Mechanisms of action of triptolide against colorectal cancer: insights from proteomic and phosphoproteomic analyses

Abstract

Triptolide is a potent anti-inflammatory agent that also possesses anticancer activity, including against colorectal cancer (CRC), one of the most frequent cancers around the world. In order to clarify how triptolide may be effective against CRC, we analyzed the proteome and phosphoproteome of CRC cell line HCT116 after incubation for 48 h with the drug (40 nM) or vehicle. Tandem mass tagging led to the identification of 403 proteins whose levels increased and 559 whose levels decreased in the presence of triptolide. We also identified 3,110 sites in proteins that were phosphorylated at higher levels and 3,161 sites phosphorylated at lower levels in the presence of the drug. Analysis of these differentially expressed and/or phosphorylated proteins showed that they were enriched in pathways involving ribosome biogenesis, PI3K-Akt signaling, MAPK signaling, nucleic acid binding as well as other pathways. Protein-protein interactions were explored using the STRING database, and we identified nine protein modules and 15 hub proteins. Finally, we identified 57 motifs using motif analysis of phosphosites and found 16 motifs were experimentally verified for known protein kinases, while 41 appear to be novel. These findings may help clarify how triptolide works against CRC and may guide the development of novel treatments.

Keywords: colorectal cancer; molecular docking; phosphoproteomic; proteomic; triptolide.

Figure 1

Global proteomic and phosphoproteomic analysis of colorectal cancer cells. (A) Schematic of the experimental workflow; LC, liquid chromatography; MS, mass spectrometry; TMT, tandem mass tags. (B) Numbers of proteins whose levels were significantly higher (red) or lower (blue) in triptolide-treated cell cultures than in control cultures. (C) Numbers of differentially expressed proteins in different subcellular compartments. (D) Numbers of sites in proteins whose phosphorylation was significantly higher (red) or lower (blue) in triptolide-treated cell cultures than in control cultures. (E) Numbers of differentially phosphorylated proteins in different subcellular compartments.

Figure 2

Differential expression levels of the quantitative proteome and their enrichment in Gene Ontology terms. (A) Heatmap of the quantitative proteome based on fold differences in expression. (B) Volcano plot of the differences in protein levels. The volcano map was drawn based on the expression of FC and P value (T-test). The significantly down-regulated proteins were blue (FC< 0.83 and P <0.05), the significantly up-regulated proteins were red (FC>1.2 and P <0.05), and the proteins with no difference were gray. (C) Classification of differentially expressed proteins based on Gene Ontology biological processes, cellular components and molecular functions.

Figure 3

Analysis of predicted interactions among differentially expressed proteins. The four most significant modules were identified by the molecular complex detection (MCODE) algorithm. (A) Enrichment of domains in differentially expressed proteins. (B) Enrichment of KEGG pathways in differentially expressed proteins. (C) Interaction network of differentially expressed proteins. (D–G) The four most significant MCODE modules.

Figure 4

Differential phosphorylation of the quantitative proteome and enrichment in Gene Ontology terms. (A) Heatmap based on differential phosphorylation levels. (B) Volcano plot of the differences in phosphorylation levels. (C) Classification of differentially phosphorylated proteins based on Gene Ontology biological processes, cellular components and molecular functions.

Figure 5

Protein-protein interaction (PPI) network analyses of PDEPs were performed, and the four most significant modules were identified by the molecular complex detection (MCODE) algorithm. (A) Enrichment of domains in differentially expressed proteins. (B) Enrichment of KEGG pathways in differentially expressed proteins. (C) Interaction network of differentially expressed proteins. (D–H) The five most significant MCODE modules.

Figure 6

Analysis of motifs differentially phosphorylated between CRC cultures treated with triptolide or vehicle. (A) Motifs whose phosphorylation is upregulated by triptolide. (B) Motifs whose phosphorylation is downregulated by triptolide. (C) Ranking of the top six motifs upregulated by triptolide. (D) Ranking of the top six motifs downregulated by triptolide.

Figure 7

Shows the binding interactions of triptolide with the CRC-related hub genes protein. Triptolide binds to AMD1(A), IMP3(B), HNRNP(C) and DHX9(D). Ball and stick represent triptolide; cartoon represents a hub target.