点击显示 收起
Centenary Institute of Cancer Medicine and Cell Biology, Newtown
Discipline of Medicine, Central Clinical School
Department of Medicine, Nepean Hospital, Western Clinical School, University of Sydney, Sydney, Australia
Background.
Stimulation of the P2X7 purinergic receptor (P2X7) in bacille Calmette-Guérin (BCG)infected human macrophages with extracellular adenosine triphosphate (ATP) leads to pore formation and killing of mycobacteria. We examined the effect of polymorphisms in the P2X7 gene (P2X7) on the capacity of macrophages to kill mycobacteria.
Methods.
Polymorphisms and mutations in P2X7 were identified by both DNA sequence analysis and determination of uptake of ethidium by time-resolved flow cytometry. Macrophages from affected subjects were infected with Mycobacterium bovis BCG. Apoptosis was determined by use of Annexin V staining, and BCG growth was determined by use of quantitative mycobacterial cultures.
Results.
Three new mutations were identified. Macrophages from subjects heterozygous for a polymorphism in P2X7 had a 50% reduction in uptake of ethidium and a 75% reduction in the number of apoptotic cells, compared with macrophages from wild-type (wt) subjects, after stimulation with interferon (IFN) and ATP. Furthermore, after stimulation with IFN- and ATP, there was a reduction in BCG growth of up to 0.5 log10 in macrophages from single-heterozygous subjects, compared with a reduction of 1.0 log10 in macrophages from wt subjects. Interestingly, BCG-infected macrophages from compound-heterozygous subjects, for different combinations of polymorphisms in P2X7, had no uptake of ethidium, failed to undergo apoptosis, and were unable to kill mycobacteria after stimulation with IFN- and ATP.
Conclusions.
Various polymorphisms in P2X7 abrogate IFN-/ATP-induced killing of mycobacteria by human macrophages and, thus, may contribute to variability in susceptibility to mycobacterial infections.
Tuberculosis (TB) remains a major global health problem. The World Health Organization estimates that approximately one-third of the world's population is infected with Mycobacterium tuberculosis, resulting in 2.2 million deaths/year [1]. Disease from M. tuberculosis in humans is most commonly caused by reactivation of dormant infection [2], and the lifetime risk of reactivation for a 25-year-old with latent TB infection is 7.3% [3]. The risk of reactivation may be at least partly explained by genetic factors, since only a minority of cases can be attributed to known causes of acquired immunosuppression, such as HIV, use of corticosteroids, aging, and use of alcohol [4]. Mendelian-inherited mutations in the genes for the interferon (IFN), interleukin (IL)12, and signal transducer and activator of transcription1 pathway confer increased susceptibility to severe mycobacterial infections [5, 6]. These mutations are rare, however, and do not explain the susceptibility to TB in the wider population. Such susceptibility is more likely to be attributed to polymorphisms, which, by definition, are genetic variations that occur in >1% of the population and are therefore deemed to be more common than mutations. Polymorphisms in HLA type and in the genes for vitamin D3 receptor, NRAMP1, IFN- promoter, and Toll-like receptor2 have all been associated with increased susceptibility to M. tuberculosis, although, individually, the attributable risk of these polymorphisms is modest [711].
Macrophages play a central role during mycobacterial infection. They are the principal host cells for intracellular replication of mycobacteria, act as antigen-presenting cells during activation of lymphocytes at the sites of infection, and are responsible for killing of mycobacteria through a number of mechanisms, including generation of reactive nitrogen intermediates (RNIs) and reactive oxygen intermediates (ROIs) and, importantly, the promotion of phagolysosomal fusion [12]. We and others have shown that extracellular ATP, through the activation of the P2X7 purinergic receptor (P2X7), induces both macrophage apoptosis and killing of intracellular mycobacteria in infected human macrophages [1315]. P2X7 is highly expressed on human macrophages and is further up-regulated by IFN- [1618]. P2X7 receptors are ligand-gated cation channels with 2 transmembrane domains and a trimeric structure in the plasma membrane [1921]. Activation of P2X7 causes an immediate opening of a cation-selective channel, which undergoes dilatation within seconds to allow entry of larger cations up to the size of ethidium and Yo-Pro [19, 20, 22]. Activation of P2X7 in macrophages stimulates a number of signalling pathways, including the caspase cascade, with resultant apoptosis, and activation of phospholipase D (PLD) [2325]. PLD promotes phagolysosomal fusion, thereby exposing mycobacteria to destructive lysosomal enzymes [26, 27].
We have identified 3 single-nucleotide polymorphisms (SNPs) within the coding region of the P2X7 gene (P2X7), each of which leads to loss of receptor function (figure 1). The most common is the 1513AC polymorphism [17], and macrophages from subjects homozygous for this polymorphism are unable to kill mycobacteria after ATP treatment of infected macrophages [15]. Homozygosity for 1513C occurs in 1%2% of the white population, whereas heterozygosity is more common, with a frequency of 15%20% in the white population [17]. The 1729TA and 946GA polymorphisms in the heterozygote state have prevalences of 2% and 1.4%, respectively [28, 29]. We describe here 3 new mutations and demonstrate that subjects with a single loss-of-function polymorphism in P2X7 have reduced ATP-induced apoptosis and killing of mycobacteria. Furthermore, in macrophages from compound-heterozygous subjects, there is complete abolition of these ATP-induced events, with a phenotype equivalent to the 1513AC homozygous subjects.
SUBJECTS AND METHODS
Subject recruitment and genotyping.
A total of 131 adult subjects from the Centenary Institute of Cancer and Cell Biology, Sydney, and the Clinical School and Hematology Department, Nepean Hospital, Sydney, were screened for P2X7 function. Those with low-level function were subsequently genotyped for loss-of-function polymorphisms by use of primers, as described elsewhere [17, 28, 29]. Blood was collected from 18 subjects with loss-of-function polymorphisms and from 10 wild-type (wt) subjects (i.e., subjects with no detectable loss-of-function polymorphisms). Subjects provided informed consent, and the study received ethics approval from the Central Sydney Area Health Service and the University of Sydney.
Human monocyte-derived macrophage cultures.
Peripheral blood was used to derive fresh monocytes. Peripheral blood mononuclear cells were resuspended at 1 × 1062 × 106 cells/mL in RPMI 1640 (Sigma-Aldrich) containing 10% heat-inactivated fetal calf serum (Life Technologies) and 2 mmol/L L-glutamine (Sigma-Aldrich). Cells were incubated for 2 h in Nunc tissue-culture flasks (Nunc A/S) and washed twice to remove nonadherent cells. The adherent monocytes were cultured for 6 days in complete medium. For measurements of ATP-induced uptake of ethidium and expression of P2X7, macrophages that had been differentiated for 6 days were cultured overnight with IFN- (100 IU/mL; Roche) and collected by mechanical scraping. For assays of killing of mycobacteria and apoptosis, macrophages that had been differentiated for 6 days were detached and plated overnight in 96-well plates, at 1 × 105 cells/well, in antibiotic-free medium with IFN- (100 IU/mL), before mycobacterial infection.
Measurement of ethidium influx by flow cytometry.
ATP-induced uptake of ethidium into IFN-activated macrophages resuspended in HEPES-buffered KCl medium (10 mmol/L HEPES, 150 mmol/L KCl, 5 mmol/L D-glucose, and 1% bovine serum albumin [pH 7.5]) and ethidium (25 mol/L; Sigma-Aldrich) at 37°C was determined as described elsewhere [28].
Immunofluorescent staining and flow cytometry.
Cells were stained for expression of P2X7 and major histocompatibility complex class II and analyzed by flow cytometry, as described elsewhere [15].
Killing of mycobacteria and apoptosis.
IFN-activated macrophages were infected with BCGgreen fluorescent protein [30] for 4 h at an MOI of 1 and then were washed twice to remove extracellular bacteria (day 0). On day 2, cells were pulsed with 3 mmol/L ATP for 20 min, washed, and incubated overnight. On day 3, one-half of the wells were lysed with 0.1% Triton-X for 30 min, to release viable bacilli. Serial dilutions of cell lysates were plated onto 7H11 agar and incubated for 34 weeks at 37°C, to determine the load of viable mycobacteria. The remaining wells were stained with phycoerythrin-conjugated Annexin V and propidium iodide, to measure apoptosis, in accordance with the manufacturer's protocol (BD Biosciences), and were analyzed by use of a FACSCalibur flow cytometer (Becton Dickinson).
Tumor necrosis factor (TNF) assay.
Supernatants from triplicate cultures of macrophages that had been differentiated for 2 days were collected, and TNF was measured by ELISA, in accordance with the manufacturer's protocol (BD Biosciences).
Statistics.
Differences in data on uptake of ethidium, expression of P2X7, killing of mycobacteria, and apoptosis were analyzed by use of the Mann-Whitney U test. The differences were considered to be significant when P < .05.
RESULTS
Identification of new P2X7 mutations.
We studied 12 subjects heterozygous for known loss-of-function SNPs (1513AC, n = 7; 1729TA, n = 1; 946GA, n = 4). Three compound-heterozygous subjects with the 946GA/1513AC, 946GA/1729TA, or 1513AC/1729TA SNP were identified. In addition, 3 other subjects were found to have mutations not previously reported. The first had an intronic mutation (151+1GT) at the 5 donor splice site of the exon 1intron 1 boundary; this subject was also heterozygous for the 1513AC SNP. The second subject had an exonic mutation (1747GA) that predicts for an amino acid change (R574H) in the trafficking domain [31]; this subject was also heterozygous for the 1513AC SNP. The third subject had an exonic mutation (699CT) that results in a stop codon. In total, 13 single-heterozygous and 5 compound-heterozygous subjects were identified, and the functional activity of P2X7 in macrophages was compared with that in macrophages from 10 randomly selected wt subjects.
Normal expression of HLA class II and TNF production in heterozygous infected macrophages.
The activation status of macrophages during BCG infection was determined by the level of expression of HLA class II and by TNF production. Polymorphisms leading to loss of P2X7 function did not affect the expression of HLA class II moleculesa similar level of expression of HLA-DR on IFN-treated macrophages was observed in all 3 subjects (table 1). Furthermore, BCG infection increased TNF production by all macrophages, regardless of P2X7 phenotype. Although the level of TNF produced varied among subjects, no significant difference was seen in TNF production between macrophages from wt, single-heterozygous, and compound-heterozygous groups (data not shown).
Reduced apoptosis in ATP-pulsed heterozygous macrophages.
ATP-induced apoptosis, but not H2O2-induced necrosis, of mycobacteria-infected macrophages leads to the death of intracellular bacilli [32]. We investigated whether the loss-of-function polymorphisms in P2X7 affected the ability of macrophages to kill mycobacteria. Adherent monocyte-derived macrophages were infected with Mycobacterium bovis BCG for 48 h and were then pulsed with ATP for 20 min. After removal of ATP and overnight incubation, the percentage of apoptotic cells was determined. In macrophages from wt subjects, stimulation with ATP led to an increase in apoptosis, compared with that in unstimulated macrophages (range, 10.1%32.4%; median, 17.87%) (figure 3). Macrophages from single-heterozygous subjects had a significant reduction in apoptosis (range, 2.2%9.8%; median, 4.70%), compared with those from wt subjects (P < .001), and the level of apoptosis was even further reduced in macrophages from compound-heterozygous subjects (range, 0%6.4%; median, 1.24%) (P < .001).
Reduced ATP-mediated killing of mycobacteria in heterozygous macrophages.
We have previously shown that treatment of infected wt macrophages with ATP was associated with a reduction in mycobacterial load of up to 90% [15]. Macrophages from single-heterozygous subjects had a significant impairment in killing of mycobacteria, with a 56% reduction in bacterial viability after stimulation with ATP (P < .001) (figure 4). Stimulation with ATP of macrophages from compound-heterozygous subjects did not induce any reduction in bacterial viability, compared with that in unstimulated macrophages (P < .001, for macrophages from compound-heterozygous subjects vs. those from wt subjects and for macrophages from compound-heterozygous subjects vs. those from single-heterozygous subjects).
DISCUSSION
The present study has provided strong evidence that a variety of genetic polymorphisms in P2X7 impair the ability of macrophages to kill mycobacteria. Macrophages from subjects homozygous for the 1513AC polymorphism in P2X7 fail to exhibit ATP-induced killing of BCG [15]. The homozygous state is rare, but we have identified individuals who are heterozygous for P2X7 loss-of-function polymorphisms or mutations, and these individuals also showed a partial impairment of ATP-mediated killing of mycobacteria. Macrophages from all but 2 subjects with 1 demonstrable loss-of-function SNP in P2X7 had some, albeit reduced, ATP-induced apoptosis and ATP-mediated killing of BCG. Macrophages from compound-heterozygous subjects, however, were completely unable to kill mycobacteria by ATP and were insensitive to the apoptosis-inducing effect of ATP, regardless of the combination of polymorphisms involved. Thus, the effect of compound heterozygosity is additive and gives rise to a failure of P2X7 function identical to that observed in 1513AC homozygosity [15].
P2X7-mediated killing of BCG involves a transient increase in cytosolic calcium, resulting in the activation of phospholipase D, phospholipase A2, and mitogen-activated protein kinase and the subsequent promotion of phagolysosomal fusion [14, 33]. ATP-mediated killing of mycobacteria by macrophages is also independent of ROI, RNI, NRAMP1, Fas ligation, and complement-mediated lysis [13]. Infection of macrophages with M. tuberculosis results in increased levels of ATP within the extracellular medium [34]. The direct stimulation of macrophages with lipopolysaccharide or endotoxin also triggers release of ATP [35, 36]. The concentration of ATP released into the immunological synapse between infected macrophages and T cells within a granuloma has not been determined but may be sufficient to stimulate killing of mycobacteria by human macrophages in vivo.
There is emerging evidence that certain genetic polymorphisms provide an increased risk of TB. We have identified several SNPs in different regions of P2X7. These SNPs lie within domains of the gene that affect protein anchoring (1513AC) [17], receptor trafficking (1729TA) [28], and ATP binding (946GA) [29]. The newly identified 1747GA mutation (R574H), like the 1729TA SNP, also lies within a previously identified trafficking domain [31]. The other 2 newly identified mutations (155+1GT and 699CT) abolish mRNA synthesis and prevent P2X7 protein synthesis. One common feature of these loss-of-function polymorphisms is that they lead to reduced expression of receptors on the macrophage surface, which may limit the response of P2X7 to ATP. Furthermore, this reduction appears to be additivemacrophages from compound-heterozygous subjects have almost no expression of P2X7and this may explain the ablation of ATP responsiveness in macrophages from compound-heterozygous subjects. Two subjects with a single identified loss-of-function polymorphism had a phenotype identical to that of compound-heterozygous subjects. This suggests that the effect of their loss-of-function polymorphism is completely ablative or that they may have another, as yet unidentified loss-of-function polymorphism or mutation.
The overall impact of SNPs on susceptibility to TB in a population depends on the gene frequency for that allele in the population studied. The cumulative frequency of these P2X7 polymorphisms may be sufficiently high to influence susceptibility to clinical TB in a particular population. To date, the identified genetic associations with TB have all been modest in effect [711]. Genetic susceptibility to TB is, however, likely to be polygenic in nature, in which the balance between the effects of susceptibility and protective genes may act in conjunction with environmental factors to determine the risk of active disease. A study of NRAMP polymorphisms in a Gambian population showed that the presence of 2 susceptibility alleles at this locus conveyed an additive effect on the risk of disease [8]. The effect of having >1 nonfunctional P2X7 SNP may not only be additive in reducing ATP-induced intracellular killing of mycobacteria but may also confer an additive risk of development of clinical TB after infection.
Another study examined the association of 5 SNPs in the putative promoter region upstream of exon 1 of P2X7, along with the 1513AC SNP (which we previously identified [17]), with the development of pulmonary TB in Gambians. A significant protective effect against TB was found for 1 SNP in the promoter region but not for the 1513AC SNP in that population. The importance of individual SNPs within genes may vary markedly between different racial groups. For example, an additive risk of pulmonary TB was identified for SNPs in NRAMP1 in a Gambian population [8], but no such susceptibility was found in a linkage study in Brazil [37]. Furthermore, polymorphisms in the gene for mannose-binding lectin were associated with protection against TB in South African [38] and Gambian [39] populations but were associated with increased susceptibility to TB in a southern Indian population [40]. The present study has demonstrated that genetic variation in P2X7 is associated with significant differences in the capacity of human macrophages to kill mycobacteria. Further study of the association of these P2X7 polymorphisms and the development of clinical TB in different populations is warranted.
Acknowledgments
We thank Nathan Field, for his technical assistance, and Dr. Stephen Fuller, for collection of blood from some subjects. We also thank all the subjects who donated blood for this study.
References
1. Kaufmann SH. How can immunology contribute to the control of tuberculosis Nat Rev Immunol 2001; 1:2030. First citation in article
2. Horsburgh CR Jr. Priorities for the treatment of latent tuberculosis infection in the United States. N Engl J Med 2004; 350:20607. First citation in article
3. Marks GB, Bai J, Simpson SE, Sullivan EA, Stewart GJ. Incidence of tuberculosis among a cohort of tuberculin-positive refugees in Australia: reappraising the estimates of risk. Am J Respir Crit Care Med 2000; 162:18514. First citation in article
4. Flynn JL, Chan J. Immunology of tuberculosis. Annu Rev Immunol 2001; 19:93129. First citation in article
5. Doffinger R, Dupuis S, Picard C, et al. Inherited disorders of IL-12- and IFNgamma-mediated immunity: a molecular genetics update. Mol Immunol 2002; 38:9039. First citation in article
6. Dupuis S, Dargemont C, Fieschi C, et al. Impairment of mycobacterial but not viral immunity by a germline human STAT1 mutation. Science 2001; 293:3003. First citation in article
7. Ogus AC, Yoldas B, Ozdemir T, et al. The Arg753Gln polymorphism of the human Toll-like receptor 2 gene in tuberculosis disease. Eur Respir J 2004; 23:21923. First citation in article
8. 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; 338:6404. First citation in article
9. Bellamy R, Ruwende C, Corrah T, et al. Tuberculosis and chronic hepatitis B virus infection in Africans and variation in the vitamin D receptor gene. J Infect Dis 1999; 179:7214. First citation in article
10. Cooke GS, Hill AV. Genetics of susceptibility to human infectious disease. Nat Rev Genet 2001; 2:96777. First citation in article
11. 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; 361:18712. First citation in article
12. Schaible UE, Collins HL, Kaufmann SH. Confrontation between intracellular bacteria and the immune system. Adv Immunol 1999; 71:267377. First citation in article
13. Lammas DA, Stober C, Harvey CJ, Kendrick N, Panchalingam S, Kumararatne DS. ATP-induced killing of mycobacteria by human macrophages is mediated by purinergic P2Z(P2X7) receptors. Immunity 1997; 7:43344. First citation in article
14. Kusner DJ, Adams J. ATP-induced killing of virulent Mycobacterium tuberculosis within human macrophages requires phospholipase D. J Immunol 2000; 164:37988. First citation in article
15. Saunders BM, Fernando SL, Sluyter R, Britton WJ, Wiley JS. A loss-of-function polymorphism in the human P2X7 receptor abolishes ATP-mediated killing of mycobacteria. J Immunol 2003; 171:54426. First citation in article
16. Hickman SE, el Khoury J, Greenberg S, Schieren I, Silverstein SC. P2Z adenosine triphosphate receptor activity in cultured human monocyte-derived macrophages. Blood 1994; 84:24526. First citation in article
17. Gu BJ, Zhang W, Worthington RA, et al. A Glu-496 to Ala polymorphism leads to loss of function of the human P2X7 receptor. J Biol Chem 2001; 276:1113542. First citation in article
18. Falzoni S, Munerati M, Ferrari D, Spisani S, Moretti S, Di Virgilio F. The purinergic P2Z receptor of human macrophage cells: characterization and possible physiological role. J Clin Invest 1995; 95:120716. First citation in article
19. Surprenant A, Rassendren F, Kawashima E, North RA, Buell G. The cytolytic P2Z receptor for extracellular ATP identified as a P2X receptor (P2X7). Science 1996; 272:7358. First citation in article
20. Rassendren F, Buell GN, Virginio C, Collo G, North RA, Surprenant A. The permeabilizing ATP receptor, P2X7: cloning and expression of a human cDNA. J Biol Chem 1997; 272:54826. First citation in article
21. Nicke A, Baumert HG, Rettinger J, et al. P2X1 and P2X3 receptors form stable trimers: a novel structural motif of ligand-gated ion channels. Embo J 1998; 17:301628. First citation in article
22. Wiley JS, Gargett CE, Zhang W, Snook MB, Jamieson GP. Partial agonists and antagonists reveal a second permeability state of human lymphocyte P2Z/P2X7 channel. Am J Physiol 1998; 275:C122431. First citation in article
23. Ferrari D, Los M, Bauer MK, Vandenabeele P, Wesselborg S, Schulze-Osthoff K. P2Z purinoreceptor ligation induces activation of caspases with distinct roles in apoptotic and necrotic alterations of cell death. FEBS Lett 1999; 447:715. First citation in article
24. Humphreys BD, Rice J, Kertesy SB, Dubyak GR. Stress-activated protein kinase/JNK activation and apoptotic induction by the macrophage P2X7 nucleotide receptor. J Biol Chem 2000; 275:267928. First citation in article
25. el-Moatassim C, Dubyak GR. A novel pathway for the activation of phospholipase D by P2z purinergic receptors in BAC1.2F5 macrophages. J Biol Chem 1992; 267:2366473. First citation in article
26. Kusner DJ, Barton JA. ATP stimulates human macrophages to kill intracellular virulent Mycobacterium tuberculosis via calcium-dependent phagosome-lysosome fusion. J Immunol 2001; 167:330815. First citation in article
27. Fairbairn IP, Stober CB, Kumararatne DS, Lammas DA. ATP-mediated killing of intracellular mycobacteria by macrophages is a P2X7-dependent process inducing bacterial death by phagosome-lysosome fusion. J Immunol 2001; 167:33007. First citation in article
28. Wiley JS, Dao-Ung LP, Li C, et al. An Ile-568 to Asn polymorphism prevents normal trafficking and function of the human P2X7 receptor. J Biol Chem 2003; 278:1710813. First citation in article
29. Gu BJ, Sluyter R, Skarratt KK, et al. An Arg307 to Gln polymorphism within the ATP-binding site causes loss of function of the human P2X7 receptor. J Biol Chem 2004; 279:3128795. First citation in article
30. Triccas JA, Berthet FX, Pelicic V, Gicquel B. Use of fluorescence induction and sucrose counterselection to identify Mycobacterium tuberculosis genes expressed within host cells. Microbiology 1999; 145:292330. First citation in article
31. Smart ML, Gu B, Panchal RG, et al. P2X7 receptor cell surface expression and cytolytic pore formation are regulated by a distal C-terminal region. J Biol Chem 2003; 278:885360. First citation in article
32. Molloy A, Laochumroonvorapong P, Kaplan G. Apoptosis, but not necrosis, of infected monocytes is coupled with killing of intracellular bacillus Calmette-Guerin. J Exp Med 1994; 180:1499509. First citation in article
33. Stober CB, Lammas DA, Li CM, Kumararatne DS, Lightman SL, McArdle CA. ATP-mediated killing of Mycobacterium bovis bacille Calmette-Guerin within human macrophages is calcium dependent and associated with the acidification of mycobacteria-containing phagosomes. J Immunol 2001; 166:627686. First citation in article
34. Sikora A, Liu J, Brosnan C, Buell G, Chessel I, Bloom BR. Cutting edge: purinergic signaling regulates radical-mediated bacterial killing mechanisms in macrophages through a P2X7-independent mechanism. J Immunol 1999; 163:55861. First citation in article
35. Ferrari D, Chiozzi P, Falzoni S, Hanau S, Di Virgilio F. Purinergic modulation of interleukin-1 beta release from microglial cells stimulated with bacterial endotoxin. J Exp Med 1997; 185:57982. First citation in article
36. Sperlagh B, Hasko G, Nemeth Z, Vizi ES. ATP released by LPS increases nitric oxide production in raw 264.7 macrophage cell line via P2Z/P2X7 receptors. Neurochem Int 1998; 33:20915. First citation in article
37. Shaw MA, Collins A, Peacock CS, et al. Evidence that genetic susceptibility to Mycobacterium tuberculosis in a Brazilian population is under oligogenic control: linkage study of the candidate genes NRAMP1 and TNFA. Tuber Lung Dis 1997; 78:3545. First citation in article
38. Hoal-Van Helden EG, Epstein J, Victor TC, et al. Mannose-binding protein B allele confers protection against tuberculous meningitis. Pediatr Res 1999; 45:45964. First citation in article
39. Bellamy R, Ruwende C, McAdam KP, et al. Mannose binding protein deficiency is not associated with malaria, hepatitis B carriage nor tuberculosis in Africans. Qjm 1998; 91:138. First citation in article
40. Selvaraj P, Narayanan PR, Reetha AM. Association of functional mutant homozygotes of the mannose binding protein gene with susceptibility to pulmonary tuberculosis in India. Tuber Lung Dis 1999; 79:2217. First citation in article