Evaluation of T cell immune responses in multi-drug-resistant tuberculosis (MDR-TB) patients to Mycobacterium tuberculosis total lipid antigens

Abstract

Mycobacterium tuberculosis lipid antigens produce significant T cell responses in healthy tuberculin reactor [purified protein derivative (PPD-positive] individuals. In the present study, proliferation and interferon (IFN)-γ/interleukin (IL)-4 responses were analysed to M. tuberculosis total lipid antigens in T lymphocytes from 25 patients with multi-drug-resistant tuberculosis (MDR-TB). The obtained results were compared with those of 30 asymptomatic healthy PPD-positive and 30 healthy tuberculin skin test negative (PPD-negative) subjects. Peripheral blood mononuclear cells (PBMCs) and T cells (CD4⁺ and CD8⁺) were stimulated using autologous immature dendritic cells. Proliferation responses were assessed using 3–{4,5-dimethylthiazol-2-yl}–2,5 diphenyl tetrazolium bromide (MTT). IFN-γ/IL-4 concentrations in the supernatant of the CD4⁺ and CD8⁺T cells were measured by enzyme-linked immunosorbent assay. Proliferation assay showed that the peripheral blood mononuclear cells and CD4⁺ T cells from the MDR-TB patients responded significantly less to the M. tuberculosis total lipid antigens than to the CD4⁺ T cells in the PPD-positive subjects. Total lipid antigen-specific proliferative responses in the CD8⁺ T cells from the MDR-TB patients were minimally detected and the responses were similar to those of the PPD-positive subjects. IFN-γ production by the CD4⁺ T cells stimulated by total lipid antigens from the MDR-TB patients was decreased significantly compared with the PPD-positive individuals, whereas IL-4 production in the patients was elevated. IFN-γ and IL-4 production in the CD8⁺ T cells of the MDR-TB patients was similar to those of the PPD-positive subjects. In conclusion, it is suggested that stimulated CD4⁺ T cells by M. tuberculosis total lipid antigens may be shifted to T helper 2 responses in MDR-TB patients.

Introduction

Mycobacterium tuberculosis is the causative agent of pulmonary tuberculosis (TB) in humans and leads to an estimated of 2–3 million deaths worldwide each year [1,2]. The appearance of strains of M. tuberculosis resistance to current antibiotics is a growing problem, both in the third world and in the developed countries. The strains of M. tuberculosis resistance to both isoniazid and rifampicin with or without resistance to other drugs have been termed multi-drug-resistant (MDR) strains [3–5]. On the other hand, the high prevalence of HIV-1 infection in the MDR-TB patients due to the suppressed immune system is of concern [6].

Substantial studies have indicated that antigen-specific T cells play an important role in developing and maintaining immunity to M. tuberculosis[7]. It has also been shown that T cells and major histocompatibility complex (MHC) class I and class II molecules are involved in the protective immune response to M. tuberculosis protein antigens [8]. However, in recent years it has been proved clearly that CD1 molecules are also involved in the generation of cell-mediated immune responses to mycobacterial pathogens [9–12]. Human CD1 markers are a family of antigen-presenting molecules that bind lipids and present them to T cells. These molecules are expressed constitutively on professional antigen-presenting cells and can be induced in immature dendritic cells derived from peripheral blood monocytes by treatment with granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin (IL)-4 [13,14]. CD1-restricted T cells can contribute to protective immunity by production of high levels of interferon (IFN)-γ and IL-4 [15,16]. Administration of M. tuberculosis lipid vaccine to guinea-pigs created a considerable immune response [17]. It has been revealed recently that the type of lipids contributes to the severity of M. tuberculosis strain infection [18,19].

The precise role and relative importance of this novel pathway for antigen recognition in generating protective immunity to M. tuberculosis, especially in TB and MDR-TB patients, remains poorly understood. The present study was undertaken to determine the role of CD4⁺ and CD8⁺ T cells in MDR-TB patients against M. tuberculosis total lipid and total sonicate antigens in comparison with newly smear-positive TB patients and healthy (positive tuberculin reactor and non-reactor) subjects.

Materials and methods

Preparation of total sonicate antigens

M. tuberculosis strain H37Rv was obtained from the Mycobacteriology Research Centre, Masih Daneshvari Hospital (Tehran, Iran). Preparation of M. tuberculosis total sonicate antigens was carried out using the method described previously [17]. Briefly, the bacteria were resuspended in Tris buffer and centrifuged. The supernatant was removed and the bacteria pellet resuspended in sterile phosphate-buffered saline (PBS). To inactivate the bacteria, the pellet was resuspended in ethanol and allowed to stand overnight. Ethanol was removed by evaporation and the remaining bacteria were lyophilized until desiccated. Finally, a suspension of M. tuberculosis with a concentration of 20 µg/ml in RPMI-1640 was prepared. This suspension was sonicated using sonicator (Dr Hielscher, GmbH, Germany; up to 400 s) on ice for 5 min with 50% maximum power. The total sonicate was centrifuged and the supernatant was filtered and stored at −20°C as the total sonicate antigens. The cells were stimulated by 20 µg/ml concentration of the total sonicate antigens.

Preparation of total lipid antigens

The Folch procedure, with some modifications, was used to produce M. tuberculosis total lipid antigens, as described previously [20]. Briefly, the dried bacteria were resuspended in chloroform: methanol (2 : 1) and sonicated with 50% maximum power for 5 min on ice and mixed overnight on a platform rocker. The suspension was then centrifuged and the supernatant was removed. The insoluble bacterial particles were removed by centrifugation and re-extracted by chloroform : methanol, as described above. Supernatants from the centrifugation steps containing soluble total lipid antigens were collected and dried under nitrogen gas stream. The extracted lipid antigens were weighed and appropriate concentrations were prepared in chloroform : methanol (2 : 1) and stored at −20°C. The cells were stimulated by 20 µg/ml concentration of the total lipid antigens.

Evaluation of total lipid antigens

To evaluate the possible contamination of the protein in the lipid extracted solution, the Lamelli sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) procedure was carried out on the dried lipids. Electrophoresis was performed under reducing conditions, as described previously [21]. For this purpose, an SDS-PAGE sample buffer was added to the dried lipids, boiled for 5 min and run on 12% polacrylamide gel. Standard protein marker (Sigma, St. Louis, MO, USA) was also loaded. To visualize the protein bands, the silver staining method was used according to the procedure described [22].

The extracted lipid was also analysed by thin-layer chromatography (TLC), as described previously [23], with modifications in which phosphatidylcholine, phosphatidylinositol and phosphatidylethanolamine mixture (Avanti, Polar Lipids, Alabaster, AL, USA) was used as standard. TLC was carried out on a silica gel F60 plate (Merck, Kronbery, Germany) using petroleum benzene : diethyl ether: acetic acid (8 : 2 : 1) as the solvent system. The lipid bands were visualized by 10% sulphuric acid spray followed by heating at 140°C.

Limulus amoebocyte lysate (LAL) assay

Endotoxin contamination in M. tuberculosis total sonicate and total lipid antigen extracts was carried out using a LAL assay (Cambrex Bio Science, Walkersville, MD, USA) according to the manufacturer's instructions. Briefly, lyophilized purified endotoxin from Escherichia coli strain 0111.B4 was used as control standard. After reconstitution of the endotoxin vial with 1·0 ml reagent water, a serial twofold dilution up to 0·125 EU/ml concentration was prepared. LAL reagent was used as negative control. The LAL clot assay was performed in test tubes to which 250 µl of diluted antigen samples was added and incubated for 60 ± 2 min at 37°C ± 1°C. The test tubes were examined by 180 inversion for the presence of a stable clot, which was considered positive.

Human subjects

Four groups, including MDR-TB, TB, healthy tuberculin reactor [purified protein derivative (PPD)-positive] and healthy tuberculin skin test negative (PPD-negative) subjects, were selected from the Masih Daneshvari Hospital, National Research Institute of Tuberculosis and Lung Disease (NRITLD), Beheshti University of Medical Sciences (Tehran, Iran). This study was approved by the Institutional Review Board (IRB) and Ethical Review Board (ERB) of NRITLD. All the patients and healthy volunteers consented to take part in the study.

Healthy tuberculin reactor (PPD-positive) donors

Thirty healthy PPD-positive donors (14 females, 16 males, mean age: 35·7 years) were selected from the Masih Daneshvari Hospital personnel and patients accompanied by family members. All individuals had a history of contact with TB patients. M. tuberculosis infection was confirmed in the subjects using the tuberculin skin test. A positive tuberculin skin test was confirmed if the diameter of induration at the site of injection was > 10 mm. These individuals did not show any clinical TB symptoms.

Healthy tuberculin skin test-negative (PPD-negative) subjects

Thirty healthy PPD-negative donors (13 females, 17 males, mean age: 33·5 years) were selected from the control population. None of them had any history of contact with TB patients in the family or in the workplace. Tuberculin skin tests were carried out according to the standard method [24]. Briefly, the tuberculin skin test was considered negative if an initial injection of 1 U of PPD (Razi Institute, Karaj, Tehran) and a follow-up injection of 10 U of PPD were both negative (defined as ≤ 10 mm of induration) 2 weeks later. Eleven individuals had received bacille Calmette–Guérin (BCG) vaccine and had had a negative PPD skin test, while the remaining individuals had not received BCG vaccine.

Active TB patients

Thirty new smear-positive cases (18 males, 12 females, mean age: 47·7 years) were selected from the Masih Daneshvari Hospital. Microscopic and culture examinations of their sputum specimens were positive for acid-fast bacilli. All of them were diagnosed clinically as tuberculosis (TB). Their blood samples were obtained prior to the initiation of anti-TB drug therapy.

Multi-drug-resistant tuberculosis (MDR-TB) patients

Twenty-five MDR-TB patients (12 males, 13 females, mean age: 44·2 years) were obtained from the Masih Daneshvari Hospital. They had the following inclusion criteria: they had a history of at least one previous period of TB treatment under the centre's direct observation (6 months documentation), two positive sputum smear tests and a positive sputum culture. Their susceptibility testing showed resistance to isoniazid and rifampin, and their chest X-ray and clinical symptoms were compatible with pulmonary TB. Duration of infection in all the patients was less than 3 years. Characteristics of the patients are showed in Table 1.

Table 1.

Profiles of the multi-drug-resistant tuberculosis (MDR-TB) patients.

PN§	Sex/age	CS*	ID**	TD (months)†	CXR††	DS‡	AT‡‡
P1	M/42	3 +	1·5	10	C	A	2
P1	M/43	2 +	2	14	I	M	> 2
P3	M/38	3 +	1	12	I	M	2
P4	F/45	1 +	3	16	C	A	> 2
P5	F/61	2 +	2	14	I	M	> 2
P6	F/39	2 +	1	6	I	M	2
P7	F/46	3 +	3	11	C	A	> 2
P8	M/34	3 +	2·5	10	I	M	> 2
P9	M/57	1 +	2	12	I	M	> 2
P10	M/49	1 +	3	15	C	A	> 2
P11	F/43	2 +	1	7	I	M	> 2
P12	M/48	1 +	1	8	I	M	2
P13	M/56	2 +	2	11	I	M	2
P14	F/34	3 +	1	10	I	M	2
P15	M/51	2 +	3	17	I	M	2
P16	F/43	2 +	2	14	I	M	2
P17	F/41	1 +	1	10	C	A	> 2
P18	M/44	2 +	1	8	I	M	> 2
P19	M/38	3 +	2	7	C	A	> 2
P20	M/56	3 +	3	18	C	A	> 2
P21	F/42	3 +	2	14	C	A	2
P22	F/47	2 +	2	16	I	M	> 2
P23	F/44	3 +	3	20	I	M	> 2
P24	M/53	3 +	3	19	C	A	> 2
P25	M/36	2 +	2	12	C	A	> 2

^§Patient number (PN).

^*Culture sputum (CS) test result.

^**Diagnosed infection (years), confirmed by positive sputum culture.

^†Duration of treatment (DT) in months.

^††Chest X-ray results. C, cavity; I, infiltration without cavity.

^‡Disease stage (DS): A, advanced; M, moderate.

^‡‡Anti-biogram test (AT) results: 2, resistance to isoniazid and rifampcin: >2 resistance to other antibiotics in addition to isoniazid and rifamcin.

Exclusion criteria for all the subjects were: human immunodeficiency virus (HIV), hepatitis C virus (HCV) antibody-positive, hepatitis B surface antigen (HBsAg)-positive, any known concurrent infection, allergy and asthma, graft organ-implanted individuals and age < 14 and > 70 years.

From all subjects, 15–20 ml of heparinized whole blood was obtained by venipuncture at the same time each day.

Preparation of cells

Peripheral blood mononuclear cells (PBMCs) were isolated by Lymphodex (Inno-Train, Germany) density centrifugation according to the standard protocol. Viability of the PBMCs was determined by Trypan blue (0·4%) and counted. The cells were resuspended in complete RPMI-1640 medium (10 mM HEPES buffer, 200 mm l-glutamine, 50 U of streptomycin-penicillin/ml, all from Gibco, Auckland, New Zealand) and cultured in plastic tissue culture flasks (Nunclon, Nunc A/S, Kamstrupvej, Denmark) for 2 h at 37°C to allow firm adherence of the monocytes. Then non-adherent PBMCs were isolated and the monocytes were cultured in complete medium supplemented with 10% AB serum (Sigma, Germany) and 200 µ/ul IL-4 (R&D Systems, Minneapolis, MN, USA) and 400 µ/ul GM-CSF (Roche, Mannheim, Germany). On the third day, the same doses of IL-4 and GM-CSF were added, and on the fifth day the immature dendritic cells were detached by 5 mM ethylenediamine teraacetic acid (EDTA) in PBS and irradiated with 5000 rads. Fluorescence activator cell sorter (FACSCaliber) flow cytometry (Becton-Dickinson, San Jose, CA, USA) and cellquest software were used for the analysis of CD14, CD1a, -b, -c and human leucocyte antigen D-related (HLA)-DR (DakoCytomation, Glostrup, Denmark A/S) expressions on the immature dendritic cells.

Magnetic cell sorting

For the enrichment of the CD4⁺ and CD8⁺ T cells, non-adherent PBMCs were incubated with anti-CD4 and anti-CD8 magnetic microbeads (Miltenyi-Biotec GmbH, Gladbach, Germany). The PBMCs were washed by PBS-0·05% EDTA and loaded onto mini-magnetic cell sorting (MACS) columns (Miltenyi-Biotec), placed in the magnetic field of a MACS Separator (Miltenyi-Biotec). The separated cells were analysed for CD4 and CD8 markers using anti-CD4, CD8 and anti-CD3 antibodies (DakoCytomation). Purity of the CD4⁺ and CD8⁺ T cells was determined by flow cytometry and resuspended in freezing medium (65% complete medium with 20% AB serum, 15% dimethylsulphoxide) and cryopreserved in liquid nitrogen.

3–{4,5-dimethylthiazol-2-yl}–2,5 diphenyl tetrazolium bromide (MTT) assay

Immature dendritic cells were cultured in flat-bottomed microtitre plates at 30 000 cells/well in 0·20 ml of a complete medium containing 10% AB serum. The cells werestimulated by M. tuberculosis total sonicate and total lipid antigens at 20 µg/ml concentration and incubated for 24 h. Autologus PBMCs and CD4⁺ and CD8⁺ T cells with a density of 100 000 cells/well were added to the immature dendritic cells and incubated for 72 h. The supernatants were collected for IL-4 and IFN-γ measurements. To the remaining cells, 0·1 ml of the medium and then 10 µl of labelling reagent MTT (Roche) at a final concentration of 0·5 mg/ml was added and incubated for 4 h. The cells were then mixed with solubilization solution (Roche) and incubated at 37°C overnight. Optical density was read at 570 nm wavelength using 650 nm wavelength references.

Cytokine assay

After 72 h incubation, supernatant fluids from the CD4⁺ and CD8⁺ T cell cultures were collected and cytokine concentrations in pg/ml were measured with Quantikine human IL-4 and IFN-γ immunoassay kits (R&D Systems) according to the manufacturer's instructions. The enzyme-linked immunosorbent assay (ELISA) assay was performed in duplicate for each sample and the cytokine concentrations were calculated using standard curves.

Statistical methods

Statistical analysis was performed by non-parametric analysis and Mann–Whitney U-test. A value of P < 0·05 was considered significant.

Results

Antigen evaluation

M. tuberculosis H37Rv total sonicate and total lipid antigens were prepared according to Materials and Methods. Whole lipid extracts were analysed by SDS-PAGE and silver staining to assess protein contamination in the extracted solution. As shown in Fig. 1a, no protein bands were observed. The extracted lipid was shown by thin-layer chromatography (TLC) (Fig. 1b). Based on the LAL assay with sensitivity of 0·125 EU/ml, lipopolysaccharide (LPS) was undetectable in M. tuberculosis total sonicate and total lipid extracts.

Fig. 1.

Biochemical analysis of M. tuberculosis total lipid extracts. (a) SDS-PAGE analysis of M. tuberculosis lipid extracts. Lipid extracts from M. tuberculosis H37Rv were analyzed by SDS-PAGE and silver staining to assess protein contamination. In Lane 1, 10 μg wide molecular protein marker and in Lane 2, 100 μg lipid extract were loaded. (b) Thin layer chromatography (TLC) analysis. M. tuberculosis total lipid extract was evaluated using TLC according to the Materials and Methods. In Lane 1, 100 μg M. tuberculosis total lipid extract and in Lane 2, 40 μg phosphadidylcholin (PC), phosphatidylinositol (PI) and phosphatidylethanolamine (PE) mixture were loaded.

Cell analysis

In order to study the effect of conventional T cell subtypes in induction of immune responses to M. tuberculosis lipidantigens, CD3⁺ CD4⁺ and CD3⁺ CD8⁺ T cells were separated from the non-adherent PBMCs by magnetic cell sorting according to Materials and Methods. More than 95% of the obtained cells from the magnetic cell sorting expressed CD4 and CD8 markers according to the flow cytometry analysis documents (Fig. 2).

Fig. 2.

Analysis of the separated CD3⁺ CD4⁺ and CD3⁺ CD8⁺ T cells: Nonadherent PBMCs were separated to the CD3⁺ CD4⁺ and CD3⁺ CD8⁺ T cells by magnetic cell sorting according to the Materials and Methods. Purity of the CD4 and CD8 T cells were analyzed by flowcytomery. (a) CD3⁺ CD4⁺ T cells analysis (FL1-H = Anti-CD3 FITC, FL2-H = Anti-CD4 PE). (b) CD3⁺ CD8⁺ T cell analysis (FL1-H = Anti-CD3 FITC, FL2-H = Anti-CD8 PE). Purity of the CD3⁺ CD4⁺ T and CD3⁺ CD8⁺ T cells was 98·04% and 98·06%, respectively.

Proliferation responses

For comparative analysis of proliferation responses in the MDR-TB patients with the other groups, the PBMCs and T (CD4⁺ and CD8⁺) lymphocytes were cultured with autologous immature dendritic cells primed by M. tuberculosis totalsonicate and total lipid antigens. After 72 h of stimulation, the proliferation responses were assessed using MTT. Proliferative responses of the PBMCs to total sonicate antigens revealed that there was no significant difference in the healthy donors in this regard, but there was a significant decrease in the TB and MDR-TB patients (P = 0·00). In contrast, the PBMC proliferative responses to total lipid antigens in the PPD-positive donors were significantly greater than in the other groups (P = 0·001), while the responses in the PPD-negative, TB and MDR-TB patients were similar (Fig. 3). Accordingly, PBMC proliferation responses to the total sonicate and total lipid antigens in the MDR-TB patients were lower than those in the healthy PPD-positive donors.

Fig. 3.

Proliferation responses of the PBMCs, CD4⁺ and CD8⁺ T cells to M. tuberculosis total sonicate and total lipid antigens. Proliferative responses of the PBMCs to both the total sonicate and total lipid antigens in the PPD-positive donors were significantly higher than in the other groups (P = 0·00), while the responses to total lipid antigens in the PPD-negative subjects, TB and MDR-TB patients were similar (P = 0·233). Proliferative responses of the CD4⁺ T cells to total sonicate and total lipid antigens in the PPD-positive subjects were higher than in the TB and MDR-TB patients (P = 0·00), while the responses to total lipid antigens in the PPD-negative, TB and MDR-TB patients were similar (P = 0·033) to each others. Also proliferative responses of the CD8⁺ T cells to the total sonicate and total lipid antigens in the all subjects were similar (P > 0·005).

For elucidation of T cell subtype importance in M. tuberculosis infection, the CD4⁺ and CD8⁺ T cell proliferative responses to the antigens in the four groups were analysed. Our findings showed that proliferative responses of the CD4⁺ T cells to the total sonicate antigens in the PPD-positive (OD = 0·501 ± 0·071) and PPD-negative (OD = 0·453 ± 0·074) healthy donors were greater than in the TB (OD = 0·408 ± 0·076) and MDR-TB (OD = 0·393 ± 0·060) patients (P = 0·002). On the other hand, these responses in the TB and MDR-TB patients showed no differences (Fig. 3). According to Fig. 3, proliferative responses of the activated CD4⁺ T cells by the total lipid antigens in the PPD-positive donors (OD = 0·417 ± 0·060) showed a significant increase relative to the PPD-negative healthy subjects (OD = 0·228 ± 0·066), TB (OD = 0·190 ± 0·051) and MDR-TB patients (OD =0·197 ± 0·042) (P = 0·001). To elucidate the role of the CD8⁺ T cells in M. tuberculosis infection, we also investigated the proliferative responses of these cells in all the groups. The results showed that proliferative responses of the CD8⁺ T cells to the total sonicate antigens had no significant differences in the four groups (P = 0·243), but the responses were weak compared to those of the CD4⁺ T cells. Proliferative responses of the CD8⁺ T cells stimulated by the total lipid antigens were very weak in the study groups and, in some cases, no responses were observed (Fig. 3). Taken together, proliferative responses of the CD4⁺ T cells to the total lipid antigens in the MDR-TB and TB patients were significantly decreased comparing to the PPD-positive donors. On the other hand, CD8⁺T cells responses in this regard were similar in all the groups.

IFN-γ production by CD4⁺ and CD8⁺ T cells

The CD4⁺ and CD8⁺ T cells of the MDR-TB, TB patients and healthy donors were stimulated by M. tuberculosis total sonicate (20 µg/ml) and total lipid antigens (20 µg/ml). After 72 h incubation, the supernatants were collected and their IFN-γ and IL-4 concentrations were measured by ELISA. Mean IFN-γ concentrations in each group were compared with those of the other groups by Mann–Whitney U-test.

IFN-γ

The results showed that mean IFN-γ concentrations in the MDR-TB and TB patients were significantly lower than their values in the PPD-positive healthy donors (P = 0·004). In addition, IFN-γ levels in the PPD-negative donors were lower than in the PPD-positive subjects (P = 0·063) (Fig. 4a).

Fig. 4.

IFN-γ production by the CD4⁺ and CD8⁺ T cells. (a) IFN-γ production by the CD4⁺ T cells. The mean of IFN-γ concentration in response to the total sonicate and total lipid antigens in the TB and MDR-TB patients was significantly lower than in the healthy donors (P = 0·004). IFN-γ titer in the TB and MDR-TB patients also were similar (P = 0·123). IFN-γ production in response to the total lipid antigens in the PPD-positive donors was higher than in the other groups (P = 0·00). In the TB and MDR-TB patients, level of IFN-γ production was lower than in the PPD-negative donors (P = 0·002). (b) IFN-γ production by the CD8⁺ T cells. IFN-γ production in the CD8⁺ T cells stimulated by total sonicate antigens was higher in the PPD-positive subjects than in the other groups, but statistically it was not significant (P = 0·123). Differences in IFN-γ response to the total lipid antigens in the four groups were not significant (P = 0·134). : Cells without antigenic stimulation, □: Cells stimulated with PHA, : Cells stimulated by total sonicate antigens, : Cells stimulated with total lipid antigens.

Evaluation of IFN-γ production by the CD4⁺ T cells in response to the total lipid antigens showed that the mean IFN-γ concentration in the PPD-positive subjects was significantly higher than in the PPD-negative healthy donors (P = 0·003). On the other hand, in the TB and MDR-TB patients, the mean IFN-γ concentration was lower than in the healthy PPD-negative and PPD-positive donors (P = 0·00) (Fig. 4a). Based on our findings, there was no significant difference in the mean IFN-γ concentration in the CD8⁺ T cell culture supernatant in response to the total sonicate antigens among the different groups. Besides, only a weak IFN-γ production response to the total lipid antigen was detected (P = 0·012) (Fig. 4b). According to these results, IFN-γ production in the stimulated CD4⁺ T cells by total lipid antigens in the TB and MDR-TB patients was suppressed in comparison to the healthy donors.

IL-4

IL-4 concentration in the supernatants of the CD4⁺ and CD8⁺ T cells were assessed simultaneously with IFN-γ, according to Methods and methods. In the MDR-TB and TB patients, production of IL-4 by CD4⁺ T cells in response to total sonicate antigens were significantly greater than its corresponding values in the PPD-positive and PPD-negative healthy donors (P = 0·004) (Fig. 5a), while it was similar between the PPD-positive and PPD-negative subjects. In response to lipid antigens, IL-4 concentration in the TB and MDR-TB patients was higher than in the healthy donors (P = 0·00), while it was similar between the PPD-positive and PPD-negative subjects (P = 0·233) (Fig. 5a). In response to the total sonicate antigens, production of IL-4 by the CD8⁺ T cells in the MDR-TB and TB patients was significantly higher in the healthy donors (P = 0·00). The CD8⁺ T cells stimulated by total lipid antigens did not produce significant IL-4 in all the groups (Fig. 5b). Collectively, the stimulated CD4⁺ T cells by total lipid antigens produced higher levels of IL-4 in the MDR-TB and TB patients than in the healthy donors.

Fig. 5.

IL-4 production by the CD4⁺ and CD8⁺ T cells. (a) IL-4 production by the CD4⁺ T cells. Mean of IL-4 concentration in response to the M. tuberculosis total sonicate antigens in the TB and MDR-TB patients was significantly greater than those in the healthy donors (P = 0·004). IL-4 concentration in the PPD-positive and PPD-negative donors was similar (P = 0·233). IL-4 production in response to the M. tuberculosis total lipid antigens in the TB and MDR-TB patients was higher than in the other groups (P = 0·00). (b) IL-4 production by the CD8⁺ T cells: IL-4 production by the CD8⁺ T cells stimulated by the M. tuberculosis total sonicate antigenn in the TB and MDR-TB patients was higher than in the healthy subjects (P = 0·003). IL-4 response in the M. tuberculosis total lipid antigens in the four group was not significant (P = 0·154). : Cells without antigenic stimulation, □: Cells stimulated by PHA, : Cells stimulated by total sonicate antigens, : Cells stimulated by total lipid antigens.

Discussion

MDR-TB is now a serious problem in the control and management of M. tuberculosis infection even in the developed countries [1,2]. The high prevalence of HIV infection in MDR-TB patients is of great concern. Therefore, further researches into the recognition of immune responses mechanisms and aetiology of MDR-TB is essential. In the present study, antigen preparation was carried out with LPS-free solutions and instruments. On the other hand, in M. tuberculosis, LPS exists in undetectable amounts [24]. Moreover, the total lipid antigens were used for stimulation of the cells in the same conditions in all the patients and healthy subjects. Several researchers have investigated the role of M. tuberculosis protein antigens in PBMCs and T cell stimulation [25–27]. While presentation of lipids and glycolipids antigens by CD1 molecules is an important pathway in T cell immune responses [15,16,28], the function of T lymphocytes against M. tuberculosis lipid antigens in TB and MDR-TB patients is less investigated. In the present study, for the first time, the function of CD4⁺ and CD8⁺ T cells in MDR-TB patients against M. tuberculosis total lipid antigens was assessed.

In agreement with previous studies [16,29–32], our results show that proliferation responses of the PBMCs cells to M. tuberculosis total sonicate antigens in the PPD-positive donors were significantly greater than in the TB and MDR-TB patients. It was also elucidated that proliferation responses to the CD4⁺ T cells from PPD-positive donors to total lipid antigens increased significantly in comparison to the other groups, whereas the same number of CD4⁺ and CD8⁺ T cells (100 000 cells/well) was used in MTT and cytokine assays. It seems that, in PPD-positive donors, total lipid antigens recruit T cells memory immune responses. On the other hand, it was revealed that such responses in the TB and MDR-TB patients were suppressed. It is possible that lipid-specific T cells were accumulated in the inflammatory tissues of the lung in TB and MDR-TB patients. Stenger et al. showed that expression of CD1 molecules on monocytes infected with M. tuberculosis was decreased [33]. Also, our data demonstrate (data not shown) that stimulated immature dendritic cells by total lipid antigens in the MDR-TB patients up-regulated IL-10 production.

Proliferation responses to the total lipid antigens were also observed in PPD-negative subjects who have not been infected previously with M. tuberculosis. It seems that there was cross-reactivity between M. tuberculosis lipid antigens and the other bacteria lipids. Hence, further studies on the mitogenic properties of M. tuberculosis lipid antigens are suggested. Our results, in agreement with the other studies [16], confirmed that CD8⁺ T cell proliferation responses to total lipid antigens was very weak, and the role of these cells in the pathogenesis of the TB and MDR-TB patients was not critical.

In analysis of IFN-γ and IL-4 production, our results show that IFN-γ production by CD4⁺ T cells in the PPD-positive donors in response to M. tuberculosis total sonicate and total lipid antigens increased significantly compared to the other groups. Other researchers have shown that the activated CD4⁺ T cells by M. tuberculosis lipid antigens from the PPD-positive healthy donors produced more IFN-γ than in the PPD-negative subjects [16]. According to other studies, IFN-γ production by the PBMCs in response to the 30 kDal M. tuberculosis antigen in the MDR-TB and TB patients was suppressed [34,35]. Similar studies have confirmed that the decrease of IFN-γ production in MDR-TB and TB patients was not due to the increase of IL-10 [35,36]. In addition, a high frequency of T cell apoptosis, a decrease in IL-12Rβ1, IL-12Rβ2 expression and up-regulation of TGF-β1 and IL-10 simultaneously in mycobacterial infection could be the cause of IFN-γ down-regulation in CD4⁺ T cells from the TB and MDR-TB patients [30–32,36,37]. However, the obvious reason for the depression of IFN-γ production by CD4⁺ T cells against M. tuberculosis lipid antigens in TB and MDR-TB patients is not well known. It is postulated that defects in CD1 expression on antigen-presenting cells in mycobacterial infection decreased lipid antigen presentation to the T cells.

Our results show that IFN-γ production in the stimulated CD8⁺ T cells by M. tuberculosis total lipid and total sonicate antigens in the TB and MDR-TB patients had no significant differences with the healthy donors. The results of a study by Smith et al. [38] have shown that intracellular IFN-γ production in the CD8⁺ T cells stimulated by M. tuberculosis in the TB patients was decreased significantly in comparison to the PPD-positive donors.

Based on our findings, CD4⁺ T cells of the TB and MDR-TB patients activated by M. tuberculosis total sonicate antigens produced high levels of IL-4 compared with the PPD-positive donors. A study by Zhang revealed that IL-4 production by T cells infected with live M. tuberculosis in TB patients and PPD-positive donors were similar [29]. A study by Bai et al. showed that IL-10 and IL-4 expression on the biopsies of granuloma tissues in pulmonary TB was reduced compared to PPD-positive subjects [39]. Based on the majority of studies, expression of IL-4 is elevated in T cells stimulated by M. tuberculosis protein antigens [36,38,40]. It seems that M. tuberculosis protein antigens and lipid antigens have similar behaviour in stimulation of CD4⁺ T cells.

Further to the present study, the CD8⁺ T cells from the TB and MDR-TB patients produced prominent IL-4 compared with the healthy donors. In agreement with our results, another study showed that CD8⁺ T cells in the TB patients shifted to IL-4 production [42]. According to other studies, increased IL-4 expression in the lymphocytes infected to M. tuberculosis promoted CD30 expression which, in turn, sensitized the lymphocytes to TNF-α-mediated apoptosis [24]. Therefore, IL-4 can suppress cell-mediated immunity in TB and MDR-TB patients. In another study, IL-4 knock-out mice developed large granulomas accompanied by a significant increase in lung colony-forming units (CFU) of M. tuberculosis[41]. Overall, these data show that IL-4 may also have a beneficial effect in the prevention of lung tissue destruction induced by T helper 1 (Th1) cells. Accordingly, investigation of the exact role of IL-4 in the pathogenesis of TB and MDR-TB patients is strongly suggested.

It is concluded that proliferation responses and IFN-γ concentrations in T lymphocytes stimulated by total lipid antigens in the TB patients were slightly decreased compared to MDR-TB patients, but the differences were not statistically significant. IL-4 concentration in the TB patients was also increased in comparison to MDR-TB patients, and not statistically significant (P = 0·01). The responses may be related to the clinical status and chemotherapy diet in the patients. Other studies also showed that 2 or more weeks after chemotherapy of the immune system functions were improved in the TB patients [18].

According to our results, CD4⁺ T cell proliferation responses and IFN-γ production to M. tuberculosis lipid antigens in the PPD-positive healthy donors were significantly higher than in the MDR-TB patients. Further, lipid antigens in all the patients promoted IL-4 expression in the CD4⁺ T cells. Therefore, it appears that M. tuberculosis lipid antigens, like protein antigens, play an important role in the specific immune response. Taken together, the role of M. tuberculosis lipid antigens should be considered in the designation of effective vaccines and treatment protocols.