Foxp3 Expression and Nitric Oxide Production in Peripheral Blood Mononuclear Cells of Communicants with Pulmonary Tuberculosis


Correspondence to: V. R. A. Pereira, Departa-mento de Imunologia, Centro de Pesquisa Aggeu Magalhães/FIOCRUZ, Av. Professor Moraes Rego s/n, Cidade Universitária, Campus da UFPE, 50670-420, Recife, PE, Brazil, CEP. E-mail:


The understanding of the mechanisms involved in the immune response is of significant relevance to the control of tuberculosis (TB), especially in individuals living with patients with TB. To characterize the nitric oxide (NO) production and the Foxp3 marker expression in this population, peripheral blood mononuclear cells of intradomiciliary contacts of individuals with pulmonary tuberculosis with (CTb, susceptible) and without (STb, resistant) previous history of active infection were stimulated in vitro with Mycobacterium tuberculosis antigen (TbAg) and with the mitogen Concanavalin A for 24 and 48 h. The groups analysed did not present significant difference in the Foxp3 mRNA expression nor in the NO production. Negative correlation (P = 0.09) between NO and Foxp3 after a 48-h stimulation with TbAg was observed in the STb group. In this group, after a 24-h culture stimulated with TbAg (P = 0.03), this same correlation was observed. In comparison with the cytokines previously studied by our group (Cavalcanti et al., 2009), a positive correlation was observed between IL-10 and Foxp3 after a 48-h culture of cells from communicants susceptible to tuberculosis (STb) stimulated with TbAg (P = 0.04). Evaluating the entire population, a positive correlation was observed between the cytokine TNF-α and the Foxp3 marker in the cultures stimulated for 24 (P = 0.03) and 48 (P = 0.02) hours with TbAg. Therefore, considering the similarity in the exposure and the individual capacity of responding to the contact with M. tuberculosis, the present study contributes to the comprehension of the immune regulation in individuals living with patients with TB.


Pulmonary tuberculosis (TB) is a global public health problem. In Brazil, it presents a high incidence, with nearly 46 cases per 100.000 population [1]. Clinical and epidemiological investigation of the contact of patients with bacilipherous pulmonary tuberculosis has been one of the aims of the TB combat and control plans [2, 3]. In the intradomiciliary contact, the frequency and the intensity of the exposure to the infection source contribute to the transmission of tuberculosis, being the detection of secondary cases and early treatment, some of the measures to the disease prevention [4-6].

It is estimated that a bacilipherous patient may infect between 10 and 15 individuals per year [7], but the factors determining the illness of individuals are not yet completely elucidated. It is known that the development of tuberculosis is associated with the immune state of the infected individual and that the host protective immune response against this pathogen is mediated by the cellular immunity, in which some cytokines and Th1 cells have a critical role [8-11]. In a previous study, our group has shown that there were no differences in the production of the cytokines TNF-α, IL-10 and IFN-γ in peripheral blood mononuclear cells between communicants of intradomiciliary bacilipherous resistant or susceptible to the disease, that is, with or without history of active secondary tuberculosis [12]. We suggested that circulating memory T cells respond to the exposure of mycobacterial antigens in vitro in a similar manner in individuals, apart from their susceptibility to the development of tuberculosis [12].

In patients with active tuberculosis, the increase in CD4+CD25+ regulatory T cells in the infection sites may influence the occurrence of high levels of IL-10 and TGF-β in the serum of these patients [13-16]. The regulation of the Th1 and Th2 cytokine production profile by Treg cells is considered an important stage to the control of infection, while the transcription factor Foxp3 is the better studied effector molecule of Treg cells [17, 18]. The Th1 immune response profile in tuberculosis involves the production of pro-inflammatory mediators, such as IFN-γ and TNF-α, and production of IFN-γ-dependent nitric oxide (NO) by macrophages with microbicidal functions [19-22]. However, little is known about the Foxp3 expression and nitric oxide production in response to the exposure to M. tuberculosis and its relationship with the development of active tuberculosis.

Thus, the present study aimed to analyse the Foxp3 expression and NO production between intradomiciliary communicants who did or did not develop secondary tuberculosis after the exposure to the same source of infection, and compare the results with the levels of the cytokines TNF-α, IL-10 and IFN-γ.

Materials and methods

Study population

The present study was composed of two groups: CTb, composed by communicants who had developed secondary TB from intradomiciliary contact, and STb, composed by intradomiciliary communicants who did not develop TB. These individuals were exposed to M. tuberculosis, being considered intradomiciliary communicants of a patient with pulmonary tuberculosis. Details about the study group were previously published [12]. Therefore, in the moment of the samples collection, a group of communicants considered susceptible to secondary pulmonary tuberculosis was compared with a group of communicants equally exposed, who, however, did not develop pulmonary tuberculosis, being considered resistant. The exclusion criteria were: individuals with age below 14 years old, individuals with serological test positive for HIV, diabetes, and individuals who have refused to sign the ‘Term of Free and Informed Consent’. CPqAM⁄ Fiocruz Research Ethics Committee (CAAE 0028.0.000.095–09) approved the experimental protocols.

Between the individuals, 17 intradomiciliary communicants with secondary tuberculosis history and 40 healthy intradomiciliary communicants without previous history of pulmonary tuberculosis residing in the municipality of Cabo de Santo Agostinho in the State of Pernambuco, north-eastern region of Brazil, were selected. The intradermal reaction test (PPD) was performed in 31 individuals of the study group.

Preparation of the crude M. tuberculosis antigen

A M. tuberculosis culture was heat-inactivated to further obtain the antigen. The protein concentration was measured using a modified Bradford assay [23]. The antigen was separated into aliquots and stored in −80 °C until use. The electrophoretic analyses were performed using polyacrylamide gels according to Laemmli (1970) [24].

Blood collection and cell culture

In intradomiciliary visits, 15 ml of peripheral blood was collected in heparinized tubes from each individual.

Peripheral blood mononuclear cells (PBMCs) were isolated from blood samples by means of gradient density, using Ficoll–Hypaque. A total of 2 × 106 cells/well were cultured in duplicate in 24-well culture plates for 24 h at 37 °C with 5% CO2 and M. tuberculosis antigen WPE (whole-protein extract) at 10 mg/ml. Peripheral blood mononuclear cells cultured in the medium or Concanavalin A (Sigma Chemical Co., St. Louis, MO, USA) at 2.5 mg/ml were used for negative and positive controls, respectively.

After the incubation period, the plates were centrifuged (1800 RPM for 10 min, at room temperature). The precipitate containing the cells was collected and stored in 1 ml of Trizol® (Invitrogen, Carlsbad, CA, USA) at −80 °C for further analysis.

Nitric oxide determination in PBMC culture supernatants

The NO production was determined indirectly through nitrite quantification (NO2−), resultant of its oxidation in the cell culture supernatant through the Griess colorimetric reaction [25]. ELISA plates (96-well-Costar half-area plate) were filled with 25 μl of culture supernatants (two replicates), followed by the Griess reagent in the same volume. A standard curve was generated using sodium nitrite at 200 μm and submitted to serial dilution (factor 2) in RPMI medium supplemented with 2% of foetal calf serum (Cultilab, Brazil). After incubation for 10 min in the dark, the absorbance was read at 450 nm using spectrophotometer. The absorbances were compared with the standard curve (threshold set in 0.19 μm), and the results were expressed as the replicate means ± standard error, using the Microplate Manager Version 4.0 software (Bio-Rad Laboratories).

Foxp3 detection and quantification in PBMCs RNA extraction

Total RNA was obtained from 2 × 106 cells/ml of the cultures challenged using the protocol of the Trizol reagent, according to the manufacturer's recommendations. The RNA obtained was resuspended in RNase-free water and incubated for 5 min at 65 °C. Prior quantification of the RNA using the NanoDrop ND-1000 UV-Vis spectrophotometer (Thermo) was performed. RNA quality was considered acceptable when the 260/280 nm ratio was between 1.8 and 2.1. All the samples were normalized to the same concentration prior to the cDNA obtention.

cDNA obtention

For the reverse transcription, the ‘TaqMan® Reverse Transcription Reagents’ kit (Applied Biosystems, Alameda, CA, USA) was used, according to the manufacturer's protocol. RT–PCR was performed in a final volume of 50 μl, containing TaqMan RT buffer 1x, MgCl2 5.5 mm, dNTPs mix 500 μm, random hexamers 2.5 μm, RNase inhibitor 0.4 U/μl, multiscribe reverse transcriptase enzyme 3.125 U/μl and 12.375 μl of the sample. The reaction consisted in a cycle, which is divided into ten minutes at 25 °C, thirty minutes at 48 °C and 5 min at 95 °C. After this stage, samples were stored at −20 °C until its use in the real-time PCR assays (qPCR).

Evaluation of Foxp3 gene expression

Foxp3 mRNA expression to the marker Foxp3 was evaluated using real-time PCR (ABI PRISM 7500, Applied Biosystems). The reaction was carried out in a final volume of 25 μl, containing TaqMan Universal PCR Master Mix (2x), TaqMan Gene Expression Assay for Foxp3 (Assay ID: (Hs00203958_m1)), constituted of specific primers and probes (FAM fluorophore), RNase-free water and 1 μl of the sample. The reaction's endogenous control (reference gene) was the 18S ribosomal RNA, whose reaction was carried out using TaqMan Universal PCR Master Mix, forward and reverse primers to 18S ribosomal RNA, 18S ribosomal RNA probe (VIC fluorophore), RNase-free water and 1 μl of the sample. mRNA relative quantification was made using the comparison of CT (threshold cycle) of the target gene (Foxp3) with the endogenous gene (18S ribosomal RNA). Expression analysis consisted in normalizing the results, discounting the CT obtained from the endogenous gene from the CT of the target gene from each sample (ΔCT). After normalization of values of the target gene, its ΔCT was subtracted from the ΔCT of a calibrator element (ΔΔCT). It was chosen according to the situation analysed, in this case, in comparison with the control group (individuals without TB history) or between groups of patients (individuals with previous history of the disease). The formula ΔΔCt = ΔCt (sample) – ΔCt (calibrator) was used, where ΔCt (sample) corresponds to an interest parameter to be relativized. After that, ΔΔCt is applied on the following formula:

display math

thus obtaining the Foxp3 relative quantification expressed in a sample.

Statistical analysis

The Shapiro–Wilk test was used to test the supposition of normality in the variants involved in this study. For the comparative analysis of the normality between the variants, the Mann–Whitney U-test and the Student's t-test were used. For the comparative analysis between the qualitative analysis, the chi-square test or the Fisher's test was used when necessary. Moreover, Pearson's coefficient of linear correlation was used for correlation analysis. The results were considered significant when < 0.05. Excel 2003, GraphPad Prism 5.0 and R 2.10.0 were used to perform the statistical analysis.


In the population studied, 17 individuals (29.82%) were part of the CTb group and 40 individuals (70.18%) were in the STb group. The age of the population ranged from 14 to 80 years. 22 individuals were male (38.6%) and 35 female (61.4%). CTb group presented predominance of men (70.59%), while STb group presented a predominance of women.

The intradermal reaction test (PPD) was performed in 31 (54.4%) of the individuals, in which the majority have shown a result > 10 mm. Including the vaccination with BCG in the CTb group, a total of 12 individuals (70.59%) did not show a vaccination scar, while in the STb group only 14 individuals (35%) were not vaccinated against tuberculosis. Considering the familiar bond in both STb and CTb groups, 33 individuals (82.5%) were related and 7 (17.5%) were aggregated.

For 24 and 48 h, the NO production was similar between the analysed groups when the cells were stimulated with tuberculosis bacillus crude antigen (Fig. 1). Nitric oxide production was similar in the individuals resistant to the tuberculosis (STb group) independently of the PPD skin test result (<10 mm or ≥10 mm). In the same line, there was no significant difference in the NO detection between groups with or without BCG scaring (data not shown).

Figure 1.

Nitric oxide (NO) production in peripheral blood mononuclear cell cultures in intradomiciliary communicants of patients with pulmonary tuberculosis with (CTb) or without (STb) secondary tuberculosis, stimulated with 10 ug/ml M. tuberculosis antigen for 24 and 48 h. Horizontal bars represent the NO production mean in each group.

Foxp3 mRNA expression between PBMCs from individuals susceptible and resistant to the development of secondary pulmonary tuberculosis due to intradomiciliary contact exposure did not present significant difference (P > 0.05) when stimulated with Con A or with M. tuberculosis antigen (Fig. 2). Similar results were presented when comparing groups of individuals with or without BCG vaccination scars and between patients resistant to tuberculosis, the ones with higher (≥10 mm) or lower (<10 mm) response to the PPD test (data not shown).

Figure 2.

Foxp3 gene expression profile in peripheral blood mononuclear cells in communicants with (CTb) and without (STb) previous tuberculosis history without stimuli (= 0.35) and after stimulation with 2.5 μg/ml Concanavalin A (Con A, = 0.75) and 10 μg/ml M. tuberculosis antigen (Ag, P = 0.21) in a 24-h culture.

The correlation between the NO secretion and Foxp3 mRNA expression was determined through Pearson's correlation test. Cell culture of communicants resistant to tuberculosis stimulated with TbAg presented a negative correlation between NO and Foxp3 mRNA in 24 h (= 0.030). This correlation was only observed in the group of communicants susceptible to tuberculosis after 48 h of stimuli (= 0.013) (Fig. 3).

Figure 3.

Negative correlation between nitric oxide and Foxp3 in individuals susceptible (CTb group, panel A) and resistant to tuberculosis (STb group, panel B).

Comparing NO production and Foxp3 mRNA expression results with the data about cytokines already published [12], we observed a positive correlation between Foxp3 mRNA expression and IL-10 production in communicants susceptible to tuberculosis, but not in communicants resistant to tuberculosis (Table 1). Considering the total population of communicants, a positive correlation was also observed between the Foxp3 mRNA expression and TNF-α production (Table 1).

Table 1. Correlations between IFN-γ, IL-10, TNF-α, nitric oxide and Foxp3 in PBMC cultures of the study population after stimulation with TbAg for 24 and 48 h
 Communicant susceptible to tuberculosisCommunicant resistant to tuberculosisTotal communicants
CCI 95% P CCI 95% P CCI 95% P
Variants (24 h)
IL-10 × IFN-γ0.3−0.24–0.700.260.1−0.22–0.410.520.14−0.12–0.380.29
IL-10 × TNF-α−0.67−0.56–0.460.81−0.31−0.58–0.180.06−0.21−0.44–0.540.11
IL-10 × NO0.33−0.29–0.760.280.15−0.24–0.510.440.23−0.570.13
IL-10 × Foxp30.22−0.32–0.660.41−0.05−0.37–0.270.750.07−0.20–0.330.62
TNF-α × IFN-γ0.05−0.47–0.540.85−0.09−0.40–0.240.59−0.06−0.31–0.200.65
TNF-α × NO−0.47−0.82–−0.04–0.650.080.01−0.29–0.320.92
TNF-α × Foxp30.37−0.16–0.740.160.23−0.10–0.520.170.290.01–0.520.03
IFN-γ × NO0.3−0.32–0.740.33−0.06−0.44–0.330.750.02−0.28–0.320.88
IFN-γ × Foxp30.16−0.37–0.620.55−0.14−0.44–0.190.41−0.07−0.34–0.200.61
No × Foxp30.52−0.07–0.840.08−0.42−0.70–−0.030.03−0.01−0.33–0.300.92
Variants (48 h)
IL-10 × IFN-γ−0.13−0.60–0.400.630.13−0.20–0.440.420.15−0.10–0.400.24
IL-10 × TNF-α0.34−0.19–0.730.2−0.05−0.37–0.280.760.09−0.17–0.350.47
IL-10 × NO−0.16−0.69–0.480.630.16−0.20–0.480.380.05−0.24–0.340.70
IL-10 × Foxp30.510.005–0.810.040.06−0.27–0.390.690.17−0.10–0.430.22
TNF-α × IFN-γ0.26−0.29–0.680.340.19−0.14–0.490.250.12−0.14–0.370.35
TNF-α × NO−0.3−0.76–0.350.350.09−0.26–0.420.620.04−0.25–0.320.77
TNF-α × Foxp30.32−0.22–0.710.230.28−0.05–0.560.090.320.05–0.550.02
IFN-γ × NO0.55−0.06–0.860.070.27−0.08–0.560.130.21−0.08–0.470.15
IFN-γ × Foxp3−0.11−0.59–0.410.67−0.01−0.34–0.310.91−0.07−0.34–0.200.58
NO × Foxp3−0.62−0.86–0.160.0130.15−0.10–0.550.16−0.005−0.30–0.290.97


The immune mechanisms developed by the host directing M. tuberculosis infection to the progress of clinical disease or to the maintenance of the health state and contention of bacillus have been evaluated by different study groups [26-29]. It is of major importance to understand that some factors are primordial to the progress of active pulmonary tuberculosis, once that while some individuals exposed to the bacillus in the same socio-environmental conditions develop the disease, others remain healthy.

The population studied was equally exposed to the tuberculosis bacillus through the intradomiciliary contact with bacilipherous individuals. However, while they have been in the same socio-environmental conditions, some individuals have developed the disease, while others remained healthy for at least 5 years until the closure of this study. Beyond the environmental and nutritional factors, genetic factors may have predisposed some individuals to the progress of the disease. In this context, our group have shown that intradomiciliary contacts with secondary tuberculosis presented a higher frequency of HLA DRB1*04 alleles (OR = 2.44; = 0.032) with an aetiological fraction of 0.15, while intradomiciliary contacts resistant to tuberculosis presented a higher frequency of DRB1*15 alleles (OR = 0.48; P = 0.049). In this study, the presence of HLA-DRB1 alleles shared between the primary cases and their intradomiciliary communicants was a risk factor to tuberculosis (P = 0.028) [30].

Interindividual differences have also been observed in the IFN-γ production. Some studies have demonstrated a decrease in the IFN-γ production in active tuberculosis [31, 32]. Others have shown that the resistance to tuberculosis has been associated with the high production of IFN-γ during M. tuberculosis infection [33, 34]. These studies reinforce the idea that the protective response against M. tuberculosis depends not exclusively on IFN-γ [35, 36]. The role of TNF-α, for instance, in the control of infection caused by M. tuberculosis is complex. It is known that it acts in synergy with IFN-γ, inducing the formation of intermediate reactive nitrogen and oxygen species, mediating the tuberculostatic action of macrophages [37]. On the other hand, the pulmonary granulomatous lesion is associated with the exacerbated production of TNF [37, 38]. In a previous study, we investigated the immune memory of mononuclear cells of intradomiciliary communicants susceptible or resistant to tuberculosis after TbAg stimuli and we demonstrated that there was a similar IL-10, IFN-γ and TNF-α production profile [12].

In our study, a negative correlation between NO production and Foxp3 expression in the CTb and STb groups in 48 h of culture was observed. The same correlation between these groups was observed in the 24-h period. Although there was no statistical difference between the Foxp3 expression and NO production in the groups analysed, we highlight the results of correlations. The inverse correlation between the NO production and Foxp3 mRNA expression in communicants was demonstrated in the first 24 h after antigenic stimulation in individuals resistant to tuberculosis, while in individuals susceptible to tuberculosis the increase in the Foxp3 mRNA expression associated with the decrease in the production of NO and with the increase in the IL-10 was only observed after 48 h of antigenic stimulation. The positive correlation between the Foxp3 mRNA expression and IL-10 production (= 0.04) was only observed in communicants susceptible to tuberculosis, corroborating the results corresponding to the inhibitory function of IL-10 over the pro-inflammatory immune response and differentiation of the Th2 response [39]. Higher levels of CD4+CD25+Foxp3+ Treg cells in the peripheral blood of patients with TB were described in comparison with the levels observed on infected cells of a latent form with M. tuberculosis (healthy individuals) and non-infected controls when stimulated with BCG [15]. Positive correlation between Foxp3 mRNA expression and TNF-α production observed in the population of this study confirms previous data suggesting that the maintenance of the immunological memory with effective production of regulatory and protective cytokines in individuals exposed to M. tuberculosis is primordial to the immune response in these individuals [38, 40, 41].

Data found in this study suggest a dynamic participation of these mediators as primordial elements in the maintenance of the health state of the individuals exposed to M. tuberculosis [42, 43]. Thus, considering the similarity in the exposure and the individual capacity of responding to the contact with M. tuberculosis, this study contributes to the comprehension of the immune modulation in individuals living with patients with TB and their risk to develop secondary tuberculosis.


The authors are grateful to L. F. da Rocha for technical assistance and George Tadeu Diniz for support with statistical analyses. This work was supported by the State of Pernambuco Research Foundation (FACEPE), by the Brazilian National Research Council (CNPq) and by the Oswaldo Cruz Foundation (FIOCRUZ).