The TLR9 2848C/T Polymorphism Is Associated with the CMV DNAemia among HIV/CMV Co-Infected Patients

Toll-like receptors (TLRs) recognize pathogen-associated molecular patterns and are essential components of the host’s innate immune response. The aim of this study was to determine the TLR9 genotype frequency and investigate the association between TLR9 polymorphisms and cytomegalovirus (CMV) DNAemia in human immunodeficiency virus (HIV)/CMV co-infected patients. A total of 205 HIV/CMV co-infected adults were screened for the presence of the four TLR9 polymorphisms (−1237T/C, −1486T/C, 1174G/A, and 2848C/T) by using polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). Mutation presented in at least one allele of the TLR9 2848C/T single nucleotide polymorphism (SNP) was associated with the occurrence of CMV DNAemia among HIV-infected patients with CMV co-infection (p = 0.004). The level of CMV DNA was higher in patients who were homozygous recessive or heterozygous for the 2848C/T polymorphism compared with those who had a wild-type genotype for this polymorphism (p = 0.005). Mutation detected in at least one allele of this SNP was also associated with a lower interferon type β (IFN-β) concentration (p = 0.048), while no relationships between TLR9 −1237T/C, −1486T/C, and 1174G/A SNPs and CMV DNAemia were observed. Our findings suggest that the mutation present in at least one allele of the TLR9 2848C/T SNP may be associated with the active CMV infection in HIV/CMV co-infected subjects.


Introduction
Human cytomegalovirus (CMV) is a member of the Herpesviridae family and is a leading cause of morbidity and mortality in immunocompromised hosts, including patients with AIDS, organ transplant recipients, and neonates [1][2][3]. The majority of CMV infections are asymptomatic or subclinical among immunocompetent hosts. CMV infection in immunocompromised patients may lead to various clinical manifestations, such as pneumonia, nephritis, hepatitis, gastroenteritis, or central nervous system disease of severe or even fatal course [4,5]. Furthermore, CMV retinitis is the most disabling and characteristic disease in AIDS patients [4]. Hence, the identification of host molecular determinants of the innate immune response to viruses is critical to understanding the pathogenesis of HIV/CMV co-infection and developing novel therapeutic strategies.
Toll-like receptors (TLRs) are a family of transmembrane proteins, which are essential components of the innate immune response to various bacterial, parasitic, and viral pathogens. TLRs recognize and bind conserved pathogen-associated molecular patterns (PAMPs) shared by large groups of microorganisms, which leads to the production of type I interferons (IFNs) and other proinflammatory cytokines. TLR9 recognizes cytosinephosphate-guanosine (CpG) motifs in viral DNA and activates interferon regulatory factor 7 (IRF-7), which leads to type I IFNs secretion. Host-specific differences in the TLR9 gene are likely to alter the course of an adaptive immune response, especially in chronic viral infection [6][7][8]. It was found that chronic viral infections can impair TLR9 signaling and down-regulate TLR9 gene expression [9]. Several published studies have analyzed the role of TLR single nucleotide polymorphisms (SNPs) in immunocompromised patients with CMV infection [10][11][12][13][14][15][16][17][18]. A large genome-wide association cohort study in patients after hematopoietic cell transplantation revealed that the TLR9 SNPs were associated with the CMV reactivation (rs5743836) and disease (rs352140) [11]. The TLR9 rs352139 and rs352140 SNPs had a significant impact on the risk of acute graft-versus-host disease and early CMV infection in allogeneic hematopoietic recipients [12]. The findings suggest a potentially important role of host TLR9 polymorphisms in HIV disease progression [8,[19][20][21][22][23][24][25][26][27][28]. The TLR9 rs352140 GG genotype was associated with an increased risk of HIV infection, disease progression, and CD4 count [20,27,29]. Mutation present in at least one allele of the TLR9 1635A/G SNP (rs352140) was found to be associated with increased CMV acquisition in HIV-exposed infants [19]. Two TLR9 polymorphisms, −1237T/C (rs5743836) and −1486T/C (rs187084), are located within the promoter region and most likely alter the functional ability of the receptor. Functional analysis of the −1237T/C SNP showed that the C allele created a putative NF-kB binding site and increased TLR9 transcriptional activity, driven mainly by activation of NF-kB [30]. Although the 2848G/A SNP (rs352140) is a synonymous substitution, it was associated with alterations in gene expression [31].
The TLR4 Asp299Gly and Thr399Ile polymorphisms have also influenced the susceptibility of other serious infections (e.g., CMV and toxoplasmosis) in patients with advanced HIV infection [32]. However, studies focusing on the effect of TLR9 SNPs on HIV/CMV co-infection and CMV replication in adult subjects are lacking.
In the present study, we elucidated the TLR9 genotype distribution and investigated the correlation between the four specific TLR9 SNPs (−1237T/C, −1486T/C, 1174G/A, and 2848C/T) and CMV DNAemia in HIV/CMV co-infected subjects.

Study Population
The study was approved by the Ethics Committee of the Medical University of Lodz (RNN/211/06/KE and RNN/33/17/KE) and was conducted in accordance with the Declaration of Helsinki and the good clinical practice guidelines. Written informed consent was obtained from all participants and parents/legal guardians of participants (age < 18 years) before study entry. DNA isolated from the whole blood of 205 HIV/CMV co-infected patients (median age: 37.2; range 17-72 years) were included in this study (Table 1). From 2006 to 2008, and from May 2017 to March 2018, 43 HIV/CMV coinfected subjects with a CD4 + T count ≤200 cells/mm 3 and 162 individuals with a CD4 + T cell count > 200 cells/mm 3 were bled once for this study. The HIV/CMV co-infected subjects were enrolled in this study during routine HIV disease monitoring. The HIV RNA was detected in 35/205 (17.1%) subjects. The mean HIV RNA concentration in peripheral blood was 3.98 × 10 3 copies/mL ± 4.11 × 10 4 copies/mL (median: 0, range: 0-5.82 × 10 5 copies/mL). HIV viremia was determined using the RT-PCR method (Cobas AmpliPrep/Cobas TaqMan HIV-1 Test, Roche Diagnostics GmbH, Mannheim, Germany). CMV infection was confirmed by the detection of CMV DNA in the whole blood and/or specific anti-CMV IgG antibodies in the serum samples. The CMV IgG antibodies were detected in 192/205 (93.7%) of patients with HIV/CMV co-infection and undetected in 13/205 (6.3%) of subjects. Patients with negative serologic status were positive for the CMV DNAemia. No symptoms specific for cytomegaly were observed in almost all (203/205, 99.0%) patients with HIV/CMV co-infection. Among patients who developed symptoms, one patient had retinitis, while pneumonia has been reported in the second patient. All patients were Caucasian and were enrolled from the central area of Poland (Lodz).

Detection of TLR9 Polymorphisms
Genomic DNA was isolated from blood samples using a QIAamp DNA Blood Mini Kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's instructions. Genotyping for TLR9 (−1237T/C, rs5743836; −1486T/C, rs187084; 1174G/A, rs352139; and 2848C/T, rs352140) SNPs were performed by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) as we described elsewhere [10]. The reaction was performed in a Veriti 96 Well Thermal Cycler (Applied Biosystems, Foster City, CA, USA). The concentration and purity of DNA were determined using a NanoDrop 2000c UV-Vis Spectrophotometer (Thermo Scientific, Waltham, MA, USA). The digested fragments were analyzed using a QIAxcel system (Qiagen). The selected samples of each TLR9 SNP were sequenced using the 96-capillary 3730xl DNA Analyzer (Applied Biosystems) to confirm the detected genotypes.

Assessment of CMV DNAemia
The analyses of the CMV DNA copy number in DNA isolates from the whole blood samples were performed using quantitative real-time PCR (qRT-PCR) as previously described [33]. The amplification was performed using a 7900HT Fast Real-Time PCR system (Applied Biosystems) with the primer set located within the UL55 gene.

Measurement of Cytokine Levels
The concentration of serum cytokines was determined using the BD Cytometric Bead Array (CBA) Human Th1/Th2/Th17 Cytokine kit (BD Biosciences, San Jose, CA, USA) by the LSR II BD flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA). Levels of IFN-α, IFN-β (PBL Assay Science, Piscataway, NJ, USA), and IL-7 (Thermo Scientific, Frederick, MD, USA) were estimated using ELISA kits according to the manufacturer's recommendations.

Statistical Analyses
Statistical analyses were performed using GraphPad Prism 5.00 (GraphPad Software, San Diego, CA, USA) or SPSS 25.0 for Windows (SPSS Inc., Chicago, IL, USA). The comparison of genotypes and the allele frequency between the HIV/CMV co-infected patients with or without CMV DNAemia was evaluated by Fisher's exact test. The Mann-Whitney U test was used to describe the relationship between the TLR9 SNPs and CMV DNA copy number or IFN-β concentration. Hardy-Weinberg equilibrium (HWE), linkage disequilibrium (LD), and haplotype analyses were performed using the SNPStats software (http://www.snpstats.net/start.htm, accessed on 30 November 2020). The odds ratio (OR) with a 95% confidence interval (95% CI) for assessing the effect of genotype distribution and allelic frequencies on the occurrence of CMV DNAemia in co-infected subjects was calculated by logistic regression analysis. Statistical significance was set at p ≤ 0.05. Bonferroni's correction of the significance level (p c ) was applied to account for multiple testing. The p c -value was count by dividing the p-value by the number of comparisons (the total number of SNPs in the data set, n = 3). The level of significance was p c < 0.017 (0.05/3).

Heterozygous and Homozygous Recessive Genotypes of the TLR9 2848C/T SNP Are Prevalent in HIV/CMV Co-Infected Patients with CMV DNAemia
The TLR9 −1237T/C, −1486T/C, 1174G/A, and 2848C/T SNPs were genotyped in 205 HIV/CMV co-infected patients. The demographic and clinical characteristics of the examined patients are summarized in Table 1. The wild-type −1237 TT genotype was detected in almost all (199/205; 97.1%) of HIV/CMV co-infected patients ( Table 2). No significant differences for −1486C/T and 1174G/A SNPs were found. For the TLR9 2848C/T SNP, the genotype distribution was significantly different between the HIV/CMV co-infected patients with CMV DNAemia and in co-infected subjects without CMV DNA in blood samples. Mutation present in at least one allele of the 2848C/T SNP was detected more frequently in HIV/CMV co-infected patients with CMV DNAemia than in co-infected individuals without CMV DNAemia (76.0% vs. 57.2%; p = 0.006). The observed genotype frequency of the TLR9 −1486T/C SNP was not in HWE among HIV/CMV co-infected patients and was excluded from further analysis (p > 0.05). Considering the distribution of TLR9 alleles, no significant differences in the frequency of −1237T/C and 1174G/A alleles were observed ( Table 2, p > 0.05). In the case of the 2848C/T SNP, the T allele was more common in HIV/CMV co-infected patients with CMV DNAemia than in subjects without CMV DNA in the blood (52.1% vs. 37.5%, p = 0.005; Table 2).

Mutation Present in at Least One Allele of the TLR9 2848C/T SNP Occurs More Frequently in HIV/CMV Co-Infected Patients with CMV Viremia
Mutation presented in at least one allele of the TLR9 2848C/T SNP occurred at least two times more frequently in HIV/CMV co-infected patients with CMV DNAemia than in HIV/CMV co-infected subjects without CMV DNA (OR: 2.38; 95% CI: 1.30-4.34; p = 0.004, in the dominant model; Table 3). This TLR9 SNP also showed a higher rate of occurrence of viral co-infection even after Bonferroni correction for multiple testing (p c = 0.017) in the unadjusted model (Table 3). A statistically significant difference was noticed for the influence of the heterozygous genotype of the TLR9 2848C/T SNP on the CMV DNAemia among individuals undergoing antiretroviral treatment (OR: 2.44; 95% CI: 1.25-4.76; p = 0.0049, in the codominant model). In addition, the TT genotype of the TLR9 2848C/T SNP was more frequently observed in the HIV/CMV co-infected individuals with the CD4 + T cell count ≤200 cells/mm 3 than in HIV/CMV co-infected patients with the CD4 + T cell count >200 cells/mm 3 (OR: 2.96; 95% CI: 1.23-7.12; p < 0.001).

The TGT Haplotype of TLR9 −1237T/C, 1174G/A, and 2848C/T SNPs Is Prevalent in HIV/CMV Co-Infected Patients
Haplotype analysis of TLR9 −1237T/C, 1174G/A, and 2848C/T SNPs showed that the most frequent haplotype was TGT (30.4% and 42.9% for HIV/CMV co-infected patients with or without CMV DNAemia, respectively). The next most prevalent haplotype was TGC (34.1% and 21.4% for HIV/CMV co-infected patients with CMV DNAemia and HIV/CMV co-infected subjects without CMV DNAemia, respectively) and was associated with a higher rate of the occurrence of CMV viremia (OR: 1.95; 95% CI: 1.10-3.47; p = 0.023). No evidence of LD in −1237T/C, −1174G/A, and 2848C/T SNPs was found (r 2 < 0.2).

The TLR9 2848C/T SNP Is Associated with Lower Serum IFN-β Level
To further investigate the clinical relevance of TLR9 SNPs and the outcome of infection, we determined the serum concentrations of target cytokines in HIV/CMV co-infected patients. Subjects who had wild-type genotype for the TLR9 2848C/T SNP had higher serum IFN-β concentration compared with those who were heterozygous or homozygous recessive for this polymorphism (708.2 pg/mL ± 1126.7 pg/mL vs. 60.7 pg/mL ± 128.0 pg/mL; p = 0.048; Figure 2). No other cytokines showed significant associations with the TLR9 SNPs in both of the study groups. most frequent haplotype was TGT (30.4% and 42.9% for HIV/CMV co-infected patients with or without CMV DNAemia, respectively). The next most prevalent haplotype was TGC (34.1% and 21.4% for HIV/CMV co-infected patients with CMV DNAemia and HIV/CMV co-infected subjects without CMV DNAemia, respectively) and was associated with a higher rate of the occurrence of CMV viremia (OR: 1.95; 95% CI: 1.10-3.47; p = 0.023). No evidence of LD in −1237T/C, −1174G/A, and 2848C/T SNPs was found (r 2 < 0.2).

The TLR9 2848C/T SNP Is Associated with Lower Serum IFN-β Level
To further investigate the clinical relevance of TLR9 SNPs and the outcome of infection, we determined the serum concentrations of target cytokines in HIV/CMV co-infected patients. Subjects who had wild-type genotype for the TLR9 2848C/T SNP had higher serum IFN-β concentration compared with those who were heterozygous or homozygous recessive for this polymorphism (708.2 pg/mL ± 1126.7 pg/mL vs. 60.7 pg/mL ± 128.0 pg/mL; p = 0.048; Figure 2). No other cytokines showed significant associations with the TLR9 SNPs in both of the study groups.

Discussion
It was found that the common polymorphisms of various TLR genes may be associated with increased susceptibility to or protection from several infections [20,21,32,[34][35][36]. Moreover, the genetic variability in TLR signaling pathway molecules (i.e., interleukin-1 receptor-associated kinase 4 [IRAK4], an inhibitor of nuclear factor κB kinase γ [IKKγ], and an inhibitor of nuclear factor κB α [IκBα]) causes rare inherited immunodeficiencies that can also influence the susceptibility to human diseases [37,38]. Among the TLRs, intracellular TLR9 recognizes unmethylated cytosine-phosphate-guanosine (CpG) DNA motifs in bacteria and viruses [39,40] and is critically required in the process of CMV sensing [41,42]. TLR9 is mainly expressed by cells of the immune system such as dendritic cells, monocytes/macrophages, natural killer cells, and lymphocytes [43]. TLR9 triggers signaling cascades that lead to a pro-inflammatory cytokine response, while TLR ligands induce immune activation in vitro in CD4 + and CD8 + T lymphocytes derived from

Discussion
It was found that the common polymorphisms of various TLR genes may be associated with increased susceptibility to or protection from several infections [20,21,32,[34][35][36]. Moreover, the genetic variability in TLR signaling pathway molecules (i.e., interleukin-1 receptor-associated kinase 4 [IRAK4], an inhibitor of nuclear factor κB kinase γ [IKKγ], and an inhibitor of nuclear factor κB α [IκBα]) causes rare inherited immunodeficiencies that can also influence the susceptibility to human diseases [37,38]. Among the TLRs, intracellular TLR9 recognizes unmethylated cytosine-phosphate-guanosine (CpG) DNA motifs in bacteria and viruses [39,40] and is critically required in the process of CMV sensing [41,42]. TLR9 is mainly expressed by cells of the immune system such as dendritic cells, monocytes/macrophages, natural killer cells, and lymphocytes [43]. TLR9 triggers signaling cascades that lead to a pro-inflammatory cytokine response, while TLR ligands induce immune activation in vitro in CD4 + and CD8 + T lymphocytes derived from HIV patients [44]. Host-specific differences in the TLR9 gene are likely to alter the course of an adaptive immune response, especially in a chronic infection like HIV or CMV. To date, the TLR9 SNPs have been showing to modulate the risk of CMV infection in individuals with the allogeneic hematopoietic stem cell transplantation (allo-HSCT) [11,12], renal transplant recipients [13,14], and infants [10,15]. The TLR9 2848C/T SNP was associated with congenital CMV infection in fetuses and infants [10,15]. Moreover, the TLR9 1635 G (rs352140) allele has been associated with HIV load and disease progression [20,21].
In the present study, we report, for the first time, a relationship between the polymorphisms of the TLR9 gene and the incidence of CMV DNAemia among HIV/CMV co-infected subjects. The mutation detected in at least one allele of the TLR9 2848C/T SNP occurred more frequently in HIV patients with CMV viremia than in subjects without CMV DNAemia. The TLR9 2848 TT genotype seems to be associated with higher levels of CMV DNAemia and CD4 + T cell depletion, whereas the subjects who had a wild-type genotype for this SNP had higher serum IFN-β concentration. Interestingly, the results of other studies have shown that two common TLR9 polymorphisms, −1486C/T and 1635A/G, may be associated with disease progression in HIV-infected patients [20,22]. Joshi et al. demonstrated an association between TLR9 1635A/G SNP and HIV mediated immune activation in both CD4 + and CD8 + T cells as well as interferon gamma-induced protein 10 (IP10) levels [8]. The TLR9 1635 G allele reduced HIV acquisition [23], was associated with lower HIV load [21], and was protective against disease progression in HIV-infected adults [21,26]. Most notably, the TLR9 1635 A allele increased the risk of HIV acquisition and viral load in the Kenyan perinatal cohort [19], whereas in children of European ancestry no significant association with the risk of mother-to-child transmission of HIV infection was found [25]. Among African descendants, the 1635 AA genotype was also associated with higher susceptibility to HIV and hepatitis C virus co-infection [24]. Moreover, another TLR9 SNP, 1486C/T was also weakly associated with the lower CD4 + T cell count [8]. A low average CD4 + T cell count, higher HIV load, and a higher probability of disease progression were associated with the 1635 AA genotype [26]. A significant correlation between the 1635 GG genotype and low average CD4 + T cell count during the viremic period in HIV-infected patients was found in the second study [27]. The donor genotype at the rs352140 SNP was significant for CMV disease in recipients following HSCT [11]. Clinical evidence has also demonstrated that the donor homozygous recessive genotype of the TLR9 1635A/G SNP was associated with a higher incidence of CMV reactivation and yielded a protective effect against acute graft versus host disease in allo-HSCT recipients [12]. Ito et al. [40] reported that the TLR9 −1486T/C polymorphism was not associated with Vogt-Koyanagi-Harada disease, an autoimmune disorder in which the cytomegalovirus antigen has been hypothesized as a possible triggering factor for the disease. The same SNP has also been found to be significantly associated with bacterial vaginosis among HIV-infected women [28] and bronchiolitis obliterans syndrome in lung transplant recipients [45].
The function of TLR9 SNPs in host antiviral defense and development of the CMV infection remains unknown. The 2848C/T SNP occurs in exon 2 of the TLR9 gene but does not lead to any amino acid change (Pro545Pro), while −1237T/C SNP is located within the putative promoter region of the TLR9 gene and also does not induce an amino acid change. The 1486C/T and 1174G/A polymorphisms are located in the promoter region and intron 1, respectively. Several findings suggest that all polymorphisms can potentially affect TLR9 expression [31,46,47]. It was found that the 1635 G allele is associated with lower TLR9 expression, as well as the combination of the C allele at position −1486 with a G allele at position 1174 can downregulate TLR9 expression [46]. Moreover, Kikuchi et al. [47] reported that the 2848 GG genotype reduced TLR9 expression at the transcriptional level. Therefore, we speculate that the genetic variation of TLR9 that downregulates TLR9 expression could reduce the function of the innate immune response against viral infection. We suppose that the other SNPs in linkage disequilibrium might also be responsible for the functional effect of the individual SNPs that were examined. This study has some limitations. First, our study is limited by its small sample size of HIV/CMV co-infected volunteers. A second limitation is the lack of a control group consisting of age-and sex-matched immunocompetent subjects with CMV DNAemia. However, among immunocompetent individuals, an overt CMV infection is rare, as viral replication is controlled effectively by the immune system. Another limitation of the study includes the fact that HIV-infected patients with CMV co-infection received ART that may have influenced immune responses. Despite several limitations of this study, our data suggest that active CMV infection in HIV-infected individuals may be associated with the TLR9 2848C/T polymorphism. To determine whether this association occurs only in the HIV/CMV-infected group or also in other people, the authors plan to evaluate this association in immunocompromised patients not infected with HIV. Further studies using larger patient groups are needed to confirm our results. Moreover, we suggest that functional and mechanistic studies should be conducted to further understand the molecular mechanisms underlying our observations.

Conclusions
Our study demonstrated significant differences in the rate of occurrence of genetic variants of the TLR9 gene in HIV/CMV co-infected patients. We found that the 2848 CT and TT genotypes seem to be more frequent in HIV/CMV-infected patients with CMV viremia than in HIV/CMV co-infected subjects without CMV DNAemia. HIV/CMV-positive subjects that are homozygous or heterozygous carriers of the 2848C/T SNP seem to develop a higher CMV load. We suggest that genetic polymorphism may negatively regulate TLR9 signaling and contribute to reduced efficiency of antiviral response, resulting in CMV replication. These findings provide insight into the role of host TLR9 polymorphism in the frequency of CMV infection among HIV-infected subjects. Author Contributions: All authors substantially contributed to this work. Conceptualization, E.P., E.J. and A.J.; methodology, A.J., M.S. and E.P.; formal analysis, A.J. and E.P.; resources, A.J., E.J., M.S., J.K. and E.P.; writing-original draft preparation, A.J., E.P. and E.J; writing-review, A.J, E.P. and E.J. All authors have read and agreed to the published version of the manuscript. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data generated during the current study are available from the corresponding author on reasonable request.