Identification of Potential Key Genes and Pathways in Early-Onset Colorectal Cancer Through Bioinformatics Analysis

Abstract

This study was designed to identify the potential key protein interaction networks, genes, and correlated pathways in early-onset colorectal cancer (CRC) via bioinformatics methods. We selected microarray data GSE4107 consisting 12 patient’s colonic mucosa and 10 healthy control mucosa; initially, the GSE4107 were downloaded and analyzed using limma package to identify differentially expressed genes (DEGs). A total of 131 DEGs consisting of 108 upregulated genes and 23 downregulated genes of patients in early-onset CRC were selected by the criteria of adjusted P values <.01 and |log2 fold change (FC)| ≥ 2. The gene ontology functional enrichment analysis and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were accomplished to view the biological process, cellular components, molecular function, and the KEGG pathways of DEGs. Finally, protein–protein interactions (PPIs) were constructed, and the hub protein module was identified. Genes such as ACTA2, ACTG2, MYH11, CALD1, MYL9, TPM2, and LMOD1 were strongly implicated in CRC. In summary, in this study, we indicated that molecular mechanisms were involved in muscle contraction and vascular smooth muscle contraction signaling pathway, which improve our understanding of CRC and could be used as new therapeutic targets for CRC.

Keywords

colorectal cancer bioinformatics analysis differentially expressed genes

Introduction

Colorectal cancer (CRC) is one of the most common malignant diseases in the world, and its incidences increased with age. According to estimates, more than 777 000 of new cases with CRC were registered in 2015 in the developed countries,^1,2 there were about 376 000 of new CRC cases and 191 000 of death were reported in 2015 in China.³ Most CRC were related to old age and lifestyle factors, with only a fraction of cases caused by underlying genetic disorders.^4,5 Although numerous efforts has been taken to understand the genetic mechanism for initiation and progression of CRC, it remains a major challenge for researchers to prevent and treat early-onset CRC. Therefore, it is important and urgent to uncover the mechanisms of early-onset CRC and develop novel therapeutic routes.

Gene chip or expression profile is a gene-level detection technique that has been applied to scientific research during 2000. Using gene chips, integrated bioinformatic knowledge makes it possible to detect the expression of the entire genome within the same sample in a single experiment, which is particularly suitable to screen differentially expressed genes.^6,7 With the application of the gene chips, a large amount series of correlated CRC slice data have been produced, archived, and deposited in public databases. Reanalyzing and reintegrating those data sets may find some meaningful clues for new research. A series of microarray data sets have been carried out on CRC in recent years,^8,9 and a large number of differentially expressed genes (DEGs) have been obtained, which are involved in different pathways, biological process, cellular components, or molecular functions.

In this study, we downloaded the original raw data set (GSE4107) from the website of Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/), which is a public database for archiving and querying microarray data. Gene expression profiles of patients with CRC were compared to those in normal healthy control to identify the DEGs. Subsequently, the DEGs were screened using Rstudio software installed Limma packages^10,11; then gene ontology (GO) and pathway enrichment analysis were performed on the online website DAVID (https://david.%20ncifcrf.gov).¹² Through analyzing their biological functions and pathways, we may sketch out the outline of CRC development at molecular level and identify the potential candidate genes for diagnosis, prognosis, and therapeutic targets.

Materials and Methods

Microarray Data

Microarray data GSE4107¹³ were downloaded from the National Center Biotechnology Information Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/) database,^14,15 which was executed with help of GPL570 Human Genome U133 Plus 2.0 Array. GSE4107 contains 12 patients and 10 healthy control (average age: 50 or less, ethnicity: Chinese).

Data Preprocessing

After GSE4107 was downloaded, probe identification numbers were transformed into gene symbols. For multiple probes corresponding to one gene, the significant expression value was taken as the gene expression value. After that, gene expression values were normalized using the Affy package.¹⁶

Identification of DEGs

The raw data GSE4107 files used for analysis included the .CEL format files (Affymetrix). The analysis was processed and executed by Rstudio software and limma package.^17,18 We applied linear models for the assessment of differential expression and the analysis of designed experiments.^18,19 Limma package in Rstudio was applied to identify the DEGs between early-onset CRC samples and healthy control samples. Genes with |log2 fold change (FC)| ≥ 2 and adjusted P values <.01 as the cutoff criterion were selected for subsequent analysis.

Gene Ontology and Pathway Enrichment Analysis

The GO analysis is a useful method for annotating gene and gene product²⁰ and identifying characteristic biological meaning of genome and transcriptome.^21,22 The KEGG is a systematic analysis database of gene function, linking genomic information with high-level functional information.²³ Candidated DEGs functional-level enrichment were analyzed through multiple online tools. DAVID, among them, is an online website with gene annotation, visualization, and providing gene attributes. P < .05 as the cutoff criterion was considered statistically significant.^22,24,25

Integration of Protein–Protein Interaction Network and Module Analysis

First, Search Tool for the Retrieval of Interacting Genes (STRING)²⁶ database was used to demonstrate DEG-encoded proteins and protein–protein interaction (PPI) information. Second, to evaluate the interactive relationships among DEGs, we mapped the DEGs to STRING, and minimum required interaction score >0.400 (medium confidence) was selected as significant threshold. Then, PPI networks were constructed using the Cytoscape software.²⁷ The plug-in Molecular Complex Detection (MCODE), a well-known automated method to find highly interconnected subgraphs as molecular complexes or clusters in large PPI networks, was used to screen the modules or clusters of PPI network in Cytoscape. The MCODE parameters criteria were set by default, except K-core = 6. Moreover, the functional enrichment analysis was performed for DEGs in the modules with P < .05 as the cutoff criterion.

Results

Identification of DEGs

In this study, we included 12 patients with CRC and 10 healthy controls for the analysis. GSE4107 was analyzed using Rstudio software and following identifies the DEG sets. Using adjusted P values <.01 and |log2 fold change (FC)| ≥ 2 criteria, a total of 131 genes were identified; among them, 108 were upregulated and the other 23 were downregulated (Table 1).

Table 1.

131 Differentially Expressed Genes (DEGs) Were Identified From GSE4107, Including 108 Upregulated Genes and 23 Downregulated Genes in the Patients With Early-Onset Colorectal Cancer, Compared to Healthy Control.^a

DEGs	Gene Name
Upregulated	CYR61, FOS, EGR1, DUSP1, ADAMTS1, VIP, CTGF, MGP, SRPX, LINC01279, FOSB, CCDC80, TSC22D3, RSPO3, RGS1, RHOB, DPT, UCHL1, IGFBP5, ADH1B, FBLN1, PMP22, ACTA2, NR4A1, SFRP2, OGN, PLN, PTGIS, CHRDL1, C11orf96, GEM, TIMP3, TAGLN, GREM1, C7, SCGN, DCN, TNS1, ANK2, FILIP1L, ASPN, PTRF, GAL, FABP4, AOC3, FAM129A, RTN1, FHL1, SYNPO2, TUBA1A, CRYAB, RERGL, MFAP5, MAP1B, MYH11, GUCY1B3, CXCL12, CLU, AKAP12, TGFB1I1, APOLD1, MYL9, DDR2, SELM, MLLT11, RGS2, ATF3, CAV1, LMOD1, TMEM47, TUBB6, ACTG2, PEG3, COL12A1, ZAK, EBF1, TPM2, MAMDC2, SYNM, SCG2, C2orf40, FERMT2, SOCS3, SDPR, SIK1, MSRB3, SCN7A, GUCY1A3, SORBS1, PCDH7, ATP1A2, CNN1, CXCR4, MYOCD, CALD1, FLNA, GAS1, CDH19, HSPB6, REEP1, GPM6B, NEXN, KCNMB1, FN1, MEIS2, PGM5, PI15, HSPB8
Downregulated	NR1H4, SLC51A, SCIN, NETO2, ETNK1, SLC38A4, HOOK1, GBA3, MEP1B, ACOX1, METTL7B, NQO1, VAV3, CWH43, AKR1C3, DSC2, BCL2L15, EHF, LRRC31, LIMA1, MGST1, UGT2A3, ACE2

DEGs

Gene Name

Upregulated

CYR61, FOS, EGR1, DUSP1, ADAMTS1, VIP, CTGF, MGP, SRPX, LINC01279, FOSB, CCDC80, TSC22D3, RSPO3, RGS1, RHOB, DPT, UCHL1, IGFBP5, ADH1B, FBLN1, PMP22, ACTA2, NR4A1, SFRP2, OGN, PLN, PTGIS, CHRDL1, C11orf96, GEM, TIMP3, TAGLN, GREM1, C7, SCGN, DCN, TNS1, ANK2, FILIP1L, ASPN, PTRF, GAL, FABP4, AOC3, FAM129A, RTN1, FHL1, SYNPO2, TUBA1A, CRYAB, RERGL, MFAP5, MAP1B, MYH11, GUCY1B3, CXCL12, CLU, AKAP12, TGFB1I1, APOLD1, MYL9, DDR2, SELM, MLLT11, RGS2, ATF3, CAV1, LMOD1, TMEM47, TUBB6, ACTG2, PEG3, COL12A1, ZAK, EBF1, TPM2, MAMDC2, SYNM, SCG2, C2orf40, FERMT2, SOCS3, SDPR, SIK1, MSRB3, SCN7A, GUCY1A3, SORBS1, PCDH7, ATP1A2, CNN1, CXCR4, MYOCD, CALD1, FLNA, GAS1, CDH19, HSPB6, REEP1, GPM6B, NEXN, KCNMB1, FN1, MEIS2, PGM5, PI15, HSPB8

Downregulated

NR1H4, SLC51A, SCIN, NETO2, ETNK1, SLC38A4, HOOK1, GBA3, MEP1B, ACOX1, METTL7B, NQO1, VAV3, CWH43, AKR1C3, DSC2, BCL2L15, EHF, LRRC31, LIMA1, MGST1, UGT2A3, ACE2

^aThe upregulated genes were listed from the largest to the smallest of fold changes, and downregulated genes were listed from the smallest to largest.

Gene Term Enrichment Analysis

We uploaded DEGs to the online website DAVID to identify GO Terms and KEGG pathways and classified them into 3 functional categories: biological process (BP), cellular component (CC), and molecular function (MF; Figure 1A). As shown in Figure 1B and Table 2, GO analysis showed that the DEGs were most significantly enriched in muscle contraction and regulation of muscle contraction. Moreover, the upregulated DEGs were significantly enriched in biological process, including muscle system process, muscle contraction, and regulation of muscle contraction (Figure 1B and Table 2); the downregulated DEGs were enriched in organic acid transport, lipid metabolic process, and cellular lipid metabolic process (Figure 1B and Table 2).

Figure 1.

Gene ontology analysis and significant enrichment of differentially expressed genes (DEGs) in early-onset colorectal cancer (CRC). (A) Gene ontology (GO) analysis classified DEGs into BP, CC, and MF group. (B) Ranking significant enriched GO terms of DEGs.

Table 2.

The Gene Ontology Analysis of DEGs Associated With Early-Onset Colorectal Cancer.

Category	Term	Count	P Value
Upregulated DEGs
GOTERM_BP_FAT	GO:0003012∼muscle system process	23	2.84E-15
GOTERM_BP_FAT	GO:0006936∼muscle contraction	21	8.11E-15
GOTERM_BP_FAT	GO:0006937∼regulation of muscle contraction	12	1.87E-09
GOTERM_CC_FAT	GO:0043292∼contractile fiber	14	1.97E-09
GOTERM_BP_FAT	GO:0090257∼regulation of muscle system process	13	2.71E-09
GOTERM_CC_FAT	GO:0031012∼extracellular matrix	19	6.75E-09
GOTERM_CC_FAT	GO:0044449∼contractile fiber part	13	7.83E-09
GOTERM_BP_FAT	GO:0061061∼muscle structure development	19	1.41E-08
GOTERM_BP_FAT	GO:0001933∼negative regulation of protein phosphorylation	16	1.41E-08
GOTERM_BP_FAT	GO:0072358∼cardiovascular system development	23	3.91E-08
Downregulated DEGs
GOTERM_BP_FAT	GO:0015849∼organic acid transport	4	4.75E-04
GOTERM_BP_FAT	GO:0006629∼lipid metabolic process	7	4.90E-03
GOTERM_BP_FAT	GO:0044255∼cellular lipid metabolic process	6	8.37E-03
GOTERM_BP_FAT	GO:0031667∼response to nutrient levels	4	1.36E-02
GOTERM_BP_FAT	GO:2000188∼regulation of cholesterol homeostasis	2	1.63E-02
GOTERM_BP_FAT	GO:0009991∼response to extracellular stimulus	4	1.66E-02
GOTERM_BP_FAT	GO:0010565∼regulation of cellular ketone metabolic process	3	1.67E-02
GOTERM_BP_FAT	GO:0031669∼cellular response to nutrient levels	3	1.69E-02
GOTERM_BP_FAT	GO:0007584∼response to nutrient	3	1.94E-02
GOTERM_BP_FAT	GO:0016137∼glycoside metabolic process	2	2.00E-02

Abbreviation: DEG, differentially expressed gene.

Kyoto Encyclopedia of Genes and Genomes Pathway Analysis

We used online website DAVID to perform DEG functional and signaling pathway enrichment analysis. Figure 2 shows the most significantly enriched pathways of DEGs, and Table 3 lists the significantly enriched pathways of the upregulated DEGs, while there are no available significantly enriched pathways of the downregulated DEGs (Table 3). The significant signal pathway of the (upregulated) DEGs mainly enriched in vascular smooth muscle contraction and cGMP-PKG signaling pathway.

Figure 2.

Significantly enriched signal pathway of differentially expressed genes (DEGs) in early-onset colorectal cancer (CRC).

Table 3.

Signaling Pathway Enrichment Analysis of DEGs Associated With Early-Onset Colorectal Cancer.

Pathway	Term	Count	P Value	Genes
Upregulated DEGs
KEGG_Pathway: hsa04270	Vascular smooth muscle contraction	7	8.33E-05	ACTA2, MYL9, ACTG2, GUCY1B3, KCNMB1, CALD1, GUCY1A3
KEGG_Pathway: hsa04022	cGMP-PKG signaling pathway	7	5.12E-04	MYL9, RGS2, ATP1A2, GUCY1B3, PLN, KCNMB1, GUCY1A3
KEGG_Pathway: hsa05205:	Proteoglycans in cancer	6	0.007610	FLNA, TIMP3, ANK2, DCN, FN1, CAV1
BIOCARTA: h_ccr5Pathway	Pertussis toxin-insensitive CCR5 Signaling in Macrophage	3	0.012779	FOS, CXCR4, CXCR12,
KEGG_Pathway: hsa04921	Oxytocin signaling pathway	5	0.016252	MYL9, GUCY1B3, FOS, GUCY1A3, RGS2
KEGG_Pathway: hsa04540	Gap junction	4	0.017068	GUCY1B3, TUBA1A, TUBB6, GUCY1A3
Downregulated DEGs
No significant signal pathway(P value <.05) available

Abbreviation: DEG, differentially expressed gene.

Module Analysis, Key Candidate Genes, and Pathway Identification From PPI Network

Based on the STRING online database (http://string-db.org) and Cytoscape software, a total of 131 DEGs (108 upregulated and 23 downregulated genes) were filtered into the DEG PPI network complex, containing 82 nodes and 199 edges (Figure 3A), and 49 genes did not fall into the PPI network. According to the filtering of node degree ≥10 criteria, the top 10 hub genes were ACTA2, ACTG2, FOS, DCN, MYH11, MYL9, EGR1, TPM2, LMOD1, and CALD1. Based on the MCODE, the significant module (10 nodes 42 edges, Figure 3B) from the PPI network was selected, and the functional annotation of the common genes were analyzed (Table 4). Enrichment analysis showed that the genes were mainly associated with vascular smooth muscle contraction signaling pathway.

Figure 3.

Protein–protein interaction (PPI) network of differentially expressed genes (DEGs; Node color: Skyblue indicates up-regulated gene, Red indicates down-regulated gene). (A) Based on the STRING online database, 82 DEGs were filtered into DEGs PPI network. (B) The most significant module from the PPI network.

Table 4.

Pathway Enrichment Analysis of Common Genes Function.

Term	Description	Count	FDR	Genes
GO.0006936	Muscle contraction	6	3.17E-07	ACTG2, CALD1, LMOD1, MYH11, MYL9, TPM2
GO.0043292	Contractile fiber	5	9.26E-06	ACTA2, CALD1, LMOD1, MYH11, TPM2
4270	Vascular smooth muscle contraction	4	1.87E-05	ACTG2, CALD1, MYH11, MYL9
GO.0030016	Myofibril	4	0.000315	CALD1,LMOD1,MYL9,TPM2
GO.0044449	Contractile fiber part	4	0.000315	ACTA2,LMOD1,MYH11,TPM2
GO.0016459	Myosin complex	3	0.000661	ACTG2,MYH11,MYL9
GO.0008307	Structural constituent of muscle	3	0.00147	MYH11,MYL9,TPM2
GO.0015629	Actin cytoskeleton	4	0.00264	ACTG2,LMOD1,MYH11,TPM2
GO.0005829	Cytosol	7	0.00274	ACTA2,ACTG2,CALD1,LMOD1,MYH11,MYL9,TPM2
GO.0005859	Muscle myosin complex	2	0.00477	MYH11,MYL9

Discussion

The CRC is a disease of accumulated genetic, epigenetic, and environmental aberrations.²⁸ Understanding the molecular mechanism of CRC is of very importance for diagnosis and treatment. It has been known that Wnt signaling pathway was associated with the major causes of CRC.

In this study, we expected to find out the key candidate genes and signal pathway in early-onset CRC. By comparing the 12 patients’ mucosa with 10 healthy control mucosa, 108 upregulated and 23 downregulated DEGs were screened. By using GO and PPI network analysis, 7 hub genes, namely, ACTA2, ACTG2, MYH11, CALD1, MYL9, TPM2 and LMOD1, coupled with vascular smooth muscle contraction signaling pathway have been identified.

ACTA2 (smooth muscle cell alpha actin) was identified as one of the hub genes showing the highest degree of connectivity. Lee’s group identified a correlation between early brain metastasis of lung adenocarcinoma and amplification of the ACTA2 gene, and ACTA2 could be a promising diagnostic and therapeutic target for lung cancer.²⁹ The second hub gene ACTG2 (actin gamma smooth muscle 2), encoding an ACTG2 protein, was related to metastasis of hepatocellular carcinoma.^30

-33 The third hub gene MYH11 (myosin-11) is a smooth muscle myosin belonging to the myosin heavy-chain family.³⁴ The MYH11 gene may be related to cell migration and adhesion, intracellular transport, and signal transduction, and MYH11 functions as a contractile protein, converting chemical energy into mechanical energy through adenosine triphosphate (ATP) hydrolysis. Wang et al³⁵ reported that MYH11 can contribute to predicting prognosis in stage II and III CRCs. Jo YS et al³⁶ also reported an oncogenic fusion CBFB/MYH11 and frameshift mutations in CRCs. Moreover, CALD1 (Caldesmon) encodes caldesmon protein, which is a calmodulin-binding and cytoskeleton-associated protein, and caldesmon is a biomarker for differentiation of smooth muscle.^37
-39 Yokota M group revealed that CALD1 showed a poor prognosis in colon cancer⁴⁰ Myosin regulatory light polypeptide 9 (MYL9) encoded by MYL9 is a myosin light chain that may regulate muscle contraction by conducting the ATPase activity.⁴¹ The research unveiled that MYL9 expression level might be associated with the occurrence of non-small-cell lung cancer (NSCLC), which may be correlated to NSCLC metastasis.⁴² Another hub gene, TPM2 (β-Tropomyosin), encoded tropomyosin beta chain, which is roughly 32 KD in molecular weight.⁴³ Bellavance⁴⁴ suggested that TPM2 has an important role in regulating actin cable information and controlling actin nucleation in vivo. The last hub gene LMOD1 (Leiomodin 1) is expressed in most tissue, with the high expression levels in thyroid, skeletal muscle, eye muscle, and ovary.⁴⁵ Aberrant expression of LMOD1 may be associated with the disease. Comley⁴⁶ revealed that LMOD1 was a novel component of the smooth muscle actin cytoskeleton.

Module analysis of the PPI networks suggested that the early-onset CRC is associated with vascular smooth muscle contraction signaling pathway, and the vascular smooth muscle cell (VSMC) principal function is contraction.⁴⁷ The principal mechanisms that regulate the contractile state of VSMCs are changes in cytosolic Ca²⁺ concentration. Moreover, Rho/Rho kinase, PKC, and arachidonic acid have been proposed to play a pivotal role in this event.⁴⁸

In conclusion, in this study, we investigated the potential candidate gene and signal pathway of DEGs in early-onset CRC. Genes were selected by DEG, GO, KEGG, and PPI analysis. This study has improved our understanding of the pathogenesis and underlying molecular mechanism in early-onset CRC; these selected candidate genes and pathways could give us a clue to new therapeutic targets for treatment of CRC. However, further molecular biological experiments are required to confirm the function of these identified genes in CRC.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Project of National Natural Science Foundation of China (No 81770294).

ORCID iD

Zulqarnain Baloch

References

Ferlay

Soerjomataram

Dikshit

. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012[J]. Int J Cancer. 2015;136(5):E359–E386.

Siegel

R L

Miller

Jemal

. Cancer statistics, 2015[J]. CA Cancer J Clin. 2015;65(1):5–29.

Chen

Zheng

Baade

. Cancer statistics in China[J]. CA Cancer J Clin. 2016;66(2):115–132.

Organization W. Global tuberculosis report - 2014[J]. Australasian Med J. 2013;6(2).

Pmhdev. Colorectal Cancer Prevention (PDQ®)[M]. Bethesda, MD: National Cancer Institute (US); 2013.

Kirby

Heath

Shaw

Hamdy

. Gene expression assays. Adv Clin Chem. 2007;44:247–292.

Vogelstein

Papadopoulos

Velculescu

. Cancer Genome Landscapes. Science. 2013;339(6127):1546–1558.

Guo

Bao

Yang

. Identification of key candidate genes and pathways in colorectal cancer by integrated bioinformatical analysis. IJMS. 2017;18(4):272.

Isella

Terrasi

Bellomo

. Stromal contribution to the colorectal cancer transcriptome. Nat Genet. 2015;47(4):312.

10.

Racine

. RStudio: a Platform-Independent IDE for R and Sweave. J Appl Econ. 2012;27(1):167–172.

11.

Smyth

. Limma: Linear Models for Microarray Data. New York, NY: Springer; 2005.

12.

Sherman

Lempicki

. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4(1):44.

13.

Hong

Cheah

. A susceptibility gene set for early onset colorectal cancer that integrates diverse signaling pathways: implication for tumorigenesis. Clini Cancer Res. 2007;13(4):1107–1114.

14.

Barrett

Dennis

Wilhite

Ledoux

. NCBI GEO: archive for high-throughput functional genomic data. Nucleic Acids Res. 2009;37(Database issue):D885.

15.

Edgar

Domrachev

Lash

. Gene expression omnibus: NCBI gene .expression and hybridization array data repository. Nucleic Acids Res. 2002;30(1):207–210.

16.

Tsai

Chen

. Testing for differentially expressed genes with microarray data. Nucleic Acids Res. 2003;31(9):e52.

17.

Servant

Gravier

Gestraud

. EMA - A R package for easy microarray data analysis. Bmc Res Notes. 2010;3(1):277.

18.

Smyth

. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol. 2004;3(3):Article3.

19.

Phipson

Lee

Majewski

. Robust hyperparameter estimation protects against hypervariable genes and improves power to detect differential expression. Ann Appl Stat. 2016;10(2):946.

20.

Harris

Deegan

Lomax

. The gene ontology project in 2008. Nucleic Acids Res. 2008;36(Database issue):440–444.

21.

Consortium GO. The Gene Ontology (GO) project in 2006. Nucleic Acids Res. 2006;34(Database issue):322–326.

22.

Consortium

TGO

Ashburner

Ball

. Gene Ontology: tool for the unification of biology. Nat Genet. 2000;25(1):25–29.

23.

Kanehisa

Goto

. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;27(1):29–34.

24.

Blake

Christie

Dolan

. Gene Ontology Consortium: going forward. Nucleic Acids Res. 2015;43(D1):D1049–D1056. doi:10.1093/nar/gku1179.

25.

Lebrec

Huizinga

Toes

. Integration of gene ontology pathways with North American Rheumatoid Arthritis Consortium genome-wide association data via linear modeling. BMC Proc. 2009;3(suppl 7):S94.

26.

Franceschini

Szklarczyk

Frankild

. STRING v9.1: protein-protein interaction networks, with increased coverage and integration. Nucleic Acids Res. 2013;41(Database issue):D808–15.

27.

Shannon

Markiel

Ozier

. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498.

28.

Sakai

Fukuyo

Ohata

. Genetic and epigenetic aberrations occurring in colorectal tumors associated with serrated pathway[J]. Int J Cancer. 2016;138(7):1634–1644.

29.

Lee

Park

Lee

. Alpha-smooth muscle actin (ACTA2) is required for metastatic potential of human lung adenocarcinoma. Clin Cancer Re. 2013;19(21):5879–5889.

30.

Zhao

Mei

. Identification of ACTG2 functions as a promoter gene in hepatocellular carcinoma cells migration and tumor metastasis. Biochem & Biophys Res Commun. 2017;491(2):537.

31.

Miwa

Kamada

. The nucleotide sequence of a human smooth muscle (enteric type) gamma-actin cDNA. Nucleic Acids Res. 1990;18(14):4263.

32.

Szucsik

Lessard

. Cloning and sequence analysis of the mouse smooth muscle gamma-enteric actin gene. Genomics. 1995;28(2):154–162.

33.

Ueyama

Inazawa

Nishino

Han-Xiang

Ochiai

Ohkubo

Chromosomal mapping of the human smooth muscle actin gene (enteric type, ACTA3) to 2p13.1 and molecular nature of the hindIII polymorphism. Genomics. 1995;25(3):720–723.

34.

Matsuoka

Yoshida

M C

Furutani

. Human smooth muscle myosin heavy chain gene mapped to chromosomal region 16q12. Am J Med Genet. 2010;46(1):61–67.

35.

Wang

Cai

. Down-regulated MYH11 expression correlates with poor prognosis in stage II and III colorectal cancer. Asian Pac J Cancer Prev Apjcp. 2014;15(17):7223–7228.

36.

Kim

Yoo

Lee

. Somatic mutations and intratumoral heterogeneity of MYH11 gene in gastric and colorectal cancers. Appl Immunohistochem Mol Morphol. 2018:26(8):562–566.

37.

Mani

Mccubbin

Kay

. Calcium-dependent regulation of caldesmon by an 11-kDa smooth muscle calcium-binding protein, caltropin. Biochemistry. 1992;31(47):11896–11901.

38.

Wang

CL.

Caldesmon and smooth. Cell Biochemistry & Biophysics. 2001;35(3):275–288.

39.

Zheng

Sieuwerts

Luider

. Differential expression of splicing variants of the human caldesmon gene (CALD1) in glioma neovascularization versus normal brain microvasculature. Am J Pathol. 2004;164(6):2217–2228.

40.

Yokota

Kojima

Higuchi

.Gene expression profile in the activation of subperitoneal fibroblasts reflects prognosis of patients with colon cancer. Int J Cancer. 2016;138(6):1422–1431.

41.

Kumar

Mohan

Zavodny

Narula

Leibowitz

. Characterization and differential expression of human vascular smooth muscle myosin light chain 2 isoform in nonmuscle cell. Biochemistry. 1989;28(9):4027.

42.

Tan

Chen

MYLK and MYL9 expression in non-small cell lung cancer identified by bioinformatics analysis of public expression data. Tumour Biol. 2014;35(12):12189–12200.

43.

Eyre

Akkari

Wilton

. Assignment of the human skeletal muscle α-tropomyosin gene (TPM1) to band 15q22 by fluorescence in situ hybridization. Cytogenetic & Genome Res. 1995;69(1-2):15–17.

44.

Bellavance

. A novel role for Tpm2 in regulating formin-mediated actin assembly. 2015. http://hdl.handle.net/10192/30617.

45.

Fagerberg

Hallström

B M

Oksvold

. Analysis of the human tissue-specific expression by genome-wide integration of transcriptomics and antibody-based proteomics. Mol Cell Proteomics. 2014;13(2):397–406.

46.

Conley

. Leiomodin and tropomodulin in smooth muscle. Am J Physiol Cell Physiol. 2001;280(6):C1645.

47.

York

SN.

Vascular Smooth Muscle Cell (VSMC). Curr Pharmaceutical Design. 2013;20(4):625–634.

48.

Young

Jae Ho

Mina

. MLCK and PKC involvements via Gi and Rho A protein in contraction by the electrical field stimulation in feline esophageal smooth muscle. Korean J Physiol Pharmacol Official. 2010;14(1):29–35.