The immunological recognition of cancerous and virus-infected cells requires the degradation of endogenous protein into peptides via the ubiquitin/proteasome and other proteolytic pathways, which are translocated from the cytosol into the lumen of the endoplasmic reticulum (ER) by the transporter-associated with antigen processing (TAP) (Lehnert and Tampé, 2017). In the ER, the peptide binds to the classical HLA class I (HLA-I) molecules to be further transported to the membrane surface for recognition by CD8+ cytotoxic T cells (CTLs) and NK cells (Thomas and Tampé, 2021).

The presence of single nucleotide polymorphisms (SNPs) in TAP genes may influence the selection of peptides bound to HLA-I molecules, modulating the diversity of peptide transport from cytosol to ER (Colonna et al., 1992; Lehnert and Tampé, 2017). Most SNPs in TAP genes (TAP1 and TAP2) are close to the peptide-binding site, and some are in the protein transmembrane domain (Praest et al., 2018). These coding region SNPs at the TAP1 and TAP2 genes have been associated with rheumatoid arthritis (Zhang et al., 2002), systemic lupus erythematosus (Correa et al., 2003), esophageal cancer (Zou et al., 2015; Guo et al., 2016), multiple myeloma and chronic lymphocytic leukemia (Ozbas-Gerceker et al., 2013), and tuberculosis (Sunder et al., 2011; Thu et al., 2016). Additionally, TAP gene SNPs have been associated with cervical precursor lesions of cervical cancer (CC) (Einstein et al., 2009; Natter et al., 2013).

Human papillomavirus (HPV) is a sexually transmitted infection, accounting for 99.7% of CC cases (Walboomers et al., 1999). In Brazil, CC is the third most common cancer among women aged 15 to 44 (Bray et al., 2018). The host’s immune system clears most cervical HPV infections within 2 to 3 years; however, in many cases, the infection can persist longer, leading to low-grade (LSIL) and high-grade lesions (HSIL). The histological lesions are classified as cervical intraepithelial neoplasia (CIN) 1, 2, or 3, which can progress to invasive cancer (Oyervides-Muñoz et al., 2018). Primary cancer transforming factors mediated by high-risk oncogenic HPV (HrHPV), particularly E6 and E7 oncoproteins, interact with the host cells to manipulate cellular processes and promote cell transformation (Pal and Kundu, 2020). Little attention has been devoted to the relationship between HPV infection and CINs associated with the TAP gene variability.

To understand the contribution of TAP on HPV-induced cervical lesions, we analyzed five SNPs at the coding regions of TAP genes and evaluated the TAP genes expression and protein levels in cervical samples obtained from women infected or not by HPV.

The recognition, processing, and presentation of antigens to immune cells are complex and require a well-orchestrated activity of several genes to eliminate the aggressor because infectious agents have developed mechanisms to evade the immune response, adapting to the adversities of the environment and persisting in their parasitism (Estêvão et al., 2019; Mantel et al., 2022). In this study, we showed that TAP tissue levels changed in HPV infection and that TAP gene variants modulated TAP expression, which may contribute to the progression of cervical lesions.

The TAP1 and TAP2 genes are conserved with few mutations in the coding region assigned in the HLA database, with six and five isoforms identified, respectively (Robinson et al., 2020). As most of the reported mutations are in the cytosolic nucleotide-binding domains (NBDs) of the TAP heterodimeric complex, in addition to being possible changes in mRNA expression and structural stability and at the protein level, variants of the TAP gene can cause functional alterations, affecting ATP binding capacity and peptide transport efficiency (Lapinski et al., 2001; Lankat-Buttgereit and Tampé, 2002).

In the TAP1 gene, the p.Ile333Val SNP in exon-4 and the p.Asp697Gly SNP in exon-10 are the most studied due to their association with disease. The p.Ile333Val A allele was not reported in other primates, suggesting associated with human evolution, and is in LD with the Asp697Gly A allele (Tang et al., 2001). Because the p.Asp697Gly variant was mapped to the NBD region and was associated with recurrent respiratory papillomatosis in Mexicans (Palomares-Marin et al., 2020) and with cervical lesions in Americans and Europeans (Mehta et al., 2007; Deshpande et al., 2008), we focused on studying it. We showed that the frequency of the G allele for the p.Asp697Gly SNP of TAP1 in northeastern Brazilians (18%) was similar among Europeans (14.4%), Africans (23.3%), Asians (16.5%), Latin Americans with Afro-Caribbean ancestry (16.3%), and Latin Americans with European and Native American ancestry (17.6%) (Phan et al., 2020).

We demonstrated that women who carry the G allele in homozygosis (3%) have a higher risk of having aneuploidy in cervical cells ( Table 2 ). In the acute phase of HPV infection, the viral genome remains in episomal form within the infected epithelial cell; during epithelial desquamation, recognizing the HPV antigen increases the inflammatory infiltrate. Integration of the HPV genome into the human genome leads to genetic instability and aneuploidy of cervical cells (Martins et al., 2014; Li et al., 2018). Integration is also responsible for activating viral oncoproteins, which co-opt effector molecules of the immune system, contributing to the progression to malignancy (Estêvão et al., 2019).

Our findings also showed a relationship between aneuploidy and low levels of TAP-1 ( Figure 1C ). Studies have explicitly shown the downregulation of TAP1 and HLA-I expression by HPV-16 E7 and HPV-18 E7 oncoproteins (Georgopoulos et al., 2000; Vermeulen et al., 2007; Li et al., 2010). HPV-16 E5 also interacts with the heavy chain of the HLA-I molecule through the hydrophobic region of the E5 protein and retains HLA-I in the Golgi and ER (Ashrafi et al., 2006). The lower expression of TAP, resulting from the inhibition of gene transcription, protein translation, or post-translational events, decreases the presentation of viral peptides coupled to HLA-I molecules on the cell surface, decreasing the activation of cytotoxic T cells and inflammation, consequently, viral elimination. In this case, HLA-I molecules can present peptides from the membrane, cytosolic, or secreted proteins through a less efficient TAP-independent mechanism, which may also contribute to the virus evading the NK cell-mediated immune response (Mantel et al., 2022).

Unfortunately, we cannot draw strong conclusions from the results with TAP-1 because the analysis at the protein level was compromised by the small number of samples analyzed. Also, there was no significant difference associating the mRNA expression level with cervical lesions, aneuploidy, or the presence of HPV infection. As a preliminary analysis of the data, we could hypothesize that the presence of HPV increases tissue TAP-1 expression ( Figure 1D ), which is in line with the increase in protein in the inflammatory phase of the low-grade lesion ( Figure 1A ) in response to the antigenic stimulus (Gostout et al., 2003; Kelly and Trowsdale, 2019). Still, we cannot confirm the association of the mutant allele with the reduction in protein levels since the association between the allele and aneuploidy may not be associated with expression intensity but with transport efficiency; this was not assessed in this study (Praest et al., 2018). The mutant G allele encodes an aspartic acid residue at position 697 of the TAP-1 protein, close to the peptide binding site, which may influence the peptide-TAP-1 interaction and its transport to the endoplasmic reticulum (Quadri and Singal, 1998).

In this study, we also evaluated the association of TAP2 variants with aneuploidy and the severity of the cervical lesions, irrespective of the TAP1 gene. The Val379Ile SNP codes an amino acid localized in the transmembrane domain region (TMD2) of the TAP-2 protein, and the p.Ala565Thr, p.Arg651Cys, and p.Thr665Ala SNPs code for amino acids in the cytosolic nucleotide-binding domain (NBD2) of TAP-2 (Praest et al., 2018). We showed that the rare A allele for TAP2 p.Val379Ile and p.Ala565Thr SNPs, in single or double doses, were associated with the risk of cervical cell aneuploidy, considered a hallmark of cervical lesion progression (Ben-David and Amon, 2020). In addition, Brazilian carriers of the rare G allele for p.Thr665Ala SNP were at risk for developing CIN and HSIL.

Regarding the presence of HPV, carriers of the wild-type genotype for all SNPs studied in homozygosis showed increased expression of mRNA and TAP-2 protein as part of the elaboration of the immune response to the infection. However, carriers of the p.Ala565Thr mutant allele are high producers of TAP-2, suggesting that the protein is more stable, as mRNA and protein levels were not affected by the viral infection ( Figure 4B ). On the other hand, carriers of the mutant p.Val379Ile allele respond to the presence of HPV with a more expressive increase in protein levels than those produced by carriers of the wild-type allele in response to the presence of HPV ( Figure 4A ), suggesting a post-translational modulation in the expression of TAP, since the mRNA levels were not altered by the HPV infection (Pedroza-Torres et al., 2014; Lazaridou et al., 2020).

Controversies about the association of TAP1 and TAP2 gene variants with the risk of cervical cancer may arise from the ethnic diversity of the studied populations, whether the study was in precancerous or cancerous lesions, and depending on the parameter used for the comparison between groups, that is, cytological or histological changes, or abnormal cervical cell ploidy. For example, in Austrian women, the TAP1 p.Asp697Gly SNP showed no association with CIN compared to healthy controls (Natter et al., 2013). However, the G allele was associated with a reduced risk of HSIL (CIN 2 and CIN 3) in American women infected by HPV (Einstein et al., 2009); and protection against the development of cervical cancer in Chinese women (Yang et al., 2021). Another example is the lack of association between the SNPs TAP2, p.Val379Ile, p.Ala565Thr, and p.Arg651Cys, with histological and cytological alterations and cervical cancer reported in Chinese (Yang et al., 2021) and Swedish (Ivansson et al., 2008) women when compared to the control group. However, the TAP2 SNP p.Val379Ile A allele was associated with the risk of esophageal cancer in Chinese patients with HPV+ (Cao et al., 2005) and with cervical cell aneuploidy in our study population. Also, the T allele of the SNP TAP2 p.Arg651Cys was associated with an increased risk of cervical cancer in women from Bali, a region of Indonesia (Mehta et al., 2015), while in Dutch women with HPV+, the risk was attributed to the C allele (Mehta et al., 2007).

Regarding the p.Thr665Ala SNP of the TAP2 gene, the women’s ancestry also influenced cancer susceptibility, especially cervical cancer (Liu et al., 2019). We found that Brazilian carriers of the rare G allele were at risk for developing CIN and HSIL. In contrast, the A allele of the p.Thr665Ala SNP was associated with cervical cancer in Swedish women (Ivansson et al., 2008). In addition, neither allele was related to cervical cancer in New Yorkers (Einstein et al., 2009) and Austrian (Natter et al., 2013) women. A study of 1,306 Swedish women with cervical cancer showed a strong linkage disequilibrium between the TAP2 p.Thr665Ala and DQB1 alleles. The authors investigated the distribution of TAP2 alleles among women that expressed DQB1 risk (*0301, *0402, *0602) or protection alleles (*0501 and *0603), showing less difference in TAP2 alleles distribution between cases and controls among carriers of the DQB1 risk alleles, suggesting a hitchhiking effect of TAP2 with DQB1 alleles (Ivansson et al., 2008). This may explain the controversies in the association of the TAP2 p.Thr665Ala allele with the risk of cervical cancer in different populations. Moreover, the HLA system comprises a large region with many genes in linkage disequilibrium. Therefore, other genes in different populations may modulate the association between the presence of a specific allele and the severity of the cervical lesion (Dawkins et al., 1999).

Since the effect of each allele may differ, and the resultant product is more important for evaluating the impact on the evolution of the cervical lesion (Mehta et al., 2007), we considered the sequence of each haplotype observed in our population to assess its clinical significance ( Table 5 ). The lower genetic diversity may result from selecting more efficient variants for the best innate or adaptive immune response against environmental pathogens (Zeberg and Pääbo,). This is the case of the TAP2A haplotype formed by the wild-type homozygous genotypes of the polymorphic sites of the regions responsible for peptide binding and transport across the endoplasmic reticulum membrane, which corresponds to 68.6% of cases. Considering that the TAP2A and TAP2C haplotypes account for 77.1% of the identified types and because the median TAP-2 levels in tissue were similar, we assumed these levels as standard and classified them as intermediate. The TAP2A and TAP2C haplotypes differ only in the p.Val379Ile SNP. The TAP-2 levels associated with the TAP2D and TAP2E haplotypes, which correspond to 12.9% of the observed haplotypes, were higher, and carriers were at risk for HSIL and aneuploidy. The sequences of the TAP2D and TAP2E haplotypes also differ only in the p.Val379Ile SNP, suggesting that this SNP may not be directly related to the increase in TAP-2. The minor allele A of the p.Ala565Thr SNP seems to play a role in the increase of TAP-2 levels in TAP2D and TAP2E carriers, while the mutant allele G of the p.Thr665Ala SNP is involved in the decrease of the protein in TAP2B carriers, whose are also at increased risk of serious injury. Few association studies on TAP2 haplotypes and cervical cancer are available. In one of them, Natter et al. (2013) also reported that Caucasian women from Austria with TAP2C haplotype were at lower risk for the development of CIN (Natter et al., 2013).

The TAP2A/TAP2A carriers responded to HPV infection, increasing the TAP2 mRNA transcripts compared to women carriers non-infected but without a corresponding protein level, which may indicate an increase in the protein turnover or a viral posttranslational regulation. Gameiro et al. (2017) analyzed TAP1 and TAP2 expression using TCGA (The Cancer Genome Atlas) data and found higher expression in cervical carcinoma tumors in HPV+ samples than in normal control tissue (Gameiro et al., 2017). TAP1 and TAP2 expressions were also higher in EBV-associated gastric carcinoma samples than in normal control tissues or other gastric carcinoma subtypes (Ghasemi et al., 2020). Interestingly, the TAP2A/TAP2C carriers responded to the HPV infection with unchanged TAP2 expression and a concomitant increase in protein levels. The rare allele A of the p.Val379Ile SNP in the TAP2C haplotype elevates the TAP-2 level in HPV-infected tissue and CIN 3 samples.

The increase in TAP2 expression and protein level may be related to greater recognition and processing of the viral peptide, increasing the adaptive immune response (Thomas and Tampé, 2021). Nevertheless, the viral modulation of TAP2 expression, possibly mediated by p.Val379Ile SNP allele A, may contribute to cervical cancer progression from precursor lesions (Fowler and Frazer, 2004; Mpakali and Stratikos, 2021).

Some viruses have an immune response evasion mechanism based on the release of inhibitory molecules, which decrease TAP levels (Estêvão et al., 2019; Mantel et al., 2022). In the reduction of TAP, the viral peptide is less efficiently transported into the endoplasmic reticulum, and the assembly of the HLA-peptide complex is decreased. The empty HLA-I molecule is unstable and degraded, leading to its depletion at the cell surface for antigen presentation to cytotoxic T cells. However, it was observed that some HLA-B allotypes had increased expression in the absence or decrease of TAP. These allotypes are stable and highly efficient in peptide loading for a broad specificity of peptides, allowing to augment 5 to 10-fold the viral peptide presentation on the surface through a mechanism independent of TAP (Geng et al., 2018).

The main limitation of our study was the low number of biopsies, which compromised the TAP-1 and TAP-2 protein analyses, reducing the possibility of multiple comparisons. Nevertheless, the evaluation of mRNA expression in exfoliative cervical cell samples allowed the analysis of a more significant number of samples, increasing the strength of the observed association. In addition, we chose to use the DNA index in the final association analysis because aneuploidy is related to the integration of the HPV genome into cervical cells, and the increase in cervical dysplasia has a prognostic value for cervical cancer (Heselmeyer et al., 1996; Ben-David and Amon, 2020).

The results obtained with the diplotypes formed by the four SNPs in the TAP2 gene allowed the identification of low, intermediate, and high TAP-2 producers in women. We found the association of some haplotypes with aneuploidy and the presence of HSIL. We showed the increase of TAP-2 protein in HPV+ and CIN 3 cervical lesions, a physiologic mechanism for eliminating infected and damaged cells. In addition, we addressed the possible mechanism of immune evasion exerted by HPV in modulating the expression of TAP variants, which may favor viral persistence, contributing to the evolution of the disease to malignancy. Further functional studies will confirm the relationship between TAP gene variants and TAP2 expression, HPV infection, and disease outcome.

The data presented in the study are deposited in the Figshare repository, and accession links and DOIs can be found below: https://doi.org/10.6084/m9.figshare.21263343, https://doi.org/10.6084/m9.figshare.21265785, https://doi.org/10.6084/m9.figshare.21265848.v1.

The studies involving human participants were reviewed and approved by The Ethics Committee of the Instituto Aggeu Magalhães approved this study under the protocol CAAE:51111115.9.0000.5190. The patients/participants provided their written informed consent to participate in this study.

FM, NL-S conceived, designed the study, did the formal analysis, and wrote the paper. FM, MS, NS, TG, RG, FG, LP conducted the experimental work. FM, LP, MR, SW, MM followed-up patients and performed cytopathological and coloscopy evaluations. NL-S and ED applied for financial support and managed the project. All authors contributed to the article and approved the submitted version.