Discussion (Part 8) - Discussion [Corrected 2].docx

**DISCUSSION** **[8 Comprehensive Discussion of the Study Outcomes]** The findings derived from this pre-clinical investigation suggest that ATRA exhibits considerable promise as a therapeutic agent in the management of gastric-cancer. In fact, our experimental data and hypotheses support the idea...

**DISCUSSION** **[8 Comprehensive Discussion of the Study Outcomes]** The findings derived from this pre-clinical investigation suggest that ATRA exhibits considerable promise as a therapeutic agent in the management of gastric-cancer. In fact, our experimental data and hypotheses support the idea that ATRA-based therapeutic strategies could be effective in at least half of the gastric-cancer cases. It is noteworthy that ATRA-sensitive tumors do not exhibit enrichment for any of the gastric cancer sub-types that have been identified on the basis of the histological characteristics, the gene expression profiles, or the *Lauren*\'s classification. We developed a gene-expression model capable of predicting the potential sensitivity of gastric cancers to ATRA. The model consists of a platform comprising 42 genes whose basal expression levels are directly or inversely correlated with ATRA-sensitivity. This platform offers insights into the properties of gastric malignancies that exhibit a response to the retinoid. *WNT2* stands out among the genes that are directly associated with ATRA-sensitivity. In gastric-cancer cells^1,2^, high expression levels of *WNT2* are associated with an invasive and metastatic phenotype which is the result of a stimulation of the *EMT* (*Epithelial-to-Mesenchimal-Transition*) process^3^. Indeed, up-regulation of the *WNT2* gene in stomach tumors is a negative prognostic factor. Thus, activation of *EMT* and *WNT2* may contribute to the responsiveness of gastric cancer cells to ATRA. With respect to this, we observed a direct correlation between *LOXL1* (Lysyl-Oxidase-Like-1) levels and ATRA-sensitivity, which supports the hypothesis that *EMT* plays a role in determining sensitivity to the retinoid. In fact, *LOXL1* promotes the *EMT*-dependent dissemination of gastric-cancer cells in the peritoneum^4^ and it is over-expressed in these cells. **[8.1 ATRA-Induced IFN-Dependent Immune Responses and Viral Mimicry in Gastric Cancer]** A significant finding of our research relates to the fact that ATRA enhances IFN-dependent immune responses and boosts antigen presentation in gastric cancer cell lines. This finding is analogous to what has been reported by our group in breast-cancer^5^. The immune-modulatory effects triggered by ATRA in breast-cancer are linked to a \"viral mimicry\" response. This response is caused by an increased production of RNAs from endogenous retroviruses. The list of retroviral families considered for retrotransposon quantification includes the elements and the corresponding annotations presented in **Table 6**. The most interesting outcome of our research is the discovery of a limited set of genes that are induced (IRF1, CTSS, PSMB10, CYP26B1, DHRS3, and TINAGL1) and repressed (AHNAK2) by ATRA in all the retinoid-sensitive gastric-cancer cell-lines examined. Notably, ATRA directly or indirectly regulates the transcription of IRF1, CTSS, and PSMB10, which encode proteins that interact with each other to control immunity^6--8^. The IRF1 gene is expected to exert a significant impact on the immune responses and IFN-dependent reactions triggered by ATRA in gastric-cancer. Furthermore, CYP26B1 and DHRS3 up-regulation suggests that ATRA controls the metabolism of endogenous vitamin-A in gastric cancer cells. Indeed, CYP26B1 codes for a protein that promotes the oxidation of ATRA, converting it into an inactive oxo-derivative^9^. On the other hand, DHRS3 codes for a NADPH-dependent enzyme, which reduces retinaldehyde into retinol^10^ ^11^ ^12^. From a functional perspective, up-regulation of the *TINAGL1* gene is intriguing because of the role played by the corresponding protein in suppressing the development and spread of triple-negative breast-cancer. These TINAGL1-dependent effects are achieved *via* suppression of the FAK and EGFR signaling pathways^13^. In sensitive gastric-cancer cells, ATRA causes down-regulation of a single gene, *i.e. AHNAK2*. *AHNAK2* encodes a carcinogenic nucleoprotein^14^ involved in calcium signaling^15^. In addition, *AHNAK2* expression plays a role in gastric-cancer resistance to chemotherapy^16^. Therefore, the ATRA-dependent decrease in AHNAK2 levels may enhance the clinical effectiveness of certain chemotherapeutics. Nevertheless, during an exploratory bioinformatics investigation, we determined that ATRA-dependent immune-modulation in gastric-cancer cells does not involve a \"*viral-mimicry*\" response. This is because no changes have been noticed in the total expression of the many endogenous retroviral mRNAs identified (**Fig. 48**). Indeed, our preliminary results indicate that the molecular pathways responsible for the immune-modulatory effects of ATRA in gastric- and breast-cancer are different. The immunological responses triggered by ATRA in gastric-cancer have significant implications from a translational and therapeutic perspective, as they suggest that the retinoid is likely to enhance the tumor\'s susceptibility to immune-modulatory drugs, such as immune-checkpoint inhibitors ^17,18^. **[8.2 Role of IRF1 and DHRS3 in the growth inhibitory action of ATRA:]** Notwithstanding the fact that functional investigations are required for all of the aforementioned genes, we focused our first studies on IRF1 and DHRS3. Given the likely significance of IRF1 in the inflammatory/immunological/IFN-dependent responses^7,19^ and DHRS3 (*Short-Chain-Dehydrogenase/Reductase-Family-16C-Member-1* or *Retinol-Dehydrogenase-17*) in retinoid metabolism^20^, we investigated the relevance of these two genes for ATRA therapeutic effects. In a first set of studies, we carried out preliminary bioinformatic analyses to obtain evidence on the role played by IRF1 and DHRS3 in gastric-cancer. Gene expression analysis revealed that both IRF1 and DHRS3 (**Fig. 49/Fig. 50**) show higher expression levels in ATRA-treated cell lines compared to those treated with DMSO, suggesting that ATRA may stimulate the expression of these genes in gastric-cancer.  This increased expression may reflect the role of ATRA in modulating signaling pathways involving IRF1 and DHRS3, which implies their potential cooperation in mediating the antiproliferative effects of ATRA on gastric-cancer cells. To further define possible relationships between IRF1 and DHRS3, we examined the correlation between their expression levels through computational analysis. Calculation of the Pearson correlation coefficient highlighted a positive association between the expression of IRF1 and DHRS3 in all cell lines examined (**Fig. 51**). This result suggests that the expression of IRF1 and DHRS3 is coordinated, potentially reflecting a joint regulation or functional interaction between these genes in the context of gastric-cancer.  The positive correlation between IRF1 and DHRS3 expression indicates that these genes are involved in a common signaling pathway or they are regulated by similar transcriptional factors in the context of ATRA treatment. Furthermore, this correlation highlights the possibility that IRF1 and DHRS3 participate together in the mechanisms through which ATRA exerts its antitumor effects, including the modulation of the immune response and inflammation, as well as the control of retinoid metabolism. To validate the computational results obtained, we conducted functional studies on IRF1 and DHRS3 in HGC-27 cells. In an initial round of experiments, two IRF1-targeting siRNAs (*siIRF1a* and *siIRF1b*) and a control siRNA (*siCTRL*) were delivered into HGC-27 cells using a transient transfection approach. Transfected cells were exposed to vehicle (DMSO) or ATRA for 48 hours. In line with the up-regulation of the corresponding mRNA, ATRA enhances the baseline levels of the IRF1 protein in mock transfected (*no siRNA*) and *siCTRL* transfected HGC-27 cells, as evidenced by the Western-blots presented in **Fig. 52**. In contrast, the presence of the IRF1 protein cannot be detected in cells transfected with *siIRF1a* and *siIRF1b*, regardless of whether they were treated with vehicle or ATRA. Since ATRA modulates both *IRF1* and *DHRS3* mRNAs, the amounts of the *DHRS3* protein were also measured. As shown in **Fig. 52**, ATRA increases *DHRS3* protein levels in both native and *siCTRL*-transfected cells. The two siRNAs, *siIRF1a* and *siIRF1b*, decrease the basal levels of the *DHRS3* protein. Notably, *siIRF1a* and *siIRF1b* transfection suppresses retinoid-dependent *DHRS3* induction. Therefore, *IRF1* up-regulation may be responsible for the ATRA-induced increase in *DHRS3* which is observed in gastric-cancer cells. As to the functional significance of *IRF1* in ATRA-dependent effects on gastric-cancer, *siIRF1a* and siIRF1b decrease the anti-proliferative effects exerted by ATRA in both native and *siCTRL*-transfected HGC-27 cells (**Fig. 53**).  These data suggest that *IRF1* up-regulation contributes to the growth-inhibitory effect exerted by ATRA in HGC-27 cells. Overall, the preliminary findings from the knock-down experiments indicate that while *IRF1* up-regulation is necessary for the anti-proliferative effect of ATRA in gastric cancer cells. In a second set of experiments, the involvement of *DHRS3* in ATRA\'s growth-inhibitory action was investigated using an shRNA approach. HGC-27 cells were infected with lentiviral particles containing DHRS3-targeting shRNA constructs (*shDHRS3a* and *shDHRS3b*) as well as *pGR*/*shCTRL2* controls. This procedure resulted in the generation of distinct, puromycin-resistant cell populations, which were subsequently treated with vehicle or ATRA for 48 hours. In vehicle-treated *pGR* and *shCTRL2* cells, significant amounts of *DHRS3* were synthesized, whereas the protein was undetectable in *shDHRS3a* and *shDHRS3b* cells (**Fig. 54**). As predicted, ATRA increases the levels of *DHRS3* in pGR- and shCTRL2-infected cells, whereas the protein remains undetectable in retinoid-treated *shDHRS3a*- and *shDHRS3b*-infected cells. Unexpectedly, *DHRS3* silencing appears to inhibit the ATRA-dependent induction of *IRF1*, suggesting the presence of a positive feedback loop. To clarify the consequences of *DHRS3 silencing*, *pGR*-, shCTRL2-, *shDHRS3a*-, and *shDHRS3b*-infected cells were treated with vehicle or ATRA (*0.1µM/1.0µM*) for 3, 6, and 9 days. At all these time points, both concentrations of ATRA resulted in reduced growth of the *pGR*- and *shCTRL*-infected cells (**Fig. 55**).  By day 9, the ATRA-dependent anti-proliferative effect observed in *pGR*- and *shCTRL*-infected cells is decreased in the *shDHRS3a*- and *shDHRS3b*-infected counterparts. These findings indicate that *DHRS3* up-regulation may also be involved in the growth-inhibitory effects exerted by ATRA in gastric-cancer cells. However, it is important to note that the RNA-sequencing data indicate that ATRA induces *DHRS3* expression in all the gastric-cancer cell-lines examined, regardless of their relative sensitivity or resistance to the retinoid. This suggests that *DHRS3* is likely to require additional gene products, factors, or pathways, which are selectively stimulated or repressed by the retinoid only in sensitive gastric-cancer cells, to mediate the anti-proliferative action of ATRA. Our findings suggest that IRF1 and, to a lower extent, DHRS3 influence the growth-inhibitory action of ATRA in gastric-cancer cells characterized by sensitivity to the retinoid. Indeed, silencing of the two genes hinders and decreases the anti-proliferative effect of ATRA in representative cell-lines. The influence of IRF1 on gastric-cancer cells seems to be restricted to ATRA-sensitive cell lines, as indicated by our data. In conclusion, our results indicate that drugs targeting the two proteins may be used to develop new and effective strategies for the personalized treatment of gastric-cancer patients. **[9 Future Perspectives]** Future studies should focus on investigating the molecular mechanisms underlying the therapeutic action of ATRA in gastric-cancer. Given the identification of a number of genes controlled by ATRA, it is crucial to further investigate the functional roles exerted by these genes in the anti-proliferative effects of the retinoid in gastric-cancer cells. This study has the potential to uncover novel therapeutic targets, making a substantial contribution to the development of more accurate and efficient cancer therapies. One of the main results obtained support the relevance of IRF1 and DHRS3 up-regulation in the anti-proliferative effects exerted by ATRA in gastric-cancer. For this reason, future research will focus on comprehending and identifying the practical importance of these genes in the treatment of stomach cancer. To obtain further information on the Viral Mimicry responses activated by the retinoid in gastric-cancer cells, further studies are necessary to identify the individual retroviral DNAs that may be altered or modified by ATRA. Efforts should also focus on the clarification of the specific mechanisms underlying ATRA immune-modulatory activities in various forms of cancer. Specifically, research should focus on comprehending the function of ATRA in regulating the tumour microenvironment (TME)^21^, which is a crucial determinant of all the cancer associated immune responses. A number of experimental studies support the potential of ATRA in the control of the tumor microenvironment (TME), as the retinoid has been reported to influence the function of Myeloid-Derived Suppressor Cells (MDSCs)^22^. Indeed MDSCs are known to control tumor progression and treatment effectiveness. These observations provide the foundation for the development of strategies exploiting ATRA\'s ability to modify the dynamics of the tumour microenvironment (TME^23^) in combination with existing and new immune checkpoint inhibitors or chemotherapeutics. This approach has the potential to not only increase the efficacy of current medicines but also to develop effective combination therapies. The complementary use of tumor-on-chip technology and 3D cell cultures is likely to enhance our knowledge on how ATRA modulates the tumor microenvironment (TME26). We envisage that this approach will provide a more detailed understanding of the therapeutic effects of ATRA on different subtypes of gastric-cancer. Another important and future goal is the development of a Single Cell Atlas with the use of meticulous bioinformatic/computational analyses and patients' datasets available in the scientific literature. This computational approach will include re-examining and merging both untreated and ATRA-treated data to confirm the ATRA sensitivity profile that we have established in our findings. The objective is to detect distinct clusters linked to ATRA-sensitivity and facilitate further investigations. Building upon this basis, the additional aim is to enlist patient samples for internal Single Cell analysis, facilitating further validations. By adopting this strategy, we may enhance our comprehension of ATRA\'s molecular effects on cells and gain a more detailed understanding of its therapeutic potential. By incorporating extensive genomic data from various patient groups, it is possible to improve the accuracy of our models predicting ATRA sensitivity, which may eventually result in more focused and efficient treatment approaches in gastric-cancer. **CONCLUSION** This study contributes to the topic of gastric-cancer therapy by emphasizing the critical role of ATRA in facilitating the development of individualized therapeutic strategies. Through an in-depth exploration of the molecular dynamics of gastric-cancer, new insights have been discovered that highlight the intriguing function of ATRA. Our thorough pre-clinical research has yielded data that strongly support the use of ATRA in the stratified treatment of gastric-cancer. This approach takes into account the complexity and heterogeneity of the disease. Furthermore, the present study provides evidence that supports ATRA\'s therapeutic efficacy in the treatment of gastric-cancer. Within this specific framework, the gene-expression model generated will enable the creation of a predictive clinical tool for the identification of patients who may take advantages from ATRA-based therapy approaches. Our results provide support to the design and organization of clinical studies aimed at investigating the utility of ATRA in the treatment of this tumor type. Specifically, the potent immuno-regulatory responses triggered by the retinoid indicate that combining ATRA with immune checkpoint inhibitors is a rational therapeutic approach in the context of gastric-cancer. Within the broader scope of our work, the discovery of specific genes and the clarification of their functions in facilitating ATRA\'s effects provide a guiding light for future research endeavors. ATRA\'s immune-modulatory potential, together with its impact on the tumour microenvironment, demonstrates its importance as a therapeutic agent. In addition, our preliminary results on the connection between ATRA and different biological pathways provides new opportunities for combining ATRA with other therapeutic agents, including immune checkpoint inhibitors, to improve the therapeutic efficacy in gastric-cancer. This work provides a detailed understanding of how ATRA affects gastric-cancer at the genetic and cellular levels. It confirms that ATRA has the potential to be an essential tool against gastric-cancer and paves the way to future clinical-studies. These investigations, guided by our predictive gene-expression model, seek to thoroughly assess the sensitivity of ATRA in tailored treatment plans, with the ultimate goal of enhancing patient opportunities in the fight against this challenging type of cancer. **[BIBLIOGRAPHY:]** 1\. Lin, L. *et al.* Overexpression of IL-8 and Wnt2 is associated with prognosis of gastric cancer. *Folia Histochem. Cytobiol.* **60**, 66--73 (2022).2. Zhang, Z., Wang, J. & Dong, X. Wnt2 contributes to the progression of gastric cancer by promoting cell migration and invasion. *Oncol. Lett.* (2018) doi:10.3892/ol.2018.9050.3. Katoh, M. Epithelial-mesenchymal transition in gastric cancer (Review). *Int. J. Oncol.* **27**, 1677--1683 (2005).4. Hu, Q. *et al.* Potential association of LOXL1 with peritoneal dissemination in gastric cancer possibly via promotion of EMT. *PloS One* **15**, e0241140 (2020).5. Bolis, M. *et al.* All-Trans Retinoic Acid Stimulates Viral Mimicry, Interferon Responses and Antigen Presentation in Breast-Cancer Cells. *Cancers* **12**, 1169 (2020).6. Wang, Y. *et al.* IP-score correlated to endogenous tumour antigen peptide processing: A candidate clinical response score algorithm of immune checkpoint inhibitors therapy in multiple cohorts. *Front. Immunol.* **13**, 1085491 (2022).7. Zhou, H., Tang, Y.-D. & Zheng, C. Revisiting IRF1-mediated antiviral innate immunity. *Cytokine Growth Factor Rev.* **64**, 1--6 (2022).8. de Mingo Pulido, Á. *et al.* Differential Role of Cathepsins S and B In Hepatic APC-Mediated NKT Cell Activation and Cytokine Secretion. *Front. Immunol.* **9**, 391 (2018).9. Isoherranen, N. & Zhong, G. Biochemical and physiological importance of the CYP26 retinoic acid hydroxylases. *Pharmacol. Ther.* **204**, 107400 (2019).10. Gudas, L. J. Retinoid metabolism: new insights. *J. Mol. Endocrinol.* **69**, T37--T49 (2022).11. Billings, S. E. *et al.* The retinaldehyde reductase DHRS3 is essential for preventing the formation of excess retinoic acid during embryonic development. *FASEB J.* **27**, 4877--4889 (2013).12. Adams, M. K., Belyaeva, O. V., Wu, L. & Kedishvili, N. Y. The Retinaldehyde Reductase Activity of DHRS3 Is Reciprocally Activated by Retinol Dehydrogenase 10 to Control Retinoid Homeostasis. *J. Biol. Chem.* **289**, 14868--14880 (2014).13. Shen, M. *et al.* Tinagl1 Suppresses Triple-Negative Breast Cancer Progression and Metastasis by Simultaneously Inhibiting Integrin/FAK and EGFR Signaling. *Cancer Cell* **35**, 64-80.e7 (2019).14. Xu, M. *et al.* AHNAK2 is a biomarker and a potential therapeutic target of adenocarcinomas. *Acta Biochim. Biophys. Sin.* **54**, 1708--1719 (2022).15. Komuro, A. *et al.* The AHNAKs are a class of giant propeller-like proteins that associate with calcium channel proteins of cardiomyocytes and other cells. *Proc. Natl. Acad. Sci. U. S. A.* **101**, 4053--4058 (2004).16. Methylation of drug resistance-related genes in chemotherapy-sensitive Epstein-Barr virus-associated gastric cancer - PubMed. https://pubmed.ncbi.nlm.nih.gov/31736281/.17. Yoon, J., Kim, T.-Y. & Oh, D.-Y. Recent Progress in Immunotherapy for Gastric Cancer. *J. Gastric Cancer* **23**, 207 (2023).18. Shen, J. & Wang, Z. Recent advances in the progress of immune checkpoint inhibitors in the treatment of advanced gastric cancer: A review. *Front. Oncol.* **12**, 934249 (2022).19. Feng, H., Zhang, Y.-B., Gui, J.-F., Lemon, S. M. & Yamane, D. Interferon regulatory factor 1 (IRF1) and anti-pathogen innate immune responses. *PLOS Pathog.* **17**, e1009220 (2021).20. Coyle, K. M., Maxwell, S., Thomas, M. L. & Marcato, P. Profiling of the transcriptional response to all-trans retinoic acid in breast cancer cells reveals RARE-independent mechanisms of gene expression. *Sci. Rep.* **7**, 16684 (2017).21. Bi, G. *et al.* The immunomodulatory role of all-trans retinoic acid in tumor microenvironment. *Clin. Exp. Med.* **23**, 591--606 (2023).22. Li, X. *et al.* Targeting Myeloid-Derived Suppressor Cells to Enhance the Antitumor Efficacy of Immune Checkpoint Blockade Therapy. *Front. Immunol.* **12**, 754196 (2021).23. Tiwari, A., Trivedi, R. & Lin, S.-Y. Tumor microenvironment: barrier or opportunity towards effective cancer therapy. *J. Biomed. Sci.* **29**, 83 (2022).