Diverse humoral immune responses and changes in IgG antibody levels against mycobacterial lipid antigens in active tuberculosis

Yukiko Fujita1, Takeshi Doi1, Koji Sato2 and Ikuya Yano1

1 Japan BCG Central Laboratory, 3-1-5 Matsuyama, Kiyose-shi, Tokyo 204-0022, Japan
2 National Sanatorium Kumamotominami Hospital, 2338 Toyofuku Matsubase-machi, Shimomashiki-gun, Kumamoto 869-0593, Japan

Correspondence
Yukiko Fujita
y-fujita{at}bcg.gr.jp


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Humoral immune responses of active TB patients against six mycobacterial lipid antigens [trehalose 6,6'-dimycolate (TDM) from Mycobacterium bovis BCG (TDM-T) and Mycobacterium avium complex (TDM-M), trehalose 6-monomycolate (TMM) from M. bovis BCG (TMM-T) and M. avium complex (TMM-M), triacyl (PL-2) and tetraacyl (PL-1) phosphatidylinositol dimannosides] were examined by ELISA. IgG antibodies of TB patients with active disease reacted against the six lipid antigens distinctively, but heterogeneously. If tests were combined and an overall positive was scored cumulatively when any one of the six tests was positive, a good discrimination between patient and normal subject was obtained. A positive result in any one of the six tests was obtained in 91·5 % of all 924 hospitalized patients and 93·3 % of 210 patients at their first visit to the outpatient clinic. The IgG antibody response differed considerably from patient to patient, and the response patterns were grouped into several types. IgG antibody levels paralleled the bacterial burden; however, the smear-negative (culture-positive) patient group also showed high positive rates and mean ELISA {Delta}A values against the six lipid antigens. There were also marked differences in positive rate and mean {Delta}A values between cavity-positive and -negative groups, the former being higher than the latter. After anti-TB chemotherapy was initiated, IgG antibody levels decreased dramatically, paralleling the decrease in the amount of excretion of bacteria. Since multiple-antigen ELISA using particular lipid antigens was highly sensitive, and IgG antibody levels vary greatly at different stages of the disease, this technique is applicable for early diagnosis of smear-negative (and -positive) active TB and the prognosis for completion of anti-TB chemotherapy.


Abbreviations: DOTS, directly observed treatment, short course; MAC, M. avium complex; NAA, nucleic acid amplification; PL-1, tetraacyl phosphatidylinositol dimannoside; PL-2, triacyl phosphatidylinositol dimannoside; TDM-M, TDM-T, trehalose 6,6'-dimycolate from M. bovis and M. avium complex, respectively; TMM-M, TMM-T, trehalose 6-monomycolate from M. bovis and M. avium complex, respectively; TB, tuberculosis; TST, tuberculin skin test


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Tuberculosis (TB) has long been one of the most important infectious diseases of humans. During the 20th century the incidence of TB decreased, owing to the development of anti-TB chemotherapy. Nevertheless, TB remains globally uncontrolled, especially in developing countries, owing to co-infection with HIV or environmental factors such as malnutrition and poverty. The diagnosis of active TB has been done based on the detection of acid-fast bacilli by smear test or cultivation; recently, the PCR or nucleic acid amplification (NAA) technique has been introduced to diagnose early stages of the disease. However, sensitivity of the smear test is low, cultivation takes a long time and NAA techniques are expensive. The tuberculin skin test (TST) is highly sensitive and useful as a criterion for delayed-type hypersensitivity based on the cellular immune response due to the Mycobacterium tuberculosis infection. However, this test is not applicable for differential diagnosis of active disease versus latent infection or the differential diagnosis of TB versus other mycobacterial infection, including BCG vaccination. Serological testing based on the humoral immune response is simple, economical and non-invasive and it is applicable to smear-negative (and -positive), culture-negative and -positive patients diagnosed clinically. To date, many investigations or trials on the serodiagnostics of TB using crude cell sonicate (Grange et al., 1980; Kardjito et al., 1982), culture filtrates (Benjamin et al., 1984; Kiran et al., 1985), purified protein antigens (Benjamin & Daniel, 1982; Zeiss et al., 1982) and cell wall lipids (Reggiardo et al., 1980; Sada et al., 1990) as antigens have been reported (for reviews see Chan et al., 2000; Daniel & Debanne, 1987). However, the practical use of serodiagnosis for active TB has not been widely introduced, apparently because the reliability of the current tests using protein antigen is not satisfactory (Chan et al., 2000; Daniel & Debanne, 1987). If these tools are to facilitate the diagnostic process and to increase case findings, a DOTS strategy would be more effective and a decrease in transmission would be expected (WHO, 2001). We have previously reported that IgG antibody against cord factor (trehalose 6,6'-dimycolate), the most characteristic cell surface glycolipid antigen, can be detected by ELISA with 81 % sensitivity and 96 % specificity for active TB patients and 70 % sensitivity in smear-negative TB patients (He et al., 1991; Maekura et al., 1993). For a clinically reliable test to diagnose active TB, sensitivity and specificity are necessary. Recent studies show that the sensitivity of serodiagnostic testing in smear-negative TB cases is equal or rather higher than that of NAA tests (Iinuma et al., 2002; Maekura et al., 2003).

In this study we examined humoral immune responses of patients with active TB against various mycobacterial lipid antigens by ELISA. The purpose of this investigation was to determine the humoral immune responses of TB patients to representative mycobacterial lipid antigens and to improve the sensitivity of IgG antibody detection system for simple and precise diagnosis of active TB and other mycobacterioses. We found that the IgG antibody responses of the individual TB patient sera were extremely heterogeneous depending on the structure of the antigens. However, the multiple combination of selected lipid antigens gave a high sensitivity, and the IgG antibody levels against lipid antigens varied greatly according to the progression of the disease. Therefore, this test is useful as a criterion for the completion of anti-TB chemotherapy.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Serum samples and patients.
(a) Samples from hospitalized patients. A total of 924 serum samples were obtained from patients attending the National Sanatorium Tokyo Hospital, Kiyose, Tokyo. These patients were diagnosed to be suffering from pulmonary TB, based on clinical symptoms such as long-lasting coughing and sputum over 2 months, and chest X-ray examination. Smear test with acid-fast stain and culture examination with 2 % Ogawa medium were performed 1 month after admission to the hospital. The patients of this group stayed in the hospital for a further 1 month or longer during 1995 and 1996. (b) Samples from patients at their first visit to the outpatient clinic. A total of 210 patients before hospitalization were diagnosed to be suffering from TB clinically based on the typical symptoms and chest X-ray examination between 1995 and 1996 at the outpatient clinic of the National Sanatorium Tokyo Hospital. Among them, 157 were smear-positive, 22 were smear-negative, culture-positive and 31 were unknown.

Selection of healthy control subjects and determination of cut-off points for ELISA.
To select the healthy control subjects, we carefully chose individuals who had received a health examination including a chest X-ray. Among these, 70 % of individuals were BCG vaccinated when younger than 12 years old. However, the IgG antibody titres against mycobacterial lipid antigens were not elevated after BCG vaccination, although TST results were positive; hence a co-relationship between IgG antibody titre elevation and the TST results was not observed. Therefore, we simply determined that the normal range for IgG antibody titres was lower than the mean {Delta}A value +2SD of the healthy control {Delta}A.

The control subjects consisted of 85 sera from HIV-negative healthy adult individuals with no history of TB in their families, and included both TST-positive and TST-negative individuals.

Antigens.
Lipid antigens were isolated from heat-killed M. tuberculosis Aoyama B, Mycobacterium bovis BCG and Mycobacterium avium complex (MAC) serotype 16 strain. Lipids were extracted from packed cells with chloroform/methanol (2 : 1, v/v) and the solvent was evaporated off with a rotary evaporator. The lipids were first partially separated by solvent fractionation and then the acetone-soluble or chloroform-soluble, or tetrahydrofuran-soluble or -insoluble fractions obtained were further separated by thin-layer chromatography on silica gel (Uniplate, Analtech) with the solvent system chloroform/methanol/water (65 : 25 : 4, by vol.), chloroform/methanol/acetone/acetic acid (90 : 10 : 6 : 1, by vol.) or chloroform/methanol/water (90 : 10 : 1, by vol.). Antigenic lipids were purified until a single spot was obtained by repeating thin-layer chromatography, and these were identified by mass spectrometric analysis. Each component such as carbohydrates or fatty acids was analysed by gas-chromatography/mass spectrometry (GC/MS) after hydrolysis or methanolysis of the original lipids, and the molecular mass of the intact lipids was determined by matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry.

In a preliminary study, we tested the reactivity of 40 active TB patient sera (IgG antibodies) against 12 lipid antigens [TDM from M. tuberculosis Aoyama B, M. bovis BCG Tokyo 172 and BCG Connaught and MAC serotype 16 strain, sulfolipid I, triacyl (PL-2) and tetraacyl (PL-1) phosphatidylinositol dimannosides from M. tuberculosis Aoyama B, mycoside B and cardiolipin from M. bovis BCG Tokyo 172, TMM from M. bovis BCG Tokyo 172 and BCG Connaught and MAC serotype 16 strain]. None of the single antigens gave a sufficiently high positive rate to diagnose active TB, although up to 70 % of the 40 active TB patient sera tested positive against a single lipid antigen (data not shown). However, if an overall positive was scored when any one or more tests among the 12 antigens were positive, the positive rate increased to a much higher level than that obtained with a single antigen. Based on the antigenic reactivity, the stability in ELISA plates, the species specificity in mycobacteria and the availability of lipid antigens, we selected the following six antigens as most appropriate: TDM-T, TDM-M, TMM-T, TMM-M, PL-1 and PL-2.

MALDI-TOF mass analysis showed that TMM-T derived from M. bovis BCG Tokyo was a trehalose monomycolate possessing one molecule of {alpha}-, methoxy- or keto-mycolic acid ranging from C76 to C91 (the methoxy- and keto-mycolic acid subclasses predominated), while TDM-T from the same strain was a trehalose dimycolate possessing two molecules of mycolic acid from among the above subclasses or molecular species, thus showing a highly complex mixture of trehalose diesters with molecular masses ranging from 2660 to 2920 Da. TMM-M from MAC serotype 16 strain showed a different mass spectrometric pattern, showing the existence of C82 or longer wax ester-mycolic acids characteristic of the MAC, instead of the methoxy-mycolic acids of M. bovis BCG Tokyo. TDM-M from MAC serotype 16 strain also showed multiplex molecular ions in mass spectra, due to the combined {alpha}-, keto- and wax ester-mycolic acids. On the other hand, both PL-1 and PL-2 derived from M. tuberculosis Aoyama B were major components of membrane mannophospholipids possessing C16 : 0 and/or branched-chain C19 (tuberculostearic) fatty acids. Thus, TDM (-T and -M) and TMM (-T and -M) were species-specific mycobacterial cell wall antigens, although they shared part of their molecular structure and were therefore cross-reactive. PL-1 and PL-2 were essentially common antigens in the mycobacterial strains. By the combination of above six antigens, the multiple-antigen ELISA gave a highly positive rate, reliable for the diagnosis of active TB and other mycobacterioses.

Multiple-antigen ELISA microplate system.
TDM-T and TDM-M were dissolved in n-hexane (0·2 µg TDM in 50 µl n-hexane per well), while TMM-T, TMM-M, PL-1 and PL-2 were dissolved in ethanol (1·0 µg TMM, PL-1 and 2·0 µg PL-2 in 50 µl ethanol per well). One antigen was deposited in a polystyrene microplate well (Nunc-Immunoplate; Nalge Nunc International) and the plates were allowed to dry at room temperature overnight. Plates were either used immediately or sealed in aluminium foil and stored at 4 °C. We confirmed that no difference was observed in results between plates used immediately after preparation and those stored for a month (data not shown). All chemicals used were purchased from Wako Pure Chemical Industries. For multiple-antigen ELISA, non-specific binding was blocked with 150 µl per well 0·05 % Tween 20 in phosphate-buffered saline (PBS-T) adjusted to pH 7·4 with 5 M NaOH and incubated for 10 min at room temperature. The plates were then washed three times with 250 µl PBS-T per well. Serum samples tested for determination of the titre of IgG antibodies were diluted 1 : 201 with PBS-T. The diluted serum samples (50 µl per well) were added to each well, and the plate was incubated for 1 h. Horseradish-peroxidase-conjugated goat anti-human IgG (H+L) (Zymed) diluted 1 : 500 in PBS-T was used as a secondary antibody. After incubation for 1 h, the substrate, o-phenylenediamine (Sigma; 1 mg ml–1) in citrate buffer containing 0·06 % H2O2 was added. The reaction was stopped with 0·5 M H2SO4, and absorbance was measured in a microplate reader (NPR-A4i; Tosoh) at 492–600 nm. Each incubation was performed at room temperature, and after each step of the procedure the plates were washed three times with PBS-T.

Classification of types of pathological lesion based on chest X-ray examination.
The extent and severity of the pathological lesion, with or without cavity, were classified based on chest X-ray examination, according to the 1963 ‘Gakken’ classification of pulmonary TB. Extent of disease or the basic lesion including exudative, caseous, fibrotic, disseminated or far-advanced mixed was classified as follows: I, largest (larger than one side of the lung lobe); II, middle (intermediate between I and III); and III, small (less than one third of one side of the lung lobe). Side was classified as r (right lobe only), l (left lobe only) or b (both lung lobes). Severity was classified as 1 (unstable lesion without cavity confined in one third of one side of the lobe), 2 (lesion limited to one side of the lobe with or without cavity) or 3 (larger lesion with cavity in more than one side of the lobe). For example, ‘bI3’ indicates a large caseous lesion with large cavities over both lung lobes, and ‘rIII1’ indicates a small but unstable caseous and fibrotic lesion without cavity confined to one third of right lobe.

Data analysis.
The titre of serum IgG antibodies against TDM-T, TDM-M, TMM-T, TMM-M, PL-1 and PL-2 was expressed as the absorption index ({Delta}A), equal to the absorption value of the test serum sample minus that of the same serum sample in wells without coated antigen. Statistical analysis was done according to the Mann–Whitney U test.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
IgG antibody responses against single or multiple lipid antigens
We tested the reactivities of sera of 924 hospitalized patients with TB and of 210 active TB patients on their first visit to the outpatient clinic. The data for sensitivity, specificity and positive predictive values (PPV) of all these patient sera against each single antigen and six multiple antigens are summarized in Table 1. Among these six antigens, the positive rate (sensitivity) to PL-2 was the highest, at 71·8 %, and that to TDM-M was the lowest, at 53·5 %, in the total hospitalized TB patients. When the tests were combined and an overall positive was scored cumulatively, a positive result in any of the six tests was obtained for 91·5 % of all 924 hospitalized patients and 93·3 % of the 210 patients at their first visit to the outpatient clinic, including smear-positive and -negative (culture-positive) cases. In both groups, the multiple-antigen ELISA results were superior to any single-antigen tests for the clinical diagnosis of TB and other mycobacterioses.


View this table:
[in this window]
[in a new window]
 
Table 1. Sensitivities of single- and multiple-antigen ELISA in the two groups of TB patients

 
Diverse responsiveness of IgG antibody against lipid antigens
To clarify the reason for the superiority of multiple-antigen ELISA, we carefully compared the IgG antibody response patterns of each patient serum against each of the six lipid antigens.The patterns of IgG antibody responsiveness against each lipid antigen were extremely diverse in the TB patient sera (Fig. 1). Most of the smear-positive TB patient sera were more or less reactive against at least one of the six antigens, although some were only poorly reactive. The cases of Fig. 1(c) and (d) were reactive against only mycoloyl glycolipids (TDM-T, TDM-M, TMM-T and TMM-M), not against mannophospholipids (PL-1 and PL-2). In contrast, the cases of Fig. 1(e) and (f) were reactive against mannophospholipids, but not against mycoloyl glycolipids. Further, some were reactive against only TMM, but not TDM (Fig. 1g and h) and some were reactive against only TDM, but not TMM. The cases of Fig. 1(i) and (j) were highly reactive against TDM-T and TMM-T, but not against TDM-M and TMM-M, indicating recognition of the difference of the mycoloyl moiety of trehalose esters. The majority of IgG antibody titres of active TB patient sera were highly elevated, and reactive against all the six antigens (Fig. 1a and b). Some patient sera had negative IgG antibody titres at their first visit to the outpatient clinic (Fig. 1k), but after 2 or 3 weeks, the IgG antibody titres against TDM-T and PL-2 were distinctly elevated (data not shown). From the IgG antibody response patterns, we suggest that such diverse reactivity of active TB patient sera against particular lipid antigens may be partially due to the clinical background, such as the amount of excreted bacilli or the stage after onset or progression of the disease, but mainly due to the difference of individual genetic background of the patient. To reveal the immunological and clinical characteristics of each six lipid antigens, we followed up the clinical documents of each active TB patient in as much detail as we could.



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 1. Heterogeneity and diversity of IgG antibody responses at their first visit to the outpatient clinic. (a) Male, 51 years, smear (+), culture (+), multidrug resistant, relapse TB; (b) male, 45 years, smear (+), culture (+), relapse TB; (c) male, 61 years, smear (+), culture (+), primary TB; (d) male, 47 years, smear (paucibacillary), culture (+), primary TB; (e) female, 78 years, smear (+), culture (+), primary TB; (f) male, 67 years, smear (+), culture (+), primary TB; (g) male, 23 years, smear (paucibacillary), culture (+), primary TB (pleuritis); (h) male, 67 years, smear (+), culture (+), primary TB (pleuritis); (i) male, 44 years, smear (+), culture (+), primary TB (pleuritis); (j) male, 37 years, smear (+), culture (+), primary TB; (k) male, 51 years, smear (+), culture (+), primary TB (pleuritis). There are several types of IgG antibody response to lipid antigens: (a, b) reactive to all six lipid antigens; (c, d) reactive to TDM mainly and TMM weakly, but not to PL; (e, f) reactive to only PL markedly; (g, h) reactive to only TMM; (i, j) reactive to TDM-T and TMM-T, but not or slightly to TDM-M and TMM-M; (k) low-reactive to all six lipid antigens (slightly reactive to PL-2). All ELISA conditions were the same and are described in the text.

 
Time-course changes in IgG antibody titres after initiation of anti-TB chemotherapy
The time-course changes of the IgG antibody titres against each lipid antigen were examined in active TB patient sera. Fig. 2(a, b, e and f) shows the initial changes in IgG antibody titres against the six mycobacterial lipid antigens during the first month after anti-TB chemotherapy was started. In the cases of Fig. 2(a, b and e), IgG antibody titres against PL-1 and/or PL-2 were already elevated when they first visited the outpatient clinic at the initial stage of the disease, whereas in case Fig. 2(f), IgG antibody titres against TMM-T, but not TDM, were elevated. After anti-TB chemotherapy was initiated, the elevated IgG antibody titres against PL-1 and/or -2 decreased immediately (Fig. 2a and b), whereas those against TDM were sharply elevated (Fig. 2a, b, e and f). Fig. 2(c, d, g and h) shows four representative active TB cases whose smear tests were all positive (Gaffky +8 to +10) and IgG antibody titres were fully elevated against six lipid antigens at the first examination. In these cases, the initially elevated IgG antibody titres decreased altogether at once, or increased for about a month and then declined sharply until the smear test became negative. Thus, the decline in IgG antibody titres against lipid antigens paralleled the decrease in the amount of excreted acid-fast bacilli, and IgG antibody titres had settled down near to the normal level at 3–4 months after the anti-TB chemotherapy was initiated and the smear or culture test became negative. In cases with prolonged infection, a tendency for a long-lasting high level of IgG antibody titres was observed. In the chemotherapy-failed or relapsed cases also, declining of IgG antibody titres retarded longer (for 5–6 months) or the titres were elevated again. Therefore, it was concluded that even in the multibacillary active TB cases with cavity, fully elevated IgG antibody titres against lipid antigens had settled down near to the normal level, when the anti-TB chemotherapy was completed successfully.



View larger version (31K):
[in this window]
[in a new window]
 
Fig. 2. Changes in IgG antibody levels after initiation of anti-TB chemotherapy. (a) Male, 67 years, smear (+), culture (+), primary TB (bII3); (b) male, 37 years, smear (+), culture (+), primary TB (bII2) (results for PL-1 and PL-2 almost identical); (c) male, 51 years, smear (+), culture (+), relapsed TB (rII2); (d) male, 38 years, smear (+), culture (+), primary TB (pleuritis, bII2); (e) male, 39 years, smear (–), culture (+), primary TB (rII3); (f) male, 46 years, smear (+), culture (+), relapsed TB (bII2); (g) male, 67 years, smear (+), culture (+), primary TB (bII3); (h) male, 43 years, smear (+), culture (+), primary TB (pleuritis, bII1). IgG antibody titres of eight active TB patient sera were estimated against six mycobacterial lipid antigens at different time points after initiation of anti-TB chemotherapy. The arrow shows the initiation point of anti-TB chemotherapy. In (a), (b), (e) and (f), IgG antibody titres were estimated during the first month after anti-TB chemotherapy was started. The final IgG antibody titre estimation of (c), (d), (g) and (h) was carried out when the smear tests were shown to be negative. IgG antibody titres against PL-2 and/or TDM-T were elevated initially, and decreased dramatically after initiation of the anti-TB chemotherapy. In general, 16–17 weeks after the initiation of anti-TB chemotherapy, all the IgG antibody titres were decreased to near the healthy control level. In cases (b), (e) and (f), IgG antibody titres against TDM-T were low at their first visit, but those were distinctly elevated 3–4 weeks after initiation of anti-TB chemotherapy.

 
Effect of amount of excreted acid-fast bacilli on the IgG antibody responses
Since the IgG antibody levels of active TB patient sera varied considerably, we compared the positive rate and IgG antibody level expressed as mean {Delta}A values for each lipid antigen, according to the bacterial burden. As shown in Table 2, at the first examination the multibacillary group, who excreted more acid-fast bacilli (Gaffky +8 to +10), showed significantly higher IgG antibody levels to all six lipid antigens, expressed as mean {Delta}A values, in ELISA than the paucibacillary group (Gaffky +1 to +3) or moderate bacillary group (Gaffky +4 to +7). However, unexpectedly, the lowest positive rates and IgG antibody levels were found in the paucibacillary group, but not in the smear-negative group, to TDM, TMM and PL-2. When we take scores for the total six antigens cumulatively, the positive rate ranged between 82·8 % in the paucibacillary cases and 95·5 % in multibacillary cases, and the smear-negative group showed a 95·5 % positive rate, indicating that the multiple-antigen ELISA gave satisfactory results reliable for the initial diagnosis of active TB, irrespective of whether smear-positive or -negative.


View this table:
[in this window]
[in a new window]
 
Table 2. Effect of numbers of excreted mycobacteria on the IgG antibody seropositive rate in active TB patients at their first visit to the outpatient clinic

 
Effect of the severity (or extent) of the pathological lesions (cavity-positive or -negative) on the IgG antibody responses
Since the IgG antibody levels of active TB patient sera varied greatly according to the stage of the disease, we compared the positive rate for each single and total (multiple) antigens of various patient groups with the different degrees of pathological lesion as determined by chest X-ray examination. Generally, the patients with more extensive or more severe pathological lesions showed a higher positive rate (Table 3). However, the highest positive rate varied among the antigens. TDM-M, TMM and PL-2 showed the highest positive rate for b, l or rI (2 or 3) cases, while TDM-T and PL-1 showed the highest for bII (1, 2 or 3) cases. The TB patients with cavitary lesion showed a higher positive rate than those without cavity, for all six mycobacterial lipid antigens (Table 4). These results indicate that the elevation of IgG antibody titres against mycobacterial lipid antigens is not homogeneous, but varies greatly with the progression or stages of active TB, related directly or indirectly to changes in the amount of acid-fast bacilli excreted and the extent and severity of pathological lesion.


View this table:
[in this window]
[in a new window]
 
Table 3. Effect of severity or extent of pathological lesions of active TB patients at their first visit to the outpatient clinic on the IgG antibody seropositive rate

 

View this table:
[in this window]
[in a new window]
 
Table 4. Effect of existence of cavity in the pathological lesion of active TB patients at their first visit to the outpatient clinic on the IgG antibody seropositive rate

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Studies on the humoral immune response in human patients infected with M. tuberculosis have been done since Arloing (1898) first reported the agglutination test for diagnosis of pulmonary TB. A serological test would have been expected to be the ‘gold standard’ or ‘Holy Grail’ of TB diagnostics worldwide due to low cost, simplicity and rapidity, as has been already used in other fields of infectious disease diagnostics. However, no satisfactory test has been available due to the low sensitivity and specificity of the antibody response against mycobacterial antigens. The World Health Organization recommended that TB diagnostic tools for wide use should have sensitivities of over 80 % and specificities of over 95 % (WHO, 2001). Although the recent development of serodiagnosis using multiple protein antigens gives a considerable improvement in the sensitivity and specificity for the early diagnosis of TB (Lyashchenko et al., 2000), it takes much work to isolate the single-protein antigen and the test gives low sensitivity and high cross-reactivity. The overall sensitivity of multiple-antigen ELISA with 10 protein antigens has been reported to be at best 70 % (Lyashchenko et al., 1998).

Previously, we reported that anti-cord-factor IgG antibody titres were elevated significantly in active TB patient sera (He et al., 1991) and demonstrated that the detection of anti-cord factor antibody was applicable for the early diagnosis of active pulmonary TB (Maekura et al., 1993), colonic TB (Kashima et al., 1995) and later tuberculous uveitis (Sakai et al., 2001). Cord factor is one of the most characteristic and ubiquitous cell wall glycolipids and the structure of the mycoloyl moiety varies widely among mycobacterial species. It has also been reported that glycolipids are extremely stable physico-chemically on microplate ELISA (Julian et al., 2001). A TBGL (tuberculous glycolipid) antibody detection system using mainly cord factor with minor glycolipid antigens derived from M. tuberculosis H37Rv has been developed (Kawamura et al., 1997) and evaluated clinically to be useful for the diagnosis of TB at early stages of the disease by a multiple-centre study (Maekura et al., 2001). Compared to protein antigens, ELISA using lipid antigens shows better stability and reproducibility, and low cross-reactivity, if we choose appropriate antigens and antibody detection system. Chan et al. (1990) reported haemagglutination tests with three glycolipid antigens and ELISA with antigen-5 on pulmonary TB patient sera; tests with all four antigens were of similar efficiency, with 30–52 % of smear-positive patients. More recently, it has also been reported that the detection of immune complexes and IgG antibodies against the three glycolipid antigens is useful for the serodiagnosis of children with a high probability of pulmonary TB (Simonney et al., 2000). Furthermore, such immunecomplex multiple-antigen ELISA has been described to serve as a fast, cheap and easy method for the diagnosis of TB (Harrington et al., 2000). Julian et al. (2004) reported antibody responses against a combination of specific glycolipids and proteins for test sensitivity improvement in TB serodiagnosis.

In our present study, we carefully selected six antigens from among 12 lipid components obtained from mycobacterial cell walls and membranes, and evaluated the reactivities of IgG antibody in active-TB patients and healthy control subjects. A diverse patient-to-patient variation of IgG antibody responses against each of the six lipid antigens was observed; when we combined the six antigens and an overall positive was scored when any one of the six tests was positive, a markedly higher sensitivity was obtained. Some patients gave a highly positive result against all six antigens generally. However, some patients scored highly positive against only mycoloyl glycolipids derived from cell walls, whereas others were positive against only mannophospholipids from cell membranes. Some were positive to TMM, but not to TDM. Some were positive to TDM-T but not to TDM-M, or positive to TMM-T but not to TMM-M, indicating that the mycolyl moiety of the trehalose esters was strictly recognized by IgG antibody in TB patient sera, as we have already reported (Pan et al., 1999).

Recently, the lipid antigen presenting molecules group 1 CD1 (such as CD1a, CD1b, CD1c) and group 2 CD1 (CD1d and CD1e) in human dendritic cells or macrophages have been investigated (Beckman et al., 1994; Moody et al., 1997; Porcelli et al., 1992), and T-cell responses in humans infected with M. tuberculosis to CD-1 restricted lipid antigens were reported (Ulrichs et al., 2003), in which the T cell response in TB patients was the most active against the glycolipid fraction by proliferation assay. However, the lipid antigen presentation to B cells in humoral immune responses has not yet been reported. Mycobacterial lipid antigen epitopes may be presented by novel mechanisms different from the classical MHC class I or class II restricted protein antigens. Although in the case of protein antigens the diversity of T cell responses has been reported to be related to HLA type exclusively, a diversity mechanism of humoral response in the human system against lipid antigens has not yet been reported. To date, serodiagnostics have been designed for early diagnosis of TB or other mycobacterial diseases, and therefore the changes or variation of antibody titres according to the progression or stage of the disease have not been focused on. To reveal the diversity mechanism of antibody responses, we examined the time-course changes in IgG antibody titres. We found that IgG antibody levels against lipid antigens in TB patients' sera varied greatly depending on the stages of the disease. This differs from TST results, which show long-lasting positive scores, once changed to positive from negative. In most cases after the initiation of anti-TB chemotherapy, the IgG antibody titres of active TB patient sera against mycobacterial lipid antigen decreased immediately or were first elevated for a few weeks, and then decreased distinctly near to the normal healthy control level after 3–4 months, when the excretion of bacilli had ceased. The above results and our previous reports (Maekura et al., 1993, 2001), strongly suggest that in parallel with the bacterial burden, IgG antibody levels changed dramatically. To demonstrate the relationship between positive rate and/or mean {Delta}A values and loading amount of bacteria, we carefully compared the IgG antibody responses of smear-negative (Gaffky 0), paucibacillary (Gaffky +1 to +3), moderate bacillary (Gaffky +4 to +7) and multibacillary (Gaffky +8 to +10) groups. Unexpectedly, the smear-negative group did not necessarily show the lowest positive rate and IgG antibody levels, and the paucibacillary group showed the lowest. This may reflect the incompleteness of the detection technique of the smear test for acid-fast bacteria. IgG antibody responses were also related to the severity or extent of the pathological lesions determined by chest X-ray examination. In general, patients with the larger and more severe pathological lesions showed a higher positive rate against all antigens, while patients with smaller or less severe lesions showed a lower rate. As the chemotherapy for TB proceeded, excretion of bacilli was eradicated and the pathological lesions were minimized, IgG antibody levels against the lipid antigens decreased distinctly. Therefore, the serodiagnostic test can be useful for the diagnosis of the progression of the disease or as a criterion to determine whether the patient should end or continue the anti-TB chemotherapy. Although the pathological role of the humoral antibody response in TB is not known, marked elevation of IgG antibody against cell-surface lipid antigens may promote the generation of granulomatous lesions or induction of cytokine production by the cellular immune responses, as reported previously (Fujiwara et al., 1999).

Recently, the incidence of non-TB mycobacteriosis such as MAC infection has increased in TB patients being treated with anti-TB chemotherapy and in immunosuppressed hosts such as AIDS. Therefore, a differential diagnostic test is required. In the present study, we observed that TB patient sera were more reactive against TDM-T and TMM-T, while MAC patient sera were more reactive against TDM-M and TMM-M, indicating that IgG antibody recognized the mycoloyl residue of TDM and/or TMM, as reported previously (Enomoto et al., 1998; Pan et al., 1999). However, in the present study, IgG antibody responses against species-specific TDM and TMM were cross-reactive, and 100 % sensitivity or specificity was not obtained. This is probably due to the fact that natural trehalose esters consist of a mixture of mycolic acid subclasses. In contrast, phosphoinositomannnosides are widely distributed in mycobacterial species; thus PL-1 and PL-2 are essentially mycobacterial common antigens. Therefore, with the present combination of six antigens we cannot strictly differentiate between TB and MAC disease. However, if we use species-specific antigens, more precise differential diagnosis of TB and MAC disease would be possible. Our recent papers have described highly sensitive discrimination between TB and MAC disease, using ELISA with serotype-specific glycopeptidolipid (GPL) cocktail antigens (Kitada et al., 2002) or GPL core (Kitada et al., 2005) from MAC. A more comprehensive serological analysis of MAC infection using mycobacterial multi-lipid antigens including MAC-specific GPL core will be reported in a later paper.

Another use of multiple-antigen ELISA is for serological testing of high-risk contacts who might have been infected with M. tuberculosis recently and may develop active disease if chemoprophylaxis is not applied. Silva et al. (2003), using multivariate analysis, reported that the humoral response to 38 kDa protein was associated with active disease, while those to 14 kDa and ESAT-6 antigens were associated with inactive TB, namely recent M. tuberculosis infection or latent TB. Since in our study also, there were probably two types of antigens – early-responsive or infection-specific antigens such as PL-1, PL-2, TMM-T and TMM-M, and late-responsive or disease-specific antigens such as TDM-T and TDM-M – it would be possible for serological screening to use serodiagnostics with appropriate antigens. In contrast to TST, which is an infection-associated criterion based on the cellular immune response, humoral IgG antibody responses well reflected the loading amount of bacilli and were more disease specific. Therefore, TST-negative active-TB patients who have been excreting bacilli and are immunocompromised due to steroid use, diabetes mellitus or anti-TNF-{alpha} antibody treatment, and old-aged or low-protein nutritional cases, may yet give a positive IgG antibody response (data not shown). To establish whether a serological tool using a particular single antigen or multiple antigens is useful for the screening of latent infection of M. tuberculosis, it is essential to perform clinical tests on a population from an area of high TB incidence. Overall, multiple-antigen ELISA, based on the humoral immune responses, may be the most useful for the rapid diagnosis of active disease due to mycobacterial infection, diagnosis of progression of the disease and differential diagnosis of the mycobacterial species responsible for the infection.

The future interest will be to investigate the possible immunological role of IgG antibodies against each lipid antigen and the relationship between the humoral and cellular immune responses.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Arloing, S. (1898). Agglutination de bacille de la tuberculose vraie. C R Acad Sci 126, 1398–1400.

Beckman, E. M., Porcelli, S. A., Morita, C. T., Behar, S. M., Furlong, S. T. & Brenner, M. B. (1994). Recognition of a lipid antigen by CD1-restricted alpha beta+ T cells. Nature 372, 691–694.[CrossRef][Medline]

Benjamin, R. G. & Daniel, T. M. (1982). Serodiagnosis of tuberculosis using the enzyme-linked immunoabsorbent assay (ELISA) of antibody to Mycobacterium tuberculosis antigen 5. Am Rev Respir Dis 126, 1013–1016.[Medline]

Benjamin, R. G., Debanne, S. M., Ma, Y. & Daniel, T. M. (1984). Evaluation of mycobacterial antigens in an enzyme-linked immunosorbent assay (ELISA) for the serodiagnosis of tuberculosis. J Med Microbiol 18, 309–318.[Abstract]

Chan, S. L., Reggiardo, Z., Daniel, T. M., Girling, D. J. & Mitchison, D. A. (1990). Serodiagnosis of tuberculosis using an ELISA with antigen 5 and a hemagglutination assay with glycolipid antigens. Results in patients with newly diagnosed pulmonary tuberculosis ranging in extent of disease from minimal to extensive. Am Rev Respir Dis 142, 385–389.[Medline]

Chan, E. D., Heifets, L. & Iseman, M. D. (2000). Immunologic diagnosis of tuberculosis. Tuber Lung Dis 80, 131–140.[CrossRef][Medline]

Daniel, T. M. & Debanne, S. M. (1987). The serodiagnosis of tuberculosis and other mycobacterial diseases by enzyme-linked immunosorbent assay. Am Rev Respir Dis 135, 1137–1151.[Medline]

Enomoto, K., Oka, S., Fujiwara, N., Okamoto, T., Okuda, Y., Maekura, R., Kuroki, T. & Yano, I. (1998). Rapid serodiagnosis of Mycobacterium avium-intracellulare complex infection by ELISA with cord factor (trehalose 6,6'-dimycolate), and serotyping using the glycopeptidolipid antigen. Microbiol Immunol 42, 689–696.[Medline]

Fujiwara, N., Oka, S., Ide, M., Kashima, K., Honda, T. & Yano, I. (1999). Production and partial characterization of antibody to cord factor (trehalose 6,6'-dimycolate) in mice. Microbiol Immunol 43, 785–793.[Medline]

Grange, J. M., Gibson, J., Nassau, E. & Kardjito, T. (1980). Enzyme-linked immunosorbent assay (ELISA): a study of antibodies to Mycobacterium tuberculosis in the IgG, IgA and IgM classes in tuberculosis, sarcoidosis and Crohn's disease. Tubercle 61, 145–152.[CrossRef][Medline]

Harrington, J. J., 3rd, Ho, J. L., Lapa e Silva, J. R., Conde, M. B., Kritski, A. L., Fonseca, L. S. & Saad, M. H. (2000). Mycobacterium tuberculosis lipid antigens: use of multi-antigen based enzyme immunoassay for free and complex dissociated antibodies. Int J Tuberc Lung Dis 4, 161–167.[Medline]

He, H., Oka, S., Han, Y. K., Yamamura, Y., Kusunose, E., Kusunose, M. & Yano, I. (1991). Rapid serodiagnosis of human mycobacteriosis by ELISA using cord factor (trehalose-6,6'-dimycolate) purified from Mycobacterium tuberculosis as antigen. FEMS Microbiol Immunol 3, 201–204.[Medline]

Iinuma, Y., Senda, K., Takakura, S. & 8 other authors (2002). Evaluation of a commercially available serologic assay for antibodies against tuberculosis-associated glycolipid antigen. Clin Chem Lab Med 40, 832–836.[CrossRef][Medline]

Julian, E., Cama, M., Martinez, P. & Luquin, M. (2001). An ELISA for five glycolipids from the cell wall of Mycobacterium tuberculosis: Tween 20 interference in the assay. J Immunol Methods 251, 21–30.[CrossRef][Medline]

Julian, E., Matas, L., Alcaide, J. & Luquin, M. (2004). Comparison of antibody responses to a potential combination of specific glycolipids and proteins for test sensitivity improvement in tuberculosis serodiagnosis. Clin Diagn Lab Immunol 11, 70–76.[Abstract/Free Full Text]

Kardjito, T., Handoyo, I. & Grange, J. M. (1982). Diagnosis of active tuberculosis by immunological methods. 1. The effect of tuberculin reactivity and previous BCG vaccination on the antibody levels determined by ELISA. Tubercle 63, 269–274.[Medline]

Kashima, K., Oka, S., Tabata, A., Yasuda, K., Kitano, A., Kobayashi, K. & Yano, I. (1995). Detection of anti-cord factor antibodies in intestinal tuberculosis for its differential diagnosis from Crohn's disease and ulcerative colitis. Dig Dis Sci 40, 2630–2634.[Medline]

Kawamura, M., Sueshige, N., Imayoshi, K., Yano, I., Maekura, R. & Kohno, H. (1997). Enzyme immunoassay to detect antituberculous glycolipid antigen (anti-TBGL antigen) antibodies in serum for diagnosis of tuberculosis. J Clin Lab Anal 11, 140–145.[CrossRef][Medline]

Kiran, U., Shriniwas, Kumar, R. & Sharma, A. (1985). Efficacy of three mycobacterial antigens in the serodiagnosis of tuberculosis. Eur J Respir Dis 66, 187–195.[Medline]

Kitada, S., Maekura, R., Toyoshima, N., Fujiwara, N., Yano, I., Ogura, T., Ito, M. & Kobayashi, K. (2002). Serodiagnosis of pulmonary disease due to Mycobacterium avium complex with an enzyme immunoassay that uses a mixture of glycopeptidolipid antigens. Clin Infect Dis 35, 1328–1335.[CrossRef][Medline]

Kitada, S., Maekura, R., Toyoshima, N., Naka, T., Fujiwara, N., Kobayashi, M., Yano, I., Ito, M. & Kobayashi, K. (2005). Use of glycopeptidolipid core antigen for serodiagnosis of Mycobacterium avium complex pulmonary disease in immunocompetent patients. Clin Diagn Lab Immunol 12, 44–51.[Abstract/Free Full Text]

Lyashchenko, K., Manca, C., Colangeli, R., Heijbel, A., Williams, A. & Gennaro, M. L. (1998). Use of Mycobacterium tuberculosis complex-specific antigen cocktails for a skin test specific for tuberculosis. Infect Immun 66, 3606–3610.[Abstract/Free Full Text]

Lyashchenko, K. P., Singh, M., Colangeli, R. & Gennaro, M. L. (2000). A multi-antigen print immunoassay for the development of serological diagnosis of infectious diseases. J Immunol Methods 242, 91–100.[CrossRef][Medline]

Maekura, R., Nakagawa, M., Nakamura, Y. & 9 other authors (1993). Clinical evaluation of rapid serodiagnosis of pulmonary tuberculosis by ELISA with cord factor (trehalose-6,6'-dimycolate) as antigen purified from Mycobacterium tuberculosis. Am Rev Respir Dis 148, 997–1001.[Medline]

Maekura, R., Okuda, Y., Nakagawa, M. & 10 other authors (2001). Clinical evaluation of anti-tuberculous glycolipid immunoglobulin G antibody assay for rapid serodiagnosis of pulmonary tuberculosis. J Clin Microbiol 39, 3603–3608.[Abstract/Free Full Text]

Maekura, R., Kohno, H., Hirotani, A., Okuda, Y., Ito, M., Ogura, T. & Yano, I. (2003). Prospective clinical evaluation of the serologic tuberculous glycolipid test in combination with the nucleic acid amplification test. J Clin Microbiol 41, 1322–1325.[Abstract/Free Full Text]

Moody, D. B., Reinhold, B. B., Guy, M. R. & 9 other authors (1997). Structural requirements for glycolipid antigen recognition by CD1b-restricted T cells. Science 278, 283–286.[Abstract/Free Full Text]

Pan, J., Fujiwara, N., Oka, S., Maekura, R., Ogura, T. & Yano, I. (1999). Anti-cord factor (trehalose 6,6'-dimycolate) IgG antibody in tuberculosis patients recognizes mycolic acid subclasses. Microbiol Immunol 43, 863–869.[Medline]

Porcelli, S., Morita, C. T. & Brenner, M. B. (1992). CD1b restricts the response of human CD4-8-T lymphocytes to a microbial antigen. Nature 360, 593–597.[CrossRef][Medline]

Reggiardo, Z., Vazquez, E. & Schnaper, L. (1980). ELISA tests for antibodies against mycobacterial glycolipids. J Immunol Methods 34, 55–60.[CrossRef][Medline]

Sada, E., Brennan, P. J., Herrera, T. & Torres, M. (1990). Evaluation of lipoarabinomannan for the serological diagnosis of tuberculosis. J Clin Microbiol 28, 2587–2590.[Medline]

Sakai, J., Matsuzawa, S., Usui, M. & Yano, I. (2001). New diagnostic approach for ocular tuberculosis by ELISA using the cord factor as antigen. Br J Ophthalmol 85, 130–133.[Abstract/Free Full Text]

Silva, V. M., Kanaujia, G., Gennaro, M. L. & Menzies, D. (2003). Factors associated with humoral response to ESAT-6, 38 kDa and 14 kDa in patients with a spectrum of tuberculosis. Int J Tuberc Lung Dis 7, 478–484.[Medline]

Simonney, N., Bourrillon, A. & Lagrange, P. H. (2000). Analysis of circulating immune complexes (CICs) in childhood tuberculosis: levels of specific antibodies to glycolipid antigens and relationship with serum antibodies. Int J Tuberc Lung Dis 4, 152–160.[Medline]

Ulrichs, T., Moody, D. B., Grant, E., Kaufmann, S. H. & Porcelli, S. A. (2003). T-cell responses to CD1-presented lipid antigens in humans with Mycobacterium tuberculosis infection. Infect Immun 71, 3076–3087.[Abstract/Free Full Text]

WHO (2001). Modelling the impact of new diagnostic technologies in disease endemic countries. TBDI Activity Summary May 2001, 1–5.

Zeiss, C. R., Radin, R. C., Williams, J. E., Levitz, D. & Phair, J. P. (1982). Detection of immunoglobulin G antibody to purified protein derivative in patients with tuberculosis by radioimmunoassay and enzyme-linked immunosorbent assay. J Clin Microbiol 15, 93–96.[Medline]

Received 26 November 2004; revised 16 February 2005; accepted 17 February 2005.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Fujita, Y.
Articles by Yano, I.
Articles citing this Article
PubMed
PubMed Citation
Articles by Fujita, Y.
Articles by Yano, I.
Agricola
Articles by Fujita, Y.
Articles by Yano, I.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.