Human in vitro immune responses to Mycobacterium tuberculosis

Share Embed


Descrição do Produto

Tubercle and Lung Disease (1999) 79(6), 371–377 © 1999 Harcourt Publishers Ltd Article no. tuld.1999.0223

Human in vitro immune responses to Mycobacterium tuberculosis M. L. Wilsher,* C. Hagan,† R. Prestidge,† A. U. Wells,* G. Murison† *Department of Respiratory Medicine, Green Lane Hospital, Auckland 3, New Zealand † Genesis Research & Development Corporation, 1 Fox Street, Parnell, Auckland 1, New Zealand

Summary Setting T helper cells can be divided into 2 subsets on the basis of their cytokine generation. T helper 1 cells secreting gamma interferon and interleukin 2 appear to be more prominent in patients with limited tuberculous disease. Objective The purpose of this study was to evaluate human T helper cell immune responses to mycobacterial antigens in vitro and correlate these with the clinical features of patients with tuberculous infection or disease. Design We studied 51 subjects and 11 controls who were grouped according to disease involvement as follows: 1) Mantoux negative, BCG negative, no disease; 2) Mantoux positive, no disease; 3) localized extrapulmonary; 4) healed pulmonary; 5) active pulmonary; and 6) miliary/disseminated. Peripheral blood mononuclear cells were cultured with PHA, PPD or Tetanus Toxoid, proliferation assessed and the supernatant analysed using an ELISA for IFNγ. ELISA was also used to measure M. tuberculosis specific antibodies in the serum. Results Mantoux size correlated with PPD proliferation r = 0.5, P = 0.005 and γIFN production r = 0.36, P < 0.01. All groups produced abundant γIFN although there was a trend toward higher production in groups 3 and 4. M. tuberculosis specific IgA (P = 0.003) and IgG1 (P = 0.002) was higher in groups 5 and 6. Those patients with limited disease (groups 2–4) had significantly lower levels of IgG4 than patients with severe disease (groups 5 & 6) (P < 0.02). Conclusion In conclusion patients with healed or extrapulmonary disease have immune responses in vitro suggestive of a TH1 (cell mediated immune) response, whereas patients with miliary/disseminated disease have antibody production suggestive of a TH2 response, together with high γIFN production. Both TH1 and TH2 responses may be necessary for host protection if there is a high bacillary load. © 1999 Harcourt Publishers Ltd INTRODUCTION The recognition that T helper cells on the basis of their cytokine generation can be divided into TH1 or TH2 cells has offered new insights into the host response to a variety of intracellular pathogens.1 T helper 1 (TH1) cells producing interleukin 2(IL2) and gamma interferon (γIFN) predominate in tuberculoid leprosy and confer protection in experimental leishmaniasis.2–5 On the basis of γIFN production they are also more dominant in healthy tuberculin reactors compared with tuberculosis patients.6,7 TH1 cytokines are also more prominent in tuberculous pleuritis, a manifestation of tuberculous disease that often resolves without chemotherapy and patients with less radiologically advanced pulmonary tuberculosis have Correspondence to: M. Wilsher, Department of Respiratory Medicine, Green Lane Hospital, Auckland 3, New Zealand Tel.: 00 64 9 6389909; Fax: 00 64 9 6310712; Email: [email protected] Received: 1 April 1999; Revised: 25 May 1999; Accepted: 22 June 1999

bronchoalveolar lavage lymphocytes that secrete γIFN and not IL4.8,9 TH2 cells producing IL4, IL5 and IL10 are found in lepromatous leprosy which might be considered an ineffective immune response to M. leprae.2,10 To date TH2 responses have not been shown to be increased in human tuberculosis, but peripheral blood mononuclear cells from patients with active pulmonary tuberculosis do have diminished production of the TH1 cytokines IL2 and γIFN.7,11 A TH1 response has been postulated by many workers as confering protection against M. tuberculosis whereas a TH2 response is associated with severe disease. We hypothesised that patients with Mantoux responsiveness alone or localised extrapulmonary disease such as pleural or node tuberculosis or obsolete pulmonary disease have an effective host response and may have different T cell immune responses in vitro to those patients with active pulmonary or miliary disease. Our aims in this study were to identify the cytokine profile of cultured peripheral 371

372

Wilsher, Hagan, Prestidge et al.

blood mononuclear cells from patients with a spectrum of tuberculous disease and also to measure M. tuberculosis specific immunoglobulin production by these cells. METHODS Patients and controls The patients and controls were grouped as follows: 1. Healthy volunteers (preclinical medical students) with negative Mantoux (to 5TU PPD), and no known exposure to M. tuberculosis and no prior BCG (n=11). 2. Healthy volunteers (hospital clinical staff) with positive Mantoux (≥ 10 mm) but no evidence of tuberculous disease (n=10). 3. Patients with localised disease (n=11), lymph node (n=6), pleural effusion (n=3), pericardial effusion (n=1), soft tissue abcess (n=1), (4 smear positive, 9 culture positive, 2 granulomatous inflammation responded to antituberculous treatment). 4. Patients with CXR evidence of healed pulmonary disease (apical fibrocalcific changes and positive Mantoux) who had never received treatment (n=7). 5. Patients with active pulmonary disease (8 smear positive, 16 culture positive) all had typical CXR changes of pulmonary tuberculosis.16 6. Patients with miliary or extensive disease (bilateral upper and lower lobe infiltrates or multisite disease) (n=5, 4 smear positive, 5 culture positive). With the exception of two patients with miliary tuberculosis, who succumbed to their disease, all patients with tuberculous disease responded well to treatment and had a good outcome. Whereas the majority of Mantoux nega-

tive controls were European the majority of patients were of non-European race, predominantly Polynesian (Maori/ Pacific Island), Asian or Indian reflecting the impact immigration has had on the incidence of tuberculosis in New Zealand. The demographic details of patients and controls are shown in Table 1. All of the Group 2 Mantoux positive subjects had prior BCG in adolescence but all had exposure to tuberculosis working with patients in a tuberculosis ward. No patients were HIV positive. Although two patients in group 3 were culture negative both had tuberculosis contact, positive Mantoux responses (18 mm, 35 mm) and both had lymphadenopathy with granulomatous inflammation on lymph node biopsy and responded to antituberculous chemotherapy. Patients with active tuberculosis were recruited within 2 weeks of receiving treatment. All subjects gave informed consent and the project was approved by the North Health Ethics Committee. METHODS Peripheral blood mononuclear cell preparation and culture Fifty millilitres of blood was removed by venipuncture. Heparinized blood was layered onto Ficoll Hypaque (Pharmacia, Sweden) and centrifuged at 800 g for 20 min. The mononuclear layer was removed and washed twice with PBS. The cells were cultured in triplicate at 1 × 105 cells/well in a volume of 200 ul with RPMI 1640 (Gibco BRL USA) + 10% heat inactivated autologous serum supplemented with 2 mM L-glutamine and 100 U/ml penicillin/streptomycin (complete RPMI). Cells were stim-

Table 1 Characteristics of controls and subjects Group

n

Mean age

Gender Race

Mean Mantoux (mm)

Smear positive

Culture positive

1. Mantoux –ve BCG –ve 2. Mantoux +ve

11 10

21 36

6M 5M

0 13

N/A N/A

N/A N/A

3. Localized extrapulmonary disease (node, pleura pericardium, abcess)

11

32

5M

18

4

9

7

48

7M

16

0

0

16

52

7M

16

8

16

5

50

2M

7

4

5

4. Healed disease 5. Active pulmonary

6. Miliary/disseminated

11E 4E 3A 2P 7P 1E 3O 5O 1A 8A 3E 3P 2O 3A 2P 1E

Race: E = European; P = Polynesian, (Maori or Pacific Island); A = Asian; O = Other (Indian subcontinent, Middle East, Central Africa).

Tubercle and Lung Disease (1999) 79(6), 371–377

© 1999 Harcourt Publishers Ltd

Human in vitro immune response to Mycobacterium tuberculosis

ulated with PPD (Central Animal Health Laboratory, NZ) or Tetanus Toxoid (Statens Seruminstitut, Denmark) for 5 days or PHA (Gibco BRL, USA) for 3 days at 37°C, 5% CO2. The concentration of PHA and Tetanus Toxoid used in the study was 1/500 and 1/20 respectively. PPD was used at 25 µg/ml. Non-stimulated cells were included as a control. After completion of the incubation period, cells were pulsed with 0.5 uCi of 3H thymidine (925 G Bq/ mmol, Amersham, USA) and incubated for a further 18 h before being harvested and placed on a liquid scintillation counter. The results are expressed as the mean of the triplicate cultures and presented as a stimulation index (SI). The SI was calculated as the ratio of the mean proliferative response in the presence of mitogen or antigen to the mean proliferative response in the absence of mitogen or antigen. A cut off value of 2 was assigned to exclude background reactivity. Cytokine measurement Peripheral blood mononuclear cells (PBMC) were cultured at 2 × 106 cells/ml in a volume of 1 ml in cRPMI. Cells were incubated with either PHA (1/1000), Tetanus toxoid (1/20), PPD (25 µg/ml), or media alone for both 24 and 48 hour periods after which the supernatants were collected and stored at –20°C. IL-4, IL-5, and γIFN concentrations were determined by modification of the enzyme-linked immunosorbent assay (ELISA).12 Cytokines and antibodies used were supplied as follows: γIFN: recombinant human γIFN (Genzyme, USA), anti-human γIFN and anti-human γIFN-biotin (Endogen, USA). IL-4: recombinant human IL-4 (Genzyme, USA) antihuman IL-4 and rabbit anti-human IL-4 (Immunex Corp, USA) and goat anti-rabbit IgG-biotin (Dako, Denmark) IL-5: recombinant human IL-5 (R and D systems, USA), rat anti-human IL-5 and rat anti-human IL-5-biotin (Pharmingen, USA). Measurement of mycobacterial specificimmunoglobulins The levels of mycobacterial-specific serum immunoglobulins IgG, IgG1, IgG4, IgA, IgM, and IgE were determined by ELISA assays developed in our laboratory (unpublished data). Sera were obtained from the study population and stored at –20°C. Monoclonal mouse antihuman IgG1-biotin and monoclonal mouse anti-human IgG4-biotin were obtained from Sigma. Goat anti-human IgG-biotin was obtained from Biosource International. Horseradish peroxidase conjugated goat anti-human IgA, IgM, and IgE, were obtained from Sigma. Avidin-HRP conjugate was obtained from Vector laboratories. Ninety-six-well flat bottom microtitre plates were coated with 50 µl of 20 µg/ml PPD in carbonate buffer © 1999 Harcourt Publishers Ltd

373

overnight at 4°C. Plates were then blocked with PBS/T (PBS, 0.05% Tween 20) at room temperature for 1 h. Serial dilutions of human sera were made in PBS/T and 50 µl volumes added to the plates which had been washed four times with PBS/T. Samples were incubated for 1 h at room temperature followed by 6 washes with PBS/T. For the IgA, IgE, and IgM ELISAs, their respective horseradish peroxidase-conjugated antibodies were added at a predetermined concentration in PBS. Plates were then incubated at room temperature for 1 h followed by 6 washes in PBS/T and the addition of substrate buffer as described earlier. For the IgG and IgG subclass-specific antibody ELISAs their respective biotinylated antibodies were added at a predetermined concentration in PBS/T. Plates were incubated at room temperature for 1 h followed by 6 washes and the addition of 50 µl of avidin-HRP conjugate at a dilution of 1/000 in PBS. Plates were incubated at room temperature for 1 hour followed by 6 washes with PBS/T. Substrate buffer was then added as described earlier. Colour intensity was measured with a microtitre plate reader. Results are expressed as: (Optical density of sample – non specific optical density) × 1000.

Statistical analysis Group data were compared using non-parametric methods (Wilcoxon’s rank-sum test or the Kruskal–Wallis one-way equality of populations rank test, as appropriate). Univariate correlations between Mantoux reactivity and serum indices were evaluated using Spearman’s rank correlation. Statistically significant correlations were reexamined using a multivariate model, with Mantoux reactivity as the dependent variable; zero-skewness logarithmic transformation was used to normalize the distribution of individual variables and the validity of the assumptions of linear regression was confirmed using residual-versus-predictor plots and leverage-versusresidual plots. For some analyses, subjects from groups 2, 3 and 4 were combined and compared with groups 5 and 6 combined in order to compare limited disease (a good host response) with severe disease (a weak host response).

RESULTS Mantoux reactivity There was a significant difference between the 6 groups with respect to Mantoux size (P = 0.000 Kruskal–Wallis), but when the Mantoux-negative group 1 was excluded from analysis there was no significant difference between the remaining groups 2–6. There was a trend, however, for groups 4 and 5 to have larger Mantoux responses and half of group 6 (miliary disease) had negative Mantoux Tubercle and Lung Disease (1999) 79(6), 371–377

374

Wilsher, Hagan, Prestidge et al.

responses. There was no difference in Mantoux size between the various racial groups. Mantoux size correlated positively with SI (r = 0.37, P = 0.01) (Fig. 1), and γIFN (rs = 0.35, P = 0.01) (Fig. 2) but was unrelated to IgG1, IgG4, IgA or IgM levels. On multivariate analysis, Mantoux size was independently and positively related to γIFN (P < 0.001) and SI (P < 0.0005), equation R2 = 0 44; after the exclusion of normal control subjects Mantoux size was positively related to SI (P < 0.02) but not to γIFN (P = 0.14), equation R2 = 0.16. These relationships were unaltered after the exclusion of four and three outliers respectively (identified using leverage-versus-residual plots).

Fig. 3 Production of γIFN by PBMC from normal Mantoux negative controls (1), normal controls with positive Mantoux (2), patients with focal disease (3), healed pulmonary disease (4), active pulmonary disease (5), miliary/disseminated disease (6). (P = 0.04, Kruskal–Wallis). The horizontal bars represent median values.

Proliferative assays There was no significant difference in proliferation to PPD (SI) between the 6 groups (P = 0.074 Kruskal–Wallis), although the trend towards higher SI was observed in groups 2–4 and 5. As noted above there was a modest correlation between SI and Mantoux size (r = 0.37, P = 0.01) (Fig. 1). Fig. 1 Correlation (Spearman) between Mantoux reactivity and proliferation to PPD (SI) in normal controls and patients with tuberculous infection or disease (rs = 0.37, P = 0.01).

Cytokine assays There was a significant difference in γIFN production between the 6 groups (P = 0.04 Kruskal–Wallis), but when the Mantoux negative controls were excluded this difference disappeared (P = 0.2) (Fig. 3). γIFN production was highest in groups 3) and 4) and γIFN correlated with Mantoux size (r = 0.35, P = 0.01) (Fig. 2) as noted above, but not SI (r = –0.142, P = NS). IL4 and IL5 were very difficult to detect in the supernatants of the cultures. Where measurable for both it was only at the level of detection of the assay (50 pg/ml IL4, 20 pg/ml IL5) and so this data has not been reported. The only patients in whom IL4 or IL5 was detected came from groups 5 and 6. Mycobacterial-specific immunoglobulin assays

Fig. 2 Correlation (Spearman) between Mantoux reactivity and γIFN production by PBMC from normal controls and patients with tuberculous infection or disease. The clear circles represent the Mantoux negative controls (rs = 0.35, P = 0.01).

Tubercle and Lung Disease (1999) 79(6), 371–377

IgA levels were significantly higher in groups 5 and 6 (P = 0.003 Kruskal–Wallis). On comparing normal controls with patients with limited disease (groups 2–4) and severe disease (groups 5 & 6), this difference was highlighted (Fig. 4). Patients with severe disease also had higher IgG1 levels (P = 0.002) (Fig. 5) compared with © 1999 Harcourt Publishers Ltd

Human in vitro immune response to Mycobacterium tuberculosis

375

DISCUSSION

Fig. 4 Serum mycobacterial-specific IgA levels in normal controls, patients with limited disease (groups 2–4) and patients with severe disease (groups 5 & 6). (P = 0.003 Kruskal–Wallis). The horizontal bars represent median values.

Fig. 5 Serum mycobactecrial-specific IgG1 levels in normal controls, patients with limited disease (groups 2–4) and severe disease (groups 5 & 6) (P = 0.002 Kruskal–Wallis).

limited disease; the Mantoux negative controls had intermediate values. Those patients in group 6 had higher IgG4 and IgE levels than other 5 groups, although this did not reach statistical significance. However, when the patients with limited disease (groups 2–4) were compared with severe disease (groups 5 & 6) a significant difference in IgG4 production did become apparent (P < 0.02 Wilcoxon). There was a weak correlation between IgA levels and IgG1 (r = 0.26, P = 0.037), and IgA and IgG4 (r = 0.34, P = 0.008). There was no difference in IgM levels between the groups, although the highest values were seen in group 6. On multivariate analysis γIFN did not independently negatively correlate with IgG1 or IgG4. © 1999 Harcourt Publishers Ltd

These results lend some support to the hypothesis that a TH1 response confers protection against M. tuberculosis in that patients with untreated healed tuberculosis or with localized extra pulmonary disease tended to have strongly positive Mantoux responses, high PPD responsiveness and high levels of γIFN production. In addition these patients had relatively little mycobacterial-specific immunoglobulin production and less IgG1 production than Mantoux negative controls. Conversely patients with active pulmonary or miliary disease had high levels of mycobacterial-specific immunoglobulin in their serum, particularly IgG1 and IgA. These were also the only patients who had measurable levels of IL4 and IL5. Although the PPD stimulation index and Mantoux response was lower in patients with more extensive disease, they still produced abundant γIFN in vitro. Those patients with miliary/active pulmonary disease produced as much γIFN as the patients with healed or localized extrapulmonary disease. Normal controls also produced γIFN but this was clearly less than the patient groups. The high levels of γIFN in 6/11 Mantoux negative controls may reflect environmental exposure to non-tuberculous mycobacteria as none of these controls had prior known exposure to MTb. Skin testing for environmental mycobacteria is not available in our laboratory. Our study which has attempted to carefully characterize a cohort of patients with a spectrum of tuberculous disease provides results which suggest that both arms of the immune response are employed in the host response to M. tuberculosis and this is particularly evident in those patients with more extensive disease. The IgA response is interesting in that the highest production of this mycobacterial-specific immunoglobulin was seen in patients with extensive disease. Although IgG1 and IgG4 are traditionally thought to rise in a TH2 like immune response, IL10 has been reported to act with transforming growth factor β to promote IgA production in a murine model.13 Since IL10 is a known product of TH2 cells, the level of IgA production may support the hypothesis that extensive disease in human tuberculosis is associated with a TH2 cytokine response. Humoral and cell mediated immunity have long thought to be inversely related in tuberculosis and patients with advanced or miliary disease have previously been reported to have reduced Mantoux responses indicating impaired cell mediated immunity.14 Such patients can however mount vigorous specific immunoglobulin responses in spite of reported deficiencies of IL2 and γIFN production, although there does seem to be considerable heterogeneity in their immune response to M. tuberculosis.7 This is borne out by our study in which two patients with extensive disease had strongly positive Mantoux reTubercle and Lung Disease (1999) 79(6), 371–377

376

Wilsher, Hagan, Prestidge et al.

sponses, high levels of γIFN production and high levels of mycobacterial-specific immunoglobulins. Both of these patients also had measurable mycobacterial-specific IgE production. Huygen et al. described one similar patient although they did not report the extent of disease7 and Surcel et al. reported high levels of γIFN in 13 patients with active pulmonary disease and although they failed to make reference to the extent of disease, 10 patients were smear positive suggesting a high bacillary load.15 Huygen et al. did note the extent of disease in their patients and inversely correlated this with proliferation to PPD. In turn they observed an inverse correlation between lymphoproliferation and IgG antibodies to PPD. On closer examination of their data there also seems to be an inverse correlation between γIFN and IgG production. Their results are thus similar to those obtained in our study. Sharma et al. also reported a polyclonal hypergammaglobulinaemia in blood and bronchoalveolar lavage fluid from patients with miliary tuberculosis,16 although mycobacterial-specific immunoglobulins were not measured. Barnes et al. have reported that M. tuberculosis-reactive T cell clones from four healthy tuberculin responders show a heterogeneous TH1 and TH2 response based on cytokine production.17 However, they have also shown a predominent TH1 response at the site of disease in 12 patients with pleural tuberculosis.8 Given that this manifestation of tuberculous disease is associated with a good outcome, they speculated that γIFN contributes to an appropriate host response. They later demonstrated high levels of IL12 in the pleural fluid from such patients.18 In a subsequent study they showed that although peripheral blood T cells from 15 patients with active pulmonary tuberculosis produced less γIFN in response to M. tuberculosis in vitro compared with 11 healthy tuberculin reactors, there was no increase in the production of the TH2 cytokines IL4 or IL10.19 Further, an examination of mRNA expression in affected lymph nodes from 20 patients with tuberculous adenitis failed to show evidence of TH2 responses, but TH1-like γIFN producing cells were prominent, as were macrophages staining positively for IL12. These same authors concluded therefore that whilst PBMC from patients with active pulmonary tuberculosis have depressed TH1 responses in vitro there is enhanced production of γIFN at the site of disease in patients with tuberculous lymphadenitis. However, node, like pleural, tuberculosis is often associated with a good outcome. Examination of lung lymphocytes from those patients with active pulmonary tuberculosis may have yielded different results. This was done by Taha et al.20 who showed that a higher percentage of BAL cells from patients with active tuberculosis expressed mRNA for IFN and IL12 compared with those from patients with inactive disease. Condos et al.,9 also using BAL, showed that Tubercle and Lung Disease (1999) 79(6), 371–377

patients with less severe pulmonary disease had a local immune response characterized by γIFN but not IL4 production. Sanchez et al.6 using factor analysis showed an association between IL4, IgG and disease compared with PPD responsiveness, IL2 and γIFN production in 45 patients with pulmonary tuberculosis and 16 healthy tuberculin reactors. They speculated that γIFN and IL12 actively suppressed IL4 production in healthy tuberculin reactors following PPD stimulation and that in tuberculosis patients this suppressive mechanism failed. However, Lai et al. failed to show any difference in IL4 or γIFN gene expression in PBMC from 20 patients with active pulmonary tuberculosis compared with healthy tuberculin reactors.21 Clinically, their patients were thought to have mild disease. There can be little doubt from the murine work performed to date that γIFN confers very important host protection against M. tuberculosis and that the secretion of γIFN is mediated by IL12 production by macrophages. In common with the human in vitro immune response to other intracellular pathogens, e.g. M. leprae, L. major, a TH1 response seems essential for a good outcome in tuberculosis. It is possible that a TH2 response confers additional protection and that this is manifest in patients with a large bacillary load or severe disease but equally likely that a TH2 response contributes to tissue destruction and an adverse outcome. A similar situation may exist in schistosomal disease where cell mediated immunity is important in resistance but that specific IgE and IgA antibodies confer protection against reinfection.22 Murine malarial infection similarly requires a dual TH1 and TH2 response for protection.23 Our study attempted to correlate activity of disease with the in vitro immune response to M. tuberculosis. Although we like others cannot demonstrate a clear dichotomy between TH1 responses and immune protection versus TH2 responses and disease susceptibility, we have demonstrated that patients with active pulmonary disease and miliary disease can produce abundant γIFN together with high levels of mycobacterial specific antibodies suggesting that TH1 and TH2 responses are both important in the host response to a high bacillary load. ACKNOWLEDGEMENTS We thank the nursing staff of the tuberculosis ward for helping with recruitment and Margaret McKinlay and Allison Kelly for typing the manuscript. This work was supported by the Health Research Council of New Zealand and the New Zealand Lottery Grants Board.

REFERENCES 1. Mossmann T R, Coffman R L. TH1 and TH2 cells: Different patterns of lymphokine secretion lead to different functional properties. Ann Rev Immunol 1989; 7: 145–173.

© 1999 Harcourt Publishers Ltd

Human in vitro immune response to Mycobacterium tuberculosis

2. Yamamura M, Uyemura K, Deans R J et al. Defining protective responses to pathogens: cytokine profiles in leprosy lesions. Science 1991; 254: 277–279. 3. Haanen J B A G, Malefijt R de W, Res P C M et al. Selection of a human T helper type 1-like T cell subset by mycobacteria. J Exp Med 1991; 174: 583–592. 4. Kemp M, Kurtzhals J A, Bendtzen K et al. Leishmania donovanireactive Th1- and Th2-like T-cell clones from individuals who have recovered from visceral leishmaniasis. Infect Immun 1993; 61: 1069–1073. 5. Heinzel F P, Schoenhaut D S, Rerko R M, Rosser L E, Gately M K. Recombinant interleukin 12 cures mice infected with Leishmania major. J Exp Med 1993; 177: 1505–1509. 6. Sánchez F O, Rodríguez J I, Agudelo G, García L F. Immune responsiveness and lymphokine production in patients with tuberculosis and healthy controls. Infect Immune 1994; 62: 5673–5678. 7. Huygen K, Van Vooren J P, Turneer M, Bosmans R, Dierckx P, De Bruyn J. Specific lymphoproliferation, interferon production, and serum immunoglobulin G directed against a purified 32 kDa mycobacterial protein antigen (P32) in patients with active tuberculosis. Scand J Immunol 1988; 25: 187–194. 8. Barnes P F, Lu S, Abrams J S, Wang, Yamamura M, Modlin R L. Cytokine production at the site of disease in human tuberculosis. Infect Immun 1993; 61: 3482–3489. 9. Condos R, Rom W N, Liu Y M, Schluger N W. Local immune responses correlate with presentation and outcome in tuberculosis. Am J Respir Crit Care Med 1998; 157: 729–735. 10. Sieling P A, Abrams J S, Yamamura M et al. Immunosuppressive roles for IL-10 and IL-4 in human infection. In vitro modulation of T cell responses in leprosy. J Immunol 1993; 150: 5501–5510. 11. Zhang M, Lin Y, Iyer D, Gong J, Abrams J S, Barnes P F. T-cell cytokine responses in human infection with mycobacterium tuberculosis. Infect Immun 1995; 63: 3231–3234. 12. Schumacher J H, Ogarra A, Shrader B et al. The characterization of four monoclonal antibodies specific for mouse IL-5 and

© 1999 Harcourt Publishers Ltd

13.

14.

15.

16.

17.

18. 19.

20.

21.

22.

23.

377

developments of mouse and human IL-5 enzyme-linked immunosorbent. J Immunol 1988; 141: 1576–1581. Hyland L, Hou S, Coleclough C, Takimoto T Doherty PC. Mice lacking CD8+ T cells develop greater numbers of IgA-producing cells in response to a respiratory virus infection. Virology 1994; 204: 234–241. McMurray D N, Acheverri A. Cell-mediated immunity in anergic patients with pulmonary tuberculosis. Am Rev Respir Dis. 1978; 118: 827–834. Surcel H-M, Troye-Blomberg M, Paulie S et al. Th1/Th2 profiles in tuberculosis, based on the proliferation and cytokine response of blood lymphocytes to mycobacterial antigens. Immunology 1994; 81: 171–176. Sharma S K, Pande J N, Singh Y N et al. Pulmonary function and immunologic abnormalities in miliary tuberculosis. Am Rev Respir Dis 1992; 145: 1167–1171. Barnes P F, Abrams J S, Lu S, Sieling P, Rea T H, Modlin R L. Patterns of cytokine production by mycobacterium-reactive human T-Cell clones. Infect Immun 1993; 61: 197–203. Zhang M, Gately M K, Wang E et al. Interleukin 12 at the site of disease in tuberculosis. J Clin Invest 1994; 83: 1733–1739. Lin Y, Zhang M, Hofman F M, Gong J, Barnes P F. Absence of prominent Th2 cytokine response in human tuberculosis. Infect Immun 1996; 64: 1351–1356. Taha R A, Kotsimbos T C, Song U, Menzies D, Hamid Q. IFN-γ and IL-12 are increased in active compared with inactive tuberculosis. Am J Respir Crit Care Med 1997; 155: 1135–1139. Lai C K W, Ho S, Chan C H S et al. Cytokine gene expresssion profile of circulating CD4+ T cells in active pulmonary tuberculosis. Chest 1997; 111: 06–11. Chensue S W, Terebuh P D, Warmington K S et al. Role of IL-4 and IFN-γ in Schistosoma mansoni egg-induced hypersensitivity granuloma formation. J Immunol 1992; 148: 900–906. Taylor-Robinson A W, Phillips R S, Severn A, Moncada S, Liew F Y. The role of TH1 and TH2 cells in a rodent malaria infection. Science 1993; 260: 1931–1934.

Tubercle and Lung Disease (1999) 79(6), 371–377

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.