From genotype to functional phenotype: unraveling the metabolomic features of colorectal cancer

Abstract

Much effort in recent years has been expended in defining the genomic and epigenetic alterations that characterize colorectal adenocarcinoma and its subtypes. However, little is known about the functional ramifications related to various subtypes. Metabolomics, the study of small molecule intermediates in disease, provides a snapshot of the functional phenotype of colorectal cancer. Data, thus far, have characterized some of the metabolic perturbations that accompany colorectal cancer. However, further studies will be required to identify biologically meaningful metabolic subsets, including those corresponding to specific genetic aberrations. Moreover, further studies are necessary to distinguish changes due to tumor and the host response to tumor.

Figure 1

Classification of CRC by pathogenic pathway. The classical pathway involves a progressive accumulation of mutations due to chromosomal instability as an adenoma develops into adenocarcinoma. Serrated polyps, which are thought to develop from hyperplastic polyps, are generated due to microsatellite instability and/or high levels of CpG island methylation. The adenocarcinomas that emerge from that pathway have distinct clinical and pathological features.

Figure 2

Relationship of CIMP expression phenotype and microsatellite instability. There is significant overlap between MSI-H tumors and CIMP-H tumors, although microsatellite stable (MSS) tumors and tumors with a low level of MSI (MSI-L) may have high levels of CpG island methylation.

Figure 3

CRC subgroups identified by analysis of molecular data compiled from 224 tumors analyzed by The Cancer Genome Atlas Project [9]. Hypermutated CRC is highly enriched for hypermethylation, CIMP expression phenotype and BRAF mutations; it most frequently occurs in the proximal colon.

Figure 4

CRC is derived from an accumulation of genomic and epigenomic alterations. Alterations at the transcriptional level also occur due to the influence of regulatory RNAs (e.g., long noncoding RNA, miRNA, pseudogenes). Post-translational regulation and post-translational modifications further contribute to functional perturbations in tumor cells. These sequential and synchronous events contribute to the phenotype. Phenotype can further modify the genotype as well as any of the downstream events. According to this model, the metabolome represents the closest molecular representation of phenotype.

Figure 5

Examples of predicted metabolic effects of dysregulated proteins associated with CRC. The proteins are dysregulated as a consequence of modulatory molecular events at multiple levels. Alterations in metabolic function (and ultimately phenotype) result from the closely connected functional networks’ response to these upstream signals. (GDP: guanine diphosphate; GTP: guanine triphosphate; GEF: guanine nucleotide-exchange factor; GAP: GTPase-activating protein; PPP: pentose phosphate pathway) [65,66,67,68].