Polymorphisms and expression of TLR4 and 9 in malaria in two ethnic groups of Assam,northeast India

Abstract

Infectious diseases have been postulated to play an important role in exerting pressure and in selection of TLR polymorphisms. Single nucelotide polymorphisms (SNPs) of TLR4 have been reported to show unique distributions in populations from Africa, Asia and Europe, and malaria is suggested to influence these patterns. In this context, we examined association of TLR polymorphisms with the risk of malaria in two ethnic groups—the Austro-Asiatics and Tibeto-Burmans—from malaria endemic districts of Assam to understand the influence of malaria in selection of TLRs in these genetically-distinct populations. TLR9 (T-1237C) mutation was positively associated with complicated (P = 0.001) and frequent (P = 0.035) malaria in Austro-Asiatics (relative risk = 0.595 95% CI: 0.479–0.836), but not in Tibeto-Burmans. Nonetheless, these alleles were not in Hardy-Weinberg Equilibrium in Tibeto-Burmans (P < 0.001). In contrast, the TLR9 1486T/C genotype was favourable where it was negatively associated with complicated malaria (Fishers exact P = 0.014). Sequencing data revealed that the two populations differed in nucleotide diversity of the TLR9 promoter region. Enhanced expression of TLR4 (P = 0.05), but not of TLR9, was associated with complicated malaria. Austro-Asiatics appeared to have accumulated favourable genotypes of TLR9, perhaps because of their longer exposure to malaria.

Keywords

Austro-Asiatic malaria Tibeto-Burman Toll-like receptors

Introduction

Innate immunity is gaining importance in the understanding of immune responses to infectious organisms. Cells of the immune system armed with germ-line encoded receptors (pathogen recognition receptors) of limited repertoires respond to the molecular motifs on exogenous antigens by secreting cytokines and chemokines that significantly influence adaptive immunity and disease outcome.¹

TLRs are an important component of the innate immune system. Their role in bacterial and viral infections is well documented and they are becoming important in immune responses to protozoan infections.^2–4 Activation of TLRs leads not only to the robust production of pro-inflammatory mediators, but also to the production of unique effectors, which provide pathogen-tailored immune responses. In malaria, Plasmodium falciparum glycosylphosphatidylinositol (GPI) has been shown to induce the expression of proinflammatory cytokines and immune mediators in vitro.⁵ Furthermore, GPI was reported to induce signalling via both TLR2 and TLR4, whereas haemozoin (HZ), in combination with plasmodial DNA, activated dendritic cells by engaging TLR9.⁶ Changes in TLR responses also modulated the production of inflammatory cytokines and thus prevented the development of severe pathology, emphasising the role of these responses in malarial pathogenesis.⁷ Increased expression of TLR genes was also seen in P. falciparum infected subjects and it was associated with enhanced IFN-γ, TNF-α and IL-10 production.⁸

Genetic variations of TLRs can change the ability of the organism to respond to different stimuli, hence the interest in the possible associations of TLR polymorphisms with altered susceptibility to infections and inflammatory diseases is growing.² Studies have linked TLR2 (Arg753Gln) with rheumatic fever in children and TLR4 polymorphisms with increased susceptibility to bacterial infections.^9–11 TLR9 promoter polymorphisms (T-1237C and T-1486C) have been assumed to influence transcription regulation and they have been associated with asthma, Crohns disease, atopic eczema and lymphoma.^12–14 Many studies have also implicated polymorphisms of TLR2, 4 and 9 in the disease manifestation of malaria. Single nucleotide polymorphisms (SNPs) of TLR4 (Asp299Gly) and TLR9 (T-1486C), but not TLR9 (T-1237C), were associated with low birth rate, maternal anaemia and with high parasitaemia, suggesting that these SNPs play a role in the manifestation of malaria during pregnancy.^15,16 SNPs of TLR4 Asp299Gly and Thr399Ile were seen to increase the risk of severe malaria in Ghanian children.^15,17

A number of SNPs associated with inflammatory and immunologic diseases also show large frequency differences among ethnically-distinct populations.^2,18 A study in India, showed that high frequency of FcγR2A exon4 AA genotype protected the Tharus, a population living in the foothills of the Himalayas, from falciparum malaria.¹⁹

Distribution of TLR polymorphisms have been shown to differ among ethnically different populations in Asia, Africa and Europe, and malaria was suggested to be the driving force.²⁰ In this context, we have examined the association of SNPs in TLR4 and 9, and the expression of these genes in relation to malaria and ethnicity in Assam State of northeast India, in which malaria is endemic.²¹ We report the differential association of polymorphisms and expression of TLR9 with malaria in two ethnic groups. Increased expression of TLR4 gene was seen in complicated malaria. The Austro-Asiatics (plains) had the favourable genotype, conferring protection from complicated malaria.

Materials and methods

Study sites

The study was conducted at two study sites: Guabari, a village of Baksa district which lies at the foothills of Bhutan; and at Kondoli in Karbi Anglong foothills of Nagaon district of Assam. The two study sites may be classified as mesoendemic for malaria. The characteristics of the two study sites have been described previously.²²

Study participants

Patients were enrolled into the study after obtaining written informed consent. Active, as well as passive, case detection was followed which was carried out by local health workers in Guabari, while at Kondoli it was done by hospital staff. Based on linguistic group affinities, the study population was stratified into two groups, namely the speakers of Tibeto-Burman (TB) that included the Bodo-Kachari and Nepalis, the Austro-Asiatic (AA) that comprised the tea tribes who are Mundari speakers.¹⁹ Sampling was done such that the samples were drawn independently and randomly from each group, particularly at Guabari, which has an ethnically-mixed population. Though the people belonging to AA and TB groups are found in Assam at present, but their origin is very different. The AA group is believed to include the earliest inhabitants of Indian plains whereas the TBs are believed to have migrated from South China and Tibet approximately 5000–10,000 years ago.^23–25 Moreover, based on mitochondrial DNA and NRY (non-recombing region of the Y chromosome) haplotypes, the AA was found to be genetically distinct from TB.²⁵ Khasis are a group of people residing in the state of Meghalaya and they are AA because their language belongs to the Khasi-Khmic subfamily of the AAs.^26,27 In the present study, the AA group from Assam was denoted as AA (plains) and the Khasis as AA (hills). Khasi participants (n = 30) were from hilly regions (>4900 ft above sea-level) where exposure to malaria is unlikely and so served as controls for AA (plains) in SNP-malaria association studies.

Sample collection

Blood samples were obtained from individuals, irrespective of their P. falciparum-positive (Pf+) status at the time of collection. Inclusion criteria were individuals with Pf+ or with history of falciparum malaria. Exclusion criteria were children with age less than 1 year, pregnant women and individuals suffering from any other diseases. Plasmodium falciparum positivity was checked using Rapid Diagnostic Kits (FalciVax; Orchid Biomedical Laboratories, Goa, India), and confirmed by microscopy of thin and thick blood smears. Individuals were categorized according to their disease symptoms and frequency of malaria incidence, which were recorded simultaneously. Complicated malaria was defined following World Health Organization (WHO) guidelines.²⁸ Uncomplicated malaria included fever, headache, body ache, and other mild symptoms or asymptomatic cases with parasitaemia of ≤5000/µl of blood. On the basis of number of clinical episodes of malaria, the participants were classified as infrequent (≤2 episodes) or frequent malaria (>2 episodes) groups as detailed in our earlier work.²⁹ The study was approved by the Tezpur University Ethical Committee (Resolution number 3 dated 13 June 2006).

SNP typing of TLR4 and 9

Two hundred microlitres of blood were collected in EDTA-tubes and genomic DNA extraction was carried out using Qiagen blood extraction kit (Qiagen, Hilden, Germany). The TLR4 and 9 genes were amplified by PCR and then digested using restriction enzymes, as described.^10,30–32 The fragments were then analysed on 3% agarose gels pre-stained with ethidium bromide.

Sequence analysis

TLR9 promoter regions were sequenced (Sequencer; Applied Biosystems, Foster City, CA) and aligned with the sequence available in the GenBank (Accession number: NT_022517.18) using CLUSTAL X and the matrix selected for the purpose is Clustal W 1.6. The alignment file created was used for further analysis using DNASp V 5.0. TLR9 promoter sequences have been submitted to GenBank with the following accession numbers: HM231310, HM231311, HM231312 and HM231313.

TLR gene expression study

The cohort used for the TLR4 and 9 genotyping study was also used for expression study. Forty-six samples were drawn randomly from each of the three diseased groups (comprising of patients having complicated malaria, uncomplicated malaria and non-malarial fever) and from the healthy control group. Total RNA was then extracted from these blood samples and stored in RNA later, as per the manufacturer’s instructions (Ribopure; Ambion, Invitrogen BioServices, India). Reverse transcription (RT) reactions were set up by using cDNA Archive kit (Applied Biosystems, Foster City, CA). Real time PCR was performed in 7300 Real time PCR system (Applied Biosystems, Foster City, CA). TLR genes expression was calculated in relation to the expression of RnaseP, which was used as an endogenous control.³³

Data analysis

The data was analysed using Excel Stat Software, 2010 version. Variation in the frequency of the TLR4 and 9 genes with respect to disease status, frequency of disease and ethnicity was analysed by Fisher’s exact test. The allele frequencies were calculated (estimation of P values by Pearson test) to test if they were in agreement with Hardy–Weinberg equilibrium (HWE). For TLR gene expression data analysis we used the comparative CT method (ΔΔCT method). Expression of TLR genes was analysed by t-test used for comparing the means in the different groups. Expression levels of TLR genes were compared between the younger age group (≤10 years) and older age group (>10 years) who had malaria at that point in time.

Results

Overall frequency of SNPs of TLR genes

A total of 191 samples were studied, where we found median age of the participants was 26.174 (range: 2–58 years). The younger age group (2–10 years) was seen to have a higher mean parasitaemia of 2526.09 ± 1764.92 trophozoites/μl of blood than the older age group (11–58 years) which was 1816.833 ± 700.9 trophozoites/μl of blood. These samples were examined for the frequency of SNPs in TLR4 (Asp299Gly), TLR9 (T-1237C) and (T-1486C) (Table 1). The mutant alleles of TLR9 were seen predominantly in heterozygosity in the three ethnic groups. In case of TLR4, the wild genotype (A/A) was seen to dominate. The two alleles of TLR4 were not in HWE, neither in the overall population nor in each of the two ethnic groups (P < 0.001) (Table 2). The frequency of the homozygous mutant genotype of the studied TLR genes was low except for TLR9 (T-1486C) where it was seen to be present at higher frequency in the TB (Fishers exact P = 0.014; Table 1). The genotype frequencies of TLR4 and 9 were compared between AA (hills) and AA (plains), and were comparable (Table 3), suggesting that malaria may not be a factor for the variation.

Table 1.

Comparison of the genotype frequencies of the TLR4 and 9 loci in the study population. Analysis was performed to check the interactions in the three genotypes of each TLR gene by Fisher’s exact test.

	TLR4 Asp299Gly
Genotype	n	AA	GG	AG	P value
AA(plains)	79	65	4	10	0.89
TB	75	64	3	8	0.89
	TLR9 (T-1237C)
Genotype	n	TT	CC	TC	P value
AA(plains)	81	25	0	56	0.43
TB	74	21	2	51	0.43
	TLR9 (T-1486C)
Genotype	n	TT	CC	TC	P value
AA (plains)	75	32	7	36	^a0.014
TB	80	18	16	46	^a0.014

P value is significant (P < 0.05). number of samples.

AA, Austro-Asiatic; TB, Tibeto-Burman.

Table 2.

Comparison of the allele frequencies of the SNPs of TLR4 and 9 loci in the study population.

SNP	Overall population n = 191		AA n = 111		TB n = 80
SNP	W allele	M allele	W allele	M allele	W allele	M allele
TLR4 Asp299Gly	^a0.895	^a0.105	^b0.89	^b0.11	^d0.9	^d0.1
TLR9 (T-1237C)	^f0.64	^f0.36	^e0.65	^e0.35	^c0.628	^c0.372
TLR9 (T-1486C)	0.587	0.413	0.66	0.34	0.512	0.488

a–f

Allele frequencies were not in Hardy-Weinberg Equilibrium (Pearsons P < 0.05). Allele frequencies of TLR4 suggest that there is a low prevalence of the mutation in the study population.

AA, Austro-Asiatic; TB, Tibeto-Burman; W, wild; M, mutant.

Table 3.

Comparison of the genotype frequencies of the TLR4 and 9 between Austro-Asiatics (AA) (plains) and AA (hills).

	TLR4 Asp299Gly
Genotype	n	AA	GG	AG	P value
AA (plains)	79	65	4	10	0.685
AA (hills)	30	26	0	4	0.685
	TLR9 (T-1237C)
Genotype	n	TT	CC	TC	P value
AA (plains)	81	25	0	56	0.123
AA (hills)	30	5	0	25	0.123
	TLR9 (T-1486C)
Genotype	n	TT	CC	TC	P value
AA (plains)	75	32	7	36	0.20
AA (hills)	30	7	4	18	0.20

Analysis was performed to check the interactions in the three genotypes of each TLR gene by Fisher’s exact test between the Mundari speakers, AA (plains) and the Khasi-Khmic AA (hills) subfamilies of the AAs.

Sequence analysis

Differences in SNP frequencies in the two ethnic groups were further analysed by sequencing the PCR products of TLR9 promoter region. The sequences were compared with the sequence data present in Gen Bank (GenBank accession number: NT_022517.18). Analysis of the two subject populations with the sequence present in Gen Bank showed that the TB population had a maximum variation in the nucleotide sequence in the range of nucleotide from 162–340 over 178 base pair (bp) length (Figure 1A). In contrast, in the AA population, variation was seen in the later part of the sequence over a region of 180 bp, lying in range of nucleotide number 420–600 (Figure 1B).

Figure 1.

Comparison of nucleotide divergence. The figure compares nucleotide divergence (A) between Tibeto-Burman (TB) and GenBank (NT_022517.18) (B) between Austro-Asiatic (AA) and GenBank (NT_022517.18) and (C) between AA and TB. Pi, nucleotide diversity, the average number of nucleotide differences per site between two sequences; Dxy, average number of nucleotide substitution per site between populations; Da, number of net nucleotide substitutions per site between populations. Alignment of the sequences was done using CLUSTAL X and further analysis using DNASp V 5.0.

Analysis between the two populations showed a distinct pattern in the number of nucleotide variation, as shown in Figure 1C, where the TB population showed maximum variation in the first 200 bp region while the AA population was seen to show maximum variations in the 400–600 bp region.

Association of SNPs of TLR4 and 9 genes with malaria

We analysed the frequency of SNPs in relation to disease severity: complicated malaria and uncomplicated malaria and with respect to frequency of episodes.

Complicated and uncomplicated malaria

Frequency of TLR4 (Asp299Gly) was comparable, irrespective of disease status and ethnicity (Table 4). Overall frequency of TLR9 1237T/T genotype was higher in the uncomplicated group (Fisher’s exact P = 0.009), whereas T/C genotype was seen at higher frequency in the complicated group. Stratifying data on the basis of ethnicity, T/T genotype was seen at higher frequency in AA (plains) in the uncomplicated group, suggesting an association of TLR9 (T-1237C) mutation with complicated malaria (relative risk = 0.595 95% CI: 0.479–0.836; Fisher’s exact P = 0.001; Table 4). In TB too, the frequency of T/T genotype was higher in the uncomplicated malaria group but did not reach statistical significance (P > 0.05). Interestingly, this mutation was not in HWE in either AA (plains) or TB (P < 0.001), suggesting this could be under selection in these populations (Table 2). Intriguingly, for TLR9, the 1486T/C genotype was found to be protective in TB where it was negatively associated with complicated malaria (Fisher’s exact P = 0.01; Table 4).

Table 4.

Comparison of the genotype frequencies of the SNPs of TLR4 and 9 in the two ethnicities in complicated and uncomplicated malaria groups. Analysis was performed to check the interactions in the three genotypes of each TLR gene.

		AA					TB
SNP	Genotype	n	Com M	Un Com	P value	n	Com M	Un Com	P value
TLR4	AA		24	40			26	38
Asp299Gly	GG	78	2	2	0.57	75	1	2	0.87
	AG		5	5			4	4
TLR9	TT		4	21			7	14
(T-1237C)	CC	81	0	0	^a0.001	74	1	1	0.80
	TC		30	26			21	30
TLR9	TT		14	18			9	9
(T-1486C)	CC	75	1	6	0.28	77	10	6	^b0.01
	TC		13	23			11	32

a,b

P values are significant (P < 0.05).

AA, Austro-Asiatic; TB, Tibeto-Burman; Com M, complicated malaria; Un Com, uncomplicated malaria.

Frequency of the disease (number of episodes)

The frequencies of SNPs in TLR4 and 9 genes were unrelated to frequency of the disease. When the data was stratified according to ethnicity, we found that malaria occurred less frequently in absence of (T-1237C) mutation of TLR9 in AA (plains) (P = 0.035; Table 5), indicating the mutation was not favourable to malaria outcome.

Table 5.

Comparison of the genotype frequencies of the SNPs of TLR4 and TLR9 in the two ethnicities in frequent and infrequent malaria episodes.

		AA				TB
SNP	Genotype	n	F	IF	P values	n	F	IF	P values
TLR4	AA		30	34			28	36
Asp299Gly	GG	78	2	2	0.3	75	1	2	0.5
	AG		7	3			5	3
TLR9	TT		11	14			10	11
(T-1237C)	CC	81	0	0	^a0.035	74	1	1	0.3
	TC		40	16			16	35
TLR9	TT		19	13			8	10
(T-1486C)	CC	75	3	4	0.6	77	8	8	0.9
	TC		18	18			20	23

Comparison of the frequency of the given SNPs between frequent and infrequent malaria episodes in the two ethnic groups. Analysis was done using χ² test to check whether the frequencies of SNPs in TLR4 and TLR9 genes were related to frequency of the disease. Significant difference was seen only for TLR9 (T-1237C)

Between frequent and infrequent malaria episodes in AA and TB.

F, frequent; IF, infrequent; AA, Austro-Asiatic; TB, Tibeto-Burman.

SNP grouping of TLR9

Based on the combination of the presence or absence of the wild and mutant alleles for the two SNPs of TLR9 (T-1237C and T-1486C) a total of eight groups could be identified (Table 6). Group 1, with co-occurrence of the two mutations in heterozygosity, was seen to be the dominant group and occurred in 67 individuals. Interestingly, 77.77% of the individuals of this group belonged to the uncomplicated group (Kruskal-Wallis test P = 0.05), suggesting its association with protection from having complicated malaria. Group 4, which contained the T/T genotype of both mutations, was more frequent than group 7 containing C/C genotype for both the polymorphisms. In group 4, 9 individuals were of AA (plains) and 76.9% of them suffered from complicated malaria (Table 6).

Table 6.

SNP-grouping of TLR9 (T-1237C and T-1486C) in the study population. Black and white fields denote the presence and absence of wild or mutant alleles respectively.

Group	T1237C	T1237C	T1486C	T1486C	Genotype	Total (Pf+ and control)	In relation to malaria severity
Group	W	M	W	M	Genotype	Total (Pf+ and control)	CM	UCM
1					TCTC	67	10	36
2					TCTT	39	14	14
3					TCCC	24	8	9
4					TTTT	17	1	13
5					TTCC	4	1	2
6					TTTC	36	2	24
7					CCCC	1	1	–
8					CCTC	1	–	1

The participants (n = 189) positive for both the TLR9 (T-1237C and T-1486C) mutations were SNP-grouped. Groups 4, 5, 7 and 8 were absent in the AA (hills). Negligible prevalence of C/C genotype for both the mutations may be noted. The proportions of complicated malaria were higher if TLR9 (T-1237C) mutation was present but if it was present along with the (T-1486C) a larger proportion of uncomplicated malaria was observed.

CM, complicated malaria; UCM, uncomplicated malaria; Pf+, P falciparum positive.

TLR4 and 9 genes expression

Forty-six samples were analysed for the mRNA levels of TLR4 and 9 genes and data analysed in relation to disease status. TLR4 and 9 mRNA levels showed no difference between malaria and no malaria groups (P > 0.05). When the data were analysed in the context of complicated and uncomplicated malaria (Figure 2A), TLR4 gene expression was higher in the complicated group (P = 0.05), whereas in the uncomplicated group it was lower in comparison with the no malaria group.

Figure 2.

Comparison of expression of TLR4 and TLR9 genes with disease severity and age. Mean expression of TLR4 and TLR9 genes was compared between complicated and uncomplicated malaria (A) and between diseased young and diseased old (B). TLR4 expression was higher (P = 0.05) in complicated group (CM) in comparison with uncomplicated (UC) and no malaria (NM) groups. TLR9 expression was higher (P = 0.002) in younger age group with malaria (DY) than in the older malaria group (DO).

As age is an important factor in immunity to malaria, the data was analysed in the context of age. TLR9 expression was higher in the younger age group with clinical malaria (P = 0.002; Figure 2B). Expression of TLR4 gene was comparable between younger and older age groups with malaria (Figure 2B).

Discussion

Vertebrate TLRs play a crucial role in immune protection and have undergone selection in response to immune challenge, as noted in studies that have associated TLRs and their SNPs with diseases.³ Selection favouring homozygosity or heterozygosity is one of the mechanisms that maintain polymorphism and genetic variability. Our data indicated TLR9 1486T/C heterozygosity to be protective, while it was negatively-associated with complicated malaria, whereas TLR9 (T-1237C) was seen to predispose to complicated malaria in heterozygosity. Similar findings have been reported earlier with respect to association of TLR9 (T-1237C) mutation with severe malaria.³⁴ However, our observations in context of (T-1486C) polymorphism are in contrast with earlier reports where it was associated with susceptibility to severe malaria in children and also with the adverse fetal outcome in women with placental malaria.^15,17 Although TLR9 (T-1237C) mutation has been correlated with complicated malaria, in the presence of the (T-1486C) mutation which was associated with protection from complicated malaria, a larger proportion of uncomplicated malaria was seen in our study, suggesting that (T-1486C) was the determining mutation. Advantages conferred by heterozygosity are well documented, as in the case of sickle cell haemoglobinopathy, which affords protection against mortality from falciparum malaria in heterozygous individuals, balancing the severe consequences of the disease in homozygous individuals.³⁵ Similarly, it was also found that those individuals with heterozygous class I and class II HLA alleles progressed from HIV infection to an AIDS-defining illness more slowly and had the ability to clear hepatitis B infection respectively.^36,37 We did not observe an association of SNPs and TLR4, either with malaria severity or frequency of clinical episodes. In contrast, Mockenhaupt et al. reported a higher frequency of TLR4 polymorphisms Asp299Gly and Thr399Ile in malaria cases, which increased the risk of severe malaria.¹⁵ Although, the proportion of the A/A genotype of TLR4 was seen to be higher in both the ethnicities in the present study, it was, however, not associated with either protection or susceptibility to malaria. TLR4 alleles were not in HWE in the overall population, including each of the two ethnic groups. Natural selection acting on human TLR4 ascribed to the presence of excess rare non-synonymous polymorphisms has been reported by Nakajima et al., who postulated that the selection was linked to pathogen pressure.³⁸

Expression of TLR4 genes was enhanced in patients having complicated malaria in the present study, which could possibly be an outcome of binding to GPI anchors of parasite surface proteins leading to increased levels of pro-inflammatory cytokines and also resulting in the up-regulation of cell adhesion molecules (ICAM-1 and VCAM-1), leading to cytoadherence.^5,39 An earlier study also reported increased expression of TLR4 in adults with severe and mild malaria, but decreased TLR9 expression in people with no history of malaria.⁴⁰ Also, Coban et al. found that innate immune responses via TLR2 and TLR9, and adaptor molecule MyD88-dependent pathways are critically involved in the pathogenesis of cerebral malaria.⁴¹ On the contrary, Lepenies et al. demonstrated that induction of cerebral malaria was independent of TLR2, TLR4 and TLR9 gene expression.⁴² TLR9 gene expression was seen to be age-related, with higher expression in younger malaria patients. Interestingly, the proportion of uncomplicated malaria was higher in this group and may be seen as an outcome of increased TLR9 gene expression, which induced an early pro-inflammatory response resulting in containment of the disease. In a parallel study we analysed the gene expression of a panel of cytokines (IL-2, IL-4, IL-8, IL-12α, IL-12β, IL-18, IFN-γ, TNF-α and TGF-β) and found that early pro-inflammatory cytokines were up-regulated in diseased cases in comparison with healthy controls (unpublished data). Increased TLR9 expression could be induced by HZ, a malarial pigment formed after digestion of the haemoglobin by malarial parasites.⁴³ It is well documented that the younger age group, generally, have higher parasitaemia and suffer more severe consequences of malaria.^44,45 Increased infection would result in an increased concentration of HZ in the reticulo-endothelial system, which has been demonstrated to activate the immune system by binding to TLR9 and activating the downstream signaling pathway via MyD88, thereby resulting in increased TLR9 expression as one of the outcomes.⁶ Parroche et al. also reported that HZ activates the immune system similar to Coban et al., although the mechanism of activation is different.^6,43 Notably, higher parasitaemia was seen in the younger age group of our study population. However, no difference in mRNA levels was seen between control and disease, which was surprising as we had observed association between the SNPs in promoter region of TLR9 with disease. This was, however, not a unique finding, as, earlier, Fuse et al. had also found correlation of TLR9 SNPs with Japanese ulcerative colitis, but no correlation between TLR9 mRNA expression and disease.⁴⁶ Modulation of TLR9 gene expression through cis-regulatory variants offers an explanation for the lack of correlation between SNPs in the promoter region and levels of mRNA expression.⁴⁷

Genetic variations in TLRs have been linked to susceptibility to infectious diseases and these variations could be the result of selective pressure on the TLR genes in those populations living in malaria-endemic areas. Ethnicity was seen to be an important factor in the association of frequency of SNPs of TLR9 with malaria pathogenesis in the present study. The promoter region of TLR9 was seen to vary between the AAs and TBs, as confirmed by sequencing. This was not surprising as AAs (plains) in our study group belonged to tea tribes of Mundari language affinity and are genetically distinct from TBs, based on mitochondrial DNA and NRY haplotypes.^23,24 Our data suggested that AA (plains) had the favourable genotype that protected them from complicated malaria as: (i) TLR9 1237T/T was seen to confer protection from complicated malaria and from frequent episodes of malaria; (ii) there was negligible TLR9 1486C/C genotype in AA, which was implicated in complicated malaria in TB. We propose that their favourable genotype could be a reflection of their longer exposure to malaria as compared with the TBs in our study group, as the AAs are believed to be the earliest inhabitants of the Indian plains.²³ We also compared the genotype frequencies of TLR9 between the two different AA groups and found that they did not differ, emphasizing the role of ethnicity. Thus, it may be postulated that the TLR9 gene may have been under selective pressure exerted by malaria. Our observation that TLR9 (T-1237C) alleles were not in HWE in TBs perhaps reflects an ongoing selection in this group, who are believed to have migrated from South China and Tibet approximately 5000–10,000 years ago.^24,25

In conclusion, TLR9 (T-1237C) mutation was seen to be implicated in malaria pathogenesis. These two ethnic groups, namely AAs having genetic affinity with the earliest inhabitants of Indian plains and TBs whose origin leans more towards the Southeast Asians, have accumulated different SNPs in the TLR9 promoter region as indicated from our sequencing results. Our study demonstrated that AA (plains) had the favourable genotype with respect to TLR9 polymorphisms which protected them from malaria pathogenesis and this may be related to their longer exposure to malaria. While TLR9 overexpression was favourable, that of TLR4 was a risk factor for complicated malaria. These findings thus open the possibility of the application of TLR9 SNPs as markers of population at risk of severe malaria.

Footnotes

Funding

This work was supported by Department of Biotechnology, India (grant no. BT/HRD/01/002/2007) and Rajiv Gandhi National Fellowship, University Grants Commission, India (grant no. F.14-616(ST)/2008(SA-III)) to CES.

Acknowledgements

We would like to thank all the participants for their cooperation in the study. Thanks to Dr Mamta Chawla, National Institute of Cholera and Enteric Diseases, Indian Council Medical Research, Kolkata for her help in the work carried out. Special thanks to Dr Anurodh Agrawal for his help.

References

Kawai

Akira

. TLR signaling. Semin Immunol 2007; 19: 24–32.

Corr

O’Neill

LAJ

. Genetic variation in Toll-like receptor signalling and the risk of inflammatory and immune diseases. J Innate Immun 2009; 1: 350–357.

Garantziotis

Hollingsworth

Zaas

Schwartz

. The effect of Toll-like receptors and Toll-like receptor genetics in human disease. Annu Rev Med 2008; 59: 343–359.

Gazzinelli

Denkers

. Protozoan encounters with Toll-like receptor signalling pathways: implications for host parasitism. Nat Rev Immunol 2006; 6: 895–906.

Gowda

. TLR-mediated cell signaling by malaria GPIs. Trend Parasitol 2007; 23: 596–604.

Coban

Ishii

Kawai

Hemmi

Sato

Uematsu

. Toll-like receptor 9 mediates innate immune activation by the malaria pigment haemozoin. JEM 2005; 201: 19–25.

Erdman

Finney

CAM

Liles

Kain

. Inflammatory pathways in malaria infection: TLRs share the stage with other components of innate immunity. Mol Biochem Parasitol 2008; 162: 105–111.

Franklin

Parroche

Ataíde

Lauw

Ropert

de Oliveira

. Malaria primes the innate immune response due to interferon-gamma induced enhancement of toll-like receptor expression and function. Proc Natl Acad Sci 2009; 106: 5789–5794.

Berdeli

Celik

Ozyürek

Dogrusoz

Aydin

. TLR-2 gene Arg753Gln polymorphism is strongly associated with acute rheumatic fever in children. J Mol Med 2005; 83: 535–541.

10.

Montes

Asensi

Alvarez

Valle

Ocaña

Meana

. The Toll-like receptor4 (Asp299Gly) polymorphism is a risk factor for Gram-negative and haematogenous osteomyelitis. Clin Exp Immunol 2006; 143: 404–413.

11.

Lorenz

Mira

Frees

Schwartz

. Relevance of mutations in the TLR4 receptor in patients with Gram negative septic shock. Arch Int Med 2002; 162: 1028–1032.

12.

Torok

Glas

Tonenchi

Bruennler

Folwaczny

. Crohn's disease is associated with a toll-like receptor-9 polymorphism. Gastroenterology 2004; 127: 365–366.

13.

Novak

Bussmann

Maintz

Peng

Hart

. Putative association of a TLR9 promoter polymorphism with atopic eczema. Allergy 2007; 62: 766–772.

14.

Zauner

Melroe

Sigrist

Rechsteine

Dorner

Arnold

. TLR9 triggering in Burkitt's lymphoma cell lines suppresses the EBV BZLF1 transcription via histone modification. Oncogene 2010; 29: 4588–4598.

15.

Mockenhaupt

Hamann

von Gaertner

Bedu-Addo

von Kleinsorgen

Schumann

Bienzle

. Common polymorphisms of toll-like receptors 4 and 9 are associated with the clinical manifestation of malaria during pregnancy. J Infect Dis 2006; 194: 184–188.

16.

Leoratti

Farias

Alves

Suarez-Mutis

Coura

Kalil

. Variants in the Toll-like receptor signaling pathway and clinical outcomes of malaria. J Infect Dis 2008; 198: 772– 780.

17.

Mockenhaupt

Cramer

Hamann

Stegemann

Eckert

. Toll-like receptor (TLR) polymorphisms in African children: common TLR-4 variants predispose to severe malaria. Proc Natl Acad Sci USA 2006; 103: 177–182.

18.

Barnes

. Genetic determinants and ethnic disparities in sepsis-associated acute lung injury. Proc Am Thorac Soc 2005; 2: 195–201.

19.

Sinha

Arya

Agarwal

Indian Genome Variation Consortium Habib

. Genetic differentiation of populations residing in areas of high malaria endemicity in India. J Genet 2009; 88: 77–80.

20.

Ferwerda

McCalla

Alonsod

Giamarellos-Bourboulise

Mouktaroudie

Izagirred

. TLR4 polymorphisms, infectious diseases, and evolutionary pressure during migration ofmodern humans. PNAS 2007; 104: 16645–16650.

21.

Dev

Hira

Rajkhowa

. Malaria attributable morbidity in Assam, north-eastern India. Ann Trop Med Parasitol 2001; 95: 789–796.

22.

Baruah

Lourembam

Sawian

Goswami

Baruah

. Temporal and spatial variation in MSP1 clonal composition of Plasmodium falciparum in districts of Assam, Northeast India. Infect Genet Evol 2009; 9: 853–859.

23.

Reddy

Langstieh

Kumar

Nagaraja

Reddy

ANS

Meka Reddy

AAG

. Austro-asiatic tribes of northeast India provide hitherto missing genetic link between south and southeast Asia. PloS ONE 2007; 2: e1141–e1141.

24.

Xiao

Deka

Seielstad

Kangwanpong

Xiao

. Y chromosome haplotypes reveal prehistorical migrations to the Himalayas. Hum Genet 2000; 107: 582–590.

25.

Wang

Zhang

Lin

Wang

Wan

. On the origin of Tibetans and their genetic basis in adapting high-altitude environments. PLoS ONE 2011; 6: e17002–e17002.

26.

Kumar

Langsiteh

Biswas

Babu

Rao

Thangaraj

. Asian and non-Asian origins of Mon-Khmer- and Mundari-speaking Austro-Asiatic populations of India. Am J Hum Biol 2006; 18: 461–469.

27.

Diffloth G. The contribution of linguistic palaeontology to the homeland of Austro-Asiatic. In: Sagart L, Blench R, Sanchez-Mazas A (eds) The peopling of East Asia: putting together archaeology. London: Routledge Curzon.

28.

World Health Organization. Severe falciparum malaria. Trans R Soc Trop Med Hyg 2000; 94Suppl. 1S1–S90.

29.

Lourembam

Sawian

Baruah

. Differential association of KIR gene loci to risk of malaria in ethnic groups of Assam, Northeast India. Infect Genet Evol 2011; 11: 1921–1928.

30.

Hamann

Hamprecht

Gomma

Schumann

. Rapid and inexpensive real-time PCR for genotyping functional polymorphisms within the Toll-like receptor -2,-4 and -9 genes. JIM 2004; 284: 281–291.

31.

Schroder

NWJ

Hermann

Hamann

Gobel

Hartung

Schumann

. High frequency of polymorphism Arg753Gln of the Toll-like recetor-2 gene detected by a novel allele-specific PCR. J Mol Med 2003; 81: 368–372.

32.

Lorenz

Frees

Schwartz

. Determination of the TLR4 genotype using allele-specific PCR. Biotechniques 2001; 31: 22–24.

33.

Agrawal

Sarkar

Chakrabarti

Rajendran

Kaur

Mishra

. Comparative evaluation of real-time PCR and conventional RT-PCR during a 2 year surveillance for influenza and respiratory syncytial virus among children with acute respiratory infections in Kolkata, India, reveals a distinct seasonality of infection. J Med Microbiol 2009; 58: 1616–1622.

34.

Sam-Agudu

Greene

Opoka

Kazura

Boivin

Zimmerman

. TLR9 polymorphisms are associated with altered IFNγ levels in children with cerebral malaria. Am J Trop Med Hyg 2010; 82: 548–555.

35.

Parikh

Rosenthal

. Human genetics and malaria: relevance for the design of clinical trials. J Infect Dis 2008; 198: 1255–1257.

36.

Carrington

Nelson

Martin

Kissner

Vlahov

Goedert

. HLA and HIV-1: Heterozygote advantage and B*35-Cw*04 disadvantage. Science 1999; 283: 1748–1748.

37.

Thurz

Thomas

Grennwood

Hill

AVS

. Heterozygote advantage for HLA class-II type in hepatitis B virus infection. Nat Genet 1997; 17: 11–12.

38.

Nakajima

Ohtani

Satta

Uno

Akari

Ishida

Kimura

. Natural selection in the TLR-related genes in the course of primate evolution. Immunogenetics 2008; 60: 727–735.

39.

Schofield

Novakovic

Gerold

Schwarz

McConville

Tachado

. Glycosylphosphatidylinositol toxin of Plasmodium up-regulates ICAM-1, VCAM-1, and E-selectin expression in vascular endothelial cells and increases leukocyte and parasite cytoadherence via protein tyrosine kinase-dependent signal transduction. J Immunol 1996; 156: 1886–1896.

40.

Loharungsikul

Troye-Blomberg

Amoudruz

Pichyangkul

Yongvanitchit

Looareesuwan

. Expression of toll-like receptors on antigen-presenting cells in patients with falciparum malaria. Acta Trop 2008; 105: 10–15.

41.

Coban

Ishii

Uematsu

Arisue

Sato

Yamamoto

. Pathological role of Toll-like receptor signaling in cerebral malaria. Int Immunol 2006; 19: 67–79.

42.

Lepenies

Cramer

Burchard

Wagner

Kirschning

Jacobs

. Induction of experimental cerebral malaria is independent of TLR2/4/9. Med Microbiol Immunol 2007; 179: 39–44.

43.

Parroche

Lauw

Goutagny

Latz

Monks

Visintin

. Malaria hemozoin is immunologically inert but radically enhances innate responses by presenting malaria DNA to Toll-like receptor 9. PNAS 2007; 104: 1919–1924.

44.

Syafruddin

Krisin Asih

Sekartuti Dewi

Coutrier

Rozy

Susanti

. Seasonal prevalence of malaria in West Sumba district, Indonesia. Malar J 2009; 8: 8–8.

45.

Carneiro

Roca-Feltrer

Griffin

Smith

Tanner

Schellenberg

. Age-patterns of malaria vary with severity, transmission intensity and seasonality in sub-saharan Africa: a systematic review and pooled analysis. PLoS ONE 2010; 5: e8988–e8988.

46.

Fuse

Katakura

Sakamoto

Ohira

. Toll-like receptor 9 gene mutations and polymorphisms in Japanese ulcerative colitis patients. World J Gastroenterol 2010; 16: 5815–5821.

47.

Campino

Forton

Auburn

Fry

Diakite

Richardson

. TLR9 polymorphisms in African populations: no association with severe malaria, but evidence of cis-variants acting on gene expression. Malar J 2009; 8: 44–44.