Abstract
Rationale: Interferon-γ (IFN-γ) is of central interest in the study of tuberculosis. A number of single-gene mutations have been identified in the IFN-γ signaling pathway that predispose to severe mycobacterial disease, but the relevance of polymorphism within these genes to the common phenotype of tuberculosis remains unclear.
Methods: A total of 1,301 individuals were included in a large, detailed study of West African populations with pulmonary tuberculosis. We investigated disease association with the genes encoding IFN-γ and its receptor subunits (IFNG, IFNGR1, and IFNGR2).
Results: Within the IFNG gene, two promoter variants showed evidence of novel disease association: −1616GG (odds ratio [OR], 1.49; 95% confidence interval [CI], 1.11–2.00; p = 0.008) and +3234TT (OR, 1.40; 95% CI, 1.09–1.80; p = 0.009). The +874AA genotype was not significantly more frequent among cases over control subjects (OR, 1.16; 95%CI, 0.89–1.51; p = 0.25). In addition, novel disease association was also found with the −56CC genotype of the IFNGR1 promoter (OR, 0.75; 95% CI, 0.57–0.99; p = 0.041). No disease association was seen with the IFNGR2 locus.
Conclusions: These results provide evidence of a significant role for genetic variation at the IFNG locus and provide detailed understanding of the genetic mechanisms underlying this association. The disease association with IFNGR1 is novel, and together these findings support the hypothesis that genetically determined variation in both IFN-γ production and responsiveness influences the risk of developing tuberculosis.
The interferon-γ (IFN-γ) signaling pathway is of central interest in the study of tuberculosis, a disease that still kills nearly two million people a year (1). The potential importance of the pathway is highlighted by the discovery of several families with Mendelian susceptibility to mycobacterial disease (MSMD; Mendelian Inheritance in Man [MIM] 209950) who have mutations in one of two subunits of the IFNG receptor (IFNGR1 and IFNGR2) (2). Although there are no families identified with mutation in IFNG itself, recent data suggest a more common polymorphism at position +874 is associated with the risk of tuberculosis in different populations (3–5).
IFN-γ is an acid-labile protein produced by CD4+ T cells (and other cell types, including natural killer cells, CD8+ T cells, and macrophages). It acts as a regulator of gene expression through activation of a receptor complex comprising two subunits, each encoded by a different gene (IFNGR1 on chromosome 6q23-4 and IFNGR2 on chromosome 21q22.1–22.2). Homodimers of IFN-γ interact with both receptor proteins (6), leading to receptor dimerization (7), and each of the three molecules plays a nonredundant role in ligand-activated receptor signaling.
In the most comprehensive investigation of the pathway to date, we have performed a large and detailed study of polymorphism within the following three genes: IFNG, IFNGR1, and IFNGR2. All are strong candidate genes for the complex phenotype of tuberculosis. Any association within these genes would provide insight into the clinically important immune mechanisms influencing this persistent disease and provide an opportunity to assess the extent to which rare single-gene disorders can inform our understanding of complex infectious diseases.
This study was performed as part of a large case-control study of adult pulmonary tuberculosis in three West African countries described in detail previously (10, 11), and not including subjects studied previously by Bellamy and colleagues (12–14) or Awomoyi and colleagues (8, 15). The study uses population collections from the Gambia, Guinea Bissau, and the Republic of Conakry. All cases were confirmed by either two consecutive smear-positive samples or a positive Mycobacterium tuberculosis (MTB) culture. Both HIV-positive and HIV-negative cases were included, the prevalence of each differing among countries (10). This study included a total of 682 cases and 619 control subjects, providing at least 80% power to detect an odds ratio (OR) of 2 for a genotype of frequency between 4 and 50%. DNA was available from the parents of 210 affected Gambians to allow a family-based test of association and estimation of haplotype frequencies. Informed consent was obtained from patients or their parents or guardians. Ethical approval was provided by the joint Gambian Government/Medical Research Council Ethical Committee, the Ministry of Public Health (MINSAP, Guinea-Bissau), and the National Ethics Committee, Ministry of Health, Conakry, Republique de Guinee. Human experimentation guidelines of these ministries were followed.
DNA was extracted using the Nucleon BACC2 DNA extraction kit (Nucleon Bioscience, Manchester, UK). DNA concentrations were determined using the PicoGreen kit (Molecular Probes, Paisley, UK). Initial typing of the +874A/T polymorphism was performed with an amplification refractory mutation system (ARMS) reaction similar to that first described by Pravica and colleagues (16). Reaction mix included 5 μl DNA at 10 ng/μl, 5 μl H2O, 1.5 μl 10× polymerase chain reaction buffer, 1.5 μl Mg2+, 0.3 μl of 8-mM deoxynucleotide triphosphates (dNTPs), 0.3 μl of 10-μM allele-specific and common primers, and 0.3 μl of 2-μM control primers (human growth hormone [HGH]), 0.08 μl Taq Gold. Reaction conditions were 95°C for 1 min followed by 10 cycles of 95°C for 15 s, 62°C for 50 s, and 72°C for 40 s, then 20 cycles of 95°C for 20 s, 56°C for 50 s, and 72°C for 50 s. This genotyping was subsequently confirmed using the Sequenom genotyping system, which uses mass spectrometry (MALDI-TOF) to discriminate products by their absolute mass. All other polymorphisms were typed using the Sequenom system.
The available literature and publicly accessible databases were used to identify single-nucleotide polymorphisms (SNPs). In the case of IFNG, all SNPs previously associated with infectious disease were studied. No known amino acid–changing polymorphisms have yet been identified in this gene, and thus SNPs at the following positions were selected to cover the four exons: −1616; −308, referred to elsewhere sometimes as −179 (17, 18); +874 (3–5); +1987 (19); +3234; +3686; and +5173. Within IFNGR1, two noncoding SNPs were studied at positions −56 (20) and +95 (21) alongside two exonic variants, H318P (20) and L467P (22). Within IFNGR2, two nonsynonymous polymorphisms from exon 2 (T58R and R64Q [23]), were studied alongside one variant from intron 1 (GenBank NT_011512).
Analysis of case-control data was performed with SPSS version 11 (SPSS, Inc., Chicago, IL). Overall genotype frequencies were studied with a 3 × 2 χ2 test. Binary logistic regression was used to assess the effect of genotype on disease outcome, adjusted for age, sex, ethnicity, and HIV status. Disease association within the families was studied using the transmission disequilibrium test (TDT) (24), using the TRANSMIT program (25). GENEHUNTER (26) was used to generate haplotypes prior to analysis within GOLD (Graphical Overview of Linkage Disequilibrium) (27) to estimate linkage disequilibrium (LD) coefficients.
For each SNP selected above, initial genotyping was performed on 94 cases and 94 control subjects to establish allele frequencies. One SNP was excluded from further study because it appeared to be in complete LD with another polymorphism (IFNGR1 +95). The allele frequencies for each polymorphism in Gambian control subjects are shown in Table 1.
IFNG | IFNGR1 | IFNGR2 | |||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| −1616 | −308 | +874 | +1987 | +3234 | +3686 | +5173 | −56 | +95 | +1004 | +1400 | INT1 | T58R | R64Q | ||||||||||||
| Major allele | 0.49 G | 0.96 G | 0.83 A | 0.97 A | 0.55 T | 0.95 G | 0.85 A | 0.53 T | 0.53 A | 0.97 T | 0.97 A | 0.71 C | 0.84 C | 0.75 A | |||||||||||
| Minor allele | 0.51 A | 0.04 T | 0.17 T | 0.03 G | 0.45 C | 0.05 A | 0.15 G | 0.47 C | 0.47 G | 0.03 G | 0.03 G | 0.29 T | 0.16 G | 0.25 T | |||||||||||
The +874AA genotype did not show evidence of disease association in the overall population (unadjusted OR is 1.08; 95% confidence interval [CI], 0.96–1.22; p = 0.19; after adjusting for age, sex, ethnicity, and HIV status, OR is 1.16; 95% CI, 0.89–1.51; p = 0.25). Within each of the three populations, the same trend was observed toward an excess of +874AA homozygotes among cases (+874AA: Gambia: 76.3% cases, 72.1% control subjects; Guinea Bissau: 75.2% cases, 70.3% control subjects; and Conakry: 66.3% cases, 65.9% control subjects).
LD mapping suggests strong LD exists between several of these loci, and a strong block of LD exists within the gene itself (data not shown). The +874 polymorphism is in significant LD with two other polymorphisms that do show evidence of disease association, namely those at −1616 (D′ 0.79, p < 0.001) and +3234 (D′ 0.66, p < 0.001). Overall genotype distributions for the associated SNPs are shown in Table 2. The −1616 promoter polymorphism is associated with disease, in particular the −1616GG genotype (OR, 1.49; 95% CI, 1.11–2.00; p = 0.008), and a similar trend in genotype distribution is seen across the three countries studied (−1616GG: Gambia: 31.3% cases, 24.5% control subjects; Guinea Bissau: 22.2% cases, 20.8% control subjects; and Conakry: 33.0% cases and 24.8% control subjects). The +3234 polymorphism also showed evidence of disease association (+3234TT: OR, 1.40; 95% CI, 1.09–1.80; p = 0.009) and a similar trend in genotype distribution across three countries (+3234TT: Gambia: 57.3% cases, 53.1% control subjects; Guinea Bissau: 55% cases, 48.3% control subjects; and Conakry: 52.4% cases and 43.6% control subjects). Distribution of genotypes between case and control subjects was similar in the HIV-negative and the small number of HIV-positive individuals (data not shown).
IFNG Polymorphism | Genotype | TB Cases (%) | TB Control Subjects (%) | Total |
|---|---|---|---|---|
| −1616 | GG | 183 (29.3) | 139 (23.6) | 322 |
| GA | 285 (45.6) | 294 (49.9) | 579 | |
| AA | 157 (25.1) | 156 (26.5) | 313 | |
| 625 | 589 | |||
| +874 | AA | 488 (73.2) | 415 (69.9) | 903 |
| AT | 159 (23.8) | 166 (27.9) | 325 | |
| TT | 20 (3) | 13 (2.2) | 33 | |
| 667 | 594 | |||
| +3234 | TT | 373 (55.5) | 299 (49.5) | 672 |
| CT | 231 (34.4) | 244 (40.4) | 475 | |
| CC | 68 (10.1) | 61 (10.1) | 129 | |
| 672 | 604 |
In a smaller sample set of families in the Gambian population, no evidence was seen for association with disease in a TDT-based analysis, although the −1616G allele was transmitted more often than expected by chance (182 observed vs. 178 expected, p = 0.42). Haplotype analysis found that the three alleles all putatively associated with disease (−1616, +874, and +3234) are all found on the most common West African haplotype (see Table 3), which, although overtransmitted, was not significantly associated with disease in this smaller population sample.
−1616 | −308 | +874 | +1987 | +3234 | +3686 | +5173 | f |
|---|---|---|---|---|---|---|---|
| G | G | A | A | T | G | A | 0.37 |
| A | G | A | A | T | G | A | 0.18 |
| G | G | A | G | T | G | A | 0.16 |
| A | G | A | A | C | G | A | 0.12 |
| A | G | T | A | C | G | G | 0.1 |
| A | G | A | A | C | G | A | 0.03 |
| A | T | A | A | T | G | A | 0.03 |
| A | G | A | A | C | A | A | 0.02 |
Both the two coding polymorphisms studied within IFNGR1 have low heterozygosity in the populations studied (minor allele frequency, 3% for each), resulting in poor power to detect any effect on tuberculosis. There was no evidence of disease association with these variants (overall p = 0.53 and p = 0.32, respectively; see genotype frequencies in Table 4). However, a small effect is seen with the −56 promoter variant. The −56CC genotype is significantly associated with protection from disease (unadjusted OR is 0.75; 95% CI, 0.59–0.97; p = 0.02; adjusted for age, sex, ethnicity, and HIV status, OR is 0.75; 95% CI, 0.57–0.99; p = 0.041). Haplotype analysis did not find evidence for disease association in the small number of Gambian trios studied (data not shown). Given the rarity of the two nonsynonymous polymorphisms, the −56C/T polymorphism was the most informative, discriminating 94% of all haplotypes seen. Consistent with the case-control data, the −56C allele was nonsignificantly undertransmitted to affected offspring (180 observed vs. 189 expected, p = 0.14).
Polymorphism | Genotype | TB Cases (%) | TB Control Subjects (%) | Total |
|---|---|---|---|---|
| −56 | TT | 171 (25.1) | 193 (31.2) | 364 |
| CT | 331 (48.5) | 278 (44.9) | 609 | |
| CC | 180 (26.4) | 148 (23.9) | 328 | |
| 682 | 619 | |||
| H318P (+1004) | TT | 619 (94.2) | 565 (94.8) | 1,184 |
| TG | 35 (5.3) | 27 (4.5) | 62 | |
| GG | 3 (0.5) | 4 (0.7) | 7 | |
| 657 | 596 | |||
| L467P (+1400) | AA | 618 (95.5) | 557 (94.1) | 1,075 |
| AG | 29 (4.5) | 35 (5.9) | 64 | |
| GG | 0 | 0 | 0 | |
| 647 | 592 |
None of the INT1, T58R, and R64Q variants was associated with disease (overall significance levels for genotype distributions of p = 0.31, p = 0.92, and p = 0.78, respectively). Each variant had a minor allele frequency of greater than 0.15, providing 80% power in this sample set to detect an allele-specific disease risk of 1.5. None of the three variants showed evidence of association with tuberculosis either individually in TDT or haplotype analysis (data not shown).
IFN-γ–mediated immune activation appears to have an important role in immunity to intracellular pathogens, both in mice and humans, but its role in human tuberculosis still remains unclear (28). Early studies (29, 30) demonstrated that mice with disruption of the ifng gene are more susceptible to infection by MTB than wild-type strains. Macrophages from these mice have impaired activation and treating the mice with recombinant IFN-γ slows (though does not stop) their rapidly progressive disease (30). Knockout (ifng−/−) mice given bacillus Calmette-Guérin (BCG) engineered to produce IFN-γ control infection better than those given BCG alone (31), and IFN-γ production within the lung appears to be an important component of this response (Reuter and colleagues, unpublished data, and Reference 32).
Despite the accumulating data for a role in mice, it has been harder to establish the importance of IFN-γ in human tuberculosis. Studies of IFN-γ–mediated immune responses in clinical cohorts present a complex picture. Although resting macrophages are unable to kill intracellular pathogens such as MTB, activated macrophages can. IFN-γ is critical to this activation and there are reports that measurable levels of IFN-γ are lower in individuals with active tuberculosis (33, 34) than control subjects, suggesting a protective role for the cytokines. However, high interferon levels are found in many affected individuals as well as healthy purified protein derivative (PPD+) control subjects, and this has led some to conclude that IFN-γ is an unreliable correlate of protection (28). The observation that IFN-γ levels are high at sites of infection hints at the fact that responsiveness to IFN-γ, rather than its production, might be the more important determinant of disease (35). Consistent with this hypothesis is the observation that MTB can influence changes in genes “downstream” of the receptor complex (36) and thus variability “upstream” of the receptor complex might be less important in determining disease outcome.
It has been more formally suggested that IFN-γ activity is a continuous, genetically controlled trait with genetic variability in both the production of, and responsiveness to, IFN-γ (37), although until recently there was little evidence to support a role for variability in IFN-γ production.
The data above demonstrate that both the −1616GG and the +3234TT genotypes of IFNG are associated with pulmonary tuberculosis in West African populations. The size of the effect in each case is small (adjusted OR, 1.49 and 1.30, respectively), but, in contrast to many effects seen in complex genetics, the same trend is seen consistently in the three populations studied, suggesting it is unlikely to be due to chance. This provides the first evidence that genetically determined control of IFNG expression is associated with disease in these West African populations and supports the hypothesis that production of IFN-γ is a genetically controlled trait that influences disease outcome. This knowledge should inform strategies targeting the IFN-γ pathway in vaccine development (both directly and indirectly) and could lead to a re-evaluation of IFN-γ as a therapeutic adjuvant in disease (38).
The significance of these findings stands alone without a detailed understanding of the mechanisms underlying them. However, the +3234 polymorphism is adjacent to an octamer binding transcription factor 1 (OCT-1) binding site and alters a TAAA transcription motif that might be clinically relevant. More likely, the −1616A promoter allele is part of a GATA motif that is lost in the presence of the −1616G allele (GGTA), and this could have been functionally significant. GATA transcription factors are known to play a role in T-cell activity, and appear to play a role in Th1/Th2 differentiation.
The lack of disease association with the +874 polymorphism does not preclude this as a disease-causing allele. It is still possible that the association seen with both the −1616 and +3234 polymorphisms is a result of highly significant LD with the +874AT polymorphism. The ORs observed are small and even a study as large as this might not have sufficient power in these populations to detect an effect due to the +874 variant. This is not unusual; environmental factors such as malnutrition are likely to dilute any genotypic effect seen in these populations and it is well recognized that replication datasets often need to be considerably larger than those originally reporting an association. It will be important for those groups that have reported association with the +874 polymorphism to examine the other SNPs mentioned here. The size of the observed effect does have important implications for replication studies. For a study to have 80% power to detect an effect of this magnitude, even with a common genotype, it is likely to require a sample size of approximately 1,000 cases and control subjects. From our understanding of the role of IFNGR1 in severe mycobacterial infection and the evidence of a role for IFNG itself, IFNGR1 is a strong candidate gene for tuberculosis. However, before now, there was no evidence implicating the IFNGR1 polymorphism in susceptibility to the complex phenotype of tuberculosis. The only published work addressing this question is a small Croatian case-control study, which found marginal disease association with a microsatellite in the noncoding region of the gene (39), a finding that was not replicated in another Gambian study (9).
There are convincing reports of infectious phenotypes that show linkage to the IFNGR1 locus, including severe hepatic fibrosis in Schistosoma mansoni infection (40) and a study of IgG immune responses to Helicobacter pylori infection, where it was suggested that the −56C/T promoter polymorphism might play an important role (20). Weaker evidence of linkage also exists between the IFNGR1 locus and post Kala-azar dermal leishmaniasis (PKDL) (41).
Until recently, evidence for allelic association with disease within this locus has been lacking. However, in a subgroup analysis of the most common ethnic group in the Gambia, Koch and colleagues observed that heterozygosity for the −56C/T polymorphism was associated with protection against severe malaria (21). A recent smaller Gabonese study failed to replicate this finding but did present data suggesting this polymorphism alters gene expression (42), with the −56C allele producing a lower response in a luciferase assay to phytohemagglutination (PHA) stimulation.
There is evidence from an in vitro model of cell expression that constructs bearing the −56C allele produce less transcriptional activity in a standard assay system (42). If these findings can be translated into the clinical setting, we might expect individuals with the −56CC genotype to express less IFNGR1 receptor on the cell surface. It is perhaps surprising that this genotype is associated with protection from pulmonary tuberculosis in West African populations. A reduced immune response mediated by IFNGR1 could protect against pulmonary immunopathology, but given everything else that is known about IFN-γ, this explanation seems unlikely. More plausibly is that either another variant exists in LD that might explain the disease association or that the functional role of this polymorphism has not yet been fully characterized. The in vitro functional data are also not conclusive. Although the −56C allele is associated with reduced expression in a pGL3/basic vector system on stimulation with PHA, the oligonucleotide bearing the −56C allele binds more nuclear extract from B cells. The authors suggest that this might be due to inhibitory binding of AP-4 (42), particularly because the two sites are close together, but equally, the expression vector results might not be an accurate reflection of in vivo behavior.
These data are the first to find evidence of disease association between IFNGR1 and tuberculosis. The data suggest that variation in the IFNGR1 receptor promoter does play a role, albeit a small one, in the pathogenesis of tuberculosis, and previous studies may well have lacked the power needed to see these effects. These findings are consistent with the hypothesis that both production of, and responsiveness to, IFN-γ have genetic determinants. It is likely that the relative importance of each of these factors will differ among populations, dependent on the nature and frequency of variants present, but in these populations, variation within IFNG was more strongly associated with disease.
Although a small effect, the identification of an association with IFNGR1 is an important finding. It suggests that polymorphism within this gene, previously associated with monogenic susceptibility to mycobacterial disease (MSMD, MIM 209950), influences the more common, complex phenotype of tuberculosis. Although the mutation within the gene is highly penetrating, the more frequent polymorphism within the gene does not appear to exert a major effect in these populations. Nonetheless, this is important in our understanding of the IFNG pathway's role in mediating immunity to MTB and suggests one host mechanism whereby individuals' responsiveness to IFN-γ will influence the likelihood of developing tuberculosis.
There are only a few reports of association between the IFNGR2 polymorphism and disease. A study of IgE levels in atopic British whites has found an association with the R64Q variant (23). The same polymorphism is associated with an increased risk of sytemic lupus erythematosus (SLE) in a Japanese population, but only in the presence of a heterozygous Met/Val genotype at codon 14, a variant not found in West Africa (43). No evidence is found here for an association between IFNGR2 variants and tuberculosis. This is perhaps surprising because studies of families with nonfunctional IFNGR2 proteins tell us that the molecule plays a nonredundant role in immunity to nontuberculous mycobacteria. There are several potential explanations. It is possible that the sample size is too small, but there is still considerable power in this study to detect even a modest allele or genotypic effect. It might well be that none of the variants studied has a biological function. The critical ligand binding sites on the extracellular region of IFNGR2 have not yet been identified and the two coding variants might not influence the biological function of the receptor complex. In the absence of specific in vitro data addressing this question, it is impossible to know, but the fact that two of the variants studied change the protein structure, and have been associated with other disease, suggests they might have a functional role in other diseases. Interestingly, although evidence is emerging of cases of tuberculosis in individuals with IFNG and IFNGR1 mutation, none has yet been observed with IFNGR2. These IFNGR2 mutations are rarer, but this raises the possibility that IFNGR2 signaling might be redundant in immunity to MTB but not nontuberculous mycobacteria (NTM), although the data here alone cannot substantiate such a claim.
In summary, the IFNG genotype (+874AA) previously associated with disease was seen in excess in cases compared with control subjects, but not significantly so. Two other variants in strong LD with the +874 variant did show evidence of disease association, confirming a significant role of genetic variation at the IFNG locus and providing a more detailed understanding of the genetic mechanisms underlying the disease association.
The absence of association with IFNGR2 is interesting, but firm conclusions cannot be drawn without further work. The disease association with IFNGR1 is novel, and together these findings support the hypothesis that genetically determined variation in both IFN-γ production and responsiveness influence disease. Furthermore, these findings illustrate some of the limitations of the important models of MSMD (MIM 209950) in improving our understanding of the complex phenotype of tuberculosis.
The authors thank all those who participated in this study. They also thank Oliver Koch for his helpful discussions regarding the study of IFNGR1 and Lisa Bornman for preparation of DNA samples.
| 1. | Corbett EL, Watt CJ, Walker N, Maher D, Williams BG, Raviglione MC, Dye C. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch Intern Med 2003;163:1009–1021. |
| 2. | Ottenhoff TH, Verreck FA, Lichtenauer-Kaligis EG, Hoeve MA, Sanal O, van Dissel JT. Genetics, cytokines and human infectious disease: lessons from weakly pathogenic mycobacteria and salmonellae. Nat Genet 2002;32:97–105. |
| 3. | Lio D, Marino V, Serauto A, Gioia V, Scola L, Crivello A, Forte GI, Colonna-Romano G, Candore G, Caruso C. Genotype frequencies of the +874T→A single nucleotide polymorphism in the first intron of the interferon-gamma gene in a sample of Sicilian patients affected by tuberculosis. Eur J Immunogenet 2002;29:371–374. |
| 4. | Rossouw M, Nel HJ, Cooke GS, van Helden PD, Hoal EG. Association between tuberculosis and a polymorphic NFkappaB binding site in the interferon gamma gene. Lancet 2003;361:1871–1872. |
| 5. | Lopez-Maderuelo D, Arnalich F, Serantes R, Gonzalez A, Codocen R, Madero R, Vazquez JJ, Montiel C. Interferon-gamma and interleukin-10 gene polymorphisms in pulmonary tuberculosis. Am J Respir Crit Care Med 2003;167:970–975. |
| 6. | Kotenko SV, Izotova LS, Pollack BP. Interaction between the components of the interferon gamma receptor complex. J Biol Chem 1995;270:20915–20921. |
| 7. | Greenlund AC, Schreiber RD, Goeddel DV, Pennica D. Interferon-gamma induces receptor dimerization in solution and on cells. J Biol Chem 1993;268:18103–18110. |
| 8. | Awomoyi AA, Nejentsev S, Richardson A, Hull J, Koch O, Podinovskaia M, Todd JA, McAdam KP, Blackwell JM, Kwiatkowski D, et al. No association between interferon-gamma receptor-1 gene polymorphism and pulmonary tuberculosis in a Gambian population sample. Thorax 2004;59:291–294. |
| 9. | Newport MJ, Awomoyi AA, Blackwell JM. Polymorphism in the interferon-gamma receptor-1 gene and susceptibility to pulmonary tuberculosis in the Gambia. Scand J Immunol 2003;58:383–385. |
| 10. | Bennett S, Lienhardt C, Bah-Sow O, Gustafson P, Manneh K, Del Prete G, Gomes V, Newport M, McAdam K, Hill A. Investigation of environmental and host-related risk factors for tuberculosis in Africa: II. Investigation of host genetic factors. Am J Epidemiol 2002;155:1074–1079. |
| 11. | Lienhardt C, Bennett S, Del Prete G, Bah-Sow O, Newport M, Gustafson P, Manneh K, Gomes V, Hill A, McAdam K. Investigation of environmental and host-related risk factors for tuberculosis in Africa: I. Methodological aspects of a combined design. Am J Epidemiol 2002;155:1066–1073. |
| 12. | Bellamy R, Ruwende C, Corrah T, McAdam KP, Whittle HC, Hill AV. Variations in the NRAMP1 gene and susceptibility to tuberculosis in West Africans. N Engl J Med 1998;338:640–644. |
| 13. | Bellamy R, Ruwnede C, Corrah T, McAdam KP, Thursz M, Whittle H, Hill AVS. Tuberculosis and chronic hepatitis B virus infection in Africans and variation in the vitamin D receptor gene. J Infect Dis 1999;179:721–724. |
| 14. | Bellamy R, Ruwende C, Corrah T, McAdam KP, Whittle HC, Hill AV. Assessment of the interleukin 1 gene cluster and other candidate gene polymorphisms in host susceptibility to tuberculosis. Tuber Lung Dis 1998;79:83–89. |
| 15. | Awomoyi AA, Marchant A, Howson JM, McAdam KP, Blackwell JM, Newport MJ. Interleukin-10, polymorphism in SLC11A1 (formerly NRAMP1), and susceptibility to tuberculosis. J Infect Dis 2002;186:1808–1814. |
| 16. | Pravica V, Perrey C, Stevens A, Lee JH, Hutchinson IV. A single nucleotide polymorphism in the first intron of the human IFN-gamma gene: absolute correlation with a polymorphic CA microsatellite marker of high IFN-gamma production. Hum Immunol 2000;61:863–866. |
| 17. | 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;3:165–169. |
| 18. | An P, Vlahov D, Margolick JB, Phair J, O'Brien TR, Lautenberger J, O'Brien SJ, Winkler CA. A tumor necrosis factor-alpha-inducible promoter variant of interferon-gamma accelerates CD4+ T cell depletion in human immunodeficiency virus-1-infected individuals. J Infect Dis 2003;188:228–231. |
| 19. | Henri S, Stefani F, Parzy D, Eboumbou C, Dessein A, Chevillard C. Description of three new polymorphisms in the intronic and 3′UTR regions of the human interferon gamma gene. Genes Immun 2002;3:1–4. |
| 20. | Thye T, Burchard GD, Nilius M, Muller-Myhsok B, Horstmann RD. Genomewide linkage analysis identifies polymorphism in the human interferon-gamma receptor affecting Helicobacter pylori infection. Am J Hum Genet 2003;72:448–453. |
| 21. | Koch O, Awomoyi A, Usen S, Jallow M, Richardson A, Hull J, Pinder M, Newport M, Kwiatkowski D. IFNGR1 gene promoter polymorphisms and susceptibility to cerebral malaria. J Infect Dis 2002;185:1684–1687. |
| 22. | Aoki M, Matsui E, Kaneko H, Inoue R, Fukao T, Watanabe M, Teramoto T, Kato Z, Suzuki K, Suzuki Y, et al. A novel single-nucleotide substitution, Leu 467 Pro, in the interferon-gamma receptor 1 gene associated with allergic diseases. Int J Mol Med 2003;12:185–191. |
| 23. | Gao PS, Mao XQ, Jouanguy E, Pallier A, Doffinger R, Tanaka Y, Nakashima H, Otsuka T, Roberts MH, Enomoto T, et al. Nonpathogenic common variants of IFNGR1 and IFNGR2 in association with total serum IgE levels. Biochem Biophys Res Commun 1999;263:425–429. |
| 24. | Spielman RS, McGinnis RE, Ewens WJ. Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM). Am J Hum Genet 1993;52:506–516. |
| 25. | Clayton D. A generalization of the transmission/disequilibrium test for uncertain-haplotype transmission. Am J Hum Genet 1999;65:1170–1177. |
| 26. | Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES. Parametric and nonparametric linkage analysis: a unified multipoint approach. Am J Hum Genet 1996;58:1347–1363. |
| 27. | Abecasis GR, Cookson WO. GOLD: graphical overview of linkage disequilibrium. Bioinformatics 2000;16:182–183. |
| 28. | Flynn JL, Chan J. Immunology of tuberculosis. Annu Rev Immunol 2001;19:93–129. |
| 29. | Cooper AM, Dalton DK, Stewart TA, Griffin JP, Russell DG, Orme IM. Disseminated tuberculosis in interferon gamma gene-disrupted mice. J Exp Med 1993;178:2243–2247. |
| 30. | Flynn JL, Chan J, Triebold KJ, Dalton DK, Stewart TA, Bloom BR. An essential role for interferon gamma in resistance to Mycobacterium tuberculosis infection. J Exp Med 1993;178:2249–2254. |
| 31. | Moreira AL, Tsenova L, Murray PJ, Freeman S, Bergtold A, Chiriboga L, Kaplan G. Aerosol infection of mice with recombinant BCG secreting murine IFN-gamma partially reconstitutes local protective immunity. Microb Pathog 2000;29:175–185. |
| 32. | Collins HL, Kaufmann SH. The many faces of host responses to tuberculosis. Immunology 2001;103:1–9. |
| 33. | Lin Y, Zhang M, Hofman FM, Gong J, Barnes PF. Absence of a prominent Th2 cytokine response in human tuberculosis. Infect Immun 1996;64:1351–1356. |
| 34. | Zhang M, Lin Y, Iyer DV, Gong J, Abrams JS, Barnes PF. T-cell cytokine responses in human infection with Mycobacterium tuberculosis. Infect Immun 1995;63:3231–3234. |
| 35. | Kincaid EZ, Ernst JD. Mycobacterium tuberculosis exerts gene-selective inhibition of transcriptional responses to IFN-gamma without inhibiting STAT1 function. J Immunol 2003;171:2042–2049. |
| 36. | Ting LM, Kim AC, Cattamanchi A, Ernst JD. Mycobacterium tuberculosis inhibits IFN-gamma transcriptional responses without inhibiting activation of STAT1. J Immunol 1999;163:3898–3906. |
| 37. | Dupuis S, Doffinger R, Picard C, Fleschi C, Altare F, Jouanguy E, Abel L, Casanova JL. Human interferon-gamma-mediated immunity is a genetically controlled continuous trait that determines the outcome of mycobacterial invasion. Immunol Rev 2000;178:129–137. |
| 38. | Condos R, Rom WN, Schluger NW. Treatment of multidrug-resistant pulmonary tuberculosis with interferon-gamma via aerosol. Lancet 1997;349:1513–1515. |
| 39. | Fraser DA, Bulat-Kardum L, Knezevic J, Babarovic P, Matakovic-Mileusnic N, Dellacasagrande J, Matanic D, Pavelic J, Beg-Zec Z, Dembic Z. Interferon-gamma receptor-1 gene polymorphism in tuberculosis patients from Croatia. Scand J Immunol 2003;57:480–484. |
| 40. | Dessein AJ, Hillaire D, Elwali NE, Marquet S, Mohamed-Ali Q, Mirghani A, Henri S, Abdelhameed AA, Saeed OK, Magzouh MM, et al. Severe hepatic fibrosis in Schistosoma mansoni infection is controlled by a major locus that is closely linked to the interferon-gamma receptor gene. Am J Hum Genet 1999;65:709–721. |
| 41. | Mohamed HS, Ibrahim ME, Miller EN, Peacock CS, Khalil EA, Cordell HJ, Howson JM, El Hassan AM, Bercir RE, Blackwell JM. Genetic susceptibility to visceral leishmaniasis in the Sudan: linkage and association with IL4 and IFNGR1. Genes Immun 2003;4:351–355. |
| 42. | Juliger S, Bongartz M, Luty AJ, Kremsner PG, Kun JF. Functional analysis of a promoter variant of the gene encoding the interferon-gamma receptor chain I. Immunogenetics 2003;54:675–680. |
| 43. | Nakashima H, Inoue H, Akahoshi M, Tanaka Y, Yamaoka K, Ogami E, Nagano S, Arinobu Y, Niiro H, Otsaka T, et al. The combination of polymorphisms within interferon-gamma receptor 1 and receptor 2 associated with the risk of systemic lupus erythematosus. FEBS Lett 1999;18:187–190. |
