Evaluating the CRP Interactome: Insights into Possible Novel Roles in Cellular Signaling and Tumorigenicity

Abstract

C-reactive protein (CRP) is a well-known acute phase reactant and putative biomarker for advancing and chronically established inflammation. Its biological activity across its multiple isoforms plays various roles in the initiation, potentiation, and resolution of inflammation. Its molecular signaling within the tissue microenvironment regulates cell-cell communication across cell types (e.g., epithelial cells, endothelial cells, fibroblasts, adipocytes, and immune cells) and affects the development of conditions such as cancer that are subject, at least in part, to inflammatory signaling. Considering the dynamic nature of CRP in modulating disease progression, and the growing evidence of the context-dependent direct molecular activity of CRP on regulating intra- and inter-cellular signaling, it is critical to further understand how this integral molecule alters cell signaling pathways. Although the ability of CRP to directly interact with some extracellular matrix proteins involved with inflammation and disease has been reported as early as the mid-1980s, recent advances in unbiased proteomics have revealed a broader interactome of protein-protein interactions (PPIs) involving CRP. The present study evaluates the CRP PPIs identified to date and explores the potential novel regulatory capacity of CRP on multiple key cellular functions in metabolism and cell-cell signaling, offering an updated framework of the possible biological activities of CRP relevant to tumorigenic processes.

Keywords: C-reactive protein; CRP; extracellular matrix; glycoproteins; glycosaminoglycan biosynthesis; interactome; protein-protein interactions; tumor microenvironment.

Figure 1

BioPlex Interactome of C-Reactive Protein. Protein–Protein interactions (PPIs) of C-reactive protein (CRP) in HEK293T (A) and HCT116 (B) cells identified by Huttlin et al., 2021 [23], using affinity pulldown mass-spectrometry (AP-MS). Purple node indicates target protein (CRP), with green nodes indicating target protein was both bait and prey and gray nodes indicating target protein was identified as prey; arrows indicate directionality of pulldown (i.e., anti-CRP immunoprecipitation). (C) Left: CRP PPIs identified by AP-MS in HEK293T and HCT116 included 24 unique proteins and 14 unique proteins, respectively, with 19 proteins identified in both, for a total of 57 PPIs altogether. Right: Protein–Protein interactions in both HEK293T and HCT116 cell lines, with overlapping proteins identified in purple (Overlap: AGRN, B4GALT3, B4GALT4, B4GALT5, B4GALT7, CEMIP2, CHST3, CHST12, CNTNAP3, COL18A1, LACTB, LAMA5, LAMB1, LAMB2, LAMC1, LRRC24, PDIA5, QSOX1, UBR3). Data was obtained using BioPlex Explorer 3.0 (https://bioplex.hms.harvard.edu/explorer/home, accessed on 1 May 2025).

Figure 2

Visualization of KEGG Database Signaling Pathway Enrichment. Gene Set Enrichment Analysis (GSEA) was performed using the ShinyGO database (https://bioinformatics.sdstate.edu/go/, accessed on 1 May 2025). All 71 proteins with protein–protein interactions (PPIs) with CRP curated by BioGrid (https://thebiogrid.org) from published data were used and the KEGG Legacy (Kyoto Encyclopedia of Genes and Genomes) database was queried. Signaling pathways are listed by fold enrichment and ranked in descending order. Enrichment is determined based on the number of CRP PPIs input that overlap with the total number of proteins in each dataset/signaling pathway. Color indicates false-discovery rate (FDR) q-value, with red indicating highly significant (FDR q < 0.000000005) and blue indicating somewhat significant (FDR q < 0.005). Size of node indicates the number of genes, ranging from 5 to 12 CRP PPIs identified in the gene set. The top 5 most statistically significant enriched signaling pathways observed were involved in biosynthetic processes related to glycosaminoglycans (Keratan Sulfate, Chondroitin Sulfate, Heparan Sulfate) and glycosphingolipids as well as the “ECM–Receptor Interactions” dataset. The highest number of proteins associated with specific pathways included “ECM–Receptor Interactions”, “Pathways in Cancer”, “Metabolic Pathways”, “Focal Adhesions”, “Small Cell Lung Cancer”, “PI3K–Akt Signaling Pathway”, “HPV Infection”, “Amoebiasis”, and “Taxoplasmosis”.

Figure 3

Cluster Analysis of Signaling Pathways Overrepresented in CRP PPIs. All 71 proteins with protein–protein interactions (PPIs) with CRP curated by BioGrid (https://thebiogrid.org) from published data were used and the ShinyGO database (https://bioinformatics.sdstate.edu/go/, accessed on 1 May 2025) database was queried. (A) Interactions between proteins and pathways identified by CRP PPIs showing similarities and overlap between specific pathways. The number of CRP PPIs identified in each pathway is indicated by node size with opacity indicating statistical enrichment. Number of proteins overlapping (i.e., CRP PPIs present in each dataset) is indicated by line thickness. (B) Hierarchical cluster analysis of pathways as a function of representation from the CRP PPI dataset, with PPIs shared across pathways clustering closer together. Data indicates two primary clusters: (top) metabolic pathways, specifically those associated with glycosaminoglycan and glycosphingolipid synthesis, and (bottom) cancer associated pathways and molecular responses to infectious diseases.

Figure 4

Updated Model of the CRP Interactome. STRING database (https://string-db.org/, accessed on 1 May 2025) analysis of the interaction network of all CRP PPIs showing known and predicted interactions between all proteins pulled down by CRP (A) compared to the general inquiry of the STRING database only for CRP (B). Line colors and indication: cyan (database curation), magenta (experimentally determined), green (gene neighborhood), red (gene fusion), blue (gene co-occurrence), yellow (text-mining), black (co-expression), and light blue (protein homology). Absence of a line indicates that the specific protein has not been documented in the STRING database to interact with the other proteins identified as CRP PPIs. Querying the STRING database utilizing the updated CRP interactome provides numerous additional signaling networks and indicates their interrelatedness based on CRP PPIs (A), suggesting the potential broader impact of CRP on cell signaling in comparison to its traditional CRP STRING model based on traditionally described interactions (B).