**4. Discussion**

It is well known that to *M. tuberculosis*, ethiologic agent of human TB can cause a broad spectrum of effects ranging from no infection to different clinical disease phenotypes [2, 16, 23-25]. However, the reasons for individual or ethnic differences in acquiring infec‐ tion, active disease, disease severity, and different clinical outcomes have not been completely clarified. It has long been realized that many human diseases arise from the complex interplay between environmental exposures and host genetics susceptibilities [26]. In addition, several genetic factors have also been associated with different outcomes: host susceptibility *per se* the occurrence of active TB, disease severity and / or protection for the occurrence of active disease [27-33].

The establishment of an efficient immune response involves many different molecules, among which, cytokines and their receptors play an extremely important role. Thus, any genetic alteration leading to changes in the regulation of gene expression may reflect this response. It is known that the interindividual variation in the production of these molecules is directly related to the genetic "background". Literature data have clearly demonstrated that genetic variability of the genes encoding these molecules can affect the regulation of gene expression positively or negatively influencing the final yield of the molecule in question. In the last decade, several single nucleotide polymorphisms (SNPs) in the regulatory region of different cytokine genes have been described and associated with susceptibility, severity or protection for a growing number of diseases of different etiologies including tuberculosis [7, 34-35].

Table 10 shows the distribution of the 14 identified haplotypes in the different groups used for the association study. No significant difference was observed in the haplotypes frequencies

> **EPTB n=19**

 9567.9%) 82(67.8%) 0 70(72.9%) 40(69%) 2(1.4%) 1(0.8%) 0 0 0 19(13.6%) 17(14.1%) 2(10.5%) 11(11.5%) 7(12.1%) 1(0.7%) 0 1(5.26%) 0 1(1.72%) 3(2.1%) 3(2.5%) 0 4(4.2%) 2(3.4%) 7(5 %) 6(5%) 1(5.3%) 5(5.2%) 5(8.6%) 1(0.7%) 1(0.8%) 0 1(1.1%) 1(1.7%) 9(6%) 8(6.6%) 1(5.3%) 3(3.1%) 2(3.4%) 1(0.7%) 1(0.8%) 0 1(1.1%) 0 1(0.7%) 1(0.8%) 0 0 0 1(0.7%) 1(0.8%) 0 0 0 1(0.71%) 0 1(5.26%) 1(1.04%) 0

After the genotyping of all samples and evaluation of the possible association with the different TB outcomes, the most frequent polymorphisms (-376G>A; -308G>A; -244G>A and -238G>A) were tested in a stratified analysis against the demographic variables gender and age. No

It is well known that to *M. tuberculosis*, ethiologic agent of human TB can cause a broad spectrum of effects ranging from no infection to different clinical disease phenotypes [2, 16, 23-25]. However, the reasons for individual or ethnic differences in acquiring infec‐ tion, active disease, disease severity, and different clinical outcomes have not been completely clarified. It has long been realized that many human diseases arise from the complex interplay between environmental exposures and host genetics susceptibilities [26]. In addition, several genetic factors have also been associated with different outcomes: host susceptibility *per se* the occurrence of active TB, disease severity and / or protection for the

**Controls PPD+ n=96**

**Controls PPDn=58**

between groups (data not shown) and their distribution was quite homogeneous.

**PTB n=121**

**Table 10.** Frequency of *TNF*-α haplotypes in the different groups studied

significant differences were found for gender or age (data not shown).

**Haplotype**

**4. Discussion**

occurrence of active disease [27-33].

**General TB n= 140**

90 Tuberculosis - Current Issues in Diagnosis and Management

Among the possible genetic variations associated with an increased risk of developing TB, there are several polymorphisms, mainly SNPs, in genes coding for cytokines, cytokine receptors and several other molecules such as vitamin D receptor, NRAMP1 (SLC11A1), HLA genes, etc.

The immune defense against *M*. *tuberculosis* is complex and involves the interaction between T CD4+ , T CD8+ lymphocytes, macrophages, and monocytes along with the production of cytokines, such as interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α) [36].

Convincing evidence indicating the importance of IFN-γ in particular, in the control of mycobacterial infections has been found in both experimental and clinical studies [37-38].

Among the mainly important cytokines involved in TB progress after infection with *M. tuberculosis*, TNF-α plays a key role. It is also a potent proinflammatory cytokine acting in protection against intracellular pathogens [39-40].

The genetic variability of *TNF*-*α* and *IFNG* has been described in the last decade [41-46] including association studies with tuberculosis. However, the frequencies of the polymor‐ phisms already described varies according to the ethnicity of the population studied, ham‐ pering the better interpretation of the value of association studies. Unfortunately, most of these are performed in ethnically homogeneous populations, and therefore, many of the associations described for a particular allelic variant in a certain gene may not represent genetic risk factor in other populations. In Brazil, a country characterized by ethnically mixed population, there are few data regarding the frequency of single nucleotide polymorphisms in these genes (*IFNG*,*TNF*-*α*) and the few existing studies refers to one or two SNPs only. In view of the importance of the promoter region with respect to regulation of gene expression, the major goal of this work was to proceed a partial mapping of the promoter region of IFN*G* and *TNFα* genes (approximately, 800bp upstream of the transcription starting site) through PCRsequencing approach in samples from TB patients and healthy controls from Rio de Janeiro, Brazil. Subsequently, based on frequencies of the different SNPs found individually for each gene, we performed an association study with different TB outcomes.

#### **4.1. Polymorphisms in the promoter region of IF***NG* **and its association with TB**

Characterization of the important portion within the *IFNG*was firstly identified two decades ago bydeletion analysisstudies [47-48]. According to authors, it comprises a highly conserved region from positions -117 to -47 and contains two sub regions that can be complexes with proteins. The sub-proximal region (-90 to -65) shows strong homology to the IL-2 promoter [49]. Several transcription factors activate transcription of *IFNG* by binding to this region. Conversely, several others inhibit factors binds in other regions affecting transcription. Hence, the interest in investigating the polymorphic sites within *IFNG* gene promoter, particularly considering the importance of this cytokine in eliciting the immune response.

to regulatory factors instead of polymorphisms in the gene, which is consistent with our results. Only seven SNPs were found in our population, five of which were at a low frequency. The polymorphisms found with a higher frequency were the -200G>T and -599C>G, the latter being located between two putative binding sites of transcription factors. As this is not yet a SNP described in the literature, functional studies are needed to better understand their functional role. The SNP -200 is of great interest for association studies. However, this polymorphism was not found in Caucasians or Indian populations, suggesting that different selective forces may be operating in different ethnic or racial groups. These data corroborate the evidence that IFN-γ is very important in the immune response and that mutations that interfere with their production may influence the outcome of active tuberculosis as shown by authors [28,31,53-54], and therefore, a selective force lead the gene to be so conserved.

Influence of the Interferon–Gamma (IFN–γ) and Tumor Necrosis Factor Alpha (TNF–α) Gene Polymorphisms in TB...

http://dx.doi.org/10.5772/55099

93

**4.2. Polymorphisms within the promoter region of** *TNF–α* **and their association with TB**

control and production.

humans [56-58].

altered circulating levels of this cytokine.

TNF-α is a proinflammatory and immunoregulaty cytokine which plays a key role in the initiation, regulation and perpetuation of host defense against infections, but is fatal in excess. As this molecule plays an important role against a variety of pathogens involving different patterns of risks and benefits, it is expected that several genetic elements are involved in its

The levels of circulating TNF-α are regulated at transcriptional and post-transcriptional levels and several polymorphisms within the promoter region of TNF-α have been associated with

In humans, the *TNF-α* gene is located within the complex involving the human leukocyte antigens (HLA), a highly polymorphic region on chromosome 6p21.3 and hence, many of *TNFα* polymorphisms are in linkage disequilibrium with the HLA genes. Because of differences in the distribution of HLA alleles we might expect variations in associations between polymor‐

The human genome analysis showed that the level of variations in the genome is approxi‐ mately one SNP/1.71Kb [55]. However, the *TNF-α* promoter has higher density of SNPs. Despite this level of variation, the regions involved in gene regulation are highly conserved in

In our work, we perform the mapping of the first 800pb promoter region, through direct sequencing of the amplified PCR product in 500 DNA samples from individuals living in the metropolitan area of Rio de Janeiro, Brazil and found seven polymorphisms previously described in the literature, most of which well characterized. However, according allele frequencies, only the variants -376A, -308A, -238A and -244A were present in more than one percent (0.030, 0.087, 0.011 and 0.050) respectively. The influence of these SNPs in binding of transcription factors have not been fully explored, most of the studies are focused on the association of one or two SNPs with different diseases. The SNPs -376, -308 and -238 have been the most studied but, the results of functional studies performed so far for SNPs -308 and -238 are controversial. It is believed that the variant -308A is associated with an increased tran‐ scription rate, leading to an increased production of TNF-α [59] and the variant-238A with a

phisms of *TNF-α* and various conditions in different geographical areas.

Here, analysis of the generated sequences identified seven polymorphic sites, four of which were new. The transition C →T, was identified at position-787 from the transcription starting site in five subjects, all heterozygous. The second C → T transition, previously described in the data base of SNPs at position-785 was also found in five individuals, all heterozygous, however, no reference to this SNP was found in the literature. The other three SNPs not yet described, were C to G transition at position -599; C to T at position -517 and A to G at position -255. Finally, two additional SNPs, transition from G to T at position -200 and A to G at position -172, already described and well characterized [45] were found in our population.

One of the main problem found during this mapping was the confirmation of the identified SNPs based on literature data and from different SNPs data bases available online because of the lack of standardization regarding to the reference nucleotide to define the promoter region (transcription starting site nt +1). Many authors describe the SNPs identified in relation to the site of translation or use reference sequences containing sequencing errors leading to misclas‐ sification of SNPs (eg SNP-200G>T, originally described as -183 [45], later called as -179 [50] and finally, confirmed in this study as -200). The current name, confirmed in this study is based on the correction of the reference sequence used in previous studies and now available online. These types of errors greatly hampered the beginning of the sequence analysis regarding the identification of novel SNPs.

The frequency of each polymorphism was determined in the study population. As noted in Table 1, the allele frequencies for all the identified SNPs were less than one percent, except for the variants -599G and -200T, both in a frequency of 1.4%.

Functionally, it is known that polymorphisms (-200 and -172) can affect transcription of the *IFNG*. The region from -213 to -200 induces transcription factor through (AP-1) [51]. A polymorphism at this site (position -200, for example) must change the connection of AP-1 and the promoter activity in T cells. The polymorphism -172 is near to the nuclear factorsactivated T-cells site (NFAT site) (-186TAAAGGAAA-178) and should affect the stability of this region [50].

The variant IFNG -200T is highly inducible by TNF-α and binds constitutively to nuclear extracts obtained from T cells, whereas the allele -200G does not respond to TNF-α [50; 52]. The induction of transcriptional activity, when the T allele is present, increases protection against tuberculosis. Our results corroborate these data, since the *IFNG* -200T variant showed to be associated with protection the occurrence of active TB in our study group (P = 0.033, OR= 0.18, CI= 0.03 to 1.00).

According Bream,JH et al., 2002 [50], the promoter region of *IFNG* is highly conserved, suggesting that these cytokine production variations are probably due to difference in binding to regulatory factors instead of polymorphisms in the gene, which is consistent with our results. Only seven SNPs were found in our population, five of which were at a low frequency. The polymorphisms found with a higher frequency were the -200G>T and -599C>G, the latter being located between two putative binding sites of transcription factors. As this is not yet a SNP described in the literature, functional studies are needed to better understand their functional role. The SNP -200 is of great interest for association studies. However, this polymorphism was not found in Caucasians or Indian populations, suggesting that different selective forces may be operating in different ethnic or racial groups. These data corroborate the evidence that IFN-γ is very important in the immune response and that mutations that interfere with their production may influence the outcome of active tuberculosis as shown by authors [28,31,53-54], and therefore, a selective force lead the gene to be so conserved.

region from positions -117 to -47 and contains two sub regions that can be complexes with proteins. The sub-proximal region (-90 to -65) shows strong homology to the IL-2 promoter [49]. Several transcription factors activate transcription of *IFNG* by binding to this region. Conversely, several others inhibit factors binds in other regions affecting transcription. Hence, the interest in investigating the polymorphic sites within *IFNG* gene promoter, particularly

Here, analysis of the generated sequences identified seven polymorphic sites, four of which were new. The transition C →T, was identified at position-787 from the transcription starting site in five subjects, all heterozygous. The second C → T transition, previously described in the data base of SNPs at position-785 was also found in five individuals, all heterozygous, however, no reference to this SNP was found in the literature. The other three SNPs not yet described, were C to G transition at position -599; C to T at position -517 and A to G at position -255. Finally, two additional SNPs, transition from G to T at position -200 and A to G at position

One of the main problem found during this mapping was the confirmation of the identified SNPs based on literature data and from different SNPs data bases available online because of the lack of standardization regarding to the reference nucleotide to define the promoter region (transcription starting site nt +1). Many authors describe the SNPs identified in relation to the site of translation or use reference sequences containing sequencing errors leading to misclas‐ sification of SNPs (eg SNP-200G>T, originally described as -183 [45], later called as -179 [50] and finally, confirmed in this study as -200). The current name, confirmed in this study is based on the correction of the reference sequence used in previous studies and now available online. These types of errors greatly hampered the beginning of the sequence analysis regarding the

The frequency of each polymorphism was determined in the study population. As noted in Table 1, the allele frequencies for all the identified SNPs were less than one percent, except for

Functionally, it is known that polymorphisms (-200 and -172) can affect transcription of the *IFNG*. The region from -213 to -200 induces transcription factor through (AP-1) [51]. A polymorphism at this site (position -200, for example) must change the connection of AP-1 and the promoter activity in T cells. The polymorphism -172 is near to the nuclear factorsactivated T-cells site (NFAT site) (-186TAAAGGAAA-178) and should affect the stability

The variant IFNG -200T is highly inducible by TNF-α and binds constitutively to nuclear extracts obtained from T cells, whereas the allele -200G does not respond to TNF-α [50; 52]. The induction of transcriptional activity, when the T allele is present, increases protection against tuberculosis. Our results corroborate these data, since the *IFNG* -200T variant showed to be associated with protection the occurrence of active TB in our study group (P = 0.033, OR=

According Bream,JH et al., 2002 [50], the promoter region of *IFNG* is highly conserved, suggesting that these cytokine production variations are probably due to difference in binding

considering the importance of this cytokine in eliciting the immune response.

92 Tuberculosis - Current Issues in Diagnosis and Management


identification of novel SNPs.

of this region [50].

0.18, CI= 0.03 to 1.00).

the variants -599G and -200T, both in a frequency of 1.4%.

#### **4.2. Polymorphisms within the promoter region of** *TNF–α* **and their association with TB**

TNF-α is a proinflammatory and immunoregulaty cytokine which plays a key role in the initiation, regulation and perpetuation of host defense against infections, but is fatal in excess. As this molecule plays an important role against a variety of pathogens involving different patterns of risks and benefits, it is expected that several genetic elements are involved in its control and production.

The levels of circulating TNF-α are regulated at transcriptional and post-transcriptional levels and several polymorphisms within the promoter region of TNF-α have been associated with altered circulating levels of this cytokine.

In humans, the *TNF-α* gene is located within the complex involving the human leukocyte antigens (HLA), a highly polymorphic region on chromosome 6p21.3 and hence, many of *TNFα* polymorphisms are in linkage disequilibrium with the HLA genes. Because of differences in the distribution of HLA alleles we might expect variations in associations between polymor‐ phisms of *TNF-α* and various conditions in different geographical areas.

The human genome analysis showed that the level of variations in the genome is approxi‐ mately one SNP/1.71Kb [55]. However, the *TNF-α* promoter has higher density of SNPs. Despite this level of variation, the regions involved in gene regulation are highly conserved in humans [56-58].

In our work, we perform the mapping of the first 800pb promoter region, through direct sequencing of the amplified PCR product in 500 DNA samples from individuals living in the metropolitan area of Rio de Janeiro, Brazil and found seven polymorphisms previously described in the literature, most of which well characterized. However, according allele frequencies, only the variants -376A, -308A, -238A and -244A were present in more than one percent (0.030, 0.087, 0.011 and 0.050) respectively. The influence of these SNPs in binding of transcription factors have not been fully explored, most of the studies are focused on the association of one or two SNPs with different diseases. The SNPs -376, -308 and -238 have been the most studied but, the results of functional studies performed so far for SNPs -308 and -238 are controversial. It is believed that the variant -308A is associated with an increased tran‐ scription rate, leading to an increased production of TNF-α [59] and the variant-238A with a decreased rate of transcription. Regarding the SNP-376G>A different studies show that this polymorphism is located in a region of multiple interactions between proteins and DNA, and that the minor allele acts in the recruitment of proteins OCT1 for this region. According Knight et al (1999) [60] there is a significant interaction between variant -376A and the OCT-1 protein, this variant binds the proteins while the variant -376G does not. The authors report also by tests with the reporter gene system, that this mutant variant moderately increases the basal levels of TNF-α and associate the same with a relative risk of 4 to cerebral malaria. The problem is that the linkage disequilibrium is strong in this area and it is difficult to study the function of an isolated SNP. In some Caucasian populations -376A allele variant is liked to -308G and -238A [60-61], what is not observed in the African Gambia. Thus, association between the linked allelic variants on TNF-α production and diseases has been studied. According, Hajeer& Hutchinson (2001), the combined allele variants -238G, -308A and -376G are associated with high TNF-α levels [62].

studied, although it has also been found in Caucasians [78] and Asian [80]. Instead, the -376 SNP was not detected in any of the non-African analyzed and were found with the -238.

Influence of the Interferon–Gamma (IFN–γ) and Tumor Necrosis Factor Alpha (TNF–α) Gene Polymorphisms in TB...

http://dx.doi.org/10.5772/55099

95

These data demonstrate the importance of taking into account the "background" of the

Association studies of genetic factors with infectious diseases are difficult to conduct because of the multi factorial nature of these diseases that includes host, pathogen and environmental variables in different proportions for each disease. This multi factorial nature of TB stresses the importance to look for haplotypes in the association studies. The fact that our population is so mixed allowed us to find mutations that do not exist in other populations, such as the -308, for example, which is relatively rare in Asians and American Indians. Data from the genotyping of a large number of SNPs for different samples revealed that the human genome has a block structure haplotype [82-83] and configuration of a haplotype sometimes is more important than a single SNP genotype to determine phenotype [84]. Moreover, the construc‐ tion of a haplotype block is useful for identifying SNPs that isolates would not influence the

In conclusion, this study showed that the proximal part of the promoter region of *IFNG* is highly conserved, as seen in previous publications and the identified SNPs were in very low frequency. The -200T allele variant was associated with protection occurring active TB, and pulmonary TB. In addition, this variant was also associated with latent infection. Concerning *TNF-α,* the high genetic variability was confirmed, but only the -376G>A SNP showed an association with susceptibility *per se* to TB occurrence and increased risk for the occurrence of

The data presented here shows the reality of a population with characteristics of high ethnic miscegenation, provides the different SNPs identified enabling the realization of real sample calculation for any association studies that may be idealized with these targets and other conditions for this population and finally, provides haplotype that can be used in other studies

We thank all subjects involved in the study and the Platform-genome DNA sequencing

This work was supported by FAPERJ/Pronex:Proc: E-26/170.0003/2008 and FAPERJ Pensa Rio:

frequencies of SNPs of TNF-α in studies of gene-disease association.

phenotype [85].

**5. Conclusions**

extrapulmonary TB.

of association with other diseases.

**Acknowledgements**

RPT01A PDTIS/FIOCRUZ.

E-26/110.288/2007.

A large number of studies have investigated the association between polymorphisms in the promoter region of the gene for TNF-α and tuberculosis. Results vary according to the different populations studied, finding no association [63-72] or a positive association [26, 29, 73-75]. In our analysis of the single SNP association, TB was associated with the -376G>A. In this case, we observed an association of the minor allele -376A with the outcome of susceptibility *per se* the occurrence of active TB (*p*= 0.035, OR 3.57, IC 0.95 <15.72) and an increased relative risk for the occurrence of extrapulmonary TB (*p*= 0.038, OR 2.68, IC 1.22 < 5.86). The association of this allele variant with the occurrence of TB and an increased relative risk for the development of extrapulmonary TB is intriguing. Given the influence of this allele with increased expression of TNF-α, one would expect an association with the protection. One possible explanation for this observation may be the small sample size in the stratified groups. The large confidence intervals (CI) for both outcomes could be a reflection of the small sample size.

The PCR-sequencing approach (gold standard) used for the mapping these genes practically discard the possibility of genotyping errors and all mutants found for all SNPs evaluated were confirmed twice by new PCR and resequencing. Another possibility would be due to the strong linkage disequilibrium observed in this region of the gene promoter of TNF-α. It is possible that other allelic variant (eg, 238A), as opposed to the functional role of variant -376A is canceling the same level of control of gene expression.

An important aspect of this study relates to the ethnic characteristics of the studied population. Brazilian population is characterized by mixture of ethnicities and the results obtained here contribute for a global understanding of the influence of genetic factors in TB outcomes. Usually, most of the studies on this field are made with ethnically homogeneous populations. A study conducted by Baena et al., 2002 [76] clearly shows the importance of ethnic difference in the association study of SNPs in *TNF*-α promoter with disease. According authors, the -857 SNP is a marker for Amerindians. In that study, SNPs in *TNF*-α promoter were also used to identify markers of ancestry, understanding that this region was well characterized previously with primates and humans. Several studies [77-81] have shown that some polymorphisms as -238; - 244 and -308 are in association with the HLA genes and in addition, the SNPs -308 [77]; -863 and -857 [81], are markers of Caucasians. The -238 SNP was found in three populations studied, although it has also been found in Caucasians [78] and Asian [80]. Instead, the -376 SNP was not detected in any of the non-African analyzed and were found with the -238.

These data demonstrate the importance of taking into account the "background" of the frequencies of SNPs of TNF-α in studies of gene-disease association.

Association studies of genetic factors with infectious diseases are difficult to conduct because of the multi factorial nature of these diseases that includes host, pathogen and environmental variables in different proportions for each disease. This multi factorial nature of TB stresses the importance to look for haplotypes in the association studies. The fact that our population is so mixed allowed us to find mutations that do not exist in other populations, such as the -308, for example, which is relatively rare in Asians and American Indians. Data from the genotyping of a large number of SNPs for different samples revealed that the human genome has a block structure haplotype [82-83] and configuration of a haplotype sometimes is more important than a single SNP genotype to determine phenotype [84]. Moreover, the construc‐ tion of a haplotype block is useful for identifying SNPs that isolates would not influence the phenotype [85].
