Myocardial Infarction-Induced INSL6 Decrease Contributes to Breast Cancer Progression

Myocardial infarction (MI) induces early-stage breast cancer progression and increases breast cancer patients' mortality and morbidity. Insulin-like peptide 6 (INSL6) overexpression can impede cardiotoxin-induced injury through myofiber regeneration, playing a significant role in MI progression. To investigate the diverse significance of INSL6 in a variety of malignant tumors, we explored INSL6 through MI GEO dataset and multiple omics data integrative analysis, such as gene expression level, enriched pathway analysis, protein-protein interaction (PPI) analysis, and immune subtypes as well as diagnostic value and prognostic value in pancancer. INSL6 expression was downregulated in the MI group, and overall survival analysis demonstrated that INSL6 could be the prognostic biomarkers in the overall survival of breast cancer (BRCA). INSL6 expression differs significantly not only in most cancers but also in different molecular and immune subtypes of cancers. INSL6 might be a potential diagnostic and prognostic biomarker of cancers due to the high accuracy in diagnostic and prognostic value. Furthermore, we focused on BRCA and further investigated INSL6 from the perspective of the correlations with clinical characteristics, prognosis in different clinical subgroups, coexpression genes, and differentially expressed genes (DEGs) and PPI analysis. Overall survival and disease-specific survival analysis of subgroups in BRCA demonstrated that lower INSL6 expression had a worse prognosis. Therefore, INSL6 aberrant expression is associated with the progression and immune cell infiltration of the tumor, especially in KIRP and BRCA. Therefore, INSL6 may serve as a potential prognostic biomarker and the crosstalk between MI and tumor progression.

Currently, coronary artery disease (CAD) is still one of the leading causes of death in patients and contributes to about one in every seven deaths in low-and middleincome countries ( [5], p.1151). Acute myocardial infarction (MI) mortality has increased 5.6-fold in the past 30 years, and obesity has become the major cause of morbidity and mortality in patients with some chronic diseases, for instance, diabetes and CAD ( [6], p. 229). Conservatively estimated, nowadays, 330 million people develop heart diseases in China and an unacceptable burden of recurrent cardiovascular events needs to be solved. Aseptic inflammation can promote neutrophil extracellular traps (NETs) abundant in the liver, thus increasing metastases in patients with breast and colon cancers ( [7], p. 113). Koelwyn et al. reported that MI can epigenetically reprogram Ly6C high monocytes in the bone marrow reservoir to an immunosuppressive phenotype and such monocytes were increasingly recruited to tumors, promoting MI-induced early-stage breast cancer progression and increasing breast cancer patients' mortality and morbidity ([8], p. 1452). Therefore, the crosstalk between MI and tumor progression should be investigated, which may be potential biomarkers to impede tumor progression, thus reducing mortality and morbidity. However, the researches on such crosstalk are little.
In this study, the differentially expressed genes (DEGs) were investigated in MI GEO datasets. Venn diagram was used to obtain the crosstalk between first acute myocardial infarction (FAMI) A, FAMI B, and breast invasive carcinoma (BRCA) and cox regression analysis was used to evaluate the association between the screened DEG expression and BRCA overall survival using Xiantao website. Compared to control, INSL6 expression was downregulated in FAMI A and FAMI B groups, which may impede the inhibitions on tumor progression. Therefore, we examined the INSL6 expression and the diagnostic and prognostic value in pancancer. DEGs between INSL6 high-and low-expression groups in BRCA were also explored to validate whether INSL6 can be the crosstalk between MI and BRCA.

Microarray Data and Data
Processing. Using the keywords "myocardial infarction" in "Homo sapiens," GSE24519 from the Gene Expression Omnibus (GEO) database was investigated, processed with log2 transformation for normalization and analyzed using GEO2R ( [9], p. 546). There were 17 patients affected by their very first acute myocardial infarction (FAMI), without any sign of previous cardiovascular sufferance, and 4 controls in GSE24519. Platelets from patient with acute MI within 6 hours of the onset of symptoms were collected, and the blood samples at two time points were named FAMI A and FAMI B. The RNA sequencing was based on the GE Healthcare/Amersham Biosciences CodeLink Human Whole Genome Bioarray, The Cancer Genome Atlas (TCGA) database, and the Genotype-Tissue Expression (GTEx) database by UCSC XENA. The BRCA and other cancer data were all from TCGA (https://portal.gdc.cancer.gov/). The data were downloaded and analyzed using Xiantao website tool (http://www.xiantao.love). A log 2jfold change ðFCÞj > 1 and a P value < 0.05 were regarded as the cut-off criteria. The workflow of processing the datasets is shown in Figure 1.

Cox Regression Analysis and Kaplan-Meier Survival
Analysis. Using Venn diagram, DEGs were obtained between FAMI A, FAMI B, and BRCA and cox regression analysis was used to evaluate the association between the screened DEG expression and BRCA overall survival using Xiantao website. A P value < 0.05 was regarded as the cut-off criteria.     GSM604560  GSM604562  GSM604564  GSM604566  GSM604568  GSM604570  GSM604572  GSM604574  GSM604576  GSM604578  GSM604580  GSM604582  GSM604584  GSM604586  GSM604588  GSM604590  GSM604592  GSM604672  GSM604673  GSM604674  GSM604675   -   To validate the potential effects of INSL6 expressions on BRCA progression, the INSL6 expressions in subgroups were determined and overall survival analysis of subgroups was also carried out. The RNA-seq data and related clinical data in level 3 HTSeq-fragments per kilobase per million (FPKM) format were downloaded from TCGA database, converted to transcripts per million (TPM) read format, and then analyzed after log2 transformation. A P value < 0.05 was regarded as the cut-off criteria.
2.9. Coexpression Gene Analysis of INSL6 in BRCA. Top 50 coexpression genes positively and negatively related to INSL6 in BRCA were explored. GO/KEGG pathway analysis was used to investigate the enriched pathways of the top coexpression genes. A P value < 0.05 was regarded as the cut-off criteria.    Disease Markers expression group: 50-100%) in BRCA were analyzed using the deseq2 package. Utilizing Limma, a log 2jFCj > 1 and a P value < 0.05 were applied as the cut-off criteria. Then, GO/KEGG pathway analyses, as well as gene set enrichment analysis (GSEA), were applied utilizing the "clusterProfiler" package in R. PPI network analysis was used to obtain the hub genes utilizing the Cytoscape plug-in (MCODE and MCC).  Table S1).

Results
To investigate which DEG can be a prognostic biomarker, overall survival analysis was utilized and INSL6 (hazard ratio ðHRÞ = 0:64, P = 0:007) and ODAM (HR = 0:70, P = 0:031) could be the prognostic biomarkers in overall survival BRCA (Figure 2(f); Figure S1). Compared to control, INSL6 expression was downregulated in the FAMI A and FAMI B group, which may impede the inhibitions on tumor progression.

PPI Network and GO/KEGG Enrichment Analysis.
To investigate INSL6 and its protein interactions, the nodes with a combined score > 0:4 were analyzed using STRING and Cytoscape (Figure 3(c)). The list of 50 targeting binding proteins was uploaded into the Xiantao webpage for functional analysis, which were involved in hormone activity, receptor ligand activity, relaxin signaling pathway, and neuroactive ligand-receptor interaction (Figures 3(d) and 3(e); Table 1).         Table S2). After log2 transformation There were significant differences of INSL6 expression in TCGA-BRCA patients' baseline characters, such as age, histological type, and menopause status. Overall survival analysis of subgroups in BRCA was also carried out, which demonstrated that lower INSL6 expression had a worse prognosis in white patients as well as patients with N0, M0, negative ER status, infiltrating ductal carcinoma, or LumB ( Figure 9). Disease-specific survival analysis of subgroups in BRCA demonstrated that lower INSL6 expression had a worse prognosis in patients with M0 or Her2 positive ( Figure S7).

Coexpression Gene Analysis of INSL6 in BRCA.
Top 50 coexpression genes positively related to INSL6 in BRCA were explored, which were mainly involved in detection of chemical stimulus involved in sensory perception of bitter taste, GTPase activity, and purine ribonucleoside binding ( Figure 10; Table 2). Top 50 coexpression genes negatively related to INSL6 in BRCA were also investigated, which were mainly involved in integral component of endoplasmic reticulum membrane, intrinsic component of endo-plasmic reticulum membrane, proton-transporting ATP synthase activity, rotational mechanism, and phosphatase activator activity ( Figure 11; Table 3).

DEGs between INSL6
High-and Low-Expression Groups in BRCA. Using the deseq2 package, 3833 upregulated DEGs and 105 downregulated DEGs were obtained and further GO/KEGG pathway analysis was applied, which were mainly involved in mRNA 5 ′ -splice site recognition, mRNA splice site selection, DNA packaging complex, pre-mRNA binding, and RNA transport (Figures 12(a), 12(b), and 12(d); Table 4). GSEA of DEGs between INSL6 high-and low-expression groups in BRCA was explored, which were mainly enriched in REACTOME_NEURONAL_SYSTEM, REACTOME_G_ALPHA_S_SIGNALLING_EVENTS, REACTOME_OLFACTORY_SIGNALING_PATHWAY, and REACTOME_INNATE_IMMUNE_SYSTEM (Figures 12(c) and 12(e); Table 5).
Furthermore, PPI network analysis was used to obtain the hub genes utilizing the Cytoscape plug-in (MCODE and MCC). There were 4 modules in the network, including CHGB, SST, HIST1H2BB, CSN2, and DSPP ( Figure 13).          INSL6 methylation in BRCA was evaluated to investigate the association between INSL6 and methylation site, which demonstrated that INSL6 expression was correlated to cg07531356, cg26034799, cg13504907, and cg11830061 in BRCA. Of the different loci tested, Hs_INSL6_03 was identified to contain tissue-specific differential methylation ( [29], p. 3079). The specific CG sites adjacent to the CGI of the INSL6 promoter could confer DNA methylation spreading into the CGI, particularly in the setting of KRAB-factor binding ( [30], p. 7257). In addition, cisregulatory elements of INSL6 expression facilitate preferential methylation at these promoter CpG islands ( [31], p. 2023).

Disease Markers
In the next step, we will consider the use of the machine learning and deep learning methods in the research of INSL6, such as predicting lncRNA-miRNA-INSL6 or circRNA-miRNA-INSL6 interactions in MI and BRCA progression ( [36], p. 874; [37], p. 535; [38], p. bbab286) and using inferring gene regulatory networks to investigate the INSL6 effects ( [39], p. 168). Further researches about INSL6 in the crosstalk are still needed. The crosstalk between MI and tumor progression should be investigated, which may be potential biomarkers to impede tumor progression, thus reducing mortality and morbidity.

Conclusions
INSL6 is differentially expressed after myocardial infarction and also in a variety of tumors and aberrant expression is associated with the progression and immune cell infiltration of the tumor, especially in KIRP and BRCA. Therefore, INSL6 may serve as a potential prognostic biomarker and the crosstalk between MI and tumor progression.