Publication journal: BMC Biol
Publication date: 2022 May 21
doi:10.1186/s12915-022-01316-0
APOBEC3 (Lipoprotein B mRNA Editing Enzyme Catalytic Polypeptide 3) enzymes form a family of closely related cytidine deaminase enzymes that target single-stranded DNA and are characterized by producing predominantly C>T mutations, but in their preferred sequence context slightly different. APOBEC3 activity is thought to be responsible for two well-defined single base pair substitution (SBS) mutation signatures, SBS2 and SBS13. The main function of APOBEC3 enzyme is to limit viral infection and reverse transcription gene activity. Despite its well-defined role in cells, the APOBEC3 enzyme is considered a potential source of cancer initiation and progression due to its off-target effects on the host genome. Overexpression of APOBEC3A in cellular systems causes DNA fragmentation, DNA damage response, and cell cycle arrest, while APOBEC3B causes base substitutions in the host genome. The oncogenic potential of APOBEC3s has been highlighted in many different cancers, including multiple myeloma, breast cancer, lung cancer, and urothelial cancer
Previous work has focused primarily on the use of exome sequencing data Effects of APOBEC3 in individual tumor types. Here, we investigated the relationship between APOBEC3 and genomic instability (GI) using whole-genome sequencing data from 2451 primary tumors from 39 different tumor types in the Pan-Cancer Analysis of Whole Genomes (PCAWG) dataset.
Whole-genome sequencing of 2451 whitelisted primary tumor samples provided by the Pan-Cancer Analysis of Whole Genomes (PCAWG) consortium; analysis of mutation signatures was performed by the PCAWG Mutation Signatures and Processes Working Group; in 2451 Among the whitelist samples, 741 carried mutations attributed to SBS2 and SBS13. These 741 samples were used to calculate the correlation between APOBEC3 SNVs and non-APOBEC3 SNVs
(1) Mixed effects model: use ' The lme4'R package and the 'glmmTMB'R package created mixed-effects models; the results of linear and mixed-effects models were generated using the 'Stargazer' R package and the 'texreg' R package; three mixed-effects models were built to illustrate the effect of tumor type on APOBEC3 Effects of mutation number, age, and the combination of both on the total number of non-APOBEC3 mutations; six mixed-effects models were also built to examine the relationship between the number of APOBEC3 mutations and the six genomic instability indicators investigated
(2) Survival analysis
(3) Genomic instability: Genomic instability is characterized by a series of different changes at the chromosome level. Frequent changes include increased numbers of insertions, deletions, translocations, and structural variants. Changes in ploidy were assessed by investigating the Proportion of Genome Alterations (PGA), which describes the proportion of genomes deviating from copy number 2 or 4 for samples with diploid and whole-genome duplications, respectively; PGA attributable to INDEL signatures ID6 and ID8 was assessed Number of insertions and deletions (INDELS)
(4) Volcano plot: To aid visualization and to prevent division by 0 when estimating effect sizes, the number of insertions and deletions (INDELS) was calculated for each patient carrying SBS2 and SBS13 mutations or not carrying these mutations. A pseudocount of 1 was added to the median of the genomic instability measures calculated for tumors within each tumor type, except for PGA, before taking the ratio of the median. All statistical analyzes were performed on raw data without spurious counts.
The authors studied the relationship between the number of classic APOBEC3 mutations (SBS2 and SBS13) and TMB, excluding mutations attributed to SBS2 and SBS13. Among 2451 primary tumors, 741 were found to carry mutations attributed to the APOBEC3 mutation signature. Tumors harboring APOBEC3 mutations were found in 26 of the 39 tumor types in the PCAWG dataset (Figure 1) and had significantly higher mutational burdens than tumors that did not harbor APOBEC3 mutations. Furthermore, the number of APOBEC3 mutations was significantly associated with TMB in 14 of the 22 tumor types, for which at least three samples were available for calculation of the Spearman correlation (Fig. 1). After accounting for the effect of tumor type, both age and the number of classic APOBEC3 mutations were significant predictors of the number of non-APOBEC3 SNVs (mixed-effects model).
It was previously thought that the increase in overall mutation burden coinciding with the increased number of APOBEC3 mutations might result from further processing of deaminated cytosines by DNA repair enzymes, thereby producing transpositions, transitions, and doublets. Chain breaks (DSBs). Errors in the repair process of DSBs can lead to mutations and chromosomal rearrangements. Using the number of APOBEC3 mutations as an indicator of prior APOBEC3 activity, the authors studied their impact on multiple measures of genome instability.
The authors used the number of structural variants (SV), copy number (CN) segments, the percentage of the genome altered by copy number aberrations (PGA), and the number of insertions and deletions as measures of genome instability. The number of insertions and deletions (INDELs) attributed to INDEL signatures 6 and 8 (ID6 and ID8) was also studied. For all six genomic instability metrics considered by the authors, samples carrying APOBEC3 mutations had significantly higher values ??than samples without APOBEC3 mutations (Figure 2).
The authors constructed mixed-effects models to investigate whether the number of APOBEC3 mutations could be used to predict levels of instability markers, taking into account age and tumor type. The model showed that tumors carrying APOBEC3 mutations were more genomically unstable, and that the number of APOBEC3 mutations was associated with all indicators of genomic instability except the number of ID6 INDELs. Age has a significant predictive effect on the total number of INDELs and the number of structural variations.
Comparing the median of each of the six measurements in a given tumor type highlighted several tumor types in which the presence of APOBEC3 mutations had a strong impact on genomic instability (Figure 3 ). When individual measures of genomic instability were taken into account, 13 of the 24 tumor types had significant associations between the presence of APOBEC3 mutations and measures of genomic instability. Specifically, for pancreatic cancer subtypes (pancreatic endocrine neoplasia (PAEN) and pancreatic adenocarcinoma (PACA)), bone cancer (BOCA), renal papillary cell carcinoma (KIRP), and malignant lymphoma (MALY), in Higher levels of genomic instability across multiple measures were observed in tumors harboring APOBEC3 mutations. In addition, breast cancer (BRCA), lung adenocarcinoma (LUAD), renal clear cell carcinoma (KIRC), renal chromaffin cell carcinoma (KICH), gastric adenocarcinoma (STAD), uterine endometrium carcinoma (UCEC), sarcoma (SARC) ) and a single measure of genomic instability in prostate cancer (PRAD) were also significantly associated. When combining all P values ??measuring genomic instability, there were also 2 tumor types, biliary tract cancer (BTCA) and cervical squamous cell carcinoma (CESC), that showed a clear association between the presence of APOBEC3 mutations and GI.
Some studies have found that APOBEC3 protein activity is closely related to the activity of p53, and p53 is a negative regulator of APOBEC3B activity. In addition, APOBEC3 activity is also related to TP53 gene mutations. To further investigate this link, the authors built new models that added the impact of changes in TP53.
Among tumors carrying APOBEC3 mutations, the proportion of tumors carrying TP53 missense or nonsense mutations (41.6%) was significantly higher than that of tumors not carrying any APOBEC3 mutations (19.9%). Tumors harboring TP53 missense or nonsense mutations also had more APOBEC3 mutations, as well as a higher burden of non-APOBEC3 mutations.
Adding the TP53 mutation status of the tumor to the mixed effects model generated in the previous section showed that TP53 mutations are an important factor in predicting indicators of genomic instability, with the exception of the number of ID8 INDELs. Importantly, the number of APOBEC3 mutations remained a highly significant predictor throughout and also became a significant predictor of the number of ID6 INDELs. There was a significant improvement in the number of copy number fragments, the number of structural variants, and the number of ID6 INDELs (including TP53 in the model) for PGA, but not for the total number of INDELs or ID8 INDELs. The effect of age on measures of genomic instability remains unclear, except for the effect of age on the number of structural variants.
The authors constructed a Cox proportional hazards model combined with a mixed effects model, and also studied the impact of TP53 mutations and APOBEC3 mutations on overall survival, taking into account the impact of tumor type (CoxME model). When the presence of APOBEC3 mutations was considered alone, it had no apparent impact on survival. However, when TP53 mutation status was included, it was found that APOBEC3 mutations increased the hazard ratio when TP53 was not mutated, and that TP53 mutations significantly increased the hazard ratio when APOBEC3 mutations were absent, with a negative impact on survival in both conditions. . The interaction between the presence of APOBEC3 mutations and TP53 mutations was also significant, but the hazard ratio was less than 1, indicating that the coexistence of APOBEC3 mutations and TP53 mutations leads to better survival outcomes.
To address whether the model's results could be attributed to processes such as kataegis, in which APOBECs act on single-stranded DNA byproducts of DNA damage repair rather than causing strand breaks themselves, the authors reconstructed the model to exclude attribution SNVs in kataegis events involving APOBEC3 mutations. Excluding APOBEC3 mutations associated with kataegis did not significantly change the previous conclusions.
The authors found that when accounting for the effects of TP53 mutations, the number of APOBEC3 mutations (excluding those attributed to kataegis) remained a significant predictor of genomic instability. This suggests that APOBECs may play an active role in the generation of widespread and diverse genomic instabilities.
The study found that the number of classic APOBEC3 signature mutations was associated with increased mutation burden in different tumor types. Additionally, the number of APOBEC3 mutations was a significant predictor of six different measures of GI. Two GI measurements (INDEL attributed to INDEL signatures ID6 and ID8) strongly suggested the occurrence and error-prone repair of double-strand breaks, and the relationship between APOBEC3 mutations and GI remained when SNVs attributed to kataegis were excluded.
The study also provides evidence supporting a model of cancer genome evolution in which APOBEC3 acts as a causative factor in the development of diverse and widespread genomic instability through the generation of double-strand breaks. This has important implications for the treatment of cancers harboring APOBEC3 mutations and challenges the idea that APOBEC only acts opportunistically at single-stranded DNA sites.