点击显示 收起
Immunology and Genetics of Parasitic Diseases, INSERM UMR, Faculty of Medicine Timone, Université de la Méditerranée, Marseille, France
Malaria Research and Training Centre, Faculty of Medicine, Pharmacy, and Odontostomatology, University of Bamako
Pediatric Ward, Gabriel Toure Hospital, Bamako, Mali
Background.
The pathogenic mechanisms of cerebral malaria (CM) are unclear but are thought to involve cytokine-mediated inflammation enhanced by parasite sequestration in the brain microcirculation. The role that interferon (IFN) could play that would enhance inflammation but also reduce parasitemia is unclear.
Methods.
Plasma IFN- concentrations were measured by enzyme-linked immunosorbent assay in 96 children with CM and 40 children with uncomplicated malaria (UM) who had been recruited from Gabriel Toure Hospital (Bamako, Mali). We investigated the relationship between IFN- concentrations and disease by nonparametric analysis. Polymorphisms in IFNG were characterized by restriction enzyme analysis or size-determination electrophoresis. Associations between polymorphisms and CM were evaluated by the family-based association test on 240 families.
Results.
During episodes of malaria, IFN- concentrations were lower in children with CM than in children with UM (P = .007). IFNG-183T (P = .009) and IFNG-183G/T (P = .013) were found to be less frequent than expected in children with CM. A trend toward association was also observed between IFNG(CA)14/(CA)14 (P = .073) and CM. The IFNG-183G/T and IFNG(CA)14/(CA)14 genotypes were more frequent in children with UM than in children with CM (odds ratio, 0.30 and 0.34, respectively).
Conclusions.
The low plasma IFN- concentrations in children with CM and the associations between a reduced risk of CM and (1) the IFNG-183T allele (which increases gene transcription) and (2) the IFNG-183G/T genotype are consistent with the concept that IFN- protects against CM.
Cerebral malaria (CM) is a severe complication of Plasmodium falciparum infection that occurs mostly in young children and in nonimmune adults. Causing a deep coma that may be associated with respiratory distress, acidosis, and/or severe anaemia, it kills >1 million African children each year [1]. CM is associated with the sequestration of parasites in the small blood vessels of the brain [2, 3]. The local accumulation of infected erythrocytes favors the local development of cytokine-mediated inflammatory reactions, which damage endothelial cells and alter nervous system function. Tumor necrosis factor (TNF), interleukin (IL)1, and IL-6 [46] have been associated with severe human malaria, principally in children, whereas IL-12 [7, 8], IL-10 [9], and transforming growth factor (TGF)1 [7] have been associated with protection against CM. It is currently believed [7, 10, 11] that the development of severe clinical malaria is determined by the balance between proinflammatory (IL-6, TNF, and IL-1) [46] and antiinflammatory (IL-10 and TGF-) [7, 9] cytokines. Interferon (IFN), a Th1 cytokine, may increase brain inflammation and decrease the risk of severe malaria by reducing parasite load. It is, therefore, unclear whether IFN- is involved in the development of severe malaria. We studied the role that IFN- plays during CM in children; our data suggest that IFN- (IFNG) plays a protective, rather than an aggravating, role during CM.
SUBJECTS, MATERIALS, AND METHODS
Subjects.
The present study included 240 children with CM and 40 children with uncomplicated malaria (UM). All of these children were recruited from the pediatric ward of the Gabriel Toure Hospital (Bamako, Mali). The following criteria were used to define clinical phenotypes. For CM, subjects were in a coma (Blantyre coma score of <3) and had a thick blood film positive for P. falciparum. Meningitis was ruled out by lumbar puncture. Quinine treatment rapidly led to full recovery in 187 of these children but was ineffective in 53, who subsequently died. The children with CM were recruited between 1999 and 2003. They had a median age of 6 years (interquartile range , 48 years) and a mortality rate of 22% (table 1).
For UM, the subjects had a thick blood film positive for P. falciparum, a Blantyre coma score of >4, and hematocrit >21%. These children were recruited during 2003 and had never developed CM. They attended the outpatient clinic for an episode of febrile malaria. The median age of the children with UM was 8 years (IQR, 410 years).
Informed consent was obtained from the parents, and 25 mL of peripheral blood was obtained from each child with CM and from both parents. Blood was obtained from siblings if one parent was absent. A blood sample (25 mL) was also obtained from each child with UM. Venous blood was collected into Vacutainers containing ACD before treatment. Plasma and cells were separated, dispensed into aliquots, and stored at -70°C. The present study was approved by the Ethics Committee of the Faculty of Medicine, Pharmacy, and Odontostomatology, University of Bamako (Bamako, Mali).
Cytokine assay.
IFN- concentrations were determined by ELISAs with pairs of cytokine-specific monoclonal antibodies (BD Pharmingen). The detection threshold was 5 pg/mL. Data are presented as arithmetic means of duplicate values.
Genotyping.
DNA was extracted from blood leukocytes by use of Autogen NA2000 (Geneworx). When necessary, DNA was amplified by polymerase chain reaction (PCR) with random 15-mer oligonucleotides. We identified parental inconsistencies by analyzing 10 highly informative microsatellites from the ABI Prism Linkage Mapping Set (D4S415, D3S426, D3S1601, D4S391, D3S1304, D4S1592, D3S1300, D3S1569, D4S1572, and D3S1262) (Perkin Elmer, Applied Biosystems). We found parental inconsistencies in 32 of 272 families analyzed. Genetic analysis was conducted for 240 families.
The positions of all IFNG polymorphisms are given relative to the IFNG gene sequence (GenBank accession number AF375790). Human IFN- is encoded by a single gene consisting of 4 exons and 3 introns on chromosome 12q24.1. A polymorphism in the promoter region, changing the G at position -183 to a T, creates an AP1-binding site for a nuclear transcription factor and increases gene transcription [12, 13]. The first intron of IFNG contains a polymorphic CA microsatellite repeat with 9 alleles. The (CA)12 and (CA)13 alleles are associated with high and low levels of IFN- production in vitro, respectively [14, 15].
Single-nucleotide polymorphisms (SNPs) at position -183 (-183G/T) were detected by PCRrestriction fragmentlength polymorphism assay, with 5-atggtgtgaagtaaaagtgc-3 and 5-ctcctctggctgctggtatt-3 as primers. The 30-L PCR mixture contained 75 ng of DNA, 1 mmol/L dNTP, 1.5 mmol/L MgCl2, 1× Perkin Elmer buffer, 0.5 mol/L each primer, and 1.5 U of Taq DNA polymerase. The cycling conditions were as follows: 94°C for 1 min; 35 cycles of 94°C for 1 min, 61°C for 45 s, and 72°C for 45 s; and final extension at 72°C for 5 min. PCR products were digested overnight at 37°C with 2 U of AvaII (Ozyme) and were then visualized by electrophoresis on a 2% agarose gel. The G allele gave rise to 2 bands (80 and 175 bp), and the T allele gave rise to 1 band (255 bp).
CA repeats in intron 1 of IFNG were analyzed by PCR, with 5-gctgctataataatattcagac-3 carrying a HEX label at the 5 end and 5-cgagctttaaaagatagttcc-3 as primers. The 15 L of PCR mixture contained 75 ng of DNA, 250 mol/L dNTP, 2.5 mmol/L MgCl2, 1× Perkin Elmer buffer, 0.25 mol/L each primer, and 0.4 U of TaqGold polymerase (Perkin Elmer). The cycling conditions were as follows: 95°C for 15 min; 2 cycles at 94°C for 30 s and 70°C for 30 s; 2 cycles at 94°C for 30 s and 65°C for 30 s; 2 cycles at 94°C for 30 s and 60°C for 30 s; 28 cycles at 94°C for 30 s, 56°C for 30 s, and 72°C for 30 s; and final extension at 72°C for 10 min. PCR fragments were analyzed on an ABI Prism 310 analyzer, in accordance with the manufacturer's instructions (Perkin Elmer, Applied Biosystems). PCR products were separated by high-resolution fluorescent electrophoresis, and the lengths of individual fragments were determined by comparing them with the fluorescently labeled internal size marker ROX 400 (Perkin Elmer, Applied Biosystems), by use of GeneScan software. Products 121133-bp long were obtained. Random samples from homozygous individuals were sequenced, to confirm the number of repeats.
Statistical analysis.
We used the nonparametric Wilcoxon rank sum test to assess differences between 2 matched groups. Unmatched groups were compared by the Mann-Whitney U test, with P < .05 considered to be significant. Both tests were conducted by use of SPSS (version 10.1; SPSS).
We estimated deviation from Hardy-Weinberg equilibrium and linkage disequilibrium by use of Genepop (Web version 3.4), with the genotypes of 480 unrelated first-degree relatives. For the Hardy-Weinberg exact test, we used option 1 and suboption 3 of Genepop, which corresponded to the probability test [16]. For linkage disequilibrium, we used option 2. In this option, Genepop creates contingency tables for all pairs of loci in each population and then performs a probability test (or Fisher's exact test) for each table, using a Markov chain as described by Raymond and Rousset [17].
Associations between individual polymorphisms and CM were examined by the family-based association test (FBAT; version 1.4) [18, 19]. Individual polymorphisms, combined into haplotypes, were tested for excess transmission during CM by the FBAT. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by use of JavaStat.
RESULTS
Lower plasma IFN- concentrations in children with CM than in children with UM.
We measured plasma IFN- concentrations in 96 children with CM and in 40 children with UM who had been admitted to Gabriel Toure Hospital during the period from August to February of the years 1999 to 2003. Sixteen (16.7%) of the children with CM died in the hospital. In the children with CM, the median plasma IFN- concentration was significantly higher on admission to the hospital than it was 28 days later (15 vs. 1 pg/mL, respectively; P = .002) (figure 1A). On average, the children with UM had higher plasma IFN- concentrations than did the children with CM (66 and 19 pg/mL, respectively; P = .007). Among the children with CM, those who died of the complications of malaria tended to have higher IFN- concentrations than did those who recovered (41 vs. 14 pg/mL; P = .073) (figure 1B). The median IFN- concentration was 66 pg/mL in the children with UM and was 14 pg/mL (P = .002) in those who recovered from CM (figure 1B).
Associations between protection against CM and (1) the IFNG-183T allele and (2) the IFNG-183T(CA)14 haplotype, by the FBAT.
We first assessed whether each polymorphism was in Hardy-Weinberg equilibrium. In our study population, the IFNG-183 and the IFNG(CA)n genotypes were in Hardy-Weinberg equilibrium (parents of children with CM, P = .19 for IFNG[CA]n and P = .59 for IFNG-183G/T; children with UM, P = .07 for IFNG[CA]n and P = 1 for IFNG-183G/T).
We then evaluated whether certain alleles were preferentially associated with CM; transmission of the different alleles from heterozygous parents to affected children was tested by the FBAT [18, 19]. The -183G allele was the most frequent (0.960) in the study population: 37 of the 240 children with CM had at least 1 parent who was heterozygous for this allele (-183G/T) (table 2). Only 11 of these 37 children with CM born to heterozygous parents (father and/or mother) had received the -183T allele, which is significantly different (Z = -2.6; P = .009) from the number (n = 19) that would be expected when an absence of association between this allele and CM is assumed. Thus, the IFNG-183T allele is associated with a lower risk of CM. The IFNG-183G/T genotype was less likely to be transmitted to children with CM than was expected under the null hypothesis11 versus 18.5 transmissions (Z = -2.5; P = .013)showing that IFNG-183G/T is associated with a lower risk of CM.
Seven alleles of the CA repeat were observed in our subjects: (CA)12 to (CA)18. The (CA)13 and (CA)14 alleles accounted for 21% and 48%, respectively, of the CA alleles and were the most frequent. Because IFNG(CA)15, (CA)16, (CA)17, and (CA)18 were rare, they were pooled in the analysis and are referred to as "IFNG(CA)1518." A trend toward an asociation between the IFNG(CA)14 allele and CM was observed (Z = -1.8; P = .074) (table 2). A trend toward an association between CM and genotypes IFNG(CA)12/(CA)14 (Z = 1.9; P = .054) and IFNG(CA)14/(CA)14 (Z = -1.8; P = .073) was also found (table 2).
We found no evidence for linkage disequilibrium between IFNG-183 and the IFNG(CA) alleles (P = .907). Nevertheless, the (CA)14 allele was found to be preferentially combined with the IFNG-183T allele (34 of the 38 parents with IFNG-183T also had IFNG[CA]14). We therefore conducted a haplotype analysis, combining the IFNG(CA)n and IFNG-183G/T polymorphisms. In total, we observed 7 haplotypes (table 3). Biallelic testing indicated that the -183T-(CA)14 haplotype was associated with protection against CM (Z = -2.7; P = .007). Conversely, the -183G-(CA)13 haplotype was associated with an increased risk of CM (Z = 2.2; P = .027) (table 3).
We investigated the frequency of the IFNG-183G/T genotype in the children with CM and the children with UM. The IFNG-183G/T genotype was 3 times more frequent in the children with UM (6/40 [15%]) than in the children with CM (12/240 [5%]) (OR, 0.30 [95% CI, 0.110.82]). Similar results were obtained for the IFNG(CA)14/IFNG(CA)14 genotype, which was less frequent in the children with CM (45/235 [19%]) than in the children with UM (16/39 [41%]) (OR, 0.34 [95% CI, 0.170.69]). Subjects with the IFNG-183G/T genotype and/or the IFNG(CA)14/IFNG(CA)14 genotype had a 70% lower risk of CM. This finding is consistent with the result of the FBAT.
DISCUSSION
We found that IFNG-183T was associated with CM in P. falciparuminfected children. This association suggests that the tendency to develop CM is controlled either directly by this SNP or indirectly by another SNP in linkage disequilibrium with IFNG-183G/T. Several lines of evidence suggest that the IFNG-183T allele is the protective variant. First, 2 independent studies [12, 13] have shown that the GT transition at the -183 position creates a binding site for a nuclear transcription factor, resulting in an increase in gene transcription. Second, our data have shown that those who control the infection well enough to avoid coma have plasma IFN- concentrations that are 3.5 times higher than those with CM, indicating that IFN- may play a critical role in the control of CM. Third, several studies have shown that IFN- is a key cytokine in the control of P. falciparum infection, such that mutations that increase the level of IFN- gene transcription would be expected to improve the control of infection.
Analysis of the CA repeats by the FBAT also suggested a possible association between (CA)14 and CM. The frequency of the (CA)14 allele was much higher in the children with UM than in the children with CM, providing further evidence for an association between this allele and disease. This interpretation was supported by the haplotype analysis, which demonstrated undertransmission of the IFNG-183T(CA)14 combination in the children with CM.
It is difficult to define the mechanisms of control of infection and disease for malaria, because the effects of a given cytokine may vary from protective to aggravating depending on the concentration of the cytokine in question and on associations with other cytokines. Our study clearly shows that the IFNG-183T allele protects against CM. However, we cannot exclude the possibility that exaggerated IFN- production may aggravate malaria and possibly even cause death. We found that the children who died produced much more IFN- than did those who recovered, but this effect was not necessarily specific, because the concentrations of various inflammatory cytokines were also markedly higher in the plasma of children who died (data not shown). We were unable to conduct a genetic analysis of IFNG polymorphisms in the children who died of malaria, because of the low statistical power of our study (a result of the few number of cases) and the low frequency of the IFNG-183T allele. Another problem that we faced was the recruitment of control subjects for the immunological analysis: we could not be sure that none of the children with UM would go on to develop CM. We addressed this problem by using the FBAT to analyze the disequilibrium of the transmission of a given allele from parents to affected children. This analysis does not require the use of a control group. In the immunological analysis, we analyzed cytokine concentrations in the children, to enable us to define the profile of an episode of UM.
Cytokines have been shown to play important roles during severe malaria: TNF, IL-1, and IL-6 [46] have been associated with severe malaria, principally in children, whereas IL-12 [7, 8], IL-10 [9], and TGF-1 [7] have been associated with protection against CM. Moreover, it has been suggested [7, 10, 11] that the clinical outcome of P. falciparum infection depends on the balance between proinflammatory (IL-6, TNF, and IL-1 [46]) and antiinflammatory (IL-10 and TGF-) cytokines [7, 9]. However, the role played by IFN- during CM is not clearly understood, because IFN- may increase the production of proinflammatory cytokines such as TNF or decrease the risk of severe malaria by reducing parasite load. These opposite effects have been described in humans and animals: studies in children have revealed an association between severe malaria (but not CM) and low plasma IFN- and IL-12 concentrations [7, 8], and studies in experimental models of malaria have shown that IFN- can be lethal in mice [20]. Our results provide the first evidence indicating that IFN- probably protects children against CM.
Human immunity against blood-stage parasites involves NK cells, 92 T cells, T cells, and monocytes/macrophages acting as effector cells. Numerous NK and 92 T cells are produced during the first 2448 h of infection [2123]. T cells display high levels of cytotoxicity to asexual blood-stage parasites [24]. The number of T cells increases markedly 68 days after infection, and these cells relay the action of NK and 92 T cells [2224]. Activated dendritic cells recruit NK cells to lymph nodes, where they cause T helper cell polarization. NK cells, 92 T cells, and Th1 T cells produce large quantities of IFN- [21, 23]. IFN- may be directly toxic to blood-stage parasites [25]. It also potentiates the cytotoxicity of monocytes against merozoites and infected red blood cells. Finally, IFN- contributes to Th1 polarization of the T cell response. IFN- also protects against infection, as is shown by the association between high plasma IFN- concentration and the control of parasitemia [26, 27]. Thus, the protective effect of IFN- against CM may be, at least in part, due to the positive effects that this cytokine has on the control of blood-stage parasites.
Several studies have reported an association between severe malaria and alleles of genes involved in immunity. These genes include those encoding TNF [2830], IFN- receptor ligand-binding chain 1 (IFNGR1) [31], leukocyte adhesion molecules (CD36 and intercellular adhesion molecule 1) [32, 33], and inducible nitric oxide synthase (iNOS) [3438]. The associations between CM and alleles of TNFA, CD36, and ICAM-1, which mediate the adhesion of infected erythrocytes to the endothelium, are consistent with the hypothesis that localized inflammation in the brain microcirculation plays an important role during CM. It has been shown that mutations or deletions in red blood cell proteins, especially in the and globin chains, are associated with protection against severe malaria because these mutations limit parasite multiplication [39]. However, they are unlikely to reduce local brain inflammation directly. The increase in NO production associated with the -1173CT NOS2 polymorphism [38] protects against CM. This finding is consistent with the concept that IFN- has a protective effect, because this cytokine markedly up-regulates iNOS production. Plasma NO concentrations are higher in those with UM than in those with CM [40, 41], and NO decreases leukocyte adhesion to the endothelium in microvessels.
In conclusion, CM is associated with IFNG alleles that decrease IFN- gene transcription and with low plasma IFN- concentrations in children. This, together with previous observations, provides the first piece of evidence that IFN- protects children against CM.
Acknowledgments
We thank Drs. Christophe Chevillard and Laurent Argiro, for helpful advice, and Eliane Charrat, for technical assistance. We also thank the staff of the pediatric ward at the Gabriel Toure Hospital (Bamako, Mali), for their help.
References
1. Marsh K, English M, Crawley J, Peshu N. The pathogenesis of severe malaria in African children. Ann Trop Med Parasitol 1996; 90:395402. First citation in article
2. Miller LH, Baruch DI, Marsh K, Doumbo OK. The pathogenic basis of malaria. Nature 2002; 415:6739. First citation in article
3. Taylor TE, Fu WJ, Carr RA, et al. Differentiating the pathologies of cerebral malaria by postmortem parasite counts. Nat Med 2004; 10:1435. First citation in article
4. Kern P, Hemmer CJ, Van Damme J, Gruss HJ, Dietrich M. Elevated tumor necrosis factor alpha and interleukin-6 serum levels as markers for complicated Plasmodium falciparum malaria. Am J Med 1989; 87:13943. First citation in article
5. Grau GE, Piguet PF, Vassalli P, Lambert PH. Tumor-necrosis factor and other cytokines in cerebral malaria: experimental and clinical data. Immunol Rev 1989; 112:4970. First citation in article
6. Kwiatkowski D, Hill AV, Sambou I, et al. TNF concentration in fatal cerebral, non-fatal cerebral, and uncomplicated Plasmodium falciparum malaria. Lancet 1990; 336:12014. First citation in article
7. Perkins DJ, Weinberg JB, Kremsner PG. Reduced interleukin-12 and transforming growth factor1 in severe childhood malaria: relationship of cytokine balance with disease severity. J Infect Dis 2000; 182:98892. First citation in article
8. Luty AJ, Perkins DJ, Lell B, et al. Low interleukin-12 activity in severe Plasmodium falciparum malaria. Infect Immun 2000; 68:390915. First citation in article
9. Kurtzhals JA, Adabayeri V, Goka BQ, et al. Low plasma concentrations of interleukin 10 in severe malarial anaemia compared with cerebral and uncomplicated malaria. Lancet 1998; 351:176872. First citation in article
10. Day NP, Hien TT, Schollaardt T, et al. The prognostic and pathophysiologic role of pro- and antiinflammatory cytokines in severe malaria. J Infect Dis 1999; 180:128897. First citation in article
11. Dodoo D, Omer FM, Todd J, Akanmori BD, Koram KA, Riley EM. Absolute levels and ratios of proinflammatory and anti-inflammatory cytokine production in vitro predict clinical immunity to Plasmodium falciparum malaria. J Infect Dis 2002; 185:9719. First citation in article
12. Bream JH, Ping A, Zhang X, Winkler C, Young HA. A single nucleotide polymorphism in the proximal IFN-gamma promoter alters control of gene transcription. Genes Immun 2002; 3:1659. First citation in article
13. Chevillard C, Henri S, Stefani F, Parzy D, Dessein A. Two new polymorphisms in the human interferon gamma (IFN-gamma) promoter. Eur J Immunogenet 2002; 29:536. First citation in article
14. Pravica V, Asderakis A, Perrey C, Hajeer A, Sinnott PJ, Hutchinson IV. In vitro production of IFN-gamma correlates with CA repeat polymorphism in the human IFN-gamma gene. Eur J Immunogenet 1999; 26:13. First citation in article
15. Miyake K, Nakashima H, Akahoshi M, et al. Genetically determined interferon-gamma production influences the histological phenotype of lupus nephritis. Rheumatology (Oxford) 2002; 41:51824. First citation in article
16. Guo SW, Thompson EA. Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics 1992; 48:36172. First citation in article
17. Raymond M, Rousset F. GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. J Hered 1995; 86:2489. First citation in article
18. Horvath S, Xu X, Laird NM. The family based association test method: strategies for studying general genotype-phenotype associations. Eur J Hum Genet 2001; 9:3016. First citation in article
19. Wilk JB, Volcjak JS, Myers RH, et al. Family-based association tests for qualitative and quantitative traits using single-nucleotide polymorphism and microsatellite data. Genet Epidemiol 2001; 21(Suppl 1):S3649. First citation in article
20. Kremsner PG, Neifer S, Chaves MF, Rudolph R, Bienzle U. Interferon-gamma induced lethality in the late phase of Plasmodium vinckei malaria despite effective parasite clearance by chloroquine. Eur J Immunol 1992; 22:28738. First citation in article
21. Choudhury HR, Sheikh NA, Bancroft GJ, Katz DR, De Souza JB. Early nonspecific immune responses and immunity to blood-stage nonlethal Plasmodium yoelii malaria. Infect Immun 2000; 68:612732. First citation in article
22. Hensmann M, Kwiatkowski D. Cellular basis of early cytokine response to Plasmodium falciparum. Infect Immun 2001; 69:236471. First citation in article
23. Artavanis-Tsakonas K, Riley EM. Innate immune response to malaria: rapid induction of IFN-gamma from human NK cells by live Plasmodium falciparum-infected erythrocytes. J Immunol 2002; 169:295663. First citation in article
24. Troye-Blomberg M, Worku S, Tangteerawatana P, et al. Human gamma delta T cells that inhibit the in vitro growth of the asexual blood stages of the Plasmodium falciparum parasite express cytolytic and proinflammatory molecules. Scand J Immunol 1999; 50:64250. First citation in article
25. Yoneto T, Yoshimoto T, Wang CR, et al. Gamma interferon production is critical for protective immunity to infection with blood-stage Plasmodium berghei XAT but neither NO production nor NK cell activation is critical. Infect Immun 1999; 67:234956. First citation in article
26. Luty AJ, Lell B, Schmidt-Ott R, et al. Interferon- responses are associated with resistance to reinfection with Plasmodium falciparum in young African children. J Infect Dis 1999; 179:9808. First citation in article
27. Torre D, Speranza F, Giola M, Matteelli A, Tambini R, Biondi G. Role of Th1 and Th2 cytokines in immune response to uncomplicated Plasmodium falciparum malaria. Clin Diagn Lab Immunol 2002; 9:34851. First citation in article
28. Knight JC, Udalova I, Hill AV, et al. A polymorphism that affects OCT-1 binding to the TNF promoter region is associated with severe malaria. Nat Genet 1999; 22:14550. First citation in article
29. McGuire W, Hill AV, Allsopp CE, Greenwood BM, Kwiatkowski D. Variation in the TNF-alpha promoter region associated with susceptibility to cerebral malaria. Nature 1994; 371:50810. First citation in article
30. McGuire W, Knight JC, Hill AV, Allsopp CE, Greenwood BM, Kwiatkowski D. Severe malarial anemia and cerebral malaria are associated with different tumor necrosis factor promoter alleles. J Infect Dis 1999; 179:28790. First citation in article
31. Koch O, Awomoyi A, Usen S, et al. IFNGR1 gene promoter polymorphisms and susceptibility to cerebral malaria. J Infect Dis 2002; 185:16847. First citation in article
32. Kun JF, Klabunde J, Lell B, et al. Association of the ICAM-1Kilifi mutation with protection against severe malaria in Lambarene, Gabon. Am J Trop Med Hyg 1999; 61:7769. First citation in article
33. Pain A, Urban BC, Kai O, et al. A non-sense mutation in Cd36 gene is associated with protection from severe malaria. Lancet 2001; 357:15023. First citation in article
34. Burgner D, Xu W, Rockett K, et al. Inducible nitric oxide synthase polymorphism and fatal cerebral malaria. Lancet 1998; 352:11934. First citation in article
35. Kun JF, Mordmuller B, Lell B, Lehman LG, Luckner D, Kremsner PG. Polymorphism in promoter region of inducible nitric oxide synthase gene and protection against malaria. Lancet 1998; 351:2656. First citation in article
36. Levesque MC, Hobbs MR, Anstey NM, et al. Nitric oxide synthase type 2 promoter polymorphisms, nitric oxide production, and disease severity in Tanzanian children with malaria. J Infect Dis 1999; 180:19942002. First citation in article
37. Kun JF, Mordmuller B, Perkins DJ, et al. Nitric oxide synthase 2Lambaréné (G-954C), increased nitric oxide production, and protection against malaria. J Infect Dis 2001; 184:3306. First citation in article
38. Hobbs MR, Udhayakumar V, Levesque MC, et al. A new NOS2 promoter polymorphism associated with increased nitric oxide production and protection from severe malaria in Tanzanian and Kenyan children. Lancet 2002; 360:146875. First citation in article
39. Aidoo M, Terlouw DJ, Kolczak MS, et al. Protective effects of the sickle cell gene against malaria morbidity and mortality. Lancet 2002; 359:13112. First citation in article
40. Anstey NM, Weinberg JB, Hassanali MY, et al. Nitric oxide in Tanzanian children with malaria: inverse relationship between malaria severity and nitric oxide production/nitric oxide synthase type 2 expression. J Exp Med 1996; 184:55767. First citation in article
41. Lopansri BK, Anstey NM, Weinberg JB, et al. Low plasma arginine concentrations in children with cerebral malaria and decreased nitric oxide production. Lancet 2003; 361:6768. First citation in article