The emergence of HIV and its synergistic association with tuberculosis (TB) pose a great challenge to the health systems in developing countries. TB is the most life-threatening opportunistic infection among HIV patients.
Variability in the clinical outcomes of individuals coinfected with HIV-1 and TB is determined by multiple factors, like host genetic variations. Development of TB in HIV+ patients is influenced by HLA and other genetic variants that play an important role in innate and adaptive immunity.1 Within host innate immune strategies, many studies have now unveiled the crucial role of inflammasomes in different microbial infections. Inflammasomes are molecular platforms that are assembled by hetero-oligomerization of an innate immune receptor, an adaptor protein ASC, and the effector enzyme caspase-1, which leads to the cleavage of pro-interleukin (IL)-1ß into biologically active IL-1ß. Each kind of receptor (NLRP1, NLRP3, NLRC4, and AIM2) could recognize different pathogen-associated molecular patterns inducing the assembling of inflammasome, the activation of caspase-1, and finally the secretion of the proinflammatory cytokine IL-1ß. Several proteins have been described as negative regulators of inflammasome and IL-1ß secretion, such as CARD8, HSP90, and IL-1 receptor.2,3 Dysregulation of inflammasome has been associated with susceptibility to microbial infection, as reported for knockout animal models2,3 and in human association studies.4 Critical role for NLRP3 inflammasome has also been described during Mycobacterium spp. infection in mice,5–8 whereas, nowadays, limited knowledge is available about the effect of inflammasome dysregulation on TB in humans.
Recently, our research group demonstrated that variants in NLRP3 inflammasome contribute to HIV-1 susceptibility.9,10 The present study investigates the possible association between selected variants in inflammasome genes and HIV-1 and Mycobacterium tuberculosis (HIV+TB+) coinfection in a case/control cohort of Brazilian individuals.
PATIENTS AND METHODS
Patients and Controls
Ninety-six HIV-1 and TB coinfected (HIV+TB+, 62 men/39 women, mean age 37.61 years ± SD 10.65), 96 TB-positive (TB+, 60 men/36 women, mean age 27.79 years ± SD 20.18), and 192 HIV-1–positive Brazilian adults (HIV+, 54 men/138 women, mean age 36.39 years ± SD 9.06) were enrolled at the immunological ambulatory of Correia Picanço and Instituto de Medicina Integrada Prof. Fernando Figueira hospitals. The diagnosis of TB was based on clinical symptoms and radiographic findings, along with bacteriological confirmation (culture, smear, and/or polymerase chain reaction) as described by the American Thoracic Society.11
One hundred fifty healthy controls (HC, 27 men/123 women, 33.47 years ± 13.4) were also recruited from the same metropolitan area (Recife).
HIV+ individuals and healthy donors were TB negative (negative Mantoux test, showing no symptoms of TB or previous history of the disease).
Individuals were classified as European or African derived according to phenotypic characteristics and ethnicity data of parents/grandparents reported by the participants in an appropriate questionnaire.12,13 Sixty-three HC (42%), 27 HIV+TB+ (28%), 27 TB+ (28%), and 44 HIV+ (23%) were classified as European derived, whereas 87 HC (58%), 69 HIV+TB+ (72%), 69 TB+ (72%), and 148 HIV+ (77%) as African derived. Characteristics of the patients and controls are listed in the Supplemental Digital Content (see Table S1, http://links.lww.com/QAI/A406). Written informed consent was obtained according to the protocol of Health Sciences Center Ethical Committee of Federal University of Pernambuco (CEP/CCS/UFPE no. 173/11) (Recife, Brazil).
Genomic DNA was extracted from peripheral whole blood using the Qiagen genomic DNA purification kit (Qiagen, Sao Paolo, Brazil).
Single-Nucleotide Polymoprhims Selection and Genotyping
We selected 19 single-nucleotide polymoprhims (SNPs) in 8 inflammasome genes (NLRP1, NLRP3, AIM2, CARD8, CASP1, IL1B, IL1R, and HSP90) from public databases Hapmap (www.hapmap.org) and GeneBrowser (www.genome.ucsc.edu). SNPs are listed in the Supplemental Digital Content (see Figure S1, http://links.lww.com/QAI/A406). Fourteen SNPs were chosen based on previously published data10,14,15; 3 other SNPs in NLRP3 were included (rs4925659, rs12239046, and rs10754555) considering novel association results in psoriasis and chronic inflammation.16,17 Functional polymorphisms in IL1R (rs3917254) and HSP90 (rs11621560) genes were also included because of their possible impact in downregulating inflammasome activation.2,3
SNP genotyping was performed using commercially available TaqMan assays (Applied Biosystems/AB and Life Technologies, Sao Paolo, Brazil) using ABI7500 Real-Time platform (AB). Allelic discrimination was performed using the SDS v1.4 software (AB).
Allelic and genotypic SNP frequencies were calculated using Genotype Transposer software.18 The Haploview software19 was used to investigate the association and linkage disequilibrium (LD) pattern and for deriving the haplotypes. R software (www.r-project.org) was used to perform Fisher exact test and odds ratio (OR) calculation for alleles and haplotypes and for genotype association and modeling (package SNP assoc version 1.5-2). A formal Bonferroni correction for the number of SNPs analyzed would require a significance threshold of P = 0.003 (P0/N, P0 = 0.05, N = 19 tests). The post hoc statistical power analysis was performed with the G*power software (version 3.0.5), with an alpha-error probability of 0.003.
Ninety-six HIV+TB+ patients and 192 HIV+ individuals were genotyped for 19 SNPs in 8 inflammasome genes (NLRP1, NLRP3, AIM2, CARD8, CASP1, IL1B, ILR1, and HSP90). SNP allelic and genotypic frequencies were in Hardy–Weinberg equilibrium in both groups. Four variations (rs10754558, rs2043211, rs6509365, and rs1143634) were differently distributed between HIV+TB+ and HIV+, even if only rs6509365 in CARD8 resulted to be significantly associated with coinfection after Bonferroni correction. Table 1 shows genotypes' frequencies of NLRP3 rs10754558, CARD8 rs2043211 and rs6509365, and IL1B rs1143634 SNPs.
Carriers of CARD8 rs6509365 minor G allele were significantly more frequent in HIV+TB+ than in HIV+ patients (P = 5 × 10−5), suggesting a predisposing role of this variant for susceptibility to M. tuberculosis infection in HIV+ subjects [OR = 2.45, 95% confidence interval (CI): 1.56 to 3.84] (Table 2). According to the Akaike information criterion, the rs6509365 G allele behaved according to a dominant model.
LD analysis was then performed. A significant LD was found only for NLRP3 SNPs rs12239046 and rs4925659 (D′ = 1, r2 = 0.39) (see Figure S2, Supplemental Digital Content, http://links.lww.com/QAI/A406). The resulting haplotypes were differently distributed within HIV+TB+ and HIV+ groups, whereas not in a statistically significant way after Bonferroni correction (P = 0.02) (see Figure S2, Supplemental Digital Content, http://links.lww.com/QAI/A406).
The combined effect of polymorphisms within the same genes has been evaluated, and significant results are reported in Table 2. CARD8 SNPs rs2043211 and rs6509235 formed 4 allelic combinations (A-A, T-A, T-G, and A-G). In particular, the combination A-G showed a predisposing effect for coinfection (P = 9.0 × 10−4, OR = 2.60, 95% CI: 1.43 to 4.78), whereas the A-A a protective one (P = 1.5 × 10−4, OR = 0.49, 95% CI: 0.34 to 0.72). IL1B SNP combination resulted in 6 haplotypes with frequency greater than 0.05 in HIV+TB+ and HIV+ subjects. The combined haplotype rs1143643 A–rs1143634 G–rs1143629 G showed a significant higher frequency in HIV+TB+ when compared with HIV+ (0.10 versus 0.03, P = 0.002, OR = 3.4, 95% CI: 1.50 to 8.23), even if this difference did not reach the statistical threshold after Bonferroni correction.
The frequency distribution of the 19 SNPs was evaluated in our groups stratified for European or African ethnic origin.
In subjects of European origin, the rs243211 SNP has been excluded from analysis because it was not in Hardy–Weinberg equilibrium. Even if the frequency of the 2 polymorphisms rs6509365 and rs1143634 varied in HIV+TB+ and HIV+ groups, none of the SNPs resulted to be significantly associated with the coinfection after Bonferroni correction (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A406).
In African-derived group, the rs6509365 G allele was significantly associated with the susceptibility to HIV+TB+ coinfection (P = 9 × 10−5, OR = 3.46, 95% CI: 1.83 to 6.52, corrected by sex), behaving according to a dominant model. The rs1143634 SNP resulted to be associated with the coinfection according to an overdominant model, even if the P value (P = 0.010) did not reach the significant threshold after Bonferroni correction (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A406).
Considering that the variants associated with HIV+TB+ coinfection could be related to single infection, we genotyped and compared TB+ subjects and HC.
The rs10754558 polymorphism was excluded from the analysis because it was not in Hardy–Weinberg equilibrium in TB+ group (P = 0.006).
The CARD8 rs6509365 variation was more frequent in TB+ patients with respect to controls (P = 0.010), even if not in a statistically significant way after Bonferroni correction. Distribution of the other SNPs did not significantly differ between TB+ and HC groups (see Table S3, Supplemental Digital Content, http://links.lww.com/QAI/A406).
The LD analysis showed a significant result for CARD8 SNPs, rs2043211 and rs6509235 (D′ = 0.96, r2 = 0.72), and NLRP3 SNPs, rs4925659 and rs10754558 (D′ = 0.95, r2 = 0.30). The resulting haplotypes were differently distributed in TB+ and HC groups, whereas not in a statistically significant way after Bonferroni correction (see Figure S3, Supplemental Digital Content, http://links.lww.com/QAI/A406).
When the analysis was performed in European-derived TB+ case/control cohorts, the frequency of the SNPs did not vary (see Figure S3, Supplemental Digital Content, http://links.lww.com/QAI/A406). Whereas when considering African-derived individuals, NLRP1 rs3473379 and CARD8 rs2043211 and rs6509365 were differently distributed, even if not in a significant way after Bonferroni correction (see Table S3, Supplemental Digital Content, http://links.lww.com/QAI/A406).
When we considered TB form (pulmonary or extrapulmonary), both in HIV+TB+ and TB+ groups, the rs6509235 variant seemed to be less frequent in pulmonary form than in extrapulmonary manifestation in HIV+TB+ (P = 0.004, OR = 0.13, 95% CI: 0.03 to 0.59) and also in TB+ groups (P = 0.009, OR = 0.25, 95% CI: 0.08 to 0.76), but the P value did not reach the significant threshold after Bonferroni correction (see Table S4, Supplemental Digital Content, http://links.lww.com/QAI/A406).
Susceptibility to HIV/AIDS and TB development in HIV patients is multifactorial. Innate immunity and inflammation have been proposed as important factors in the susceptibility and development of HIV/AIDS and TB.
Inflammasomes are known to be involved in recognizing several pathogens and in triggering the consequent innate immune response.2,3 Recently, we hypothesized a role of NLRP3 inflammasome and IL-1ß in HIV pathogenesis.9,10 NLRP3 inflammasome has also been associated with Mycobacterium spp infection in animal model.5–8 Polymorphisms in purinergic receptor P2X7, the cationic channel activated by extracellular ATP implied in inflammasome assembling, have been reported to be associated with extrapulmonary TB.20–22
To our knowledge, this is the first time that inflammasome polymorphisms have been evaluated in association with HIV+TB+ coinfection.
Our results suggest the novel association between CARD8 gene and HIV+TB+ coinfection. In particular, HIV+ patients carrying rs6509365 minor G allele are more prone to coinfection, and this effect is even stronger when this allele is combined to CARD8 rs2043211 major allele A.
The role of CARD8 in the biology of inflammasome is far to be fully characterized. CARD8 inhibits caspase-123 and nuclear factor-κB pathway24 and regulates apoptosis.25 The loss-of-function polymorphism rs2043211 (C10X) and the consequent inappropriate strong caspase-1 response have been associated with systemic inflammatory response syndromes and autoimmunity.26 In a genome-wide study, Ko et al27 showed that the 2 CARD8 polymorphisms, rs2043211 and rs6509365, found in strong LD, were associated with Salmonella-induced cell death. The rs6509365 SNP is an intronic variant, and its functional effect is still unknown, even if it has been postulated that it could reduce CARD8 gene expression.28
According to this hypothesis, HIV+ patients, carrying rs6509365 G allele, would have a higher caspase-1 activation and cell death in immune cells and be more susceptible to develop opportunistic infection, such as TB. On the other side, the augmented activation of inflammasome could overcome the inhibition of inflammasome proteins mediated by M. tuberculosis,28 increasing the spread of bacteria throughout the body.
Whether this variation could affect the response against M. tuberculosis in healthy subjects is not fully demonstrated because in our data, the difference in rs6509365 distribution between TB+ and controls did not reach the statistical significance.
Moreover, we did not find an association between this variant and susceptibility to HIV-1 infection (data not shown); so, we could hypothesize that rs6509365 affects specifically TB development, especially in immune-compromised individuals such as HIV-1–infected patients.
Similar to the data reported for P2X7 polymorphisms,20–22 this variant rs6509365 seemed to be more associated with extrapulmonary TB.
Even if this study is a preliminary report and further investigations are needed to elucidate the role of CARD8 and inflammasome in HIV+TB+ coinfection, our results suggest that inflammasome genetics could influence HIV-1 infection and the development of opportunistic infection. We are aware that a weak point of the study is the limited size of our groups; however, a reasonable statistical power (0.70) lets us assume that our results are significant.
Finally, because African-derived patients showed a stronger association with rs6509365, a replica on other ethnic groups may disclose a population effect of this variation.
1. Raghavan S, Alagarasu K, Selvaraj P. Immunogenetics of HIV and HIV associated tuberculosis. Tuberculosis (Edinb). 2012;92:18–30.
2. Gross O, Thomas CJ, Guarda G, et al.. The inflammasome: an integrated view. Immunol Rev. 2011;243:136–151.
3. Broz P, Monack DM. Molecular mechanisms of inflammasome activation during microbial infections. Immunol Rev. 2011;243:174–190.
4. Lamkanfi M, Dixit VM. Inflammasomes and their roles in health and disease. Annu Rev Cell Dev Biol. 2012;28:137–161.
5. Mishra BB, Moura-Alves P, Sonawane A, et al.. Mycobacterium tuberculosis protein ESAT-6 is a potent activator of the NLRP3/ASC inflammasome. Cell Microbiol. 2010;12:1046–1063.
6. Wong KW, Jacobs WR Jr. Critical role for NLRP3 in necrotic death triggered by Mycobacterium tuberculosis. Cell Microbiol. 2011;13:1371–1384.
7. Dorhoi A, Nouailles G, Jörg S, et al.. Activation of the NLRP3 inflammasome by Mycobacterium tuberculosis is uncoupled from susceptibility to active tuberculosis. Eur J Immunol. 2012;42:374–384.
8. Abdalla H, Srinivasan L, Shah S, et al.. Mycobacterium tuberculosis infection of dendritic cells leads to partially caspase-1/11-independent IL-1β and IL-18 secretion but not to pyroptosis. PLoS One. 2012;7:e40722.
9. Pontillo A, Brandão LA, Guimarães RL, et al.. A 3'UTR SNP in NLRP3 gene is associated with susceptibility to HIV-1 infection. J Acquir Immune Defic Syndr. 2010;54:236–240.
10. Pontillo A, Oshiro TM, Girardelli M, et al.. Polymorphisms in inflammasome genes and susceptibility to HIV-1 infection. J Acquir Immune Defic Syndr. 2012;59:121–125.
11. American Thoracic Society. Diagnostic standards and classification of tuberculosis in adults and children. Am J Respir Crit Care Med. 2000;161:1376–l395.
12. Vargas AE, Marrero AR, Salzano FM, et al.. Frequency of CCR5delta32 in Brazilian populations. Braz J Med Biol Res. 2006;39:321–325.
13. Veit TD, Cordero EA, Mucenic T, et al.. Association of the HLA-G 14 bp polymorphism with systemic lupus erythematosus. Lupus. 2009;18:424–430.
14. Pontillo A, Catamo E, Arosio B, et al.. NALP1/NLRP1 genetic variants are associated with Alzheimer disease. Alzheimer Dis Assoc Disord. 2012;26:277–281.
15. Pontillo A, Girardelli M, Kamada AJ, et al.. Polymorphisms in inflammasome genes are involved in the predisposition to systemic lupus erythematosus. Autoimmunity. 2012;45:271–278.
16. Day TG, Ramanan AV, Hinks A, et al.. Autoinflammatory genes and susceptibility to psoriatic juvenile idiopathic arthritis. Arthritis Rheum. 2008;58:2142–2146.
17. Dehghan A, Dupuis J, Barbalic M, et al.. Meta-analysis of genome-wide association studies in >80 000 subjects identifies multiple loci for C-reactive protein levels. Circulation. 2011;123:731–738.
18. Cox DG, Canzian F. Genotype transposer: automated genotype manipulation for linkage disequilibrium analysis. Bioinformatics. 2001;17:738–739.
19. Barrett JC, Fry B, Maller J, et al.. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005;21:263–265.
20. Fernando SL, Saunders BM, Sluyter R, et al.. A polymorphism in the P2X7 gene increases susceptibility to extrapulmonary tuberculosis. Am J Respir Crit Care Med. 2007;175:360–366.
21. Li CM, Campbell SJ, Kumararatne DS, et al.. Association of a polymorphism in the P2X7 gene with tuberculosis in a Gambian population. J Infect Dis. 2002;186:1458–1462.
22. Sambasivan V, Murthy KJ, Reddy R, et al.. P2X7 gene polymorphisms and risk assessment for pulmonary tuberculosis in Asian Indians. Dis Markers. 2010;28:43–48.
23. Razmara M, Srinivasula SM, Wang L, et al.. CARD-8 protein, a new CARD family member that regulates caspase-1 activation and apoptosis. J Biol Chem. 2002;277:13952–13958.
24. Bouchier-Hayes L, Conroy H, Egan H, et al.. CARDINAL, a novel caspase recruitment domain protein, is an inhibitor of multiple NF-kappa B activation pathways. J Biol Chem. 2001;276:44069–44077.
25. Pathan N, Marusawa H, Krajewska M, et al.. TUCAN, an antiapoptotic caspase-associated recruitment domain family protein overexpressed in cancer. J Biol Chem. 2001;276:32220–32229.
26. Wellcome Trust Case Control Consortium. Genomewide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature. 2007;447:661–678.
27. Ko DC, Shukla KP, Fong C, et al.. A genome-wide in vitro bacterial-infection screen reveals human variation in the host response associated with inflammatory disease. Am J Hum Genet. 2009;85:214–227.
28. Master SS, Rampini SK, Davis AS, et al.. Mycobacterium tuberculosis prevents inflammasome activation. Cell Host Microbe. 2008;3:224–232.