1. Introduction
A
Human papillomavirus (HPV) infection of the cervix results in different outcomes depending on the type of virus, persistence of infection, and intrinsic immune properties of the host [
1]. Asymptomatic infections or infections that result in benign lesions are more common than the malignant lesions that are precursors to cervical cancer [
2]. However, this virus is necessary to cause cervical cancer [
3].
Variations in the apolipoprotein B mRNA editing enzyme catalytic polypeptide 3 (APOBEC3)-like gene, such as the APOBEC3A and APOBEC3B deletion hybrid (A3A/B), may be associated with persistent viral infections and the establishment of the somatic mutations that predispose to cellular malignancies [
4].
The variant A3A/B allele is characterized by the deletion of 29,936 base pairs (bp) between the fifth exon of A3A and the eighth exon of A3B [
5]. It results in a hybrid gene encoding a wild-type A3A-identical transcript fused to the 3’ untranslated region (UTR) of A3B [
4,
6,
7]. As a result, exclusion of the A3B protein occurs, and posttranscriptional regulation of A3A is reduced, increasing the nucleocytoplasmic levels of the protein [
4]. The A3A/B hybrid transcript is 10 to 20 times more stable and more highly expressed than the wild-type transcript [
8].
Increased expression of the A3A protein encoded by the wild-type allele and/or the variant allele favors the formation of homodimers that tend to interact with specific sites (5’TC) in RNA or single-stranded DNA (ssDNA) [
9,
10]. The most important consequence is the deamination of cytidines (C) to uridines (U) [
11], which generates double-strand breaks (DSB) or somatic mutations through transversions (C > G) or transitions (C > T) in the absence or failure of DNA repair [
12]. These A3A signature mutations are found in bladder cancer [
13], cervical cancer [
4,
14], breast cancer [
15,
16], ovarian cancer [
17], head and neck cancer [
18], and lung cancer [
19].
In cervical cancer, the variant A3A/B allele amplifies the somatic mutational burden caused by the virus, confirms the genomic instability of cervical cells, and increases viral diversity, leading to neoplastic changes in the cervix [
12,
19]. Considering that HPV infections occur worldwide [
20] and that cervical cancer is the third-most-common cancer in Brazilian women [
21], the study of possible sources of somatic mutations, such as the A3A/B polymorphism, is essential for understanding the mechanisms of the malignancy of cervical cell lesions.
The aim of this study was to investigate the association between the A3A/B polymorphism and HPV infection, the development of cervical intraepithelial lesions, and cervical cancer.
4. Discussion
In this study, we investigated for the first time in a Brazilian sample the association between A3A/B deletion and HPV infection and the development of intraepithelial lesions and cervical cancer. We also confirmed the description of the sociodemographic, sexual, and reproductive aspects associated with a greater likelihood of persistent HPV infection and the development of low-grade (LSIL) and high-grade (HSIL) squamous intraepithelial lesions and cervical cancer in the same population.
A higher frequency of patients without lesions or with LSIL than with HSIL or cervical cancer was expected in the infected group, when we considered the probability of the progression of HPV lesions [
2]. However, in our work, the proportion of women with HSIL and cervical cancer was higher than that of LSIL, possibly because the samples were from public health services. Brazilian epidemiological data show that 70% of diagnoses in women treated by public health services occur at an advanced stage of the disease, resulting in a worse prognosis [
24], which confirms the frequency found here.
The characteristics that were significantly associated with a higher frequency of HPV infection and the development of intraepithelial lesions in the studied group were age, smoking, monthly income, contraceptives use, preservatives use, age of first sexual intercourse, and educational stage.
Age is a characteristic of susceptibility to HPV infection and lesion development that has been well-established in previous studies. Women younger than 24 years or older than 55 years are more affected by infection, as shown in this study [
25,
26,
27,
28,
29]. Younger women have more intense sexual activity, which favors transmission of oncogenic HPV types [
25]. The early onset of sexual activity, sexarche, also favors infection because the immature epithelium is more exposed to the virus [
30]. At the other end of the scale is women 55 years of age and older, who become susceptible to HPV infection because of increased sexual activity due to increased life expectancy and quality of life [
31] and possible reactivation of latent viruses due to hormonal changes at the menopause [
26]. The persistence of HPV infection, immune response limited by aging or preexisting syndromes/diseases, and accumulation of mutations due to cellular senescence partially justify the increase in intraepithelial lesions and cervical cancer in a manner proportional to age [
1,
32,
33].
We have shown here that smoking is related to cancer development and HPV infection, confirming previous studies by our group [
27,
29,
34]. This association may be due to the fact that the DNA damage, immunosuppression, and epithelial dysplasia caused by cigarette components directly affect tumor development [
35].
Regarding contraceptives use, we found a significant association with HPV infection and cervical cancer development, which may be due to endometrial hyperplasia caused by hormonal imbalance, as noted in a recent meta-analysis [
36].
The use of preservatives reduces transmission and favors viral elimination [
37]. We hypothesize that the association of infection with the use of preservatives is due to the environment in which these patients are, i.e., possibly because these women are treated in a primary health care unit and have already received instructions for preservatives use. However, consistent with expectations, 80.8% of those with HPV did not use preservatives, and 80.1% of them developed significantly more severe lesions.
The association between ethnicity and susceptibility to HPV infection is complex, especially since the data were obtained based on patient self-report. The Brazilian population is highly admixed, with heterogeneous ancestry that includes both indigenous people and immigrants from Europe, Africa, and Asia [
38]. Moreover, race, as well as access to education and health services, is a determinant of socioeconomic and cultural advancement in Brazilian society. Therefore, we divided the individuals into Caucasians and non-Caucasians to not only obtain an unbiased view of the data but also take into account the social differences between these groups. Considering that Brazil is a continental country and the distribution of this heritage varies by region, it is advisable to point out that the study group is from the same region and represents a sample from the southern region, colonized mainly by European immigrants [
39].
The overexpression or variation of APOBEC3 genes has been cited as a possible source of aberrant DNA editing during replication, repair, or transcription, which is capable of inducing somatic mutations that cause genomic instability in cancer cells [
40]. Data from The Cancer Genome Atlas (TCGA) show the presence of signature mutations (C-to-T transitions or C-to-G transversions in T
CX sequences, where the underlined base is the mutated base, and X can be any base) [
41] attributed to APOBEC3A (A3A) and APOBEC3B (A3B) activity in breast, bladder, lung, head and neck, and cervical cancers [
4,
42].
The A3A/B polymorphism confers a greater susceptibility to cancer by generating a hybrid transcript between A3A and A3B 3′UTR that is more stable and more highly expressed in cells [
17]. A3A signature mutations are more common in cancer cells [
43], and the A3A protein is able to hypermutate nuclear DNA and generate double-stranded DNA breaks (DSBs), cause apoptosis, and promote mutations in cancer-causing genes, contributing to tumorigenesis [
8].
In cervical cancer, there is a microenvironment that favors the mutagenic activity of A3A: 1. The accumulation of A3A is stimulated by the production of interferons in response to viral infection [
42]; 2. the direct action of the oncoprotein E7 of HPV-HR inhibits the proteassomal degradation of A3A, making it more stable [
44]; and 3. the replication stress promoted by HPV-HR increases cellular ssDNA exposure and makes them susceptible to A3 restriction [
43].
In our study, we found no association of A3A/B polymorphism with infection, LSIL, HSIL, or cervical cancer, which is consistent with previous studies that failed to detect this association in other sample groups [
15]; however, this result is in contrast to another study that showed this association [
44].
The absence of the association of the variant allele with cancer found in our study reflects the population in southern Brazil. This result is consistent with a previous breast cancer study conducted with an independent sample from the same geographic region [
45]. The frequency of polymorphisms depends on the ancestry of the sample group. The frequency of A3A/B varies among Asian (37%), Native American (57.7%), European (6%), and African (0.9%) populations [
6,
13,
17]. Here, as mentioned previously, the study group consisted predominantly of self-identified Caucasian women with probable European ancestry, so the prevalence of this genetic variant is considerably lower. The low frequency of the variant allele in studies with independent samples may be explained by the Caucasian characteristics of the region, with the variant allele being more common in Asian populations.
Nevertheless, there is no consensus in the literature on the role of A3A and A3B in the development of lesions and cervical cancer [
46], highlighting the need for studies such as ours to investigate this association. A3B is considered a major protein responsible for mutagenesis in cancer, which may explain the lack of association of the polymorphism (lack of translated A3B region) with carcinogenesis [
16]. In contrast, other studies cite A3A as the source of somatic mutations in the host genome and confirm a significant association of the polymorphism with cervical lesions [
11,
47]. This inconsistency in defining the A3 mechanism in cervical cancer makes it difficult to interpret the role of the polymorphism in this situation.
Regarding the study limitations, sampling by convenience may have an influence on the homogeneity of the sample characteristics, mainly the sociodemographic variables. Moreover, the experiment design does not include longitudinal monitoring of the infection and lesion development in the participants, which means that a causal relation of the statistically significant variables is not possible to be assumed. Finally, the specific detection of high- and low-risk infection could be helpful to better characterize the frequency and association of the polymorphism between the groups with lesions or cancer in this population. Despite the limitations that cannot be ignored, the importance of this work is given by the discussion of a genetic variation that is so rarely studied in the different branches of research. The non-association found does not invalidate the valuable discussion about a group of genes intrinsically related to the antiviral response and provides support for future basic research.