Abstract
Purpose
Every year, more than half a million women are diagnosed with cervical cancer (CC). Individual factors may contribute to the cervical cancer development, such as immunogenetic variation. CXCL12/CXCR4 axis is involved in tumor progression and aggressiveness. In the present study, we aimed to investigate a possible association between two single-nucleotide variants (CXCL12 rs1801157 and CXCR4 rs2228014) with HPV infection and cervical cancer development.
Methods
PCR technique was used to test HPV positivity in 424 women, in which the allelic frequency of CXCL12 rs1801157 and CXCR4 rs2228014 was also assessed by PCR-restriction fragment length polymorphism.
Results
CXCL12 rs1801157 was associated with HPV infection in the allelic distribution as well in the codominant, dominant and recessive genetic models; as well with squamous intraepithelial lesions (SIL) and CC in the codominant and dominant models. CXCR4 rs2228014 was associated to HPV infection in the codominant model and allelic distribution; as well with SIL/CC in the codominant, dominant and allelic models. Independent associations were found for CXCL12 AA genotype and HPV infection, SIL and CC development, as well as, CXCR4 allele T and HPV infection and CC. The variants interaction analysis demonstrated that the presence of both polymorphisms increases the susceptibility of HPV infection in 10.1 times, SIL (2 times) and CC development in 4.2 times.
Conclusions
This is the first study demonstrating that the interaction of CXCL12 and CXCR4 variants contributes to the increased susceptibility of HPV infection, squamous intraepithelial lesions and cervical cancer development.
Keywords: Cervical cancer, rs1801157, rs2228014, Chemokine, Polymorphism
Background
Every year more than half a million women are diagnosed with cervical cancer (CC) and from these, 300 000 die due to the disease (Cohen et al. 2019). Most of the deaths, 85%, occur in less developed countries (Bray et al. 2018). High-risk human papillomavirus (HR-HPV) is the main cause of cervical cancer, between the different histological subtypes the squamous cell carcinoma and adenocarcinoma are the most common representing 70% and 25%, respectively (Cohen et al. 2019). Even though HPV is necessary to the cervical cancer development, the virus alone is not sufficient to transform the cervix cells, therefore, it is important to investigate individual intrinsic factors, such as genetic variation in immunological genes that act as important agents to susceptibility, progression and disease outcome (Torres-Poveda et al. 2016).
CXCL12 gene is located at 10q11.1 and there are three GC boxes and one CAAT box on the gene promoter region that are binding sites for the transcription factors SP1 and CTF, respectively (Shirozu et al. 1995). CXCL12 binds to specific G protein-coupled seven transmembrane receptor CXCR4, which activate a biological axis responsible for cell trafficking, activation and differentiation. CXCR4 gene is located at 2q21 and is expressed on multiple cell types, including lymphocytes, hematopoietic stem cells, endothelial cells, epithelial cells, cancer cells and stromal fibroblasts (Guyon 2014). The CXCL12 rs1801157 and CXCR4 rs2228014 single-nucleotide variants (SNV) have been associated to glioma (Chang et al. 2015), breast cancer (Guembarovski et al. 2018), hepatocellular carcinoma (Qin et al. 2018) and lymphocytic leukemia (Butrym et al. 2016).
Regarding the HPV infection, the CXCR4 upregulation in CC, promoted by the virus, was significantly associated with the histologic grade of CC (Mortezaee 2020). CC cells expressing CXCR4 (but not CXCL12) may initiate their migration and invasion towards a CXCL12 gradient (Zhang et al. 2007; Zanotta et al. 2016) and the CXCL12/CXCR4 axis may affect tumor malignant progression and clinical aggressiveness (Lecavalier-Barsoum et al. 2018). Studies involving CXCL12 and CXCR4 SNVs in CC are scarce. CXCL12 rs1801157 was previously associated to HPV infection (Okuyama et al. 2018) but not associated to the development of squamous intraepithelial lesion (SIL) (Tee et al. 2012). No data about CXCR4 rs2228014 and infection, lesions and CC may be found in the literature.
In this context, in the present study, we investigated the possible association of CXCL12 rs1801157 and CXCR4 rs2228014, and their interaction, with HPV, SIL and CC susceptibility.
Methods
Ethical approval, patients and samples
This study was approved by the Institutional Ethics Committee Involving Humans at State University of Londrina, Londrina—PR, Brazil (CEP/UEL 133/2012; CAAE 05505912.0.0000.5231). The study purpose and procedures were explained to all patients and written informed consent was obtained prior samples collection.
Between 2013 and 2018, 424 women were enrolled. They were recruited in public health services in Londrina—PR, Brazil. Biological materials [cervical secretion, blood samples or formalin-fixed-paraffin-embedded (FFPE) tumor tissues], were collected from participants who attended CC prevention programs at an ambulatory colposcopy facility of Intermunicipal Consortium of Health of the Middle Paranapanema, at University Hospital and Clinic Center of the State University of Londrina, at Basic Healthcare Units in Londrina—PR, Brazil and at Cancer Hospital of Londrina (CHL).
After sample collection, cytobrushes were stored in 2 mL TE buffer (10 mM Tris–HCl, 1 mM EDTA pH 8.0) at − 10 °C until analysis. Peripheral blood was drawn into sterile syringes containing EDTA as anticoagulant and stored at − 10 °C until analysis. The cervical FFPE tumor tissue samples were provided by CHL. A structured questionnaire was applied to all the patients to collect socio-demographic and sexual behavioral data. Participants were stratified based on the presence or absence of HPV DNA, as tested by PCR and based on SIL and CC diagnosis, as determined by cervical cytology. Clinical and pathological data of cervical cancer patients were available from CHL. Clinical staging was determined according to the International Federation of Gynecology and Obstetrics (FIGO) criteria. Pathological features analyzed included: histological classification and histopathological grade, according to World Health Organization (WHO) histological classification of tumors of the uterine cervix.
DNA extraction
Genomic DNA was obtained from: (a) cervical cytobrushes using DNAzol (Invitrogen™ Inc., Carlsbad, CA, USA), (b) peripheral blood using Biopur Mini Spin Plus Kit (Biometrix®, Curitiba—PR, Brazil), (c) FFPE tumor tissues using PureLink™ Genomic DNA Mini Kit (Invitrogen™, Carlsbad, CA, EUA). All extraction were performed according to the manufacturer’s instructions and stored at − 10 °C until use. DNA concentration was measured at 260 nm on a NanoDrop 2000c™ Spectrophotometer (Thermo Fisher Scientific, USA) and purity was assessed by A260/A280 ratio.
HPV detection
HPV was detected by PCR using the primers MY09 (5′-CGTCCMAARGGAWACTGATC-3′) and MY11 (5′-GCMCAGGGWCATAAYAATGG-3′), which are designed to amplify a conserved region of approximately 450 bp in the HPV L1 gene (GenBank Accession number: AJ236888). Reaction conditions were 190 nM of dNTPs, 500 nM of each primer, 2 mM of MgCl2, 1X of Buffer, approximately 80 ng of DNA and 1,25U of Taq polymerase (Invitrogen™), with an annealing temperature of 55 °C. β-globin gene amplification (268 bp) was performed as an internal control, using primers GH20 (5′-GAAGAGCCAAGGACAGGTAC-3′) and PC04 (5′-CAACTTCATCCACGTTCACC-3′) (Da Silva et al. 2012) under the same conditions of HPV PCR. Reactions without template DNA were used as negative control to test for contamination and DNA from HeLa cells, which are stably integrated with HPV18, was used as positive control. PCR products were electrophoresed on 10% polyacrylamide gel and stained with silver nitrate.
CXCL12 rs1801157 and CXCR4 rs2228014 polymorphisms detection and genotyping
Genomic DNA from peripheral blood and FFPE tumor tissue samples was used to amplify regions of the CXCL12 and CXCR4 genes. Primers used for the amplification of the CXCL12 gene were designed according to the nucleotide sequence deposited in GenBank which code is L36033. The primers forward 5′-CAGTCAACCTGGGCAAGCC-3′ and reverse:5′-CCTGAGAGTCCTTTTCGCGG-3′ were utilized to amplify the 3’UTR of CXCL12. For the amplification of the CXCR4 gene the following primers were used: 5′-AACTTCCTATGCAAGGCAGT-3′ (forward) and 5′-TATCTGTCATCTCTCACT-3′ (reverse). PCR reactions were conducted using 100 nM of dNTPs, 0.25 µM of each primer, 1.5 mM of MgCl2, 1X of Buffer and 1U of Taq polymerase for CXCL12 polymorphism. For CXCR4 was used 0.75 mM de MgCl2; 100 nM de dNTP, 0.2 μM of each primer, 1U of Taq polymerase and approximately 100 ng of DNA in both cases (Invitrogen™).
The CXCL12 and CXCR4 products amplification correspond to a 293 bp and 236 bp fragments, respectively. The enzymatic restriction was performed using PCR products in the presence of the restriction enzymes MspI (at 37 °C for 1 h) for CXCL12 and BccI (at 37 °C, for 1 h) for CXCR4 (New England Biolabs, Ipswich, MA, USA). For CXCL12, MspI cleaves the amplified fragment of DNA in the presence of an adenine, producing fragments of 100 bp and 193 bp and in the presence of a guanine, the fragment of 293 bp remains intact. For CXCR4, BccI cleaves in the presence of a thymine, generating fragments of 103 and 133 pb and in the presence of a cytosine, the fragment of 236 bp remains.
Statistical analysis
HPV infection’s frequency reported in this paper was compared to the overall prevalence of cervical HPV in Brazil through the one-sample general z test [16] and the exact Clopper-Pearson confidence interval for the observed proportion was calculated (Clopper-Pearson 1934). Pearson’s Chi-squared test of independence (χ2) followed by Bonferroni correction was employed in the analysis of contingency tables to identify differences in baseline features between HPV-infected and non-infected women, and those bearing SIL and CC. Allele frequency was calculated as [1(h + 2H)]/2 N, where h represents the heterozygous genotype, H is the homozygous genotype, and N is the sample size for each population. χ2 test followed by Bonferroni correction was employed in SNVs CXCL12 rs1801157 and CXCR4 rs2228014 distribution testing. Adjusted odds ratio (OR) with 95% confidence interval (95% CI) were estimated by a binary or multiple logistic regression in the forced entry method to test the association between genetic models of inheritance of the SNVs rs1801157 and rs2228014 and HPV infection, SIL and CC, as appropriate, adjusting for baseline factors that were associated with HPV infection, SIL and CC diagnosis with a significance level lower than 0.10 found in the χ2 test (in this case, just for the model adjustment purposes). These variables were included in a forward stepwise variable selection method. The gene–gene association analysis between the SNVs was also determined by logistic regression models. All tests were two-tailed, and data were analyzed in SPSS Statistics 22.0 software (SPSS Inc., Chicago, Illinois, USA), considering a significance level alpha set at 0.05.
Results
Patient baseline characteristics
Women (n = 424) were included and categorized as Control (187/56.2%), all of them HPV uninfected and negative for intraepithelial lesion and malignancy (NILM), SIL (squamous intraepithelial lesion, 63/14.8%) and CC (cervical cancer, 92/21.7%). We also grouped HPV-infected women (146/43.8%), formed by those with SIL or NILM that were positive for HPV. Sociodemographic, sexual lifestyle and gynecological and obstetric data for Control, SIL and CC are summarized in Tables 1 and 2. To provide an epidemiological perspective, we compared the frequency of cervical HPV infection found in our cohort (43.8%) (i.e., composed by women from the north of the Paraná state (BRA)) with the overall cervical HPV prevalence in Brazil (25.41%), recently reported in a meta-analysis conducted by Colpani et al. (2020) and our results significantly differ from national data (z statistic: 7.70; p < 0.001; 95% CI of observed proportion: 38.40–49.31%).
Table 1.
Sociodemographic characteristics of patients and controls
| Variable | Control | SIL | CC | p value* | |||
|---|---|---|---|---|---|---|---|
| n | (%) | n | (%) | n | (%) | ||
| Knowledge about HPV | 0.003 | ||||||
| No | 35 | (18.7) | 20 | (31.3) | 16c | (41.1) | |
| Have ever heard | 104 | (55.6) | 30 | (46.9) | 16 | (41.0) | |
| Yes | 48 | (25.7) | 14 | (21.8) | 7 | (17.9) | |
| Knowledge about ways of HPV transmission | 0.008 | ||||||
| No | 88 | (47.1) | 34 | (53.1) | 29c | (74.4) | |
| Yes | 99 | (52.9) | 30 | (46.9) | 10 | (25.6) | |
| Age (years) | < 0.001 | ||||||
| ≤ 24 | 10 | (5.3) | 13 | (20.3) | 1 | (1.1) | |
| 25–34 | 43 | (23.0) | 20 | (31.3) | 8 | (8.7) | |
| 35–44 | 46 | (24.6) | 14 | (21.8) | 27 | (29.7) | |
| 45–54 | 55 | (29.4) | 11 | (17.2) | 21 | (23.1) | |
| ≥ 55 | 33 | (17.7) | 6 | (9.4) | 34c | (37.4) | |
| Monthly incomea | 0.004 | ||||||
| < 1 minimum wage | 83 | (44.9) | 36 | (57.1) | 26c | (72.2) | |
| 1–3 minimum wages | 95 | (51.4) | 27 | (42.9) | 8 | (22.2) | |
| > 3 minimum wages | 7 | (3.8) | 0 | (0.0) | 2 | (5.6) | |
| Smoking status | 0.003 | ||||||
| No | 142 | (75.9) | 36 | (56.3) | 56 | (62.2) | |
|
Yes Former smoker |
29 | (15.5) | 24c | (37.4) | 23 | (25.6) | |
| 16 | (8.6) | 4 | (6.3) | 11 | (12.2) | ||
| Educational levelb | < 0.001 | ||||||
| Incomplete elementary | 58 | (31.4) | 26 | (22.6) | 62c | (45.1) | |
| Complete elementary | 23 | (12.4) | 6 | (14.5) | 11 | (19.1) | |
| Incomplete secondary | 26 | (14.1) | 10 | (20.9) | 4 | (9.4) | |
| Complete secondary | 59 | (31.9) | 17 | (31.0) | 9 | (13.9) | |
| Incomplete higher education | 6 | (3.2) | 3 | (7.4) | 0 | (0.0) | |
| Complete higher education | 13 | (7.0) | 1 | (3.6) | 5 | (12.5) | |
| Marital status | 0.001 | ||||||
| Single | 135 | (72.2) | 41c | (21.8) | 6 | (6.6) | |
| Married/Civil partner | 16 | (8.6) | 38 | (59.4) | 54 | (60.0) | |
| Divorced | 25 | (13.4) | 9 | (14.1) | 15 | (16.7) | |
| Widowed | 11 | (5.8) | 3 | (4.7) | 15c | (16.7) | |
*Analysis by two-sided Chi-square (Χ²) test and p < 0.05 set as significance level (bold) (SPSS Inc., Chicago, Illinois, USA). Control group is negative for HPV and intraepithelial malignancy; SIL squamous intraepithelial lesion; CC cervical cancer. Some categories did not complete the total of patients due to lack of data
aBased on Brazilian minimum wage (approximately U$ 206.00)
bBased on Brazilian educational system
c p<0.05, tested by Bonferroni post hoc test for multiple comparisons
Table 2.
Sexual behavioral and reproductive characteristics of patients and controls
| Variable | Control | SIL | CC | p value* | |||
|---|---|---|---|---|---|---|---|
| n | (%) | n | (%) | n | (%) | ||
| Contraceptive method: condom | 0.215 | ||||||
| Yes | 21 | (11.3) | 9 | (14.1) | 17 | (19.4) | |
| No | 165 | (88.7) | 55 | (85.9) | 72 | (80.6) | |
| Contraceptive method: hormonal | 0.009 | ||||||
| Yes | 54 | (29.0) | 26 | (40.6) | 42a | (47.2) | |
| No | 132 | (71.0) | 38 | (59.4) | 47 | (52.8) | |
| Number of pregnancies | 0.001 | ||||||
| 0 | 21 | (11.2) | 10 | (15.6) | 4 | (4.4) | |
| 1 | 30 | (16.0) | 8 | (12.5) | 12 | (13.6) | |
| 2 | 57 | (30.5) | 15 | (23.4) | 20 | (22.0) | |
| 3 | 41 | (21.9) | 16 | (25.1) | 17 | (18.7) | |
| 4 | 18 | (9.6) | 10 | (15.6) | 9 | (9.4) | |
| ≥ 5 | 20 | (10.8) | 5 | (7.8) | 29a | (31.9) | |
| Abortion | 0.477 | ||||||
| No | 136 | (79.1) | 40 | (72.7) | 67 | (73.6) | |
| Yes | 36 | (20.9) | 15 | (27.3) | 24 | (26.4) | |
| Age at first sexual intercourse (years) | 0.011 | ||||||
| ≤ 17 | 92 | (49.5) | 44a | (69.8) | 52 | (61.2) | |
| ≥ 18 | 94 | (50.5) | 19 | (30.2) | 33 | (38.8) | |
| Age at menarche | 0.511 | ||||||
| ≤ 11 | 42 | (22.7) | 18 | (28.6) | 9 | (23.1) | |
| 12 | 42 | (22.7) | 17 | (27.0) | 10 | (25.6) | |
| 13 | 46 | (24.9) | 16 | (25.4) | 6 | (15.6) | |
| ≥ 14 | 55 | (29.7) | 12 | (19.0) | 14 | (35.9) | |
| Sexual partners during the lifetime | 0.002 | ||||||
| 1 | 74 | (39.6) | 8 | (12.5) | 14 | (36.8) | |
| 2–3 | 62 | (33.2) | 27 | (42.2) | 11 | (28.9) | |
| ≥ 4 | 51 | (27.3) | 29a | (45.3) | 13 | (34.3) | |
| Prior exam | < 0.001 | ||||||
| No | 7 | (3.8) | 0 | (0.0) | 27a | (42.0) | |
| Yes | 179 | (96.2) | 64 | (100.0) | 51 | (58.0) | |
*Analysis by two-sided Chi-square (Χ²) test and p < 0.05 set as significance level (bold) (SPSS Inc., Chicago, Illinois, USA). Control negative for HPV and intraepithelial malignancy; SIL squamous intraepithelial lesion; CC cervical cancer. Some categories did not complete the total of patients due to lack of data
ap < 0.05, tested by Bonferroni post hoc test for multiple comparisons
A higher frequency of SIL and CC was observed in women with no knowledge about HPV and its ways of transmission (p = 0.03 and p = 0.008, respectively), 55 years old or more (p < 0.001), receiving < 1 minimum wage (p = 0.004), smokers (p = 0.005), women with incomplete elementary education (p < 0.001) and single (p = 0.001) (Table 1). Regarding the sexual lifestyle and gynecological and obstetric data, a higher frequency of SIL and CC was found in women who use hormonal contraceptive (p = 0.009), who have had sex with or under 17 years old (p = 0.011), had four or more sexual partners during lifetime (p = 0.005), who had five or more number of pregnancies and women without prior cervical exam (p = 0.001) (Table 2).
CXCL12 and CXCR4 SNVs distribution among diagnostic groups
We compared the allelic frequency of CXCL12 rs1801157 and CXCR4 rs2228014 with genomic big data from the Trans-Omics for Precision Medicine (TOPMed) Program, which used Whole-Genome Sequencing (WGS) to sequence approximately 155.000 genomes of participants from > 80 different studies with varying designs, and data are publicly available at the National Center for Biotechnology Information site (https://www.ncbi.nlm.nih.gov/). Applying a general z test for proportion comparisons, we found that the allele distribution of CXCL12 rs1801157 and CXCR4 rs2228014 was consistent with that reported by the TOPMed Program (Fig. 1).
Fig. 1.
Allelic distribution of the CXCL12 rs1801157 and CXCR4 rs2228014 SNVs in the present study compared to genomic big data from Trans-Omics for Precision Medicine (TOPMed) Program. CXCL12 rs1801157 allelic frequency of Okuyama et al. vs. TOPMed: z-statistic = 1.639; p = 0.101. CXCR4 rs2228014 allelic frequency of Okuyama et al. vs. TOPMed: z-statistic = 1.879; p = 0.060
CXCL12 rs1801157 and CXCR4 rs2228014 allelic and genotype distributions and p values for the χ2 test are shown in Table 3. Significant association was observed between HPV infection, SIL or CC development and all the CXCL12 rs1801157 genetic models tested (codominant, dominant, recessive and allelic). For CXCR4 rs2228014, the following models were associated to HPV infection, SIL and CC development: codominant, dominant and allelic.
Table 3.
Genotype and allele distribution considering HPV infection status, SIL diagnosis, cancer and inheritance models testing
| SNVs | Control | HPV infected | p value | Control | SIL | CC | p value* | |
|---|---|---|---|---|---|---|---|---|
| CXCL12 | ||||||||
| Codominant model | < 0.001 | < 0.001 | ||||||
| GG | 139 (74.3) | 69 (47.3) | 139 (74.3) | 32 (50.0) | 58 (63.7) | |||
| GA | 45 (23.6) | 58*(39.7) | 45 (24.1) | 18 (28.1) | 23 (25.3) | |||
| AA | 3 (2.1) | 19*(13.0) | 3 (1.6) | 14* (21.9) | 10* (11.0) | |||
| Dominant model | < 0.001 | 0.001 | ||||||
| GG | 139 (74.3) | 69 (47.3) | 139 (74.3) | 32 (50.0) | 58 (63.7) | |||
| GA + AA | 48 (25.7) | 77* (52.7) | 48 (25.7) | 32 (50.0) | 33 (36.3) | |||
| Recessive model | < 0.001 | < 0.001 | ||||||
| AA | 3 (1.6) | 19* (13.0) | 3 (1.6) | 14* (21.9) | 10* (11.0) | |||
| GG + GA | 184 (98.4) | 127 (87.0) | 184 (98.4) | 50 (78.1) | 81 (89.0) | |||
| Alleles | < 0.001 | < 0.001 | ||||||
| Allele G | 323 (86.3) | 196 (67.1) | 323 (86.3) | 82 (64.06) | 119 (73.46) | |||
| Allele A | 51 (16.7) | 96* (32.9) | 51 (16.7) | 46 (35.94) | 43 (26.54) | |||
| CXCR4 | ||||||||
| Codominant model | 0.064 | 0.010 | ||||||
| CC | 163 (87.2) | 114 (78.1) | 225 (83.6) | 52 (81.3) | 64 (70.3) | |||
| CT | 22 (11.7) | 27 (18.5) | 37 (13.8) | 12 (18.7) | 24* (26.4) | |||
| TT | 2 (1.1) | 5 (3.4) | 7 (2.6) | 0 (0.0) | 3 (3.3) | |||
| Dominant model | 0.028 | 0.003 | ||||||
| CC | 163 (86.9) | 14 (78.1) | 225 (83.6) | 52 (81.2) | 64 (70.3) | |||
| CT + TT | 24 (13.1) | 32* (21.9) | 44 (16.4) | 12 (18.8) | 27* (29.7) | |||
| Recessive model | 0.137 | 0.194 | ||||||
| TT | 2 (1.1) | 5 (3.4) | 7 (2.6) | 0 (0.0) | 3 (3.3) | |||
| CC + CT | 185 (98.9) | 141 (96.6) | 262 (97.4) | 64 (100.0) | 88 (96.7) | |||
| Alleles | 0.012 | < 0.001 | ||||||
| Allele C | 348 (93.0) | 225 (87.3) | 348 (93.0) | 116* (90.6) | 152 (83.5) | |||
| Allele T | 26 (7.0) | 37* (12.7) | 26 (7.0) | 12* (9.4) | 30* (16.5) | |||
Data presented as absolutely number and percentage
SNV single nucleotide variant, Control negative for HPV and intraepithelial lesion and malignancy, SIL squamous intraepithelial lesion, CC cervical cancer
*Two-sided χ2 test, with p < 0.05 considered significant (bold)
Multivariate logistic models of association between CXCL12 and CXCR4 SNVs and SIL/CC diagnosis
When the models were adjusted by age and partners during lifetime in the binary logistic regression employing CXCL12 rs1801157 or CXCR4 rs2228014 genetic models as explanatory variables, the following models were independently associated with HPV infection: CXCL12 rs1801157 codominants GA and AA; dominant GA+AA; recessive AA; and CXCR4 rs2228014 codominant CT and dominant CT+TT (Table 4).
Table 4.
Case–control multivariate analysis considering HPV infection and SIL/CC inheritance models
| Models | Case groups [OR (CI 95%)] | ||
|---|---|---|---|
| HPV infectiona | SILb | CCb | |
| CXCL12 | |||
| Codominant model | |||
| GG | 1.0 | 1.0 | 1.0 |
| GA | 3.006** (1.77–5.08) | 1.054 (0.53–2.07) | 0.739 (0.28–1.89) |
| AA | 18.027** (4.87–66.71) | 8.857* (3.24–24.20) | 5.031* (1.48–17.00) |
| Dominant model | |||
| GG | 1.0 | 1.0 | 1.0 |
| GA + A/A | 3.841* (2.32–6.35) | 1.960 (1.10–3.47) | 1.225 (0.56–2.83) |
| Recessive model | |||
| GG + GA | 1.0 | 1.0 | 1.0 |
| AA | 12.081* (3.33–43.82) | 9.425* (3.26–23.14) | 5.962* (1.89–18.72) |
| CXCR4 | |||
| Codominant model | |||
| CC | 1.0 | 1.0 | 1.0 |
| CT | 2.005* (1.03–3.87) | 0.28 (0.01 – 5.09)c | 4.755* (2.09–10.81) |
| TT | 5.287 (0.93–2.93) | 1.340 (0.62–2.85) | 2.853 (0.50–26.21) |
| Dominant model | |||
| CC | 1.0 | 1.0 | 1.0 |
| CT + TT | 2.254* (2.25–4.20) | 1.228 (0.57–2.63) | 4.544** (1.96–11.11) |
| Recessive model | |||
| CC + CT | 1.0 | 1.0 | 1.0 |
| TT | 4.716 (0.83–26.56) | 0.27 (0.02–4.71)C | 2.645 (0.44–15.76) |
*p < 0.05. **p < 0.01 (bold values indicate statistical significance); SIL squamous intraepithelial lesion, CC cervical cancer
aOR (odds ratio) and CI (confidence interval) 95% estimated by binary logistic regression with “Control group (negative for HPV and intraepithelial lesion and malignancy)” as reference and adjusted by age and sexual partners during lifetime
bOR (odds ratio) and CI (confidence interval) 95% estimated by multinomial logistic regression with “Control group” as reference and adjusted by age and sexual partners during lifetime
cCrude OR with 95% CI was calculated using Haldane’s modification, which adds 0.5 in all cells to accommodate possible zero counts
The association case–control analysis between CXCL12 rs1801157 or CXCR4 rs2228014 and SIL/CC adjusting by age, monthly income and partners during lifetime in the multiple logistic regression, demonstrated that the following genetic models were independently associated to SIL/CC development: CXCL12 codominant AA and recessive AA. For CXCR4 codominant CT and dominant CT+TT presented themselves independently associated to cervical cancer (Table 4).
Impact of interaction between CXCL12 and CXCR4 SNVs on HPV infection, SIL and CC
Through binary logistic regression adjusted for age and sexual partners during lifetime, significant interactions for HPV infection were found for the following cases: CXCL12 GA + CXCR4 CT (codominant model); CXCL12 GA + AA + CXCR4 CT + TT. For SIL and CC, multinomial logistic regression adjusted for age, monthly income and sexual partners during lifetime was performed. Significant susceptibility for SIL was evidenced by the interactions CXCL12 GA + AA + CXCR4 CC + CT. For cervical cancer diagnosis, the significant interactions were CXCL12 AA + CXCR4 CT, CXCL12 GA + AA + CXCR4 CT + TT and CXCL12 GA + AA + CXCR4 CT (Table 5).
Table 5.
CXCL12 rs1801157 and CXCR4 rs2228014 single-nucleotide variants interaction models for HPV and SIL/CC
| Modelsa,b | Case groups [OR (CI 95%)] | ||
|---|---|---|---|
| HPV infecteda | SILd | CCd | |
| CXCL12 GA + CXCR4 CT | 7.345* (2.12–25.40) | – | – |
| CXCL12 AA + CXCR4 CT | – | – | 7.634* (2.580–22.571) |
| CXCL12 GA + AA + CXCR4 CT + TT | 10.138** (3.46–29.65) | 2.068* (1.123–3.811) | 4.207* (1.641–10.753) |
| CXCL12 GA + AA + CXCR4 CT | – | – | 3.021* (1.382–10.570) |
*p < 0.05 **p < 0.001; SIL squamous intraepithelial lesion, CC cervical cancer
a OR (odds ratio) and CI (confidence interval) 95% obtained through binary logistic regression with major alleles as reference and adjusted by age, monthly income, sexual partners during lifetime
bNote that interaction analyses between only homozygous genotypes were not performed because none of the patients co-inherited the two SNVs in homozygosity
cAn exploratory analysis of heterozygous co-inheritance was carried out, in which all patients bearing any of the heterozygous genotypes for both SNVs were grouped
dOR (odds ratio) and CI (confidence interval) 95% estimated by multinomial logistic regression with “SIL group” as reference and adjusted by age, monthly income and sexual partners during lifetime
Discussion
In this study, we provided novel information about the effect of single-nucleotide genetic variants rs1801157 of CXCL12 and rs2228014 of CXCR4 on the HPV infection, SIL and CC susceptibility. Our data indicate that CXCL12 rs1801157 AA genotype is associated with the susceptibility of developing SIL and CC. A previous study from our group had already demonstrated that the CXCL12 genotype AA and allele A were associated to HPV infection susceptibility (Okuyama et al. 2018). However, previous studies did not analyze the possible association with HPV infection but verified that this SNV is not associated with CC (Rozask et al. 2015; Maley et al. 2009; Tee et al. 2012). Considering CXCR4 rs2228014 we demonstrated, for the first time in the literature, that women who were allele T carriers presented a significant increased risk of HPV infection and CC development.
Tumors are dynamic tissue masses, so requiring continuous exposure to the host cells, nurturing them into pave a path for tumor growth and metastasis. CXCL12/CXCR4 is the key signaling for such aim. Gathering knowledge about the activity within this axis would deepen insight into the utmost importance this signaling taken to attract and cross-connect multiple cells within the tumor microenvironment. A large number of studies showed that individual genetic factors can affect the incidence and progression of cancer. Among genetic factors, SNV located in the promoter or coding region of gene expression regulation have important functions (Deng et al. 2018; Mortezaee 2020).
In the present study, the SNVs CXCL12 rs1801157, located in 3′UTR region and CXCR4 rs2228014, a synonymous SNV, were investigated. Mutations in the 3′UTR are involved in many diseases because they affect gene expression. As a regulatory region, the 3′UTR is indispensable for normal gene expression. Therefore, polymorphisms in the 3′UTR can alter miRNA binding sites and affect mRNA degradation and protein translation (Sethupathy and Collins 2008). Whilst synonymous SNVs can also accelerate or decelerate the speed at which the ribosome moves along the mRNA, thus changing the dynamics of translation, and the subsequent protein structure and function. They may also result in different mRNA secondary structures and protein secondary structures such as α helix and β folding (Deng et al. 2016).
SNVs interaction analysis was performed to evaluate possible significant combination between CXCL12 rs1801157 and CXCR4 rs2228014. It was observed that among the SNVs interactions, models with the presence of genotypes AA and CT from CXCL12 and CXCR4 SNVs, respectively, strongly indicated a greater risk of developing SIL and CC. There are no previous studies evaluating the interaction between both SNVs and HPV, SIL and CC. Data so far have shown a significant interaction for oral squamous cell carcinoma (OSCC) in the genotypic combination of CXCL12 and CXCR4 SNVs, revealing that CT + TT genotypes exerted an increased risk in patients with one or two risk habits (tobacco smoking and alcohol consumption) (Huang et al. 2019). The axis CXCL12/CXCR4 is extensively studied in breast cancer (BC). Lin et al. (2009) evidenced a genetic interaction between CXCL12 and CXCR4 SNVs in which wild type homozygotes reduced BC susceptibility, but failed in finding any association of both SNVs in isolation. These data are in accordance with Guembarovski et al. (2018) which also observed no SNV association in isolation but showed a significant increased susceptibility in the interaction between heterozygous genotypes for both single-nucleotide variants. These data were also demonstrated by Fu et al. (2016) using an analysis called multifactor dimensionality reduction (MDR), a bioinformatic technique to provide a non-linear model associated with disease.
Keratinocytes immortalized by oncogenic HPV16 or HPV18 upregulate CXCL12 and CXCR4 in a manner dependent upon expression of E6 and E7. Autocrine signaling activated by CXCL12-engagement of its receptor could control responses as motility and survival of the infected cells, and could also increase proliferation and migration of adjacent keratinocytes. The paracrine activation would enhance cell permissiveness to viral genome replication and production of the E6 and E7 proteins. Consequently, E6 and E7 expressed in the suprabasal layers would then upregulate CXCL12 and CXCR4, which further enhance cell proliferation and viral DNA replication, possibly explaining the basis of HPV associated oncogenesis (De Oliveira et al. 2011). Furthermore, CXCL12 proximal promoter in its 5′flanking and 5′UTR region contains six protein 1 transcription factor (Sp1) binding sites, and Sp1 transcription factor seems to be the major positive regulator of CXCL12 expression. After basal cells HPV infection, E6 and E7 oncoproteins are expressed, and may bind to Sp1. The E6-Sp1 and E7-Sp1 complexes can migrate into the nucleus and probably induce the CXCL12 gene expression (García-Moruja et al. 2005). Considering CXCR4 SNV, the site of the mutation is a target of RNA polymerase II and may alter motifs for SOX, RXRA and NRSF transcription factors, what has been associated to a plethora of cancers (Deng et al. 2018; Halstead et al. 2017; Kumar et al. 2019).
Moreover, SNVs can affect proteins levels. For Chang et al. (2009), expression of CXCL12 rs1801157 mRNA in GA/AA fibroblasts was three times that in GG fibroblasts. In prostate cancer patients, CXCL12 rs1801157 allele A carriers presented a protein expression significantly higher than GG genotype carriers. Among the CXCL12 allele A carriers the protein expression was also significantly higher compared to those with the GG genotype (Hirata et al. 2007). CXCR4 mRNA relative expression did not differ regarding the presence or absence of T allele and even though the expression was higher in BC patients, there was no correlation with patient clinicopathological features (Okuyama et al. 2015). CXCL12 mRNA expression in BC demonstrated that allele A carrying patients presented smaller expression compared to GG patients and the ones with positive estrogen receptor with allele A showed a significantly lower expression of CXCL12 in peripheral blood than GG hormone positive patients (De Oliveira et al. 2011).
The small number of low-grade squamous intraepithelial lesion (LSIL) patients made an analysis stratified in lesions levels difficult: LSIL and HSIL (high-grade squamous intraepithelial lesion), however, a study strength is the robust statistical analysis, controlling the potential confounding factors, minimizing the effects of possible biases; Currently we are conducting analyses to better elucidate the influence of CXCL12 rs1801157 and CXCR4 rs2228014 SNVs on gene expression and protein levels in cervical cancer development. To the best of our knowledge, this is the first time that CXCR4 rs2228014 is associated to the HPV infection susceptibility and CC. In this work, we also demonstrated that CXCL12 rs1801157 is strongly associated to SIL and CC. An expressive number of patients and a strong interaction analysis between CXCL12 rs1801157 and CXCR4 rs2228014 showed that patients carriers of allele A of CXCL12 and allele T of CXCR4 simultaneously, are more susceptible to HPV infection, squamous intraepithelial lesions and cervical cancer, highlighting these variants in the CXCL12/CXCR4 axis as promising candidates to susceptibility and prognosis molecular markers in HPV infection and cervical cancer development.
Acknowledgements
All the authors would like to thank the Intermunicipal Consortium of Health of the Middle Paranapanema, Londrina Cancer Hospital, Clinic center of the Londrina State University, Experimental Pathology Postgraduate Program. Fundação Araucária, CAPES—Coordenação de Aperfeiçoamento de Pessoal de Nível Superior and CNPq—Conselho Nacional de Desenvolvimento Científico e Tecnológico.
Author contributions
NCMO participated in study design and acquisition of data, experimental procedures, performed statistical analysis and interpretation, and drafted the manuscript. FCS, participated in sample collect and medical records, DNA extraction, statistical analysis. KPT participated in study design, acquisition of data and statistical analysis. AE, participated in sample collect and medical records. RLG critically revised the data. JOCF made possible sample and data collection at Londrina Cancer Hospital. MAEW participated in the design study. KBO participated in the design of the study, interpretation of data and given final approval of the version to be published. All authors read and approved the final manuscript.
Funding
This work was supported by Fundação Araucária—Programa Pesquisa para o SUS (34935.406.36850.19112012) and CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) financial code 001.
Availability of data and material
All data generated or analyzed during the current study are included in this published article.
Declarations
Conflict of interest
The authors declare that they have no competing interests.
Ethics approval and consent to participate
This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by Institutional Ethics Committee Involving Humans at State University of Londrina, Londrina-PR, Brazil (CEP/UEL 133/2012; CAAE 05505912.0.0000.5231).
Consent for publication
Not applicable.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- Altman DG (1991) Practical statistics for medical research. Chapman and Hall, London, p 1991 [Google Scholar]
- Balkwill F. Cancer and the chemokine network. Nat Rev Cancer 4(7):540–50. [DOI] [PubMed]
- Bray F, Ferlay J, Soerjomataram I et al (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68(6):394–424 [DOI] [PubMed] [Google Scholar]
- Butrym A, Gebura K, Iwaszko M et al (2016) Dual role of the CXCL12 polymorphism in patients with chronic lymphocytic leukemia. HLA 87(6):432–438 [DOI] [PubMed] [Google Scholar]
- Chang SC, Lin PC, Yang SH et al (2009) SDF-1alpha G801A polymorphism predicts lymph node metastasis in stage T3 colorectal cancer. Ann Surg Oncol 16(8):2323–2330 [DOI] [PubMed] [Google Scholar]
- Chang SF, Li SL, Yang B et al (2015) CXCL12 G801A polymorphism and susceptibility to glioma: a case-control study. Genet Mol Res 14(4):17399–17405 [DOI] [PubMed] [Google Scholar]
- Clopper C, Pearson ES (1934) Pearson ES (1934) The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika 26:404–413 [Google Scholar]
- Cohen PA, Jhingran A et al (2019) Cervical cancer. Lancet 393(10167):169–182 [DOI] [PubMed] [Google Scholar]
- Colpani V, Falcetta FS, Bidionotto AB et al (2020) Prevalence of human papillomavirus (HPV) in Brazil: a systematic review and meta-analysis. PLoS ONE 15(2):e0229154 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Da Silva MC, Martins RPH, De Souza JL et al (2012) Prevalence of HPV infection and genotypes in women with normal cervical cytology in the state of Paraná, Brazil. Arch Gynecol Obstet 286(4):1015–1022 [DOI] [PubMed] [Google Scholar]
- De Oliveira KB, Guembarovski RL, Oda JMM et al (2011) CXCL12 rs1801157 polymorphism and expression in peripheral blood from breast cancer patients. Cytokine 55(2):260–265 [DOI] [PubMed] [Google Scholar]
- Deng N, Zhou H, Fan H et al (2016) Single nucleotide polymorphisms and cancer susceptibility. Oncotarget 8(66):110635–110649 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Deng P, Zuo Y, Feng S et al (2018) Knockdown of NRSF inhibits cell proliferation of ovarian cancer via activating hippo pathway. Life Sci 215:73–79 [DOI] [PubMed] [Google Scholar]
- Fu OY, Chang WH, Lin YD et al (2016) Breast cancer-associated high-order SNP-SNP interaction of CXCL12/CXCR4-related genes by an improved multifactor dimensionality reduction (MDR-ER). Oncol Rep 6(3):1739–1747 [DOI] [PubMed] [Google Scholar]
- García-Moruja C, Alonso-lobo JM, Rueda P et al (2005) Functional characterization of SDF-1 proximal promoter. J Mol Biol 348(1):43–62 [DOI] [PubMed] [Google Scholar]
- Guembarovski A, Guembarovski RL, Hirata BKB et al (2018) CXCL12 chemokine and CXCR4 receptor: association with susceptibility and prognostic markers in triple negative breast cancer. Mol Biol Rep 45(5):741–775 [DOI] [PubMed] [Google Scholar]
- Guyon A (2014) CXCL12 chemokine and its receptors as major players in the interactions between immune and nervous systems. Fronti Cell Neurosci 8:65 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Halstead AM, Capadia CD, Bolzenius J et al (2017) Bladder-cancer-associated mutations in RXRA activate peroxisome proliferator-activated receptors to drive urothelial proliferation. Elife 16:1–2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hirata H, Hinoda Y, Kikuno N et al (2007) CXCL12 G801A polymorphism is a risk factor for sporadic prostate cancer susceptibility. Clin Cancer Res 13(17):5056–5062 [DOI] [PubMed] [Google Scholar]
- Huang SJ, Tseng Y-K, Lo Y-H et al (2019) Association of SDF-1 and CXCR4 polymorphisms with susceptibility to oral and pharyngeal squamous cell carcinoma. Anticancer Res 39(6):2891–2902 [DOI] [PubMed] [Google Scholar]
- Kumar P, Mistri TK (2019) Transcription factors in SOX family: potent regulators for cancer initiation and development in the human body. Semin Cancer Biol 18:1–9 [DOI] [PubMed] [Google Scholar]
- Lazennec G, Richmond A (2010) Chemokines and chemokine receptors: new insights into cancer-related inflammation. Trends Mol Med 16(3):133–144 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lecavalier-Barsoum M, Chaudary N, Han K et al (2018) Targeting the CXCL12/CXCR4 pathway and myeloid cells to improve radiation treatment of locally advanced cervical cancer. Int J Cancer 143(5):1017–1028 [DOI] [PubMed] [Google Scholar]
- Lin GT, Tseng H-F, Yanf CH et al (2009) Combinational polymorphisms of seven CXCL12-related genes are protective against breast cancer in Taiwan. OMICS J Integr Biol 13(2):165–172 [DOI] [PubMed] [Google Scholar]
- Maley SN, Schwartz SM, Johnson LG et al (2009) Genetic variation in CXCL12 and risk of cervical carcinoma: a population-based case-control study. Int J Immunogenet 3(6):367–375 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mortezaee K (2020) CXCL12/CXCR4 axis in the microenvironment of solid tumors: a critical mediator of metastasis. Life Sci 249:117534 [DOI] [PubMed] [Google Scholar]
- Okuyama KM, De Oliveira KB, Ariza CB et al (2015) Genetic polymorphism and expression of CXCR4 in breast cancer. Anal Cell Pathol 289510:1–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Okuyama NCM, Cezar-dos-Santos F, Pereira ER et al (2018) Genetic variant in CXCL12 gene raises susceptibility to HPV infection and squamous intraepithelial lesions development: a case-control study. J Biomed Sci 25:1–10 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Qin L-F, Qin J-M, Zhang J-Q et al (2018) CXCL12 and CXCR4 polymorphisms and expressions in peripheral blood from patients of hepatocellular carcinoma. Future Oncol 14(13):1261–1271 [DOI] [PubMed] [Google Scholar]
- Roszak A, Misztal M, Sowińska A et al (2015) Stromal cell-derived factor- 1 G801A polymorphism and the risk factors for cervical cancer. Mol Med Rep 11:4633–4682 [DOI] [PubMed] [Google Scholar]
- Sethupathy P, Collins FS (2008) MicroRNA target site polymorphisms and human disease. Trends Genet 24(10):489–497 [DOI] [PubMed] [Google Scholar]
- Shirozu M, Nakano T, Inazawa J et al (1995) Structure and chromosomal localization of the human stromal cell-derived factor 1(SDF1) gene. Genomics 28:495–500 [DOI] [PubMed] [Google Scholar]
- Susek KH, Karvouni M, Alici E et al (2004) The role of CXC chemokine receptors 1–4 on immune cells in the tumor microenvironment. Front Immunol 25(9):2159 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tee YT, Yang S-F, Wang P-H et al (2012) G801A polymorphism of human stromal cell-derived factor gene raises no susceptibility to neoplastic lesions of uterine cervix. Int J Gynecol Cancer 22(8):1297–1302 [DOI] [PubMed] [Google Scholar]
- Torres-Poveda K, Burguete-García AI, Bahena-Román M et al (2016) Risk allelic load in Th2 and Th3 cytokines genes as biomarker of susceptibility to HPV-16 positive cervical cancer: a case control study. BMC Cancer 24(16):330 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zanotta N, Tornesello ML, Annunziata C et al (2016) Candidate soluble immune mediators in young women with high-risk human papillomavirus infection: high expression of chemokines promoting angiogenesis and cell proliferation. PLoS ONE 11(3):1–14 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhang JP, Lu WG, Ye F, Chen HZ, Zhou CY, Xie X (2007) Study on CXCR4/ SDF-1 axis in lymph node metastasis of cervical squamous cell carcinoma. Int J Gynecol Cancer 17:478–483 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data generated or analyzed during the current study are included in this published article.