Polymorphism of genes associated with infectious lung diseases in Northern Asian populations and in patients with community-acquired pneumonia

The innate immune system is the first to respond to invading pathogens. It is responsible for invader recognition, immune-cell recruitment, adaptive-immunity activation, and regulation of inflammation intensity. Previously, two single-nucleotide polymorphisms of innate-immunity genes – rs5743708 (Arg753Gln) of the TLR2 gene and rs8177374 (Ser180Leu) of the TIRAP gene – have been shown to be associated with both pneumonia and tuberculosis in humans, but the data are contradictory among different ethnic groups. It has also been reported that rs10902158 at the PKP3-SIGGIR-TMEM16J genetic locus belongs to a haplotype race-specifically associated with tuberculosis. Meanwhile, a gradient of its frequency is observed in Asia. The aim of this work was to assess the effect of selection for the genotypes of the above-mentioned SNPs on the gene pools of populations living in harsh climatic conditions that contribute to the development of infectious lung diseases. We estimated the prevalence of these variants in white and Asian (Chukchis and Yakuts) population samples from Northern Asia and among patients with community-acquired pneumonia (CAP). Carriage of the rs5743708 A allele was found to predispose to severe CAP (odds ratio 2.77, p = 0.021), whereas the GG/CT genotype of rs5743708/rs8177374 proved to be protective against it (odds ratio 0.478, p = 0.022) in white patients. No association of rs10902158 with CAP (total or severe) was found among whites. Stratification of CAP by causative pathogen may help eliminate the current discrepancies between different studies. No significant difference in rs5743708 or rs8177374 was found between adolescent and long-lived white samples. Carriage of the alleles studied is probably not associated with predisposition to longevity among whites in Siberia. Both white and Asian populations studied were different from Western European and East Asian populations in the variants’ prevalence. The frequency of the rs8177374 T (Ser180Leu) variant was significantly higher in the Chukchi sample (p = 0, χ2 = 63.22) relative to the East Asian populations. This result may confirm the hypothesis about the selection of this allele in the course of human migration into areas with unfavorable climatic conditions.


Introduction
Innate immunity constitutes the first barrier against microor ganisms and viruses by destroying infected cells and activating adaptive immunity. Nonetheless, an excessive nonspecific immune reaction (inflammation) may be life threatening be cause it can completely disrupt the functioning of vital organs. Communityacquired pneumonia (CAP) and pulmonary tu berculosis (PTB) are infectious diseases characterized by high mortality, and according to WHO, are ranked consistently among the top 10 leading causes of death in the world (https:// www.who.int/en/newsroom/factsheets/detail/thetop10 causesofdeath).
Pneumonia is an inflammatory lowerrespiratorytract di sease caused by viruses, bacteria, fungi, and parasites. In addition, it may be due to noninfectious processes or have a combined cause. For a long time, Streptococcus pneumoniae infection has been considered the main cause of CAP; how ever, it was shown recently that CAP develops mainly as a result of viral infections (influenza A and B viruses, para influenza viruses, adenovirus, respiratory syncytial virus, or coronaviruses) (Choi et al., 2012;Hong et al., 2014;Self et al., 2017). Streptococcus pneumoniae, Haemophilus inf luenzae, Staphylococcus aureus, Mycoplasma pneumoniae, Chlamydophila pneumoniae, Legionella pneumophila and other microbes may be causative agents of bacterial pneumonia (Choi et al., 2012;Hong et al., 2014;Self et al., 2017). After invasion of the airway epithelium by pathogens, these cells start to produce reactive oxygen species, cytokines, and other mediators to recruit immune cells. Being most abundant in lungs, alveolar macrophages ingest bacteria and apoptotic cells and can present antigens on MHC II to other immune cells. Proinflammatory M1 macrophages produce cytokines TNFα, IL6, IL1β, IL12, and IL23 to enhance inflammation for elimination of the invaders. Antiinflammatory M2 macro phages produce cytokines IL4, IL13, and TGFβ to induce completion of the inflammatory reaction and remodeling of damaged tissue (Moldoveanu et al., 2009;Arango Duque, Descoteaux, 2014;Kumar, 2019).
Depending on the set of present chemokines and cytokines, different cells responsible for humoral and cellular immunity are attracted to the site of infection (Kumar, 2019). Severe pneumonias are more likely to develop in coinfections; it has been demonstrated that a viral infection (in particular influenza) facilitates the development of pneumococcal infec tion by damaging the epithelium and reducing the amount of a surfactant (McCullers, 2014;Aguilera, Lenz, 2020). Human respiratory syncytial virus, metapneumovirus, adenovirus, and influenza viruses A and B prefer a cold season (Price et al., 2019). The seasonal increase in the incidence of pneumococcal pneumonia coincides with seasonal outbreaks of influenza; S. pneumoniae, H. inf luenzae, and S. aureus infections have been reported to be associated with significant influenza pandemics (McCullers, 2014;Bystritskaya, Bilichenko, 2017;Morris et al., 2017).
PTB is a pulmonary infectious disease caused mainly by Mycobacterium tuberculosis (Mtb). According to the WHO, approximately onequarter of the world's population is es timated to be infected by Mtb, and 5-15 % of these people will fall ill with active tuberculosis. In Russia, most of these patients (95 %) have PTB (https://minzdrav.gov.ru/ministry/ 61/22/stranitsa979/statisticheskieiinformatsionnyemate rialy/statisticheskiysbornik2018god). The pathogenesis of pulmonary tuberculosis is based on Mtb survival after phagocytosis by alveolar macrophages. These bacteria can modulate a host immune response to protect the infected cells, change their metabolism, induce IL10, suppress IL12 and TNFα synthesis, and to inhibit MHC II expression and antigen presentation. Mtb makes macrophages unresponsive to interferon (IFN) γ and inhibits autophagy. It allows the mycobacteria to establish a persistent or latent infection in macrophages. Mycobacteria are believed to use the general mechanism of negative feedback regulation that restricts excessive inflammation (Harding, Boom, 2010;Richardson et al., 2015;Gopalakrishnan, Salgame, 2016). With the loss of immunitydriven control over mycobacterial reproduc tion, foamy macrophages accumulate in granulomas, and lung tissue necrosis begins (Liu C.H. et al., 2017). Vitamin D deficiency is known to negatively affect the effectiveness of the immune response in tuberculosis (Wilkinson et al., 2000;Aibana et al., 2019).
These data suggest that in Northern Asia, a region with low temperature and reduced insolation during most of the year, signs of purifying selection for genes associated with lung infections may be noticeable. For many genes of innate immu nity, an association with viral, bacterial, and autoimmune dis eases has been proven. Despite differences in the pathogenesis between CAP and PTB, it has been shown that minor alleles of rs5743708 (the TLR2 gene) and of rs8177374 (the TIRAP gene) can have a pathogenic and protective effect, respec tively, in both of these lung diseases in humans. Nevertheless, data obtained by different research groups are contradictory.
Tolllike receptors (TLRs) play a pivotal role in host de fense. Being membraneanchored (TLRs 1, 2, 4-6, and 10) or endosomal (TLRs 3 and 7−9) in human immune cells (e. g., macrophages, monocytes, dendritic cells, and some leuko cytes), they are involved in the recognition of structurally conserved surface molecules of microorganisms and viruses as well as viral nucleic acids (Barbalat et al., 2009;Kawai, Akira, 2010;Kumar, 2019). TLR2 participates in the recognition of a large number of diverse lipoproteins and peptidoglycans of grampositive and gramnegative bacteria, fungi, and virusinfected cells. After ligand binding to the receptor, TIR (Tollinterleukin 1 receptor) domains of TLR2 and TLR1 or TLR2 and TLR6 dimerize via the formation of an extensive hydrogenbonding network and hydrophobic interactions (Jin et al., 2007;Takeda, Akira, 2015). Homodimerization of the cytoplasmic domains of TLR2 does not induce TNFα production in vitro in murine macrophages, and the forma tion of the TLR2-TLR2 dimer is not detectable even in the presence of an agonist (Ozinsky et al., 2000;Shukla et al., 2018). Therefore, the existence of TLR2-TLR2 homodimers in vivo is being questioned. After ligand binding, reorientation of the TIR domains and triggering of a cascade of intracellular reactions lead to the activation of proinflammatory NFκB and MAPK pathways, synthesis and a release of proinflammatory cytokines (IL1β, IL12, TNFα, and IL6) and various che mokines into extracellular space, and the development of an inflammatory response at the pathogen entry site (Liu C.H. et al., 2017;Tapader et al., 2018). In inflammatory monocytes, TLR2 induces type I IFN production in response to a viral ligand (Barbalat et al., 2009). It is reported that prolonged stimulation of TLR2 (more than 24 h) causes PI3K/Akt pathway activation in alveolar macrophages. It limits the production of NFκB, TNFα, and IL12 and activates the synthesis of antiinflammatory IL10. This mechanism is as sumed to prevent excessive inflammation (Richardson et al., 2015;Liu Y. et al., 2016).
The TLR2 gene is located in 4q31.3, has five exons, and expresses few splicing isoforms, but all of the coding se quences are contained within exon 3. The protein consists of 784 amino acid residues (aa) and includes extracellular leucinerich repeat domains, which are primarily responsible for ligand recognition (aa 54-524), followed by the leucine rich repeat Cterminal domain (aa 525-579) and intracellular TIR domain (aa 639-782), which mediates downstream signal ing (https://www.uniprot.org/uniprot/O60603). It is expressed constitutively on macrophages and dendritic cells and can be induced in epithelial cells or Bcells. Its overexpression in patients with pneumococcal disease had been documented (Siebert et al., 2018).
Singlenucleotide polymorphism (SNP) rs5743708 (of the TLR2 gene) causing the Arg753Gln substitution is located in the TIR domain of the protein. This SNP is associated simultaneously with resistance to Lyme disease (Schröder et al., 2005) and with predisposition to tuberculosis (Guo, Xia, 2015;Patarčić et al., 2015), whereas the association with predisposition to PTB is racespecific (Caws et al., 2008;Guo, Xia, 2015;Hu et al., 2019). It is believed that TLR2 signaling may be nonessential to control acute tuberculosis but impor tant during chronic tuberculosis (Gopalakrishnan, Salgame, 2016). A metaanalysis has shown the TLR2 rs5743708 minor allele to be associated with CAP, Legionnaires' disease, and pneumococcal disease; however, the data obtained in different studies are contradictory (Moens et al., 2007;Patarčić et al., 2015;Smelaya et al., 2016).
The TIRAP (TIR domaincontaining adaptor protein) gene also known as Mal (MyD88 adapterlike) encodes one of the five adapter proteins that are involved in signal transduction from activated TLRs to protein kinases at the plasma mem brane (Bonham et al., 2014). It is located in 11q24.2, consists of six exons, and encodes a protein of 221 aa. The TIRAP protein includes an Nterminal PEST domain (aa 15-35) re sponsible for binding to special sites in the plasma membrane, followed by an ABloop mediating MyD88 and TLR4 binding. A binding site for TRAF6 (TNF receptorassociated factor 6) is located within the region aa 188-193 (Bernard, O'Neill, 2013). TIRAP is expressed in many cell types (Narayanan, Park, 2015), and its isoforms resulting from alternative splic ing have unknown functions. There are different opinions about whether TIRAP forms a complex with the TIR domain of TLR6 for signal transmission; however, it has been proven that TIRAP mediates TLR2 and TLR4 signaling by facili tating the recruitment of the MyD88 adaptor protein to the TLRs (Nagpal et al., 2009;Bernard, O'Neill, 2013).
Activation of NFκB, MAPK1, MAPK3, and JNK results in cytokine secretion and an inflammatory response. SNP rs8177374 (the TIRAP gene) is located in exon 5 and repre sents the Ser180Leu substitution in the encoded protein. It is located close to the TLRbinding site of TIRAP. In carriers of this substitution, modulation of TLR1, TLR2, TLR4, and TLR6 but not TLR9 signaling has been shown (Khor et al., 2007;Ferwerda et al., 2009;Siebert et al., 2018). Ser180Leu in a heterozygous state has a protective effect against PTB and invasive pneumococcal disease in white and African samples and against malaria in African and Asian samples (Khor et al., 2007;Panda et al., 2016). Carriage of heterozygous Ser180Leu protects children from pneumococcal lowerrespiratorytract infections, whereas carriers of the homozygous 180Leu poly morphism alone or in combination with some TLR1 and TLR6 polymorphisms may be susceptible to recurrent pneumococ cal infections (Siebert et al., 2018). Simultaneous carriage of the TIRAP 180Leu variant and some SNPs in the TLR4 gene as well as 180Leu homozygosity increases susceptibility to severe hospitalacquired infections (Kumpf et al., 2010).
The opposite results have been obtained as well. The rs8177374 T allele (180Leu) increases the risk of PTB in a sample of Iranian population (Naderi et al., 2014). A meta analysis of nine published casecontrol studies did not reveal a significant association of 180L with tuberculosis risk (Miao et al., 2011). There are controversial opinions about the mecha nism behind the observed protective effect of Ser180Leu heterozygosity. They are based on differences in observed effects of the SNP at the level of proinflammatory cytokines. Depending on the model used, some research groups showed an increased level (Ferwerda et al., 2009;Panda et al., 2016) and others a decreased level (Khor et al., 2007;Kumpf et al., 2010;Siebert et al., 2018) of cytokines after their induction in 180 Leu/Leu carriers. Accordingly, homozygosity of the minor variant of rs8177374 is thought to cause either an excessive inflammatory reaction or the absence of an adequate immune response. It is supposed that selection pressure on the TIRAP gene provides a balance between protection against excessive inflammation and effective defense during infectious diseases (Khor et al., 2007;Ferwerda et al., 2009).
Besides the polymorphisms in genes TLR2 and TIRAP, in this paper, we focused on the PKP3-SIGGIR-TMEM16J gene region. An association of its haplotypes with diffe rent types of tuberculosis has been shown among children in Vietnam and South Africa (Horne et al., 2012;Gupta et al., 2016). It is believed that the impact of the haplotypes on immunity is determined by SIGIRR (single immunoglobulin interleukin 1 receptor; synonym: IL1R8), which is a negative regulator of TLR signaling (Molgora et al., 2016). Carriage of rs10902158 GG and rs7111432 AA in introns of PKP3 and TMEM16J, respectively, acts additively with a vitamin D deficiency and "pathogenic" genotypes of rs5743708 (TLR2) and rs8177374 (TIRAP) on tuberculosis predisposition (Horne et al., 2012;Gupta et al., 2016). rs10902158 located in intron 2 of the PKP3 gene has been analyzed. Encoded desmosomal plaque protein plakophilin 3 is involved in intracellular ad hesion (Gurjar et al., 2018). Of note, rs10902158 has a fre quency gradient in Asia; according to the Genome Aggrega tion Database (GnomAD) (https://gnomad.broadinstitute. org/), it is absent in South Asia and is found with a frequency of ~50 % in Southeast Asia. Nonetheless, functional significance of genetic variants in noncoding parts of the PKP3-SIGGIR-TMEM16J gene region, including rs10902158, is not clear.
In this work, we analyzed the frequencies of rs5743708, rs8177374, and rs10902158 (for which conflicting data on the association with respiratory infections have been reported previously) in white and Asian samples from Northern Asia and among CAP patients. According to the statistics of the Ministry of Health of the Russian Federation, Novosibirsk Oblast and Yakutia are characterized by an increased inci dence of PTB, whereas Chukotka Autonomous Okrug is the leader in both pneumonia and PTB morbidity in Rus sia (https://minzdrav.gov.ru/ministry/61/22/stranitsa979/ statisticheskieiinformatsionnyematerialy/statisticheskiy sbornik2017god; https://minzdrav.gov.ru/ministry/61/22/ stranitsa979/statisticheskieiinformatsionnyematerialy/ statisticheskiysbornik2018god). According to the WHO, the highest death rate from pneumonia is observed before the age of 5 and after 75-80 years (https://www.who.int/medicines/ areas/priority_medicines/Ch6_22Pneumo.pdf). Therefore, we assumed that longlived people of the Siberian Federal District may differ from adolescents in the frequency of rs5743708 and rs8177374, and we assessed the prevalence of the pathogenic variants in the sample of longlived people.

Materials and methods
The study protocol was approved by the local Ethics Commit tee of the Institute of Internal and Preventive Medicine (branch of the Institute of Cytology and Genetics of the Siberian Branch of the Russian Academy of Sciences, Novosibirsk, Russia;approval No. 22.06.2008). Written informed consent to be examined and to participate in the study was obtained from each patient. For individuals younger than 18 years, the informed consent was signed by a parent or legal guardian.
Genomic DNA was extracted from peripheral blood leuco cytes by the standard phenol-chloroform method (Sambrook, Russell, 2006). Genotyping was performed using polymerase chain reaction (PCR) with an analysis of restriction fragment length polymorphism by electrophoresis in a 5 % polyacryl amide gel after visualization with an ethidium bromide so lution.
The rs5743708 SNP (TLR2) was identified by amplifica tion of DNA with primers 5′GCCATTCTCATTCTTCTGG* AGC3′ and 5′GGGAACCTAGGACTTTATCGCA3′ (* de notes a nucleotide changed for restriction site creation). The 168bp PCR product was digested with the Pst I restriction endonuclease (SibEnzyme, Novosibirsk) for 2 h at 37 °C. The rs5743708 A (753Q) allele was revealed by the presence of fragments of 20, 45, and 103 bp, whereas the rs5743708 G allele by fragments of 20 and 148 bp.
Primers 5′TGGCAAGGATTGGAGAACTC*C*TGTC3′ and 5′CAGGGCCAGTGCCTCCCC3′ (* denotes nucle o tides changed for restriction site creation) were used for the amplification of the PKP3 intron 2 sequence containing rs10902158. The resulting 192bp amplicon was digested with the BstEN I restriction endonuclease (Sibenzyme, Novosi birsk) for 2 h at 65 °C. In the presence of the rs10902158 А allele, the PCR product was not cut, whereas in the case of the rs10902158 G allele, fragments of 24 and 168 bp were observed.
Statistical analysis was performed in the SPSS 16.0 soft ware.

Results
Genotype distributions were consistent with the Hardy-Wein berg equilibrium among all the population samples (data not shown). Minor allele frequencies for rs5743708, rs8177374, and rs10902158 are represented in Table 1.
For the frequency of the rs5743708 A (Arg753Gln) al lele, there was a tendency for a decrease in Russian settlers in Yakutia and among longlived people compared with the Novosibirsk sample.
The rs8177374 T (Ser180Leu) frequency did not differ among the studied white samples.
In the sample of Russian settlers of Yakutia, males and females differed in the frequency of rs10902158 ( p = 0.037, χ 2 = 4.86). Moreover, the frequency of this SNP among males was closer to that observed in nonFinnish Europeans accord ing to GnomAD data, and the frequency among females was closer to that observed in the Novosibirsk sample. Perhaps there were sex differences during recent migration to Yakutia from different regions of Russia. Genetic analysis of a larger sample and estimation of this SNP's frequency in western regions of Russia are required for explaining the observed differences.
The two analyzed Asian samples differed from each other and from GnomAD East Asian cohorts. Among Yakuts, the rs5743708 A (Arg753Gln) variant, which is very rare in other Asian populations, was found at a frequency of 0.015 ± 0.007 (mean ± SD). Chukchis differed significantly from GnomAD East Asians in rs10902158 allele frequency ( p = 0, χ 2 = 63.22) (see Table 1).
Frequencies of polymorphisms rs5743708, rs8177374, and rs10902158 were not different among adolescents and total white CAP patient samples. By contrast, after the sample was divided into patients with severe and nonsevere CAP, differences were found for rs5743708 ( p = 0.021, χ 2 = 6.24). Next, genotype frequencies were estimated for rs5743708, rs8177374, and rs10902158 in the examined samples (except for longlived people regarding rs10902158). For the latter SNP (in the PKP3 gene), no difference in frequency was de tectable within any group (data not shown). The observed distribution of rs5743708 and rs8177374 genotypes among the studied samples is presented in Table 2. The frequencies of genotypes of rs5743708 and rs8177374 did not differ among the following samples: adolescents of Novosibirsk, longlived people of Siberia, all patients with CAP, and all Russians in Yakutia. Possibly, the carriage of the studied alleles is not associated with predisposition to longe vity in Siberia and does not significantly affect the probability of resettling of migrants in more unfavorable climatic condi tions at present. This notion is consistent with WHO findings that in Eastern Europe, in contrast to Western and Central Europe, the pneumonia mortality rate does not increase sig nificantly after age 80 (https://www.who.int/medicines/areas/ priority_medicines/Ch6_22Pneumo.pdf). As for pneumonia, carriage of the rs5743708 A allele predisposed to severe CAP (AG/CC + AG/CT vs all: odds ratio 2.77, 95 % confidence interval 1.227-6.272, p = 0.021). The heterozygous geno type of rs8177374 in combination with the GG genotype of rs5743708 had a protective effect against severe CAP (GG/CT vs all: odds ratio 0.478, 95 % confidence interval 0.251-0.909, p = 0.022).

Discussion
It was shown here that carriage of none of the three studied SNPs, rs5743708, rs8177374, and rs10902158, is associated with the predisposition to CAP in total. By contrast, we found that the Arg753Gln variant of TLR2 predisposes to severe CAP, and the heterozygous Ser180Leu variant of TIRAP in combination with the 753 Arg/Arg variant of TLR2 has a protective effect against it in the white population. These data partially explain the contradictions in the data from different researchers. Most likely, the contribution of the alleles of genes TLR2 and TIRAP to CAP predisposition is determined by pneumonia etiology. A substantial proportion of severe pneumonia cases are known to be caused by combined viral and bacterial infections (McCullers, 2014;Morris et al., 2017;Aguilera, Lenz, 2020). TLR2 is responsible mainly for the recognition of bacteriaassociated molecular patterns; Ser180Leu of the TIRAP gene modulates signal transduction only from TLR2 and TLR4 recognizing molecular patterns of bacteria as well (Nagpal et al., 2009). Most likely, combined and bacterial but not viral pneumonias are associated with TLR2 and TIRAP gene variants.
The studied Asian ethnogeographical groups showed an in creased frequency of the protective rs8177374 T ( Ser180Leu) variant of TIRAP relative to neighboring East Asian popula tions. In the Chukchi sample, the difference was significant ( p = 0, χ 2 = 63.22). It may be a consequence of the natural selection that has promoted protection from excessive inflam mation during pulmonary diseases. The hypothesis about the selection of the Ser180Leu variant along with the outof Africa migration to a harsh environment has been advanced earlier (Khor et al., 2007;Ferwerda et al., 2009). An increased frequency of the Arg753Gln variant of TLR2 and a decreased frequency of Ser180Leu of TIRAP as compared to nonFin nish Europeans may indicate higher genetic predisposition of the Siberian white population to PTB and severe CAP. Never theless, there are a lot of genes associated with CAP and PTB independently. Apparently, during the settlement of peoples in Northern Eurasia, the formation of gene pools had been determined by the selection that facilitated adaptation to specific infections (Lime disease among them), parasites, and the climate. It would be interesting to determine why two mutations changing the same TLR2 signaling have opposite effects on the predisposition to severe CAP. One possible ex planation is the difference in the structure and functions of heterodimers TLR2-TLR1 and TLR2-TLR6 (see the Figure).
Existing data on the roles of TLRs 1, 2, and 6 in the activa tion of proinflammatory and antiinflammatory signaling are conflicting. Overexpression of TLR2 carrying the Arg753Gln variant has been demonstrated to cause a significantly stronger impairment of cytokine induction by TLR2/TLR1 ligands as compared with TLR2/TLR6 ligands in the HEK293 cell line (Schröder et al., 2005). In later papers, it has been shown that the ArgtoGln substitution at position 753 of TLR2 changes the size, charge, and hydrophobic properties of this site and reduces the ability of TLR2 to form a heterodimer with TLR6 (Basith et al., 2011;Xiong et al., 2012). This SNP significantly alters agonistinducible association of TLR2 with adaptor proteins TIRAP and MyD88 and impairs NFκB signaling and IL8 mRNA expression in the HEK293 cell line (Xiong et al., 2012). Genes TLR1 and TLR2 have different expression activators (Lancioni et al., 2011). In the TLR2-TLR1 dimer, TLR1 and TLR2 are responsible for NFκB and MAPK pathways and for PI3K pathway activation, respectively; therefore, up regulation of proinflammatory cytokines is TLR1dependent, whereas upregulation of type I IFN is TLR2dependent (Raieli et al., 2019). TLR2-TLR6 binding disruption possibly causes an increase in the number of TLR2-TLR1 dimers in which TLR1 drives the activation of the NFκB inflammatory signal ing cascade. On the contrary, the Ser180Leu variant of TIRAP weakens proinflammatory signal transmission. Besides, TIRAP acts as an adaptor protein for TLR4 homodimers. This receptor is primarily responsible for the recognition of lipopolysaccharides of gramnegative bacteria and fungi (Takeda, Akira, 2015). It is believed that TLR4 takes part not only in MyD88dependent proinf lammatory signaling but also in MyD88independent antiinf lammatory signaling. Weaken ing of TLR4 signaling through TIRAP probably enhances the signaling through adapter proteins TRIF and TRAM, causing the secretion of antiinf lammatory cytokines (Li et al., 2013). It remains unclear why the SNPs in TLR2 and TIRAP have similar effects on the predisposition to or protection against acute (CAP) and chronic (PTB) lung infections. Perhaps this phenomenon is due to an impact on inflammation in both di seases. The severity of CAP is determined by lifethreatening acute inflammation; in PTB, the development of chronic in flammation masks the infection from the host immune system (Liu C.H. et al., 2017).

Conclusions
In Northern Asian populations, the observed difference in rs8177374 frequency may reflect consequences of natural selection during the settlement of peoples on territories with unfavorable climatic conditions. As for pneumonia, carriage of the rs5743708 A allele (the TLR2 gene) predisposes to severe CAP; the heterozygous genotype of rs8177374 (the TIRAP gene) in combination with the GG genotype of rs5743708 (the TLR2 gene) has a protective effect against it. Stratification of CAP by causative pathogen may help to eliminate the current discrepancies among research groups. Regional differences in a set of pathogens along with the genetic characteristics of ethnogeographical groups can determine the associations of various genetic variants of innate immunity with the preva lence and severity of pneumonia. Ser180Leu (rs8177374) Locations of the Arg753Gln substitution in TLR2 and Ser180Leu in TIRAP on the protein complexes.