Abstract
Polymorphism in the Toll-like receptor (TLR) 9 gene has been shown to have a significant role in some diseases; however, little is known about its possible role in the natural history of human papillomavirus (HPV) infections. We investigated the association between a single-nucleotide polymorphism (SNP) (rs5743836) in the promoter region of TLR9 (T1237C) and type-specific HPV infections. Specimens were derived from a cohort of 2462 women enrolled in the Ludwig–McGill Cohort Study. We randomly selected 500 women who had a cervical HPV infection detected at least once during the study as cases. We defined two control groups: (i) a random sample of 300 women who always tested HPV negative, and (ii) a sample of 234 women who were always HPV negative but had a minimum of ten visits during the study. TLR9 genotyping was performed using bidirectional PCR amplification of specific alleles. Irrespective of group, the WT homozygous TLR9 genotype (TT) was the most common form, followed by the heterozygous (TC) and the mutant homozygous (CC) forms. There were no consistent associations between polymorphism and infection risk, either overall or by type or species. Likewise, there were no consistently significant associations between polymorphism and HPV clearance or persistence. We concluded that this polymorphism in the promoter region of TLR9 gene does not seem to have a mediating role in the natural history of the HPV infection.
Introduction
Persistent human papillomavirus (HPV) infection is a key early event in cervical carcinogenesis (Schlecht et al., 2001). Infection acquisition and persistence are dependent on the host immune system. Toll-like receptors (TLRs) are expressed constitutively in host cells and permit the recognition of pathogen-associated molecular patterns (PAMPs), which are essential in host defence mechanisms. The clinical interest in TLRs in terms of cancer therapies is increasing. Agents that activate TLRs, either alone or more likely in conjunction with other therapies, including radiotherapy and chemotherapy, may have great potential as new cancer treatments. Similarly, the inhibition of certain TLRs in tumours associated with inflammation may lead to new therapies (Corr & O’Neill, 2009).
It has been shown that HPV modulates the expression of TLR at the transcriptional level, and this can be an important feature to avoid host defences (Hasan et al., 2007). The promoter sequence of a gene determines its transcription, and a single-nucleotide polymorphism (SNP) can alter the gene expression and consequently the host’s susceptibility or resistance to diseases. A number of genetic association studies suggest that TLR polymorphisms can alter the response to PAMPs and may be associated with susceptibility to different diseases (Misch & Hawn, 2008; Netea et al., 2012).
The rs5743836 polymorphism on the promoter region of TLR9 gene (T1237C) is related to susceptibility to allergic aspergillosis (Carvalho et al., 2008), non-Hodgkin’s lymphoma (Carvalho et al., 2012), Helicobacter pylori-induced gastric disease (Ng et al., 2010), atopic eczema (Novak et al., 2007) and chronic kidney disease (Lu et al., 2011). This SNP creates a putative NFκB-binding site, increasing the gene transcription due to the action of this transcriptional factor (Ng et al., 2010). The activity of NFκB is inhibited by HPV16 oncoproteins E6 and E7; however, when activated, NFκB inhibits proliferation and immortalization (Vandermark et al., 2012). HPV persistence seems to be associated with low IL-6 levels (Rosa et al., 2012), and the rs5743836 polymorphism creates a novel IL-6-binding site, increasing the transcription and creating a loop of TLR9/IL-6 signalling amplification (Carvalho et al., 2011).
Furthermore, the expression of E6 and E7 oncoproteins from high-risk HPV types, but not from low-risk HPVs, inhibits the activation of TLR9 by CpG motifs in human keratinocytes and reduces production of TLR9 mRNA (Hasan et al., 2007). Studies that relate the mRNA expression of TLRs and HPV infection have observed that HPV16 infections that regressed were significantly associated with increased expression of TLR2, TLR3, TLR7, TLR8 and TLR9 (Daud et al., 2011). This suggests that the modulation of immune responses triggered by TLR9 may play an important role in the immune response against HPV and may affect infection risk or clearance. With this rationale in mind, we investigated whether the rs5743836 SNP polymorphism in the promoter region of TLR9 mediates the natural history of HPV infection among women enrolled in the Ludwig–McGill Cohort Study.
Results
The cumulative risk of type-specific HPV infection according to the different TLR9 genotypes is shown in Table 1. Among the 500 ever-positive HPV cases, the following HPV genotypes were the most common: HPV16 (22.8 %), HPV53 (15.2 %), HPV51 (11.4 %) and HPV84 (11.0 %). Alphapapillomavirus 9 (46.4 %) was the most common species followed by Alphapapillomavirus 3 (27.4 %) and 7 (26.6 %). Although there was considerable variation in prevalence of TLR9 genotypes among cases, there were no noteworthy variations in type- or species-specific HPV prevalence among these genotypes.
Table 1. Cumulative risk of type-specific HPV infection in the entire cohort and among cases, and according to TLR9 genotype among cases.
| Genotype/species | Overall positivity in whole cohort (%) (n = 2462) | Overall positivity among cases (%) (n = 500) | HPV positivity by TLR9 genotype (%) | |||
| TT (%) (n = 307) | TC (%) (n = 165) | CC (%) (n = 17) | na (%) (n = 11) | |||
| HPV genotype | ||||||
| 6/11 | 109 (4.43) | 46 (9.2) | 22 (7.2) | 21 (12.7) | 3 (17.7) | 0 (0) |
| 16 | 286 (11.6) | 114 (22.8) | 72 (23.5) | 35 (21.2) | 5 (29.4) | 2 (18.2) |
| 18 | 90 (3.7) | 35 (7.0) | 25 (8.1) | 10 (6.1) | 0 (0) | 0 (0) |
| 26 | 19 (0.8) | 6 (1.2) | 2 (0.7) | 3 (1.8) | 1 (5.9) | 0 (0) |
| 31 | 106 (4.3) | 37 (7.4) | 23 (7.5) | 13 (7.9) | 0 (0) | 1 (9.1) |
| 32 | 10 (0.4) | 2 (0.4) | 1 (0.3) | 1 (0.6) | 0 (0) | 0 (0) |
| 33 | 52 (2.1) | 21 (4.2) | 10 (3.3) | 9 (5.5) | 2 (11.8) | 0 (0) |
| 34 | 2 (0.1) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| 35 | 82 (3.3) | 30 (6.0) | 12 (3.9) | 16 (9.7) | 1 (5.9) | 1 (9.1) |
| 39 | 49 (2.0) | 18 (3.6) | 10 (3.3) | 8 (4.9) | 0 (0) | 0 (0) |
| 40 | 44 (1.8) | 20 (4.0) | 12 (3.9) | 8 (4.9) | 0 (0) | 0 (0) |
| 42 | 32 (1.3) | 13 (2.6) | 7 (2.3) | 6 (3.6) | 0 (0) | 0 (0) |
| 44 | 11 (0.5) | 4 (0.8) | 2 (0.7) | 1 (0.6) | 1 (5.9) | 0 (0) |
| 45 | 85 (3.5) | 34 (6.8) | 21 (6.8) | 9 (5.5) | 3 (17.7) | 1 (9.1) |
| 51 | 161 (6.5) | 57 (11.4) | 36 (11.7) | 17 (10.3) | 3 (17.7) | 1 (9.1) |
| 52 | 115 (4.7) | 47 (9.4) | 30 (9.8) | 13 (7.9) | 2 (11.8) | 2 (18.2) |
| 53 | 184 (7.5) | 76 (15.2) | 43 (14.0) | 28 (17.0) | 1 (5.9) | 4 (36.4) |
| 54 | 76 (3.1) | 37 (7.4) | 19 (6.2) | 17 (10.3) | 1 (5.9) | 0 (0) |
| 55 | 94 (3.8) | 38 (7.6) | 19 (6.2) | 16 (9.7) | 3 (17.7) | 0 (0) |
| 56 | 71 (2.9) | 27 (5.4) | 12 (3.9) | 14 (8.5) | 1 (5.9) | 0 (0) |
| 57 | 6 (0.2) | 4 (0.8) | 1 (0.3) | 3 (1.8) | 0 (0) | 0 (0) |
| 58 | 119 (4.8) | 43 (8.6) | 29 (9.5) | 13 (7.9) | 1 (5.9) | 0 (0) |
| 59 | 69 (2.8) | 26 (5.2) | 13 (4.2) | 10 (6.1) | 1 (5.9) | 2 (18.2) |
| 61 | 65 (2.6) | 26 (5.2) | 19 (6.2) | 3 (1.8) | 4 (23.5) | 0 (0) |
| 62 | 55 (2.2) | 23 (4.6) | 11 (3.6) | 11 (6.7) | 1 (5.9) | 0 (0) |
| 64 | 2 (0.1) | 1 (0.2) | 0 (0) | 1 (0.6) | 0 | 0 (0) |
| 66 | 64 (2.6) | 22 (4.4) | 11 (3.6) | 9 (5.5) | 1 (5.9) | 1 (9.1) |
| 67 | 9 (0.4) | 5 (1.0) | 5 (1.6) | 0 (0) | 0 (0) | 0 (0) |
| 68 | 73 (3.0) | 27 (5.4) | 13 (4.2) | 10 (6.1) | 3 (17.7) | 1 (9.1) |
| 69 | 6 (0.2) | 3 (0.6) | 2 (0.7) | 1 (0.6) | 0 (0) | 0 (0) |
| 70 | 46 (1.9) | 19 (3.8) | 12 (3.9) | 6 (3.6) | 1 (5.9) | 0 (0) |
| 71 | 40 (1.6) | 18 (3.6) | 14 (4.6) | 4 (2.4) | 0 (0) | 0 (0) |
| 72 | 17 (0.7) | 9 (1.8) | 4 (1.3) | 4 (2.4) | 0 (0) | 1 (9.1) |
| 73 | 62 (2.5) | 29 (5.8) | 23 (7.5) | 5 (3.0) | 0 (0) | 1 (9.1) |
| 81 | 39 (1.6) | 20 (4.0) | 11 (3.6) | 8 (4.9) | 1 (5.9) | 0 (0) |
| 82 | 17 (0.7) | 4 (0.8) | 4 (1.3) | 0 (0) | 0 (0) | 0 (0) |
| 83 | 54 (2.2) | 20 (4.0) | 11 (3.6) | 9 (5.5) | 0 (0) | 0 (0) |
| 84 | 122 (5.0) | 55 (11.0) | 30 (9.8) | 21 (12.7) | 3 (17.7) | 1 (9.1) |
| 89 | 25 (1.0) | 8 (1.6) | 2 (0.7) | 5 (3.0) | 1 (5.9) | 0 (0) |
| Any HPV | 1194 (48.5) | 500 (100) | 299 (97.4) | 162 (98.2) | 17(100) | 11 (100) |
| Alphapapillomavirus species* | ||||||
| Species 3 | 322 (13.1) | 137 (27.4) | 80 (26.1) | 48 (29.1) | 7 (41.2) | 2 (18.2) |
| Species 5 | 194 (7.9) | 68 (13.6) | 42 (13.7) | 21 (12.7) | 4 (23.5) | 1 (9.1) |
| Species 6 | 287 (11.7) | 114 (22.8) | 62 (20.2) | 44 (26.7) | 3 (17.7) | 5 (45.5) |
| Species 7 | 341 (13.9) | 133 (26.6) | 76 (24.8) | 47 (28.5) | 6 (35.3) | 4 (36.4) |
| Species 9 | 586 (23.8) | 232 (46.4) | 143 (46.6) | 73 (44.2) | 10 (58.8) | 6 (54.6) |
| Species 10 | 199 (8.1) | 81 (16.2) | 41 (13.4) | 35 (21.2) | 5 (29.4) | 0 (0) |
na, Not applicable.
Species 3 (HPV61, -72, -62, -81, -83, -84 and -89); species 5 (HPV82, -26, -69 and -51); species 6 (HPV82, -26, -69 and -51); species 7 (HPV18, -39, -45, -59, -68 and -70); species 9 (HPV16, -31, -33, -35, -52, -58, -67); species 10 (HPV6, -11, -44 and -55).
Table 2 shows the age-adjusted odds ratios (ORs) of cumulative risk of selected type- and species-specific HPV infections according to putative TLR9 phenotypes and control group contrast. There were significant increases in risk for HPV6/11, -35 and -89 with the dominant model. With the recessive model, the only significant result was seen with HPV61. The only type with reduced risk was HPV73, observable with sufficient precision only in the dominant model. The magnitude of the associations did not change materially when contrasting with the more restricted control group, which had greater HPV detection opportunity because of the longer follow-up. This was also the situation when comparing with both control groups pooled together (data not shown). Among HPV species, risk of Alphapapillomavirus 10 HPVs was significantly elevated because of the preponderance of the effect due to HPV6/11. An analysis of the subset of cases that developed cervical cytological abnormalities did not reveal any associations with risk (data not shown).
Table 2. Age-adjusted ORs of cumulative risk of selected type-specific HPV infections according to putative TLR9 phenotype.
Significant associations are highlighted in bold.
| Genotype/species | HPV-positive vs random sample of HPV-negative women | HPV-positive vs HPV-negative who completed a minimum of ten visits | ||
| Dominant model* | Recessive model† | Dominant model* | Recessive model† | |
| OR (95 % CI) | OR (95 % CI) | OR (95 % CI) | OR (95 % CI) | |
| HPV genotype | ||||
| 6/11 | 1.82 (0.99–3.32) | 1.53 (0.44–5.27) | 1.88 (1.02–3.44) | 1.59 (0.46–5.56) |
| 16 | 0.88 (0.58–1.34) | 1.02 (0.38–2.71) | 0.87 (0.57–1.33) | 1.05 (0.39–2.84) |
| 18 | 0.63 (0.30–1.34) | na | 0.63 (0.30–1.34) | na |
| 31 | 0.89 (0.44–1.79) | na | 0.92 (0.45–1.86) | na |
| 35 | 2.38 (1.12–5.06) | 0.77 (0.10–5.88) | 2.37 (1.11–5.04) | 0.79 (0.10–6.06) |
| 45 | 0.91 (0.44–1.88) | 2.26 (0.65–7.89) | 0.92 (0.44–1.92) | 2.35 (0.67–8.31) |
| 51 | 0.89 (0.50–1.56) | 1.24 (0.36– 4.23) | 0.89 (0.50–1.57) | 1.29 (0.37–4.42) |
| 52 | 0.79 (0.42–1.51) | 1.01 (0.23–4.36) | 0.79 (0.42–1.51) | 1.03 (0.24–4.53) |
| 53 | 1.11 (0.68–1.83) | 0.29 (0.04–2.17) | 1.09 (0.66–1.79) | 0.30 (0.04–2.21) |
| 54 | 1.58 (0.82–3.07) | 0.60 (0.08–4.55) | 1.55 (0.80–3.01) | 0.62 (0.08–4.66) |
| 55 | 1.56 (0.86–3.20) | 1.96 (0.57–6.75) | 1.66 (0.86–3.20) | 2.03 (0.58–7.05) |
| 56 | 2.10 (0.97–4.55) | 0.85 (0.11–6.47) | 2.05 (0.95–4.46) | 0.87 (0.11–6.65) |
| 58 | 0.77 (0.40–1.48) | 0.50 (0.07–3.78) | 0.76 (0.40–1.47) | 0.51 (0.07–3.89) |
| 59 | 1.37 (0.60–3.10) | 0.92 (0.12–7.06) | 1.40 (0.61–3.17) | 0.95 (0.12–7.33) |
| 61 | 0.59 (0.24–1.42) | 4.43 (1.43–13.72) | 0.58 (0.24–1.40) | 4.60 (1.48–14.36) |
| 68 | 1.66 (0.76–3.62) | 3.07 (0.87–10.80) | 1.63 (0.75–3.58) | 3.17 (0.89–11.24) |
| 73 | 0.34 (0.13–0.91) | na | 0.34 (0.13–0.89) | na |
| 84 | 1.33 (0.76–2.33) | 1.34 (0.40–4.54) | 1.30 (0.75–2.28) | 1.38 (0.40–4.70) |
| 89 | 4.90 (0.98–24.48) | 3.14 (0.37–26.41) | 4.96 (0.99–24.78) | 3.24 (0.38–27.36) |
| Any HPV | 0.94 (0.70–1.27) | 0.63 (0.31–1.27) | 0.87 (0.63–1.20) | 0.62 (0.29–1.32) |
| Alphapapillomavirus species‡ | ||||
| Species 3 | 1.15 (0.79–1.68) | 1.27 (0.54–2.99) | 1.12 (0.77–1.65) | 1.32 (0.55–3.14) |
| Species 7 | 1.16 (0.78–1.70) | 1.08 (0.43–2.68) | 1.15 (0.78–1.71) | 1.11 (0.44–2.81) |
| Species 9 | 0.92 (0.67–1.27) | 1.02 (0.48–2.20) | 0.90 (0.65–1.25) | 1.06 (0.48–2.34) |
| Species 10 | 1.67 (1.05–2.66) | 1.50 (0.56–4.02) | 1.69 (1.06–2.71) | 1.56 (0.57–4.26) |
OR, odds ratio; CI, confidence interval; na, Not applicable.
Dominant model: mutant homozygote (CC)+heterozygote (TC) versus WT homozygote (TT).
Recessive model: WT homozygote (TT)+heterozygote (TC) versus mutant homozygote (CC).
Species 3 (HPV61, -72, -62, -81, -83, -84 and -89); species 7 (HPV18, -39, -45, -59, -68 and -70); species 9 (HPV16, -31, -33, -35, -52, -58 and -67); species 10 (HPV6, -11, -44 and -55).
Another potential effect for a mediating role of TLR9 polymorphism would be via a delayed or accelerated clearance of HPV infections among women who had prevalent or incident infections with given HPV types. Table 3 shows the hazard ratios (HRs) of type-specific HPV clearance by putative TLR9 phenotypes according to two different levels of stringency in defining clearance of infection events. In the dominant model, the only significant result was seen with the most stringent definition (two consecutive negative visits following HPV positivity) and HPV59 showing a decreased clearance with HR = 0.32 (95 % CI 0.10–0.98). In the recessive model, a significant five- to sevenfold increase in rate of clearance was seen with both definitions for HPV51. Accelerated clearance was also seen for HPV56 and -58 with the recessive model. There were no significant associations when grouping types within HPV species. The HPV clearance with any HPV type and the TLR9 genotypes are shown in Fig. 1.
Table 3. Type-specific HPV clearance and TLR9 genotypes using two different definitions for clearance.
Significant associations are highlighted in bold.
| Genotype/species | Clearance definition 1* | Clearance definition 2† | ||||||
| Dominant model‡ | Recessive model§ | Dominant model‡ | Recessive model§ | |||||
| Events | HR | Events | HR | Events | HR | Events | HR | |
| HPV genotype | ||||||||
| 6/11 | 24 | 1.54 (0.79–2.99) | 3 | 1.46 (0.44–4.81) | 20 | 1.42 (0.72–2.78) | 2 | 3.41 (0.74–15.64) |
| 16 | 37 | 1.09 (0.71–1.69) | 5 | 1.27 (0.51–3.19) | 36 | 0.99 (0.63–1.55) | 5 | 1.31 (0.52–3.28) |
| 18 | 8 | 1.95 (0.82–4.63) | 0 | na | 5 | 1.47 (0.52–4.14) | 0 | na |
| 31 | 10 | 0.90 (0.39–2.09) | 0 | na | 8 | 1.07 (0.44–2.59) | 0 | na |
| 33 | 9 | 0.95 (0.34–2.65) | 1 | 0.55 (0.07–4.41) | 8 | 0.92 (0.32–2.63) | 1 | 0.46 (0.06–3.75) |
| 35 | 16 | 0.63 (0.28–1.44) | 1 | 1.37 (0.18–10.52) | 14 | 1.05 (0.41–2.71) | 1 | 5.41 (0.60–48.64) |
| 45 | 9 | 0.76 (0.33–1.78) | 3 | 0.62 (0.18–2.11) | 8 | 0.83 (0.34–1.99) | 3 | 0.85 (0.24–2.97) |
| 51 | 17 | 0.94 (0.49–1.80) | 3 | 5.26 (1.58–17.53) | 15 | 0.77 (0.38–1.56) | 3 | 6.70 (1.75–25.65) |
| 52 | 14 | 1.58 (0.75–3.34) | 2 | 1.68 (0.40–7.16) | 13 | 1.41 (0.69–2.89) | 2 | 2.74 (0.62–12.09) |
| 53 | 25 | 0.76 (0.33–1.78) | 3 | 0.62 (0.18–2.11) | 24 | 1.33 (0.76–2.34) | 1 | 0.71 (0.10–5.23) |
| 54 | 17 | 1.13 (0.55–2.317) | 1 | 1.12 (0.15–8.45) | 16 | 1.11 (0.53–2.31) | 1 | 1.78 (0.23–13.76) |
| 55 | 16 | 1.04 (0.50–2.17) | 3 | 0.72 (0.21–2.40) | 12 | 0.84 (0.34–2.10) | 2 | 0.47 (0.10–2.12) |
| 56 | 15 | 1.31 (0.57–3.02) | 1 | 2.09 (0.26–16.54) | 14 | 1.39 (0.58–3.35) | 1 | 10.78 (0.97–119.70) |
| 58 | 11 | 1.30 (1.70–229.99) | 1 | 19.79(1.70–230.0) | 11 | 1.56 (0.66–3.65) | 1 | 8.71 (0.92–82.98) |
| 59 | 10 | 0.57 (0.21–1.56) | 1 | 4.97 (0.34–72.82) | 8 | 0.32 (0.10–0.98) | 0 | na |
| 61 | 7 | 0.80 (0.30–2.15) | 4 | 1.17 (0.38–3.60) | 6 | 0.90 (0.32–2.54) | 3 | 0.90 (0.26–3.19) |
| 68 | 11 | 1.39 (0.58–3.38) | 3 | 1.21 (0.32–4.57) | 9 | 1.29 (0.49–3.40) | 3 | 1.37 (0.36–5.21) |
| 73 | 3 | 2.52 (0.64–9.88) | 0 | na | 3 | 2.67 (0.70–10.22) | 0 | na |
| 89 | 4 | … | 1 | 0.47 (0.03–6.42) | 3 | na | 1 | na |
| Any HPV | 158 | 0.98 (0.80–1.20) | 16 | 1.18 (0.71–1.94) | 133 | 0.93 (0.75–1.16) | 14 | 1.30 (0.76–2.22) |
| Alphapapillomavirus species || | ||||||||
| Species 3 | 48 | 1.11 (0.76–1.63) | 7 | 0.78 (0.36–1.68) | 40 | 0.96 (0.64–1.45) | 6 | 0.85 (0.37–1.96) |
| Species 7 | 44 | 0.91 (0.62–1.34) | 6 | 0.88 (0.38–2.01) | 34 | 0.80 (0.52–1.23) | 5 | 0.65 (0.26–1.63) |
| Species 9 | 74 | 0.99 (0.73–1.33) | 9 | 1.29 (0.66–2.52) | 69 | 1.09 (0.80–1.50) | 9 | 1.56 (0.79–3.06) |
| Species 10 | 38 | 1.37 (0.85–2.21) | 5 | 0.96 (0.39–2.41) | 31 | 1.16 (0.69–1.94) | 3 | 0.72 (0.22–2.38) |
na, not applicable.
Definition 1: One negative follow-up visit after an HPV-positive test.
Definition 2: Two negative follow-up visits after an HPV-negative test.
Dominant model: mutant homozygote (CC)+heterozygote (TC) versus WT homozygote (TT).
Recessive model: WT homozygote (TT)+heterozygote (TC) versus mutant homozygote (CC).
||Species 3 (HPV61, -72, -62, -81, -83, -84 and -89); species 7 (HPV18, -39, -45, -59, -68 and -70); species 9 (HPV16, -31, -33, -35, -52, -58 and -67); species 10 (HPV6, -11, -44 and -55).
Fig. 1.
HPV clearance with any HPV type and the TLR9 genotypes [WT homozygote (TT), heterozygote (TC) or mutant homozygote (CC)] in the Kaplan–Meier analysis.
Except for a significant protection against Alphapapillomavirus 9 species infections with the dominant model (HR = 0.56, 95 % CI 0.34–0.95), there were no associations between TLR9 phenotypes and type-specific HPV persistence (data not shown).
In the analyses restricted to women with a high likelihood of exposure to HPV, the only significant type-specific association was a protective effect for HPV73 (OR = 0.22; 95 % CI 0.05–0.96) in the dominant model; no significant associations between TLR9 genotype and HPV clearance were observed (data not shown).
Discussion
We analysed the role of a SNP (rs5743836) in the promoter region of the TLR9 gene (T1237C) and its association with risk of acquisition and clearance of type-specific HPV infection using a case–control sampling strategy within the Ludwig–McGill Cohort Study. A previous study found a different TLR9 variant (SNP rs352140) to be a risk factor for cervical cancer (Roszak et al., 2012), whilst another failed to detect such an association (Pandey et al., 2011). Other studies on TLR9 and cervical HPV have focused on expression levels of TLR9 among HPV16-positive pre-cancerous or cancerous cervical lesions (Daud et al., 2011; Hasimu et al., 2011; Lee et al., 2007; Pandey et al., 2011; Weng et al., 2011).
The most common TLR9 genotype (T1237C) was TT followed by TC and CC. We found no evidence of substantial variation among these genotypes with respect to type- or group-specific HPV infection risk. Admittedly, the low prevalence of some types and the rarity of the CC genotype precluded a more robust assessment. The only significant result with our analyses restricted to women with high likelihood of exposure to HPV was on the protective effect of HPV73 in the dominant model, but the overall HPV73 positivity among the cases was low, being only 5.8 %. However, the findings with HPV73 or with any other HPV type could not be corroborated in the analyses of HPV clearance or persistence. Therefore, one might speculate that the association with HPV73 was mostly due to chance because of the high number of associations that were examined. We also did not observe an association between cases of cervical cytological abnormalities and the TLR9 polymorphism (data not shown), which fails to support a mediating role for this SNP after HPV acquisition. We did not see an association between the ethnic background and the TLR9 gene (T1237C) (data not shown), which was seen in a previous study showing an increased risk for asthma with Caucasians but not African Americans (Lazarus et al., 2003).
Our study has important limitations. We chose a hypothesis-driven, focused view on only one polymorphism in the TLR9 promoter; our findings do not preclude an effect for the entire TLR9 gene or from other innate immunity genes that mediate PAMP responses. Although we had a large sample of women who developed cervical HPV infection over a period of many years and suitable control groups based on follow-up and exposure opportunity, our sample sizes for individual HPV types were admittedly small. Although we did not observe any associations with the subset of cases that developed cervical lesions, there were less than 100 of the latter and only a few would be informative at the level of HPV type. Sample size requirements would be substantially greater than those fulfilled by the present study if TLR gene responses are mediated via specific HPV types.
HPV16 may evade innate host immunity by downregulating the expression of TLR9, thus becoming persistent (Hasan et al., 2007). This was also seen with incident HPV16 infections that cleared and were associated with increased expression of TLR9 compared with women who did not clear the virus (Daud et al., 2011). However, another study showed the opposite, with an increase in TLR9 expression in cervical neoplasia samples (Lee et al., 2007). In view of our findings, it seems unlikely that the polymorphism (T1237C) in the promoter region of TLR9 gene plays a key role in determining the course of cervical HPV infections. In contrast, our findings do not preclude a possible mediating mechanism via viral interference on the expression levels of the TLR9 receptors.
Methods
Subjects.
Recruitment and follow-up for the Ludwig–McGill Cohort Study took place between 1993 and 2005 in a population of low-income women in São Paulo, Brazil. The women were followed for up to 10 years. Eligible women were: between 18 and 60 years of age, had an intact uterus, were not pregnant or planning to become pregnant in the next 12 months and had not been treated for cervical disease in the last 6 months prior to enrolment. The study methods have been described in detail elsewhere (Franco et al., 1999). The study was approved by the ethical review boards of the participating institutions in Brazil and Canada. Informed consent was obtained from all participants prior to enrolment.
Altogether, 2462 eligible women were enrolled into the study, among whom three different groups were derived for the present analyses for the purpose of evaluating the association between the SNP (rs5743836) polymorphism on the promoter region of TLR9 and the risk of HPV infection. The first group (cases) comprised 500 randomly selected women from among those (n = 1215) who had a cervical HPV infection (of one or more HPV types) detected at least once at enrolment or at any point during follow-up. Two control groups were selected, as follows: control group 1 was a random sample of 300 women who always tested HPV negative during the entire study (out of the remainder of 1247 women), irrespective of how many follow-up visits were completed; control group 2 consisted of all 234 women who were always HPV negative as above but had a minimum of ten visits completed (approx. 7 years) in the study, including the enrolment. Expectedly, there was an overlap between these two control groups: 56 women fulfilled the criteria for inclusion in both control groups. In the analyses, these two control groups were treated independently.
Samples.
DNA was extracted previously (Rabachini et al., 2010) from exfoliated cervical cells and purified by spin column chromatography using a GFX Genomic Blood DNA Purification kit (Amersham Pharmacia Biotech) or organic solvent extraction (phenol/chloroform method) and 1 µl DNA extract was used for PCRs in a 50 µl final volume.
TLR9 genotyping.
TLR9 genotype determination was performed using bidirectional PCR amplification of specific alleles (Bi-PASA) (Liu et al., 1997), using the following primers (Carvalho et al., 2008): 0.4 µM 5′-TCATTCAGCCTTCACTCAGA-3′ (forward outer primer); 0.4 µM 5′-CACATTCAGCCCCTAGAGGG-3′ (reverse outer primer); 0.05 µM 5′-GGCGGCGGGGGCCTGCTGTTCCCTCTGCCTGA-3′ (inner WT primer); 0.1 µM 5′-GGGCCGGGGGCCATGAGACTTGGGGGAGTTTC-3′ (inner mutant primer). We used AmpliTaq Gold DNA polymerase with 1× PCR Buffer II (Applied Biosystems). Products were analysed by gel electrophoresis in a 1 % agarose gel to distinguish between WT homozygotes (TT), heterozygotes (TC) and mutant homozygotes (CC). The amplimers were: the whole fragment, amplified with the outer primers (644 bp), the WT allele-specific amplimer (395 bp) and the mutant-specific amplimer (275 bp). Validation of the Bi-PASA method was done by sequencing several of the generated fragments to confirm the different polymorphisms (data not shown).
Statistical analyses.
All statistical analyses were carried out with Stata v12.1 (STATA Corp.). We compared the rates of type-specific HPV positivity for the different TLR9 genotypes among cases. We calculated the age-adjusted ORs and respective 95 % CIs using logistic regression to measure the magnitude of the associations between TLR9 genotype and infection risk, defined as overall risk of any type, individual types or grouped types. To estimate the association between TLR9 genotype and infection, subjects who tested positive for the presence of a particular TLR9 allele were compared with those who lacked that allele. This resulted in two models of phenotype attribution: dominant, defined as CC+TC versus TT, and recessive, defined as TT+TC versus CC.
Analyses of the different outcome variables with respect to TLR9 polymorphism were carried out for the two control groups separately. We omitted HPV types from some analyses if their prevalence was not at least 5 % among cases. We also used proportional hazards regression to evaluate the effect of TLR9 variation on type-specific HPV clearance among cases using two different definitions for a cleared infection episode: (i) one negative follow-up visit following HPV positivity, and (ii) two consecutive negative visits following HPV positivity. A third outcome was persistent HPV infection. Persistence was defined as women (among cases) who tested positive for the same HPV type in two or more consecutive visits.
To test whether the effect of certain alleles depended on the level of exposure opportunity to HPV, a separate analysis was carried out based on the women’s sexual history to define a subgroup with a high likelihood of exposure to HPV. This type of analysis has been carried out previously (Ferguson et al., 2011; Mahmud et al., 2007) and was discussed methodologically by Halloran & Struchiner (1995). This subgroup included 51.2 % of the cases and controls and consisted of women who had their first sexual intercourse at age 17 or younger, or who had had more than five lifetime sexual partners. This covariate combination was chosen because the prevalence of HPV is highest in women with such a sexual behaviour profile (Trottier & Franco, 2006).
Acknowledgements
The Ludwig–McGill Cohort Study team members affiliated with the Ludwig Institute for Cancer Research in Sao Paulo, Brazil, are Maria Luiza Baggio, Lenice Galan, João Simão Sobrinho, José Carlos Mann Prado, Silvaneide Ferreira, Lara Termini, Maria Cecília Costa, Romulo Miyamura, Andrea Trevisan, Patricia Thomann, João Candeias, Laura Sichero, Paula Rahal, Antonio Ruiz, Jane Kaiano, Monica Santos, Patricia Savio, Paulo Maciag, Tatiana Rabachini and Luisa Villa (Co-Principal Investigator); members affiliated with McGill University in Montreal, Canada, are Marie-Claude Rousseau, Salaheddin Mahmud, Nicolas Schlecht, Helen Trottier, Alex Ferenczy, Thomas Rohan, Myriam Chevarie-Davis, Karolina Louvanto, Joseph Tota, Agnihotram Ramanakumar, Eliane Duarte, Sophie Kulaga, Juliette Robitaille, Robert Platt and Eduardo Franco (Principal Investigator). We thank Nuno Miguel Sampaio Osório and Fernando José dos Santos Rodrigues of the Instituto de Investigação em Ciências da Vida e da Saúde (ICVS), University of Minho, Braga Portugal, for stimulating discussions. Financial support for this study was provided by the Ludwig Institute for Cancer Research (intramural grant to L. L. V. and E. L. F.), the US National Cancer Institute (grant CA70269 to E. L. F.), the Canadian Institutes of Health Research (operating grant 49396 and team grant 83320 to E. L. F.), a FAPESP grant 2008/03232-1, a PhD fellowship to L. B. O. (2010/18388-7) and a Sigrid Jusélius Foundation fellowship to K. L.
References
- Carvalho A., Pasqualotto A. C., Pitzurra L., Romani L., Denning D. W., Rodrigues F. (2008). Polymorphisms in Toll-like receptor genes and susceptibility to pulmonary aspergillosis. J Infect Dis 197, 618–621 10.1086/526500 [DOI] [PubMed] [Google Scholar]
- Carvalho A., Osório N. S., Saraiva M., Cunha C., Almeida A. J., Teixeira-Coelho M., Ludovico P., Pedrosa J., Pitzurra L. & other authors (2011). The C allele of rs5743836 polymorphism in the human TLR9 promoter links IL-6 and TLR9 up-regulation and confers increased B-cell proliferation. PLoS ONE 6, e28256 10.1371/journal.pone.0028256 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carvalho A., Cunha C., Almeida A. J., Osório N. S., Saraiva M., Teixeira-Coelho M., Pedreiro S., Torrado E., Domingues N. & other authors (2012). The rs5743836 polymorphism in TLR9 confers a population-based increased risk of non-Hodgkin lymphoma. Genes Immun 13, 197–201 10.1038/gene.2011.59 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Corr S. C., O’Neill L. A. (2009). Genetic variation in Toll-like receptor signalling and the risk of inflammatory and immune diseases. J Innate Immun 1, 350–357 10.1159/000200774 [DOI] [PubMed] [Google Scholar]
- Daud I. I., Scott M. E., Ma Y., Shiboski S., Farhat S., Moscicki A. B. (2011). Association between Toll-like receptor expression and human papillomavirus type 16 persistence. Int J Cancer 128, 879–886 10.1002/ijc.25400 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ferguson R., Ramanakumar A. V., Richardson H., Tellier P. P., Coutlée F., Franco E. L., Roger M. (2011). Human leukocyte antigen (HLA)-E and HLA-G polymorphisms in human papillomavirus infection susceptibility and persistence. Hum Immunol 72, 337–341 10.1016/j.humimm.2011.01.010 [DOI] [PubMed] [Google Scholar]
- Franco E., Villa L., Rohan T., Ferenczy A., Petzl-Erler M., Matlashewski G., Ludwig–McGill Study Group (1999). Design and methods of the Ludwig–McGill longitudinal study of the natural history of human papillomavirus infection and cervical neoplasia in Brazil. Rev Panam Salud Publica 6, 223–233 10.1590/S1020-49891999000900001 [DOI] [PubMed] [Google Scholar]
- Halloran M. E., Struchiner C. J. (1995). Causal inference in infectious diseases. Epidemiology 6, 142–151 10.1097/00001648-199503000-00010 [DOI] [PubMed] [Google Scholar]
- Hasan U. A., Bates E., Takeshita F., Biliato A., Accardi R., Bouvard V., Mansour M., Vincent I., Gissmann L. & other authors (2007). TLR9 expression and function is abolished by the cervical cancer-associated human papillomavirus type 16. J Immunol 178, 3186–3197 [DOI] [PubMed] [Google Scholar]
- Hasimu A., Ge L., Li Q. Z., Zhang R. P., Guo X. (2011). Expressions of Toll-like receptors 3, 4, 7, and 9 in cervical lesions and their correlation with HPV16 infection in Uighur women. Chin J Cancer 30, 344–350 10.5732/cjc.010.10456 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lazarus R., Klimecki W. T., Raby B. A., Vercelli D., Palmer L. J., Kwiatkowski D. J., Silverman E. K., Martinez F., Weiss S. T. (2003). Single-nucleotide polymorphisms in the Toll-like receptor 9 gene (TLR9): frequencies, pairwise linkage disequilibrium, and haplotypes in three U.S. ethnic groups and exploratory case–control disease association studies. Genomics 81, 85–91 10.1016/S0888-7543(02)00022-8 [DOI] [PubMed] [Google Scholar]
- Lee J. W., Choi J. J., Seo E. S., Kim M. J., Kim W. Y., Choi C. H., Kim T. J., Kim B. G., Song S. Y., Bae D. S. (2007). Increased Toll-like receptor 9 expression in cervical neoplasia. Mol Carcinog 46, 941–947 10.1002/mc.20325 [DOI] [PubMed] [Google Scholar]
- Liu Q., Thorland E. C., Heit J. A., Sommer S. S. (1997). Overlapping PCR for bidirectional PCR amplification of specific alleles: a rapid one-tube method for simultaneously differentiating homozygotes and heterozygotes. Genome Res 7, 389–398 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lu K. C., Yang H. Y., Lin Y. F., Kao S. Y., Lai C. H., Chu C. M., Wu C. C., Su S. L. (2011). The T-1237C polymorphism of the Toll-like receptor-9 gene is associated with chronic kidney disease in a Han Chinese population. Tohoku J Exp Med 225, 109–116 10.1620/tjem.225.109 [DOI] [PubMed] [Google Scholar]
- Mahmud S. M., Robinson K., Richardson H., Tellier P. P., Ferenczy A. S., Roger M., Coutlee F., Franco E. L. (2007). HLA polymorphisms and cervical human papillomavirus infection in a cohort of Montreal University students. J Infect Dis 196, 82–90 10.1086/518612 [DOI] [PubMed] [Google Scholar]
- Misch E. A., Hawn T. R. (2008). Toll-like receptor polymorphisms and susceptibility to human disease. Clin Sci (Lond) 114, 347–360 10.1042/CS20070214 [DOI] [PubMed] [Google Scholar]
- Netea M. G., Wijmenga C., O’Neill L. A. (2012). Genetic variation in Toll-like receptors and disease susceptibility. Nat Immunol 13, 535–542 10.1038/ni.2284 [DOI] [PubMed] [Google Scholar]
- Ng M. T., Van’t Hof R., Crockett J. C., Hope M. E., Berry S., Thomson J., McLean M. H., McColl K. E., El-Omar E. M., Hold G. L. (2010). Increase in NF-κB binding affinity of the variant C allele of the Toll-like receptor 9 −1237T/C polymorphism is associated with Helicobacter pylori-induced gastric disease. Infect Immun 78, 1345–1352 10.1128/IAI.01226-09 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Novak N., Yu C. F., Bussmann C., Maintz L., Peng W. M., Hart J., Hagemann T., Diaz-Lacava A., Baurecht H. J. & other authors (2007). Putative association of a TLR9 promoter polymorphism with atopic eczema. Allergy 62, 766–772 10.1111/j.1398-9995.2007.01358.x [DOI] [PubMed] [Google Scholar]
- Pandey S., Mittal B., Srivastava M., Singh S., Srivastava K., Lal P., Mittal R. D. (2011). Evaluation of Toll-like receptors 3 (c.1377C/T) and 9 (G2848A) gene polymorphisms in cervical cancer susceptibility. Mol Biol Rep 38, 4715–4721 10.1007/s11033-010-0607-z [DOI] [PubMed] [Google Scholar]
- Rabachini T., Trottier H., Franco E. L., Villa L. L. (2010). Validation of dot blot hybridization and denaturing high performance liquid chromatography as reliable methods for TP53 codon 72 genotyping in molecular epidemiologic studies. BMC Genet 11, 44 10.1186/1471-2156-11-44 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rosa M. I., Moraes M. V., Vuolo F., Petronilho F., Bozzetti M. C., Medeiros L. R., Igansi C. N., Silva F. R., Dal-Pizzol F., Rosa D. D. (2012). Association of interleukin-6 in women with persistence of DNA-HPV: a nested case–control study. Arch Gynecol Obstet 285, 143–148 10.1007/s00404-011-1925-7 [DOI] [PubMed] [Google Scholar]
- Roszak A., Lianeri M., Sowińska A., Jagodziński P. P. (2012). Involvement of Toll-like receptor 9 polymorphism in cervical cancer development. Mol Biol Rep 39, 8425–8430 10.1007/s11033-012-1695-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schlecht N. F., Kulaga S., Robitaille J., Ferreira S., Santos M., Miyamura R. A., Duarte-Franco E., Rohan T. E., Ferenczy A. & other authors (2001). Persistent human papillomavirus infection as a predictor of cervical intraepithelial neoplasia. JAMA 286, 3106–3114 10.1001/jama.286.24.3106 [DOI] [PubMed] [Google Scholar]
- Trottier H., Franco E. L. (2006). The epidemiology of genital human papillomavirus infection. Vaccine 24 (Suppl. 1), S1–S15 10.1016/j.vaccine.2005.09.054 [DOI] [PubMed] [Google Scholar]
- Vandermark E. R., Deluca K. A., Gardner C. R., Marker D. F., Schreiner C. N., Strickland D. A., Wilton K. M., Mondal S., Woodworth C. D. (2012). Human papillomavirus type 16 E6 and E7 proteins alter NF-κB in cultured cervical epithelial cells and inhibition of NF-κB promotes cell growth and immortalization. Virology 425, 53–60 10.1016/j.virol.2011.12.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Weng Y., Wang Y., Shi Y., Zhou W., Wang H., Wang C. (2011). TLR9 expression and its role in chemosensitivity to DDP in human cervical cancer cells in vitro. J Huazhong Univ Sci Technolog Med Sci 31, 550–554 10.1007/s11596-011-0488-y [DOI] [PubMed] [Google Scholar]