesenchymal cells with clinical variants revealed a significant increase with advanced tumor size and stage, and lymph node metastasis, rendering CXCL1 a prognostic indicator of OSCC. Similarly, Wan et al. (15) and Yu et al. (16) detected high marker expression, in large tumors, high stages, and tumors with nodal metastasis in cervical and lung cancer respectively. Moreover, the expression positively correlated with tumor stage and nodal infiltration in breast cancer (37).

In our study, the correlation of DNMT3A was statistically significant with epithelial and mesenchymal CXCL1. The results were supported by the postulate of cytokine-derived epigenetic reprogramming. The modulation of DNMTs under the effect of pro-inflammatory cytokines during cancer initiation and progression was previously determined in literature (38). IL-6 upregulated DNMT-induced gene methylation, and tumor suppressor gene silencing (11). TGF β was acknowledged to be involved in activating DNA methyltransferases including DNMT3A in ovarian and liver cancers (23, 39). In their study on breast cancer cell lines, Mathot et al. (40) claimed that stromal cells-secreted factors were linked to the activation of genes characterized by DNA methylation patterns. They also reported that the change in methylation could be related to stromal-cancer cell interaction.

The aforementioned studies clarified the role of stromal elements in the epigenetic modification of cancer cells; nevertheless, the autocrine effect of cytokines produced by tumor cells could not be detected earlier. It is recommended to follow this immunohistochemical study with further cell line studies for the detection of the exact causative relation between CXCL1 and DNMT3A in the initiation and progression of OSCC.

CONCLUSION

Our data showed that DNMT3A positively correlated with epithelial and mesenchymal CXCL1 in OSCC. Each of DNMT3A, epithelial, and mesenchymal CXCL1 were well associated with tumor histological grades, advanced tumor features, and lymph node metastasis, rendering them potential prognostic markers for OSCC.

AUTHOR CONTRIBUTION STATEMENT

Conceptualization and design: B.A.A.

Interpretation of data: B.A.A., M.H.M.

Writing and revising the work: B.A.A., M.H.M.

References

1. Hema K.N., Smitha T., Sheethal H.S., Mirnalini S.A. Epigenetics in oral squamous cell carcinoma. JOMFP. 2017; 21 (2): 252. https://doi.org/10.4103/jomfp.jomfp_150_17

2. Cheng Y., He C., Wang M., Ma X., Mo F., Yang S., Wei X. Targeting epigenetic regulators for cancer therapy: mechanisms and advances in clinical trials. Signal Transduct Target Ther. 2019; 4 (1): 1-39. https://doi.org/10.1038/s41392-019-0095-0

3. Bollati V., Baccarelli A. Environmental epigenetics. Heredity. 2010; 105 (1): 105-12. https://doi.org/10.1038/hdy.2010.2

4. Smith Z.D., Meissner A. DNA methylation: roles in mammalian development. Nat Rev Genet. 2013; 14 (3): 204-20. https://doi.org/10.1038/nrg3354

5. Gacem R.B., Hachana M., Ziadi S., Abdelkarim S.B., Hidar S., Trimeche M. Clinicopathologic significance of DNA methyltransferase 1, 3a, and 3b overexpression in Tunisian breast cancers. Hum Pathol. 2012; 43 (10): 1731- 8. https://doi.org/10.1016/j.humpath.2011.12.022

6. Leonard S., Pereira M., Fox R., Gordon N., Yap J., Kehoe S., Luesley D., Woodman C., Ganesan R. Over-expression of DNMT3A predicts the risk of recurrent vulvar squamous cell carcinomas. Gynecol Oncol. 2016; 143 (2): 414-20. https://doi.org/10.1016/j.ygyno.2016.09.001

7. Daniel F.I., Rivero E.R., Modolo F., Lopes T.G., Salum F.G. Immunohistochemical expression of DNA methyltransferases 1, 3a and 3b in oral leukoplakias and squamous cell carcinomas. Arch Oral Biol. 2010; 55 (12): 1024-30. https://doi.org/10.1016/j.archoralbio.2010.08.009

8. Adhikari B.R., Uehara O., Matsuoka H., Takai R., Harada F., Utsunomiya M., Chujo T., Morikawa T., Shakya M., Yoshida K., Sato J. Immunohistochemical evaluation of Klotho and DNA methyltransferase 3a in oral squamous cell carcinomas. Med Mol Morphol. 2017; 50: 155-60. https://doi.org/10.1007/s00795-017-0156-9

9. Liu C.Y., Xu J.Y., Shi X.Y., Huang W., Ruan T.Y., Xie P., Ding J.L. M2-polarized tumor-associated macrophages promoted epithelial–mesenchymal transition in pancreatic cancer cells, partially through TLR4/IL-10 signaling pathway. Lab Invest. 2013; 93 (7): 844-54. https://doi.org/10.1038/labinvest.2013.69

10. Li W., Zhang X., Wang J., Li M., Cao C., Tan J., Ma D., Gao Q. TGFβ1 in fibroblasts-derived exosomes promotes epithelial-mesenchymal transition of ovarian cancer cells. Oncotarget. 2017; 8 (56): 96035. https://doi.org/10.18632/oncotarget.21635

11. Tuong Z.K., Lewandowski A., Bridge J.A., Cruz J.L., Yamada M., Lambie D., Lewandowski R., Steptoe R.J., Leggatt G.R., Simpson F., Frazer I.H. Cytokine/chemokine profiles in squamous cell carcinoma correlate with precancerous and cancerous disease stage. Sci Rep. 2019; 9 (1): 17754. https://doi.org/10.1038/s41598-019-54435-0

12. Koontongkaew S., Amornphimoltham P., Yapong B. Tumor-stroma interactions influence cytokine expression and matrix metalloproteinase activities in paired primary and metastatic head and neck cancer cells. Cell Biol Int. 2009; 33 (2): 165-73. https://doi.org/10.1016/j.cellbi.2008.10.009

13. Peltanova B., Raudenska M., Masarik M. Effect of tumor microenvironment on pathogenesis of the head and neck squamous cell carcinoma: a systematic review. Mol Cancer. 2019; 18 (1): 1-24. https://doi.org/10.1186/s12943-019-0983-5

14. Kasashima H., Yashiro M., Nakamae H., Kitayama K., Masuda G., Kinoshita H., Fukuoka T., Hasegawa T., Nakane T., Hino M., Hirakawa K. CXCL1-Chemokine (CXC Motif) Receptor 2 Signaling Stimulates the Recruitment of Bone Marrow–Derived Mesenchymal Cells into Diffuse-Type Gastric Cancer Stroma. Am J Pathol. 2016; 186 (11): 3028 -39. https://doi.org/10.1016/j.ajpath.2016.07.024

15. Wan X., Hong Z., Mao Y., Di W. Correlations of AKIP1, CXCL1 and CXCL2 expressions with clinicopathological features and survival profiles in cervical cancer patients. Transl Cancer Res. 2020; 9 (2): 726-34. https://doi.org/10.21037/tcr.2019.11.47

16. Yu S., Yi M., Xu L., Qin S., Li A., Wu K. CXCL1 as an unfavorable prognosis factor negatively regulated by DACH1 in non-small cell lung cancer. Front Oncol. 2020; 9: 1515. https://doi.org/10.3389/fonc.2019.01515

17. le Rolle A.F., Chiu T.K., Fara M., Shia J., Zeng Z., Weiser M.R., Paty P.B., Chiu V.K. The prognostic significance of CXCL1 hypersecretion by human colorectal cancer epithelia and myofibroblasts. J Transl Med. 2015; 13 (1): 1-2. https://doi.org/10.1186/s12967-015-0555-4

18. Wei L.Y., Lee J.J., Yeh C.Y., Yang C.J., Kok S.H., Ko J.Y., Tsai FC, Chia JS. Reciprocal activation of cancer-associated fibroblasts and oral squamous carcinoma cells through CXCL1. Oral Oncol. 2019; 88:115-23. https://doi.org/10.1016/j.oraloncology.2018.11.002

19. Wei Z.W., Xia G.K., Wu Y., Chen W., Xiang Z., Schwarz R.E., Brekken R.A., Awasthi N., He Y.L., Zhang C.H. CXCL1 promotes tumor growth through VEGF pathway activation and is associated with inferior survival in gastric cancer. Cancer Lett. 2015; 359 (2): 335-43. https://doi.org/10.1016/j.canlet.2015.01.033

20. Wang D., Sun H., Wei J., Cen B., DuBois R.N. CXCL1 is critical for premetastatic niche formation and metastasis in colorectal cancer. Cancer Res. 2017; 77 (13): 3655-65. https://doi.org/10.1158/0008-5472.can-16-3199

21. Lee C.H., Syu S.H., Liu K.J., Chu P.Y., Yang W.C., Lin P., Shieh W.Y. Interleukin-1 beta transactivates epidermal growth factor receptor via the CXCL1-CXCR2 axis in oral cancer. Oncotarget. 2015; 6 (36): 38866. https://doi.org/10.18632/oncotarget.5640

22. Tiwari N., Tiwari V.K., Waldmeier L., Balwierz P.J., Arnold P., Pachkov M., Meyer-Schaller N., Schübeler D., van Nimwegen E., Christofori G. Sox4 is a master regulator of epithelial-mesenchymal transition by controlling Ezh2 expression and epigenetic reprogramming. Cancer Cell. 2013; 23 (6): 768-83. https://doi.org/10.1016/j.ccr.2013.04.020

23. Cardenas H., Vieth E., Lee J., Segar M., Liu Y., Nephew K.P., Matei D. TGF-beta induces global changes in DNA methylation during the epithelial-to-mesenchymal transition in ovarian cancer cells. Epigenetics. 2014; 9 (11): 1461-72. https://doi.org/10.4161/15592294.2014.971608

24. Klymenko Y., Nephew K.P. Epigenetic crosstalk between the tumor microenvironment and ovarian cancer cells: a therapeutic road less traveled. Cancers. 2018; 10 (9): 295. https://doi.org/10.3390/cancers10090295

25. Almangush A., Mäkitie A.A., Triantafyllou A., de Bree R., Strojan P., Rinaldo A., Hernandez-Prera J.C., Suárez C., Kowalski L.P., Ferlito A., Leivo I. Staging and grading of oral squamous cell carcinoma: An update. Oral Oncol. 2020; 107: 104799. https://doi.org/10.1016/j.oraloncology.2020.104799

26. Thike A.A., Chng M.J., Tan P.H., Fook-Chong S. Immunohistochemical expression of hormone receptors in invasive breast carcinoma: correlation of results of H-score with pathological parameters. Pathology. 2001; 33 (1): 21-5. https://pubmed.ncbi.nlm.nih.gov/11280603/

27. Park S., Kim J., Jang W., Kim K.M., Jang K.T. Clinicopathologic significance of the delta-like ligand 4, vascular endothelial growth factor, and hypoxia-inducible factor-2α in gallbladder cancer. J Pathol Transl Med. 2023; 57 (2): 113-22. https://doi.org/10.4132/jptm.2023.02.01

28. Lyko F. The DNA methyltransferase family: a versatile toolkit for epigenetic regulation. Nat Rev Genet. 2018; 19 (2): 81-92. https://doi.org/10.1038/nrg.2017.80

29. Miyake M., Hori S., Morizawa Y., Tatsumi Y., Nakai Y., Anai S., Torimoto K., Aoki K., Tanaka N., Shimada K., Konishi N. CXCL1-mediated interaction of cancer cells with tumor-associated macrophages and cancer-associated fibroblasts promotes tumor progression in human bladder cancer. Neoplasia. 2016; 18 (10): 636-46. https://doi.org/10.1016/j.neo.2016.08.002

30. Daniel F.I., Alves S.R., Vieira D.S., Biz M.T., Daniel I.W., Modolo F. Immunohistochemical expression of DNA methyltransferases 1, 3a, and 3b in actinic cheilitis and lip squamous cell carcinomas. J Oral Pathol Med. 2016; 45 (10): 774-9. https://doi.org/10.1111/jop.12453

31. Choi M.S., Shim Y.H., Hwa J.Y., Lee S.K., Ro J.Y., Kim J.S., Yu E. Expression of DNA methyltransferases in multistep hepatocarcinogenesis. Hum Pathol. 2003; 34 (1): 11-7. https://doi.org/10.1053/hupa.2003.5

32. Lees-Murdock D.J., Shovlin T.C., Gardiner T., De Felici M., Walsh C.P. DNA methyltransferase expression in the mouse germ line during periods of de novo methylation. Dev Dyn: an official publication of the American Association of Anatomists. 2005; 232 (4): 992-1002. https://doi.org/10.1002/dvdy.20288

33. Yang J., Wei X., Wu Q., Xu Z., Gu D., Jin Y., Shen Y., Huang H., Fan H., Chen J. Clinical significance of the expression of DNA methyltransferase proteins in gastric cancer. Mol Med Rep. 2011; 4 (6): 1139-43. https://doi.org/10.3892/mmr.2011.578

34. Miyake M., Lawton A., Goodison S., Urquidi V., Gomes-Giacoia E., Zhang G., Ross S., Kim J., Rosser C.J. Chemokine (CXC) ligand 1 (CXCL1) protein expression is increased in aggressive bladder cancers. BMC Cancer. 2013; 13 (1): 1-7. https://doi.org/10.1186/1471-2407-13-322

35. Yuan M., Zhu H., Xu J., Zheng Y., Cao X., Liu Q. Tumor-derived CXCL1 promotes lung cancer growth via recruitment of tumor-associated neutrophils. J Immunol Res. 2016; 2016: 6530410. https://doi.org/10.1155/2016/6530410

36. Chen X., Jin R., Chen R., Huang Z. Complementary action of CXCL1 and CXCL8 in pathogenesis of gastric carcinoma. Int J Clin Exp Pathol. 2018; 11 (2): 1036-45. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6958037/

37. Wang N., Liu W., Zheng Y., Wang S., Yang B., Li M., Song J., Zhang F., Zhang X., Wang Q., Wang Z. CXCL1 derived from tumor-associated macrophages promotes breast cancer metastasis via activating NF-κB/SOX4 signaling. Cell Death Dis. 2018; 9 (9): 1-18. https://doi.org/10.1038/s41419-018-0876-3

38. Rokavec M., Öner M., Hermeking H. Inflammation-induced epigenetic switches in cancer. Cell Mol Life Sci. 2016; 73 (1): 23-39. https://doi.org/10.1007/s00018-015-2045-5

39. Martin M., Ancey P.B., Cros M.P., Durand G., Le Calvez-Kelm F., Hernandez-Vargas H., Herceg Z. Dynamic imbalance between cancer cell subpopulations induced by transforming growth factor beta (TGF-beta) is associated with a DNA methylome switch. BMC Genomics. 2014; 15: 435. https://doi.org/10.1186/1471-2164-15-435

40. Mathot P., Grandin M., Devailly G., Souazé F., Cahais V., Moran S., Campone M., Herceg Z., Esteller M., Juin P., Mehlen P., Dante R. DNA methylation signal has a major role in the response of human breast cancer cells to the microenvironment. Oncogenesis. 2017; 6 (10): e390-e390. https://doi.org/10.1038/oncsis.2017.88