Genetic variation in toll like receptors 2, 7, 9 and interleukin-6 is associated with cytomegalovirus infection in late pregnancy

Background

Seroprevalence of cytomegalovirus (CMV) amongst women of reproductive age ranges from 40 to 65% in the developed world and can reach 100% in developing countries [1, 2]. CMV infection in pregnancy, in the setting of both primary infection and reinfection, can be potentially transmitted to the foetus and or neonate, resulting in congenital CMV (cCMV). The consequences of CMV range from asymptomatic viraemia to potentially life changing conditions which include mental retardation and congenital sensorineural hearing loss. Studies have implicated maternal demographics, socioeconomics and HIV status among the strongest determinants of the biased occurrence and vertical transmission of CMV [3–7]. Furthermore, maternal immune responses to CMV infection and/or reactivation actively modulate CMV related disease outcomes [8]. Thus, variation in genes that encode components of the immune system that are directly or indirectly involved in the pathogenesis of CMV have been implicated in CMV infection outcomes [9]. However, the genetic variants, like seroprevalences and the factors influencing CMV epidemiology are heterogenous among populations hence research findings are equivocal.

Toll-like receptors (TLR) are crucial in the detection of viruses in circulation and the subsequent elicitation of an antiviral response [10, 11]. TLRs act as pattern recognition receptors of non-methylated viral CpG-containing DNA which signals the presence of CMV infection [12]. TLR2 and TLR4 are cell surface receptors while TLR3, − 7 and − 9 are endosomal receptors [13, 14]. TLRs facilitate viral attachment and entry resulting in CMV-elicited signalling antiviral responses such as type 1 interferon activation of nuclear factor kappa β (NF-k β) and pro-inflammatory cytokine gene expression [12, 15]. Activation of the type 1 interferon producing cascade and production of cytokines form the major cellular antiviral mechanisms against CMV [16–18]. Single nucleotide polymorphisms (SNPs) in the TLR2, TLR4, TLR7 and TLR9 genes were inconclusively reported to be associated with CMV infection [19–23].

In response to TLR activation, chemokine (interleukin and interferon) genes signal immediate secretion of ILs from cells such as macrophages and T-helper cells. Chemokines that trigger an immune cascade by signalling direct growth, development, maturation, activation and increased life-span of immune cells. In the case of CMV infection, chemokines signal: maturation of B-lymphocytes into plasma cells which produce anti-CMV antibodies, and activation of cytotoxic T cells for destruction of CMV infected cells [24, 25]. The differential response to CMV exposure with some but not all exposed individuals developing CMV-related diseases suggests a possible role of host genetic variation in immune response. A study by Sezgin et al. [26] showed that human interleukin-10 receptor variants potentially interfere with IL-10 binding and signal transduction influencing susceptibility to CMV retinitis. In a large Swiss HIV Cohort Study, the effect of IFNL3 TT/−G substitution, the variant allele was associated with occurrence of CMV retinitis [27]. The same allele was also associated with susceptibility to CMV replication in transplant patients [28].

Detection of host genetic variants which may confer resistance to CMV infection and reactivation could reveal potential therapeutic targets against pregnancy related CMV disease. Furthermore, host genetic determinants of CMV disease outcomes could be used as predictors of adverse outcomes of maternal CMV. While the host genetics of CMV have been studied in other populations, a glaring gap in knowledge exists among Africans. The differences in genomic variation between Africans and other populations cannot be over-emphasised, hence findings from other populations may not be an accurate reflection in Africans.

The aim of the present study was to determine if single nucleotide polymorphisms in genes that encode components of the immune system are associated reactivation of CMV in late pregnancy.

Methods

Results

Discussion

The outcome of an infection is determined, in part, by the intensity of the inflammatory response [32], which varies between individuals and can be regulated at the genetic level [33]. In this study, we hypothesised the possible contribution of genetic variation to the biased occurrence of CMV infection among pregnant women. SNPs may influence the rate and regulatory dynamics of gene transcription, stability of mRNA as well as production and biological activity of resultant protein. We therefore investigated possible association between CMV infection and SNPs in 19 genes which encode proteins that are or may be involved in the immune reaction cascade against CMV. The departure from HWE in polymorphic SNPs is due to their association with CMV infection mainly because the departure is being observed when cases and controls are separated but HWE is maintained when the two groups are combined. We report a significant association between each of; rs10499563, rs179008, rs1816702 and rs352139 SNPs and CMV DNA status. To our knowledge, this is the first report on SNPs and CMV infection in an African setting.

To minimise the confounding effects of age and HIV status, which are directly related to immune function, enrolled participants were age and HIV status matched. The observation that overweight women were less likely to be CMV+ contradicts findings from previous studies where CMV infection was associated with metabolic syndrome, higher BMI and or obesity [34, 35]. Our findings could be due to none of the participants having any form or history of metabolic syndrome. Hence, we were unlikely to observe any significant associations. The observation that CMV positivity is significantly associated with low systolic blood pressure contrasts with previous findings which have shown increasing systolic blood pressure with CMV positivity [36, 37]. It is worth noting that the previous studies were carried out in non-pregnant adults, hence discrepancy in findings could be due to the well documented effects of pregnancy on fluctuations in blood pressure [38, 39] masking the effects of CMV infection.

We found an association between SNP rs10499563 (− 6331 T > C), located within the promoter region of IL-6 gene which regulates the rate of IL-6 gene transcription [40] and CMV DNA status. Individuals carrying the C allele were less likely to be CMV infected, hence likelihood of being CMV DNA positive decreased with genotypes T/T>> > T/C> > C/C. Individuals heterozygous (T/C) and homozygous (C/C) for the variant allele were significantly less likely to be CMV infected than individuals homozygous for the T allele (T/T). The IL-6 gene codes for IL-6, a versatile inflammatory cytokine whose function is related to its expression in the tissue. Smith et al. previously reported higher level of serum IL-6, in individuals with wildtype T/T genotype compared to individuals with C/C genotype, among coronary artery bypass patients (Smith et al. [41]).

Our findings could at least in part, be explained by results from the Smith et al. study. Being a pro-inflammatory cytokine, abundance of IL-6 in circulation could promote CMV activation. In contrast, the low levels of IL-6 associated with the rs10499563C allele would disfavour the occurrence of CMV infection. Serum IL-6 levels were reported to be significantly higher among the CMV infected pregnant women compared to the CMV uninfected in a Chinese cohort [42].

We also report an association between CMV DNA status and rs179008, a non-synonymous A > T (Gln11Leu) polymorphism within exon 3 of the TLR7 gene [43]. The resulting glycine to leucine change has been suggested to code for a functionally impaired TLR7 protein [44, 45]. In the present study, the T allele was associated with significantly lower odds of CMV positivity. Individuals homozygous for the variant allele T/T were significantly less likely to be CMV infected compared to individuals homozygous for the wildtype allele A/A.

Upon recognising pathogen associated molecular patterns (PAMP), TLR7 activate a signalling cascade which activates type I IFN, dendritic cells (DCs) and B lymphocytes [46]. Activated type 1 IFN, DCs and B cells are responsible for pathogen clearance, antigen recognition and antibody production. The induced immune cascade is critical in CMV clearance. In the presence of the T allele which results in a less potent protein, an insufficient signal is mounted by TLR7, hence carriers of the rs179008 T allele are at a greater risk of CMV infection. The rs179008 T allele has been linked with unfavourable outcomes in HIV and other viral infections. The variant was associated with increased susceptibility to HIV-1 and decreased IFNα production in HIV uninfected women [47]. The T allele has also been previously associated with a higher risk of hepatitis C infection and cCMV. Our findings are therefore contrasting with previous reports suggesting that the rs179008A > T SNP could be in linkage disequilibrium with another functional SNP or epistatic gene which masks the effects of rs179008A > T.

CMV DNA status was also associated with rs1816702C > T, a SNP located in intron 2 of the TLR2 gene. The C variant was significantly more prevalent in cases than in controls which means that participants with the rs1816702 C/C genotype were at a higher risk of being CMV+ than those with rs1816702 T/T genotype. TLR2 recognise CMV glycoproteins B (gB) and gH in a process which facilitates entry of CMV into immune cells [15, 48]. The rs1816702T allele is associated with significantly elevated levels of inflammatory monocytes expressing CD14+/TLR2+ receptors than rs1816702C allele [49]. This could explain our findings of a higher risk of CMV among rs1816702C/C carriers because their immune response against CMV is impaired due to lower TLR2 expression compared to the T/T. Homozygosity for the rs1816702C allele has also been associated with increased odds of Mycobacteria leprae infection and inflammatory bowel disease which were attributable to altered NFκB-mediated inflammatory response [50, 51].

The intronic SNP rs352139T > C in the TLR9 gene was also associated with CMV DNA status. Homozygous rs352139C/C individuals were at a significantly higher risk of being CMV+ compared to homozygous T/T carriers. The effect of the C allele on risk of CMV infection was also observed in the dominant and recessive models where the significance of the compound heterozygous (T/C) and homozygous (C/C) genotypes had a greater risk than the homozygous (C/C) alone, relative to the T/T genotype in both cases. The higher risk of CMV positivity in homozygous carriers of the C allele suggest that the polymorphism results in a less potent protein compared to the T allele. Since the polymorphism is intronic, it likely creates an alternative splicing site thus, affecting mRNA transcription and the final protein product. A less potent protein would have decreased ability to form dimers that are required to illicit an immune reaction. Individuals who are homozygous T/T have impaired immune responses against CMV infection, hence are more likely to experience CMV infection or reactivation. The HIV rapid progressor phenotype has been linked to homozygosity for rs352139T allele also due to reduced TRL9 potency [52].

Conflicting findings were reported reduced risk of cCMV associated with the rs352139T/T genotype among infants in Poland [53]. The conflicting effect of rs352139T variant have also been reported in bacterial infection studies in Indonesia and Mexico, perhaps due to ethnic differences [54, 55]. We suggest that rs352139 could be in linkage disequilibrium (LD) with a polymorphic regulatory region that controls TLR9 expression or serves as a functional region SNP. LD patterns differ with level of genetic diversity among different ethnic groups, hence the effects of one SNP may vary from one population to another. Minor allele frequencies for these SNPs which seem to affect CMV infection risk were compared to other populations. TLR2 rs4696480A and TLR4 rs1075993T alleles, respectively, have lower frequencies among Zimbabweans (0.31 and 0.20) and other African populations (0.37 and 0.21) when compared to European (0.52 and 0.77) and Asian (0.57 and 0.86) populations. On the other hand, TLR4 rs7856729T and TLR9 rs5743836G, respectively, are proportionally higher in Zimbabwean (0.38 and 0.36) and other Africans (0.33 and 0.42) when compared to European (0.13 and 0.13) and Asian (0.10 and 0.00) populations. These differences in the distribution of risk alleles of world populations, is likely to lead to differential responses upon exposures to infectious pathogens. Indeed, the adaptive immune responses to the β-coronaviruses, MERS-CoV and SARS-CoV, are that can cause fatal lower respiratory tract infections, are marshalled by T cells, CD4+ T cells, and CD8+ T cells, through among other processes, activate other downstream cytokine and chemokine cascades, such as IL-1, IL-6, IL-8, IL-21 and TNF-β [56]. The molecular patterns displayed by viruses are then sensed by different immune cellular pathogen recognition receptors, including toll-like receptors (TLR:2, 3, 4, 7, 8, and 9) [57]. Whether this genetic heterogeneity among populations plays an active role in the differential prevalence of CMV is unclear and is an area of further research which should also consider the strong influence of environmental factors.

Conclusions

We conclude that TLR2, − 7, − 9 and IL-6 genetic polymorphisms are associated CMV status in late gestation among the black Zimbabweans. TLRs and ILs modulate immune responses to CMV, hence polymorphisms in genes encoding the receptors and cytokines could interfere with the immune mechanisms, hence their association with CMV status. We recommend that future studies consider evaluating the profiles of immune response genes and the polymorphisms in these genes on their possible effects in viral infections. With respect to CMV, we recommend a mother-infant longitudinal approach that will seek to factor in the effect of these immunogenetic profiles in congenital CMV and its possible sequelae.

Supplementary information

Authors’ contributions

DM conceived the ideas, carried out some laboratory experiments, analysed data and drafted the manuscript; KM reviewed the manuscript draft, assisted with laboratory experiments; AK analysed data and reviewed the manuscript draft; MM and AW assisted with analysis and interpretation of the data and read the manuscript draft; CD conceived the ideas, supervised all components as principal investigator and reviewed the manuscript draft. All authors contributed to the final version of the article. The authors read and approved the final manuscript.

Funding

The National Research Foundation (NRF) South Africa and the South African Medical Research Council (SAMRC) funded all the laboratory work through grants to Professor Collet Dandara.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on request or in the dbSNP repository, http://www.ncbi.nlm.nih.gov/SNP/snp_viewTable.cgi?handle=HUMGEN_PHARMGX

Ethics approval and consent to participate

The current study received ethical clearance from the Medical Research Council of Zimbabwe (MRCZ/A/2177) as well as the University of Cape Town Institutional Review Board (HREC628/2017). The study was carried out in accordance with the guidelines of the Helsinki Declaration of 2008 ethics clearance. All participants signed an informed consent form prior to enrolment into the study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Supplementary information

Supplementary information accompanies this paper at 10.1186/s12881-020-01044-8.