miRNAs Copy Number Variations Repertoire as Hallmark Indicator of Cancer Species Predisposition

Abstract

Aging is one of the hallmarks of multiple human diseases, including cancer. We hypothesized that variations in the number of copies (CNVs) of specific genes may protect some long-living organisms theoretically more susceptible to tumorigenesis from the onset of cancer. Based on the statistical comparison of gene copy numbers within the genomes of both cancer-prone and -resistant species, we identified novel gene targets linked to tumor predisposition, such as CD52, SAT1 and SUMO. Moreover, considering their genome-wide copy number landscape, we discovered that microRNAs (miRNAs) are among the most significant gene families enriched for cancer progression and predisposition. Through bioinformatics analyses, we identified several alterations in miRNAs copy number patterns, involving miR-221, miR-222, miR-21, miR-372, miR-30b, miR-30d and miR-31, among others. Therefore, our analyses provide the first evidence that an altered miRNAs copy number signature can statistically discriminate species more susceptible to cancer from those that are tumor resistant, paving the way for further investigations.

Keywords: DNA copy number variation; comparative study; miRNAs.

Figure 1

CNV landscape comparisons: (A) Boxplot of the distribution of significant gene CNVs in cancer-prone vs. cancer-resistant species. (B) Boxplot of the distribution of significant microRNA CNVs in cancer-prone vs. cancer-resistant species. Cancer-resistant species are highlighted in green, cancer-prone species in red. In the boxplots, the Y-axis scale has been changed to log one. The boxplots were built considering the average number of copies of each gene in the two different target groups. (C) Heatmap representing the microRNA CNV repertoires within the nine analyzed species—(Hg): Heterocephalus glaber; (Ng): Nannospalax galili; (Dn): Dasypus novemcinctus; (La): Loxodonta Africana; (Ml): Myotis lucifugus; (Mm): Mus musculus; (Rn): Rattus norvegicus; (Cf): Canis familiaris; (Hs): Homo sapiens. Hg, Ng, Dn, La and Ml have been previously described as cancer-resistant species. Mm, Rn, Cf and Hs are known to be cancer-prone species. Phylogeny was inferred from VertLife [35], created and visualized through the Interactive Tree of Life web-tool [36]. (D) PGLS correlating the cancer incidence rate with the total number of significant microRNAs copies across the nine species included in the analysis. The blue line represents a positive correlation between the two variables (adjusted R² = 0.5173; p-value = 0.01746).

Figure 2

(A) PCA based on the CNVs of all the significant genes. (B) PCA based on the CNVs of the significant microRNAs subset. Both plots show a dichotomy between cancer-resistant (blue) and cancer-prone species (red). (C) Heatmap of the significant microRNAs, clustered with Euclidean distance and complete linkage. (D,E) Bar and box plots of the significant microRNAs CNVs in cancer-prone species, cancer-resistant species and Loxodonta africana. The microRNAs repertoire of Loxodonta africana seems to reflect the cancer-prone miRNAs copy number alteration landscape, rather than the one typical of the cancer-resistant organisms. In the box plots, the Y-axis scale was changed to log one. The boxplots are built considering the average number of copies of each gene in the two different target groups.