Colorectal Cancer Chemotherapy Drug Bevacizumab May Induce Muscle Atrophy Through CDKN1A and TIMP4

Abstract

The muscle in the organism has the function of regulating metabolism. Long-term muscle inactivity or the occurrence of chronic inflammatory diseases are easy to induce muscle atrophy. Bevacizumab is an antiangiogenic drug that prevents the formation of neovascularization by inhibiting the activation of VEGF signaling pathway. It is used in the first-line treatment of many cancers in clinic. Studies have shown that the use of bevacizumab in the treatment of tumors can cause muscle mass loss and may induce muscle atrophy. Based on bioinformatics analysis, this study sought the relationship and influence mechanism between bevacizumab and muscle atrophy. The differences of gene and sample expression between bevacizumab treated group and control group were studied by RNA sequencing. WGCNA is used to find gene modules related to bevacizumab administration and explore biological functions through metascape. Differential analysis was used to analyze the difference of gene expression between the administration group and the control group in different muscle tissues. The key genes timp4 and CDKN1A were obtained through Venn diagram, and then GSEA was used to explore their biological functions in RNA sequencing data and geo chip data. This study studied the role of bevacizumab in muscle through the above methods, preliminarily determined that timp4 and CDKN1A may be related to muscle atrophy, and further explored their functional mechanism in bevacizumab myotoxicity.

Keywords: CDKN1A; TIMP4; bevacizumab; muscles; sarcopenia.

Figure 1

Validation of sequencing data. (A) PCA analysis of data distribution in 12 samples; (B) correlation analysis of genes in each sample; (C) heat map of gene distribution in samples; (D) normalization results of samples.

Figure 2

Selection of key gene modules. (A) Determination of soft threshold; (B) Heat map of module feature vector clustering; (C) Heat map of correlation between gene modules and traits.

Figure 3

Function clustering analysis of genes in key modules. (A) Co-expression network diagram of 280 key genes; (B) Rich term network of genes, showing the top 20 functional terms.

Figure 4

Differential expression analysis results. Differential expression analysis of genes in gastrocnemius muscle, with (A) volcano plot showing the differential expression results of genes, (B) heat map showing the distribution of up- and down-regulated genes; Differential expression analysis of genes in soleus muscle, with (C) volcano plot showing the differential expression results of genes, and (D) the heat map showing the distribution of up- and down-regulated genes.

Figure 5

Map3k6 expression. (A) Venn map showing the intersection of different datasets; correlation of (B) Timp4 and (C) Cdkn1a with Vegfa; (D) Timp4 and (E) Cdkn1a expression in drug-treated and control groups; (F) Timp4 and (G) Cdkn1a expression in different groups in GSE38417; (H) Timp4 and (I) Cdkn1a expression in different groups in GSE6011. *p < 0.05,***p < 0.001.

Figure 6

Functional pathways of Timp4 and Cdkn1a. (A, C) KEGG pathway and (B, D) Hallmark pathway of Timp4 and Cdkn1a in the bevacizumab-administered group; (E, G) KEGG pathway and (F, H) Hallmark pathway of Timp4 and Cdkn1a in the combined dataset of GSE6011 and GSE38417.