Taking into account genetic modifications within the
TLR2 gene, previous studies also reported contribution of
TLR2 2258 G>A coding non-synonymous SNP to HCMV infection, although the mutated homozygotes were found in some studied populations only [
23,
24,
26]. In a cohort of liver transplant recipients treated at the Mayo Clinic, Minnesota, United States, some relationship was determined between homozygosity in the analyzed
TLR2 polymorphism and HCMV infection, especially in tissue-invasive disease [
24]. An in vitro study of the transfected human embryonic kidney (HEK) 293 cells, that were exposed to HCMV gB, showed
TLR2 2258 SNP to have been involved in TLR2 signaling impairment [
23]. Another study, performed in liver transplant recipients with chronic hepatitis C, treated also at the Mayo Clinic, showed a certain association of
TLR2 polymorphism with HCMV load [
26]. The homozygotic status in the analyzed SNP was correlated with cytomegaly, as well as with an increased risk for the disease after adjusting for patienťs age, HCMV serostatus and allograft rejection [
26]. Taking into account our results and the previous literature data,
TLR2 2258 G>A SNP may be involved in congenital infection with HCMV in Polish fetuses and neonates. Considering HCMV DNA loads determined in fetal and neonatal body fluids, no association was observed with genotypic variability within the analyzed polymorhism. Before our study, no attempt had been reported to investigate the function of
TLR2 2258 G>A polymorphism in congenital infection with HCMV. A study performed in 88 infants and 63 adults infected postnatally with HCMV, and in 28 healthy neonates and 50 healthy adults, may suggest a possible protective role of CT heterozygotic status in
TLR2 2029 C>T locus against the infection development among adult patients [
13]. However, similarly to our outcomes, the mentioned study showed that all the analyzed infants were CC homozygotes in the range of 2029 C>T polymorphism [
13]. Additionally, we also found the same distribution of genotypes within
TLR2 1350 T>C SNP among both HCMV infected and uninfected fetuses and neonates, although the polymorphism was reported to be significantly associated with congenital cytomegaly among Japanese children [
25]. Since the C allele in
TLR2 1350 T>C SNP is more frequent among the Japanese than in the European populations (25.48% vs. 6.36%, see
http://www.ncbi.nlm.nih.gov/variation/tools/1000genomes/?q=rs3804100), the geographical origin might be the major reason of differential role of the analyzed polymorphism in susceptibility to congenital infection with HCMV. It seems that further studies with larger groups of patients, congenitally infected with the virus, would be an interesting challenge. Considering
TLR2 2258 G>A SNP, the individual cases of GA heterozygotes and AA recessive homozygotes were observed among HCMV infected infants, and not among the uninfected offsprings [
13]. Similarly, in our study, the prevalence rate of the minor allele in
TLR2 2258 G>A locus was higher among the HCMV infected infants than in the uninfected ones. Moreover, the paper by Jabłońska et al. reported the occurrence of AA homozygotes among the infected adults, but not among uninfected patients [
13]. The GA heterozygotes were not found among non-HCMV infected infants [
13]. In our study, we observed minor alleles in
TLR2 2258 SNP only as heterozygotes, whereas no AA homozygotes were found. It should be emphasized that in our reported study, we evaluated fetuses and neonates on the day of birth, with congenital HCMV infection. In contrast, Jabłońska et al. explored infants, aged 1–12 months, with postnatal HCMV infection, qualified by clinical symptoms or evidence of the infection confirmed by viral DNA detection in whole blood/urine samples after 3 weeks of life, and by the presence of HCMV-specific antibodies [
13]. It is possible that the obtained results are different, since the two different groups of patients were examined. Additionally, the prevalence rates of genotypes and alleles, located within the analyzed
TLR2 polymorphisms, as well as their associations with the occurrence of congenital HCMV infection, estimated in the current research, might have been due to the small sample size cohort of the studied offsprings. Likewise to our results, the prevalence rates of A allele in
TLR2 2258 G>A SNP was determined to be low and a lack of AA homozygotes was also reported in other study groups, such as German, Finnish and Caucasian adults, or Turkish children [
32,
48‐
50]. The heterozygotic status and A allele in
TLR2 2258 G>A polymorphism were significantly more frequently identified among Turkish children with tuberculosis (TB) than in control cases [
32]. In addition, the prevalence rate of
TLR2 2258 SNP was reported to be increased in patients with pulmonary TB alone, as well as with definitive pulmonary plus extrapulmonary TB, as compared to cases with latent TB infection [
32]. In German adult patients with
Candida sepsis, the heterozygotic status in
TLR2 2258 G>A SNP was correlated with altered cytokine release, including increased plasma concentrations of TNF-α and decreased levels of IFN-γ and IL8 [
33]. Considering genotypic variability in
TLR2 2258 locus, it should also be noticed that some populations, living in Barbados, the South-Western USA, Bangladesh, China, Nigeria, Texas, Gambia, Japan, the United Kingdom, Vietnam, Kenya, Sierra Leone, Los Angeles in the USA, as well as in Lima, Peru (see
http://www.ncbi.nlm.nih.gov/variation/tools/1000genomes/?q=rs5743708) do not possess the minor A allele within the reported region. Hence the polymorphism, analyzed in this study, can plausibly be involved in the development of congenital infection with HCMV only in some populations. Given the previous papers on the role of TLR2 molecule in the immune response against HCMV, the altered
TLR2 gene in the range of 2258 G>A SNP may also be involved in the development of congenital infection with the virus in populations, carrying the minor allele. Additionally, previous papers from studies, performed in fetuses and neonates, as well as in children with congenital HCMV infection, also showed some contribution of other
TLR polymorphisms to the occurrence of infection [
14,
25]. Our recent study in fetuses and neonates, with and without congenital HCMV infection, presented that
TLR4 and
TLR9 SNPs were associated with the development of congenital cytomegaly [
14]. Considering the multiple-SNP analysis, performed in the current study for
TLR2,
TLR4 and
TLR9 SNPs, a correlation was found between the presence of AA complex variants for
TLR2 2258 and
TLR9 2848 G>A SNPs and the occurrence of HCMV congenital infection. Moreover, among the infected offsprings, the AA multiple-SNP variants were significantly associated with higher viral loads, estimated in maternal blood and urine samples. The observed increased viral levels in fetal amniotic fluids and maternal urine samples were also correlated with the occurrence of GA complex variants for the analyzed
TLR2 and
TLR9 SNPs among congenitally infected fetuses and neonates. Regarding these two polymorphisms, a previous study, performed for Polish infants with postnatal or unproven congenital HCMV infection, showed both heterozygotes and recessive homozygotes in
TLR9 -1486 T>C and 2848 G>A SNPs to be at almost 4-fold increased risk of HCMV disease in an adjusted model, including HCMV DNA loads [
51]. Considering our outcomes, the common contribution of
TLR2 2258 and
TLR9 2848 G>A polymorphisms to the development of congenital infection seems to be particularly possible, since
TLR9 2848 G>A SNP was also previously reported to be involved in an increased risk of HCMV infection among fetuses and neonates, although the polymorphism is not associated either with amino acid changes of TLR9 molecule or with alterations of the regulatory site of
TLR9 gene [
14,
52]. Hence, the presented data suggest some role of different TLR molecules, as well as of various genetic modifications, located within
TLR genes, in the occurrence of congenital HCMV infection.
Since
TLR2 2258 G>A SNP coding Arg753Gln non-synonymous change is located within a group of highly conserved amino acids at the C-terminal cytoplasmic Toll-interleukin 1 receptor domain of the TLR2 molecule, its contribution to the receptor-induced signal pathways was reported as plausible [
53,
54]. Accordingly to molecular modeling studies of
TLR2 Arg753Gln variation, discrete main and side chain differences were reported, affecting the analyzed residue itself [
54]. The alteration was suggested to be associated with interactions between the intracellular signaling of Toll-IL-1R (TIR) 2 and TIR1 domains [
54]. In addition, the polymorphism was determined to change the electrostatic potential of the DD loop and αD region, related to the Arg753Gln polymorphism, causing a slight movement of the residues, participating in protein-protein interactions [
54].
TLR2 2258 G>A SNP was reported as correlated with impaired agonist-induced tyrosine phosphorylation, heterodimerization of TLR2 with TLR6, and recruitment of Mal and MyD88 adapter proteins [
54]. Out of them, Myd88 is the key molecule involved in the transmission of TLR2 induced signaling pathways of non-specific anti-HCMV response [
54]. Hence, it is plausible that
TLR2 2258 G>A polymorphism, may be involved in the occurrence and development of congenital infection with HCMV through affected TLR2/Myd88 signaling, caused by altered conformation and the electrostatic potential of the TIR2 domain [
54]. In transfected HEK293 cells, treated with
Mycobacterium tuberculosis or mycobacterial components, the presented signal alterations were associated with a decreased phosphorylation of p38, NF-κB activation and IL8 transcription [
54]. Another in vitro study, performed with transfected HEK293 cells, challenged with a tripalmitoylated hexapeptide (Pam3CSK4), showed
TLR2 2258 G>A SNP to have been associated with a substantially reduced activity of TLR2, as well as with 50% decreased activity of the NF-κB-driven reporter gene [
55]. In other studies, the heterozygotic status in
TLR2 2258 G>A locus was also reported to be associated with the molecule hypo-responsiveness upon stimulation with the synthetic TLR2 ligand Pam3CysSK4, as well as reduced signaling via the TLR2/TLR1-complex [
38,
56]. Considering the previous data on the function of TLR2 and its 2258 G>A polymorphism, in the related signaling pathways, as well as the outcomes of our study, we suggest that, in Polish fetuses and newborns with congenital HCMV infection, GA heterozygotic status in the analyzed region may cause hypo-responsiveness of the produced TLR2 molecule to infection with HCMV. The altered function of TLR2 may result from impaired heterodimerization of TLR2 with TLR1 molecule, the affected TLR2/Myd88 signaling, the changed TLR2-dependent NF-κB signaling and, in result, the inflammatory cytokine release. It seems possible that in GA heterozygotes, the presence of only one correct G allele in
TLR2 2258 G>A locus might be insufficient for complete immune response against HCMV. Considering all the recently obtained results for
TLR2,
TLR4 and
TLR9 SNPs, the
TLR2 2258 G>A polymorphism seems to be an important genetic factor, correlated with an increased risk of HCMV congenital infection among fetuses and neonates. However, further studies would be justified to investigate in more detail the molecular mechanism which underlies
TLR2 polymorphism involvement in the development of congenital HCMV infection.