Effect of C-to-T transition at CpG sites on tumor suppressor genes in tumor development in cattle evaluated by somatic mutation analysis in enzootic bovine leukosis

Abstract

Oncogenic transformation of normal cells is caused by mutations and chromosomal abnormalities in cancer-related genes. Enzootic bovine leukosis (EBL) is a malignant B-cell lymphoma caused by bovine leukemia virus (BLV) infection in cattle. Although a small fraction of BLV-infected cattle develops EBL after a long latent period, the mechanisms for oncogenesis in EBL cattle remain largely unknown. In this study, we analyzed the types and patterns of somatic mutations in cancer cells from 36 EBL cases, targeting 21 cancer-related genes. Various somatic mutations were identified in eight genes, TP53, KMT2D, CREBBP, KRAS, PTEN, NOTCH1, MYD88, and CARD11. In addition, TP53 gene was found to be mutated in 69.4% of EBL cases, with most being biallelic mutations. In some cases, associations were observed between the ages at which cattle had developed EBL and somatic mutation patterns; young onset of EBL possibly occurs due to high impact mutations affecting protein translation and biallelic mutations. Furthermore, nucleotide substitution patterns indicated that cytosine at CpG sites tended to be converted to thymine in many EBL cases, which was considered to be the result of spontaneous deamination of 5-methylcytosine. These results demonstrate how somatic mutations have occurred in cancer cells leading to EBL development, thereby explaining its pathogenic mechanism. These findings will contribute to a better understanding and future elucidation of disease progression in BLV infection.IMPORTANCEEnzootic bovine leukosis (EBL) is a malignant and lethal disease in cattle. Currently, there are no effective vaccines or therapeutic methods against bovine leukemia virus (BLV) infection, resulting in severe economic losses in livestock industry. This study provides a renewed hypothesis to explain the general mechanisms of EBL onset by combining the previous finding that several integration sites of BLV provirus can affect the increase in survival and proliferation of infected cells. We demonstrate that two additional random events are necessary for oncogenic transformation in infected cell clones, elucidating the reason why only few infected cattle develop EBL. Further exploration of somatic mutation and BLV integration sites could support this hypothesis more firmly, potentially contributing to the development of novel control methods for EBL onset.

Keywords: TP53; enzootic bovine leukosis; mutation signature; somatic mutation.

Fig 1

A scheme for somatic mutation analysis in cattle with EBL development. Thirty-six EBL cattle collected in previous studies (19, 20) were used for the mutation analysis. Blood and tumor samples derived from the same EBL cases were examined to identify variants in their genome using amplicon sequencing that targeted 21 cancer-related genes. Following variant calling, variant lists were compared between blood and tumor samples, and the variants existing only in tumor samples were identified as somatic mutations. All identified somatic mutations were used for mutation pattern analysis.

Fig 2

Somatic mutation analysis in 36 EBL cases at various ages in months. (A) Identified somatic mutations of 8 of 21 targeted cancer-related genes are shown in ascending order of age for EBL cases. Genes in which a somatic mutation are not identified at all are omitted. The top panel indicates mutation type, and the middle panel indicates detailed classification of multiple mutations and presence or absence of loss of heterozygosity (LOH). The bottom panel indicates breed, sex, and age in months of each case. Information on identified somatic mutations based on each targeted gene or each EBL case is listed in Table 2 or Table S3, respectively. (B) Number of cases and frequencies and types of somatic mutations are shown. (C–E) Ages of cattle are compared among 36 EBL cases based on (C) presence of TP53 mutation, (D) intensities of mutation impact, and (E) presence of biallelic mutation. Red bars indicate medians. TP53 wt, wild-type TP53; TP53 mut, mutated TP53. High mutation impact means stop-gained or frameshift mutation. Biallelic mutation includes multiple mutations or mutation with LOH. Significant differences between the two groups were evaluated by Mann–Whitney’s U test (P < 0.05).

Fig 3

Analysis of mutation patterns in identified single nucleotide variants (SNVs). (A) Rates of nucleotide substitutions in SNVs from 36 EBL cases are shown with a relative color scale. Identified SNVs are put together in all cases or in two groups divided based on cattle age in months. The identical somatic mutation observed in different cases was counted independently. ≥36-m-old, group of EBL cases over 36 months old; <36-m-old, group of EBL cases less than 36 months old. (B) Mutation spectra based on single base substitution (SBS) signatures are drawn using the MutSignatures R package. Identified SNVs are divided into 96 different contexts and their mutation frequencies calculated. The top three tri-nucleotide mutations in each group are marked in the spectra. Mutation spectra in the ≥36-m-old group and the <36-m-old group are defined as Signatures A and B, respectively. (C) Heatmap shows similarity between Signature A/B and known COSMIC signature 1–21. Cosine distances between each pair of signatures were calculated and are shown by color intensity.

Fig 4

Hypothesis of the general mechanisms of EBL onset. There are three events for BLV-infected cells to be transformed into neoplasms. First, infected cell clones whose proviruses are integrated near cancer driver genes acquire long life and reproductive activity (8). Other clones in which the proviruses do not integrate near cancer driver genes cannot overcome this first obstacle, and thus they are not able to be candidate future cancer cells. Second, single somatic mutation occurs accidentally mainly at CpG sites in cancer-related genes, e.g., TP53 and CREBBP, as a first “hit.” Finally, an additional mutation or loss of heterozygosity (LOH) occurs on the left normal allele as a second “hit.” These three random events need to occur in the same infected clone and be completed during the lives of infected animals, which is considered to be the reason why only a small fraction of BLV-infected cattle develops EBL after a long latent period.