WO1998031387A1 - Mycobacterium antigens and methods for their detection - Google Patents

Abstract

Serum from patients with active tuberculosis has been found to contain antibodies specific for two mycobacterial antigens, the significance of which is reported herein for the first time. The titer of these antibodies changes in patients whose disease is no longer active. This invention provides assays for detection of active tuberculosis, based on determination of antibodies specific for these antigens in patient samples.

Description

MYCOBACTERIUM ANTIGENS AND METHODS FOR THEIR DETECTION

BACKGROUND Field of the Invention

This invention relates to two mycobacterial antigens, the significance of which is reported herein for the first time. Serum from patients with active tuberculosis has been found to contain antibodies specific for these antigens, but the titer of these antibodies changes in patients whose disease is no longer active. Therefore, this invention is directed to assays for detection of active tuberculosis, based on determination of antibodies specific for these antigens in patient samples.

Review of Related Art

Tuberculosis (TB) is responsible for more deaths worldwide each year than any other single infectious disease. In the United States, a resurgence in TB over the past decade, complicated by the ongoing AIDS epidemic, has placed considerable strain on our health care system. A critical component of any successful TB control program is the availability of a rapid and low cost method for identifying persons with active TB disease so that they may be treated and reduce future spread. Diagnosis of TB is usually based on clinical suspicion and demonstration of Mycobacterium tuberculosis (Mtb) in sputum samples utilizing procedures that may take up to 8 weeks. Delays in establishing a diagnosis add to costs of care by necessitating isolation of patients and often further complicate management by delaying initiation of appropriate treatment and allowing further spread of disease.

The other conventional diagnostic test, diagnostic skin testing using purified protein derivative (PPD, a crude extract derived from heat- killed Mtb), is indicative of previous infection, but cannot discriminate between infection and active disease. The skin test is also hampered by a lack of sensitivity and specificity, especially in immunocompromised individuals. In many cases, persons with skin reactivity to PPD and radiological evidence of pulmonary lesions are placed into isolation and treated with antimicrobial drugs while awaiting the demonstration of Mtb in sputum samples. Thus, there is a need for an inexpensive diagnostic test for active TB which is rapid and selective.

Mycobacterial antigens. Although TB is predominantly controlled by cell-mediated immunity and delayed-type hypersensitivity responses, there is substantial activation of B lymphocytes by mycobacterial antigens (reviewed in Collins, 1994, Vet. Microbiol., 40:95). Initial identification of immunodominant antigens was based on serological studies using antisera from TB patients. There has been steady progress in identifying and characterizing several classes of molecules from Mtb that may function as key protective immunity-inducing antigens.

Mtb possesses a complex mixture of antigenic determinants which include of proteins, carbohydrates, lipids, and nucleic acids. Much effort has been directed into the identification of Mtb protein antigens, which have been described primarily based upon their relative molecular weights. These include proteins of 65 kDa (Shinnick, 1987, J. Bacteriol., 169: 1080), 32 kDa (Borremans, et al., 1989, Infect, and Imm n., 57: 3123), 38 kDa (Chandramuki, et al., 1989, J. Clin. Microbiol., 27: 821), as well as proteins of 10 kDa, 16 kDa, 24 kDa, and 30 kDa (Torres, et al., 1994, Clin. Exp. Immunol., 96: 75). The functions of most Mtb proteins have not been identified to date, although some have been shown to serve as heat shock proteins, based on their homology to eukaryotic heat shock proteins (Garcia, et al., 1989, Infect, and Immun., 57: 204). Efforts to develop vaccines have focused on the identification of mycobacterial antigens recognized by T cells. An important distinction between T cell and B cell antigens is that T cell antigens are predominantly proteins, whereas B cell antigens can be proteins, carbohydrates, lipids, and nucleic acids. In addition, antigenic epitopes recognized by T cells are typically chemical in nature, while antigenic epitopes recognized by antibodies can be either chemical or conformational in nature. In the case of carbohydrate antigens, the epitopes are predominantly chemical in nature, although the presence of repeating sugar moieties usually enhances the antigenicity of the molecule.

Mtb possesses antigenic determinants that are shared by nontuberculous environmental mycobacteria, and exposure to these organisms can generate a false positive PPD skin test. Some of these shared antigens have been identified as mycobacterial stress proteins, including a variety of heat shock proteins. Evidence for the ability of heat shock proteins to serve as antigens was provided by the observation that a large number of T cell clones react with a 65 kDa mycobacterial heat shock protein (HSP-65). Approximately 20% of T cell clones generated after immunization of mice with Mtb were found to react with HSP-65 (Kaufinann, et al., 1987, Eur. J. Immunol., 17:351). The observation that a large percentage of human reactive T cell clones recognized HSP-65 led to the idea that this molecule was an immunodominant Mtb antigen. In contrast, more recent data collected using live infection animal models indicate that HSP-65 is not a major target of the early protective immune response to Mtb (Orme, et al., 1993, J. Infect. Dis., 167: 1481).

A mycobacterial product that is a potent B cell antigen is the cell wall glycolipid lipoarabinomannan (LAM). LAM is a complex glycolipid containing multiple linear and branched arabinose and mannose groups linked to a phosphatidylinositol unit at its reducing end. The phosphatidylinositol moiety of LAM has been proposed to play a role in "anchoring" LAM to the membrane (Hunter, et al., 1986, J. Biol. Chem., 261: 12345, Hunter, et al., 1990, J. Biol. Chem., 265: 9272). The majority of arabinan moieties within LAM have been shown by Chatterjee and colleagues to exist in the fiiranosyl (five-membered ring), not pyranosyl (six membered ring) conformation (Chatterjee, et al., 1991, J. Biol. Chem., 266: 9652). Four major oligosaccharide structural motifs containing varying numbers of linear or branched arabinose residues with alpha 1-3, alpha 1-5, or beta 1-2 linkages were identified (Chatterjee, et al., 1991). LAM has been shown to be secreted from macrophages infected with Mtb in vitro (Sturgill-Koszycki, et al., 1994, Science, 263:678), and the presence of anti-LAM antibodies in sera from patients with active TB suggests that LAM is also released from infected macrophages in vivo ( Sada, et al., 1990, J. Clin. Microbiol., 2:2587). In addition to its antigenicity, LAM has been shown to have multiple effects on the immune system, including suppression of T cell activation (Kaplan, et al., 1987, J. Immunol., 138: 3028), and induction of the release of tumor necrosis factor (Moreno, et al., 1989, Clin. and Exper. Immunol., 76: 240).

Serodiagnostic assays for TB.

There is potentially a substantial diagnostic benefit from serology tests which reliably detect the presence of anti-mycobacterial antibodies, particularly if the presence of antibodies against mycobacterial antigens in the serum of TB patients could be shown to be an indicator of active disease. The sensitivity of such a diagnostic test is defined as the fraction of individuals with positive test results among those with culture- confirmed TB, and specificity is defined as the fraction of patients with negative test results among tested individuals without disease. In general, the sensitivity and specificity of these approaches are limited by the particular antigen selected. For example, many persons who were previously vaccinated with BCG, or who had been infected with Mtb and successfully contained the pathogen without developing disease, still possess antibodies against a variety of mycobacterial antigens in their sera. Therefore, selection of an appropriate antigen(s) is an important factor in developing a sensitive and specific assay for anti-mycobacterial antibodies.

Several diagnostic approaches have used the presence of antibodies against mycobacterial antigens in the serum of TB patients as an indicator of active disease. Solid phase ELISA assays have been employed for this purpose, utilizing crude mycobacterial antigens or purified proteins. TB ELISAs using sonicates or filtrates from Mtb or BCG have achieved a range of sensitivities from approximately 50% to 90% and specificities approaching 93% (Benjamin, et al., 1984, J. Med. Microbiol, 18: 309, Chandramuki, et al., 1989, J. Clin. Microbiol,. 27: 821). Serodiagnostic assays using cruder protein fractions as antigen, such as the A-60 fraction of PPD (Daftary, et al., 1995, Indian J. Med. Sci., 48:39) and a 55-67 kDa phospholipid-associated protein fraction from Mtb (Kaushik, et al., 1993, Med. Microbiol. Immunol, 182:317), have been reported to give greater than 90%) sensitivity. Sensitivities using a variety of purified mycobacterial protein antigens (MW 10 kDa, 16 kDa, 24 kDa, 30 kDa, 38 kDa, and 70 kDa) have demonstrated lower sensitivity, with values reported to range between 29-51%>. Rapid detection of antibodies which recognize a 38 kDa mycobacterial protein antigen was reported in 84% of patients with smear- positive TB and 82% of patients with smear-negative TB (Bothamley, et al., 1992, Thorax, 47:270). In addition, 4% of patients with a firm alternative diagnosis gave a positive result. Sensitivities using a variety of other mycobacterial protein antigens have been reported to range between 29-51% (Verbon, et al., 1993, Amer. Rev. Resp. Dis., 148:378).

Detection of antibodies against the non-protein antigens, such as LAM and cord factor (trehalose-6,6'-dimycolate), have also been used as the basis for serodiagnostic assays. Assays based on LAM or cord factor (trehalose-6,6'-dimycolate) have demonstrated a range of sensitivities from 70-95% (Sada, et al., 1990, J. Clin. Microbiol., 2: 2587; (Park, et al., 1993, Tubercule Lung Dis., 74:317; Maekura, et al., 1993, Amer. Rev. Resp. Dis., 148:997). One report suggested that serodiagnostic assays based on di- or tri-acylated trehaloses could achieve sensitivities of over 91% and specificities of 96% (Escamilla, et al., 1996, Am. J. Respir. Crit. Care Med, 154:1864-1867). However, no serodiagnostic test which exhibits both high (>95%) sensitivity and high specificity has been developed to date.

The overall sensitivity of these assays in humans can be limited by significant differences in the specificity of the antibody repertoire between various clinical forms of TB (Elsaghier, et al., 1991, Immunol. Infect. Dis., 1:323; Chandramuki, et al., 1989, J. Clin. Microbiol., 27:821). Persons who have been exposed to Mtb, but fail to develop disease, often raise a persistent humoral response against mycobacterial antigens. This can substantially decrease the specificity of any serodiagnostic assay, thus lowering its predictive value. This has been shown in the case of anti-LAM responses, where anti-LAM antibodies can be detected in the sera of patients more than one year after the successful completion of therapy (Roche, et al., 1993, Internal J. Leprosy and Other Mycobact. Dis., 61:501). Antibodies to acylated trehaloses were also found in treated patients (Escamilla, et al., 1996). AIDS patients are often co-infected with the non-tuberculous mycobacterium M. avium, where this pathogen substantially increases patient mortality and morbidity. The ability to discriminate between infection with M. tuberculosis and M. avium in these patients remains a major unsolved challenge. In addition, co-infection of TB patients with HIV can dramatically decrease the sensitivities of these assays due to suppressed humoral responses (Daniel, et al., 1994, Tubercle Lung Dis., 75:33).

Thus, none of the serological assays proposed in the literature provide the desired sensitivity and specificity appropriate for use in a clinical setting. There remains a need for a rapid, selective diagnostic assay to detect the presence of active TB in a patient.

SUMMARY OF THE INVENTION This invention provides a mycobacterial antigen of about 40 kDa apparent molecular weight measured by SDS-PAGE which is specifically bound by antibodies from patients suffering from active tuberculosis. The antigen is resistant to digestion by proteinase K as determined by retention of the antigenicity of the antigen as well as its mobility on SDS gels after the digestion. The antigen according to this invention is also substantially free of mycobacterial antigens of about 6 kDa which are resistant to proteinase K digestion and which are also specifically inummoreactive with antibodies from patients suffering from active tuberculosis. In a particular embodiment, the mycobacterial antigen of this invention is retained by a 30 kDa cutoff ultrafiltration membrane after proteinase K digestion, is resistant to alkanolysis by 0.1 N NaOH, as measured by antigenicity, can be stained with periodate-Schiff reagent, but not ethidium bromide, and is not soluble in either phenol or chloroform. Preferably, the mycobacterial antigen of this invention is extracted from lysed cells of Mycobacterium sp., and these species may be selected from M. avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M. phei, M. simiae, M. smegmatis, M. tuberculosis, and ulcer ans. More preferably, the antigen is extracted from M. avium, M. bovis, M. kansasii, or M. tuberculosis. In a particularly preferred embodiment, the mycobacterial antigen of this invention is substantially free of other mycobacterial antigens. "Substantially free of other mycobacterial antigens" as used herein means that other mycobacterial antigens provide less than 10% of the total immunoreactivity, or that at least about 90% of all the antibodies in sera which bind to an antigen preparation, bind to the antigen of this invention. In another embodiment, this invention provides a method for obtaining an antigen reactive with antibodies present in patients suffering from active tuberculosis. This method comprises the steps of obtaining an extract from lysed cells of Mycobacterium sp., digesting the extract with proteinase K, ultrafiltering the digested extract using an ultrafiltration membrane with size cutoff of 20 kDa or greater, the antigen being retained in the ultrafiltration residue, applying the ultrafiltrate residue to an anion exchange column, washing the column with low ionic strength buffer and eluting the antigen with buffer of ionic strength greater than 500 mM, and mixing the eluate with an excess of polar organic solvent, the antigen being recovered in the resulting precipitate. In preferred embodiments of this method, the anion exchange column has diethylaminoethyl groups, and the polar organic solvent is isopropanol, added at 2: 1 v/v. Preferably the Mycobacterium sp. is selected from avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M. phei, M. simiae, M. smegmatis, M. tuberculosis, and M. ulcerans; more preferably, the Mycobacterium sp. is M avium, M. bovis, M. kansasii, M. leprae, and M tuberculosis. In yet another embodiment, this invention provides a mycobacterial antigen having an apparent molecular weight of about 6 kDa measured by Western blot developed using antisera from an individual with active tuberculosis (TB), wherein said antigen retains antigenicity and mobility in SDS PAGE after digestion by proteinase K, the 6 kDa antigen being substantially free of a second mycobacterial antigen of 40 kDa apparent molecular weight (measured by Western blot developed using antisera from an individual with active tuberculosis), the second mycobacterial antigen likewise being retained by a 30 kDa cutoff ultrafiltration membrane after digestion by proteinase K. In a preferred embodiment, the mycobacterial antigen is retained by a 30 kDa cutoff ultrafiltration membrane after digestion by proteinase K, is resistant to alkanolysis by 0.1 N NaOH, as measured by antigenicity, can be stained with periodate-Schiff reagent, but not ethidium bromide, and is not soluble in either phenol or chloroform. Preferably, the mycobacterial antigen of this invention is extracted from lysed cells of Mycobacterium sp., and these species may be selected from M. avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M. phei, M. simiae, M. smegmatis, M. tuberculosis, and M. ulcerans. More preferably, the antigen is extracted fromM avium, M. bovis, M. kansasii, or M tuberculosis. In a particularly preferred embodiment, the mycobacterial antigen of this invention is substantially free of other mycobacterial antigens.

In still another embodiment, this invention provides a method for obtaining an antigen reactive with antibodies present in patients suffering from active tuberculosis. This method, comprises the steps of obtaining an extract from lysed cells of Mycobacterium sp., digesting the extract with proteinase K, ultrafiltering the digested extract using an ultrafiltration membrane with size cutoff of 20 kDa or greater, the antigen being retained in the ultrafiltration residue, applying the ultrafiltrate residue to an anion exchange column, washing the column with moderate ionic strength buffer and eluting the antigen with buffer of ionic strength greater than 500 mM, and mixing the eluate with an excess of polar organic solvent and discarding the resulting precipitate, the antigen being retained in the aqueous/solvent mixture. In preferred embodiments of this method, the anion exchange column has diethylaminoethyl groups, and the polar organic solvent is isopropanol, added at 2: 1 v/v. Preferably the Mycobacterium sp. is selected from M avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M. phei, M. simiae, M. smegmatis, M. tuberculosis, and M. ulcerans; more preferably, he Mycobacterium sp. is M avium, M. bovis, M. kansasii, M. leprae, and M. tuberculosis.

In yet another embodiment, this invention provides a diagnostic method for determining the presence of active tuberculosis in a patient comprising detecting, in a sample from the patient, antibodies specifically immunoreactive with the 40 kDa antigen according to this invention or antibodies specifically immunoreactive with the 6 kDa antigen according to this invention. In a particularly preferred embodiment, this invention provides a diagnostic method to aid in diagnosis of tuberculosis in a patient comprising (1) determining the presence of absence of antibodies which specifically bind a mycobacterial antigen having an apparent molecular weight of about 6 kDa measured by Western blot developed using antisera from an individual with active tuberculosis (TB), said antigen being resistant to digestion by proteinase K; and (2) determining the presence of absence of antibodies which specifically bind a mycobacterial antigen having an apparent molecular weight of about 40 kDa measured by Western blot developed using antisera from an individual with active tuberculosis (TB), said antigen being resistant to digestion by proteinase K, wherein the presence of antibodies which bind the 6 kDa antigen and the absence of antibodies that bind the 40 kDa antigen is indicative of past tuberculosis disease in the patient.

In still another embodiment, this invention provides a diagnostic kit for detecting active tuberculosis, containing a first mycobacterial antigen of about 40 kDa apparent molecular weight measured by SDS-PAGE, this first antigen being specifically bound by antibodies from patients suffering from active tuberculosis and resistant to digestion by proteinase K as deteirnined by retention of antigenicity and mobility on SDS gels; and a second mycobacterial antigen having an apparent molecular weight of about 6 kDa measured by Western blot developed using antisera from an individual with active tuberculosis (TB), this second antigen being specifically bound by antibodies from patients suffering from active tuberculosis and resistant to digestion by proteinase K as determined by retention of antigenicity and mobility on SDS gels, wherein the first antigen is substantially free of the second antigen, and the second antigen is substantially free of the first antigen. Preferably, the first antigen and the second antigen in this diagnostic kit are each immobilized on a solid support.

A serodiagnostic test has been developed which detects the presence of serum antibodies against a novel, protease-resistant cell- associated antigen with an apparent MW by SDS PAGE of about 40 kDa (hereinafter termed "the 40 kDa antigen"), which is expressed by Mtb. The overwhelming majority of sera obtained from patients with active TB contained antibodies against this 40 kDa antigen and a co-purifying antigen with an apparent molecular weight by SDS PAGE of about 6 kDa (hereinafter termed "the 6 kDa antigen"). Sera from healthy and BCG- vaccinated persons generally did not contain antibodies which recognized the 40 kDa or 6 kDa antigens. A TB serodiagnostic kit based on the 40 kDa antigen can be used by untrained personnel to screen large numbers of samples outside of a hospital setting, because the chemical features of this antigen indicate that such a kit would not require special equipment or storage conditions. A serodiagnostic test based on the 6 kDa antigen can provide a clinical marker of previous disease. BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows Western blots of Mtb antigen and purified LAM developed with a monoclonal antibody specific for LAM.

Figure 2 shows the effect of mild alkaline treatment on the antigenicity of LAM and Mtb antigen.

Figure 3 shows Western blots of Mtb antigen and LAM developed with TB-positive serum.

Figure 4 shows Western blots of Mtb antigen and LAM developed with serum from a TB patient post-treatment. Figure 5 shows Western blots of Mtb antigen and LAM developed with serum from another post-treatment TB patient.

Figure 6 shows dot blot assays using normal and TB sera collected from various donors.

Figure 7 shows dot blot assays using secondary antibodies specific for human IgG and IgM.

DETAILED DESCRIPTION OF THE EMBODIMENTS

TB skin tests involve intracutaneous injection of PPD to detect cell mediated immunity manifested by delayed-type hypersensitivity reactions. Such tests have several limitations. Dermal sensitivity to PPD fails to discriminate between persons with active disease, healthy persons who have been previously infected with Mtb but do not have active disease, and many persons who have been vaccinated with BCG. Furthermore, many TB patients, whether or not HIV-infected, fail to generate a positive skin reaction to PPD. T cells from healthy individuals living in the same household with TB patients show stronger recognition responses to mycobacterial antigenic peptides than do T cells from the patients, although the opposite trend characterizes their antibody and B cell responses to these antigens (Falla, et al, 1991, Infect. Immun., 59:2265). The antigen or antigens in PPD responsible for recognition of TB infection is/are unknown. In light of our current knowledge that living Mtb play an active role in modulating host immune responses, it is likely that long-term protective immunity is generated by secreted mycobacterial antigens produced by phagocytosed viable Mtb. Selection of appropriate antigen(s) is the key to developing a sensitive and specific assay that is diagnostic for active TB disease based on detecting the presence of specific antibodies in serum. Initial efforts in identifying novel mycobacterial antigens were based on the premise that TB patients can generate humoral responses against non-protein antigens. This presumption was supported by the observation that sera from patients with TB contained antibodies which recognized the mycobacterial glycolipid LAM (Sada, et al., 1990, J. Clin. Microbiol,. 2:2587).

The present invention is based on the discovery that antibodies present in the serum of patients with active TB (as opposed to TB infected persons with no disease or uninfected persons) recognize a cell-associated, heat-stable, protease-resistant antigen present in Mtb, having an apparent MW of 40 kDa. Studies show that over 90% of sera from TB patients tested contained either IgG and/or IgM antibodies against this antigen. Serum from patients infected with M. avium (MAI serum), a mycobacterial pathogen often seen in patients with AIDS,, also contained antibodies which recognized the 40 kDa antigen. Concurrent infection with HIV did not necessarily eliminate the ability of either TB or MAI patients to raise a humoral response against the antigen, although the sensitivity of diagnostic tests based on this antigen appeared lower in HIV-infected individuals. Sera from normal (PPD negative) and BCG- vaccinated persons did not contain antibodies which recognized this antigen.

Similar results may be obtained in patients suffering from other mycobacterial diseases using analogous antigens from the respective mycobacteria. A patient, as discussed herein, is an animal infected with or suspected of infection with a mycobacterium, particularly an animal suffering from mycobacterial disease. A human patient suffering from a mycobacterial disease will exhibit clinical symptoms associated with that disease. This active disease state may be differentiated from mere infection by mycobacteria, which may or may not result in clinically recognizable disease state, for example as discussed above concerning PPD-positive individuals who have never exhibited active TB. Patients may include human adults or children, either hospitalized and unhospitalized individuals, including infants. Animal patients include mammals, such as goats, pigs, sheep, horses, cats, dogs, alpacas, non-human primates, rabbits, and especially cattle, elk, deer, bison, camels or llamas.

Antigens described herein may be obtained from the crude soluble fraction of autoclaved Mtb cultures. Autoclaved cultures are centrifuged to remove insoluble debris, and the supernatant is digested with Proteinase K. The crude antigen preparation may be further fractionated by SDS-PAGE, ion exchange chromatography, and/or fractional precipitation. Protease treatment is effective at eliminating most of the mycobacterial proteins, as judged by SDS-PAGE and staining with Coomassie Blue or silver nitrate. The antigen preparation was determined not to contain DNA or RNA based on the absence of an absorbance peak at 260 nm and the inability to stain with ethidium bromide, but the antigens gave positive results in various carbohydrate assays.

Characterization of the Mtb Antigen.

Extracts of autoclaved Mtb were prepared from two Mtb strains, H37Ra (an avirulent strain in humans) and H37Rv (a human virulent strain), and analyzed by SDS-PAGE. Identical SDS-PAGE gels were prepared, and one gel was stained for protein with Coomassie blue, while the other gel was electrophoretically transferred onto nitrocellulose membranes. The SDS-PAGE protein profiles generated from each Mtb extract were virtually identical. (Virulence factors, which might be differentially expressed in the two strains, would not be evident in these crude analyses.) Analysis of these Mtb extracts by immunoblotting clearly identified a 40 kDa antigenic determinant which was recognized by antibodies present in the sera of patients with active TB, but not present in normal serum. Western blot analysis was performed on extracts prepared from autoclaved Mtb using sera from one TB patient, and these results are representative of over 10 different sera tested by Western blotting. Nitrocellulose membranes blots from SDS-PAGE of the Mtb extracts were incubated with serum obtained from TB patients or with normal serum, and bound human serum antibodies were detected using a goat anti-human IgG antibody conjugated to alkaline phosphatase. A broad band at about 40 kDa was stained by the TB serum but not the normal serum. The 40 kDa antigen was not observed in filtrates from viable Mtb cultures, suggesting that the antigen is not a secreted Mtb product.

The 40 kDa antigenic determinant does not represent a major fraction of the Coomassie blue-stainable material in extracts prepared from autoclaved Mtb cultures. Crude Mtb extracts were digested with Proteinase K (1 U/ml) for 15-60 min at 55 °C, concentrated 10-20 fold by ultrafiltration using a 30 kDa cut-off membrane, and then analyzed by immunoblotting. Low molecular weight Proteinase K-digested contaminants were removed by repeated dilution and ultrafiltration/concentration steps, whereas high molecular weight contaminants were insignificant as determined by SDS- PAGE analysis of the antigen concentrates. Immunoblot analysis of the Proteinase K-digested antigen preparations demonstrated that antibodies in the sera from TB patients still recognized a molecule with an apparent molecular size of 40 kDa, as judged by SDS-PAGE. The 40 kDa antigen does not appear to contain protein material susceptible to proteinase K.

There are several advantages to the use of a non-protein antigens such as carbohydrates as the basis for a serodiagnostic assay. One advantage is that immunoreactivity of carbohydrates is defined by a linear arrangement of sugar groups, whereas the immunoreactivity of many protein epitopes is determined by the conformational arrangement of amino acids which are in proximity only due to folding of the polypeptide chain. While the conformational epitopes on protein antigens are destroyed by heat denaturation, the antigenicity of non-protein molecules is not heat labile. Thus, autoclaved cultures of virulent Mtb bacilli could be safely and easily used as a crude source of non-protein antigens. Demonstration that the antigenicity of the 40 kDa antigen survives autoclaving and proteinase K digestion, suggests that this antigen may offer the stability desirable in serodiagnostic assays.

In addition, carbohydrate antigens are generally more potent B cell stimuli compared with protein antigens. This is most likely due to the repeating chemical structure of carbohydrates which can effectively crosslink Ig molecules on the B cell surface, thus generating a stronger activation signal than that which would be generated by proteins which would be unable to crosslink multiple surface Ig molecules.

Physicochemical Characterization

Supernatants of autoclaved Mtb cultures digested with Proteinase K (I U/ml) in the presence of 0.1%o SDS at 55 C° for 18 hr produce a crude fraction which can be concentrated by ultrafiltration in an

Amicon positive pressure concentrator using 30 kDa-cutoff membranes. The lysate retained by the 30 kDa-cutoff membrane can be dialyzed in phosphate buffered saline (PBS) using 30 kDa-cutoff membranes and stored at 4 C°.

This ultrafiltration method retains antigenic material that appears at about 6 kDa as well as 40 kDa on Western blots developed with TB-positive serum. Attempts to separate the material in these two bands by other size-based separations were not successful. Both antigens are retained by 30 kDa and 50 kDa cut-off dialysis membranes, with or without the addition of denaturants. On a molecular sizing HPLC column in non- denaturing buffer, the antigens traveled with the void peak having over 100 kDa apparent molecular weight. Together the results from Western Blots and HPLC columns suggest that in the absence of a denaturant the 40 kDa and 6 kDa antigens are complexed into large macromolecular aggregates.

Britton and colleagues found 40 kDa and 6 kDa carbohydrate antigens in soluble sonicates of M leprae (Britton, et al., 1985, J. Immunol. 135: 4171), although it is unclear whether the Mtb antigens described herein are similar or identical to the antigens from M. leprae. A study of the antigens fromM leprae suggested that their 40 kDa antigen and their 6 kDa antigen co-purify because they are tightly bound in a noncovalent manner (Britton, et al., 1986, Internal J. Leprosy and Other Mycobact. Dis., 54:545). Alternatively, the apparent molecular weight of the 6 kDa antigen may simply reflect the aberrant mobility of carbohydrates on SDS-PAGE, and the true molecular size of this antigen may be greater. Ciude antigen preparations (not fractionated on DEAE-Sephadex) contain contaminating protein (especially at approximately 65 and 10 kDa, as judged by silver stained SDS gels) and nucleic acid (as judged by absorbance at 260 nm). DEAE-purified antigen preparations are free of contaminating protein (as judged by silver-stained SDS-PAGE gels and absorbance at 280 nm) and nucleic acids (as judged by ethidium bromide-stained gels and absorbance at 260 nm). Furthermore, the antigenicity of the both 40 kDa and 6 kDa antigens is unaffected by RNAse A or DNAse I treatment.

A sugar-specific staining procedure was used to determine if the 40 kDa antigen contained, or consisted entirely of, carbohydrate. Purified antigen (prepared by crystallization as described below) was fractionated by SDS-PAGE, and the gel was fixed, then stained with periodate-Schiff stain as recommended by the manufacturer (Sigma). Gels stained by this method revealed a major band at 40 kDa, indicating that the antigen is, or contains, carbohydrate. Furthermore, the Western blotting showed a broad band of immune reactivity in the region of 30-40 kDa, suggesting the presence of multiple, closely-spaced bands; such a pattern is frequently observed with carbohydrate antigens.

The 40 kDa and 6 kDa antigens both are bound by the lectin concanavalin A (Con A), like arabinomannan and LAM, but unlike arabinogalactan. The antigenicity of the 40 kDa and 6 kDa antigens is destroyed by periodic acid treatment. Both antigens are lysozyme and mannosidase resistant. The 40 kDa and 6 kDa antigens are freely soluble in aqueous buffers, unlike cord factor and acylated trehaloses. Cord factor and acylated trehaloses do not bind to diethylaminoethyl groups (DEAE), in contrast to the antigens of this invention. The antigens of this invention are released from a DEAE-Sephadex column at > 500 mM NaCl, unlike LAM or deacylated LAM (chemically similar to arabinomannan) which eluted at < 100 mM NaCl. This indicates that the antigens are acidic polysaccharides, unlike LAM or arabinomannan which are neutral polysaccharides. Several mycobacterial carbohydrate antigens, such as LAM, also contain lipids. The 40 kDa and 6 kDa antigens are not extractable in phenol or chloroform, unlike acylated trehaloses and cord factor, although the 40 kDa antigen can be selectively precipitated by 2 volumes of isopropanol. The antigenicity and mobility of the 40 kDa antigen and 6 kDa antigen on SDS-PAGE are essentially unaffected by mild alkalinolysis, unlike LAM, cord factor, or acylated trehaloses. Limited NaOH treatment (0.1 N NaOH, 2 hr, 37 °C) of the antigen was performed in order to remove any esterified lipid moieties. This treatment failed to alter either the antigenicity or mobility of the antigen on SDS-PAGE and suggests that the 40 kDa antigen is not an ester-linked glycolipid. In contrast, a previous report showed that LAM was sensitive to NaOH treatment (Hunter, et al., 1986, J. Biol. Chem., 261: 12345), thus further distinguishing the 40 kDa antigen from LAM.

Immunological Characterization

The 40 kDa and 6 kDa antigens are recognized by both serum IgG and IgM in serum from patients with active TB. In patients who previously had TB and do not have active disease, serum antibodies can be found against the 6 kDa antigen, but not the 40 kDa antigen. An additional antigen is observed at 15-20 kDa using sera from some patients or cows, or when using Protein G is used instead of anti-IgG and -IgM reagents to detect bound serum antibodies. This 20 kDa antigen, like the 40 kDa antigen is both eluted form DEAE Sephadex at > 500 mM NaCl and precipitated by 2 vol. isopropanol. However, the 20 kDa antigen does not bind well to Concanavalin A.

Antigenic determinants recognized by antibodies present in serum from TB patients were retained by a 30 kDa cut-off ultrafiltration membrane, but glycoproteins, lipoproteins and nonpeptidic molecules migrate aberrantly on SDS-PAGE, unlike unmodified proteins. Previous studies have shown that a nonpeptidic Mtb antigen, the 17 kDa glycolipid LAM, is also recognized by antibodies present in the sera of patients with active TB (Sada, et al., 1990, J. Clin. Microbiol, 2:2587). Britton et al. (1986) report in the discussion section of their paper that their anti-40 kDa antigen monoclonal antibody (designated L9) could also recognize LAM. In the same paper, Britton et al., also state that "even though the . . . L9 defined determinants were present in M. tuberculosis, the [titers of anti-L9 antibodies] in tuberculosis sera tested were not significantly raised above that of the controls". Although the apparent molecular size of LAM has been reported to be smaller than 40 kDa {i.e. LAM migrates with an apparent molecular size of approximately 30 kDa on SDS-PAGE (Prinzis, et al., 1993, J. Gen. Microbiol, 139:2649), it was important to determine if the 40 kDa antigenic determinant of this invention was similar or identical to LAM. The inventors have found that a monoclonal antibody against LAM (CS-35) did not recognize either the 40 kDa or 6 kDa Mtb antigens, although sera from active TB patients recognized both LAM and the 40 kDa antigen. These data demonstrate that the 40 kDa antigen is antigenically distinct from LAM.

Using immobilized 40 kDa Mtb antigen, the isotype of antigen- bound serum antibodies was determined. Enzyme-conjugated secondary antibodies which were specific for either human IgM or IgG were used to determine which isotypes were responsible for the reactivity in human sera. Example 6 below describes the results of dot blots using dilutions of the 40 kDa Mtb antigen applied to a membrane. Membranes were incubated with either TB patient (TB) or normal (N) sera, and bound human serum antibodies were detected using enzyme-conjugated secondary antibodies (2nd Ab) specific for human IgM (M) or IgG (G). In more than 80% of sera tested, the 40 kDa antigen was recognized by both IgM and IgG antibodies present in the sera of TB patients. In some cases, sera contained only IgM antibodies which recognized the antigen (see Figure 9). Because early humoral responses are dominated by an IgM response, this observation may reflect the history of TB in a patient. For example, primary TB may preferentially elicit an IgM response whereas reactivation TB may elicit both and IgM and an IgG response. Preparation of the Antigen

Antigen maybe extracted from cells of Mycobacteria sp., such as M avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M. phei, M simiae, M. smegmatis, M. tuberculosis, and M. ulcerans. While antigen extracted from Mtb is described herein as representative of this invention, using other mycobacteria may be preferable, because Mtb cultures should be maintained in a biosafety level 3 (P-3) laboratory facility. Mtb or other mycobacteria can be grown according to standard procedures for mycobacteria. Mycobacteria cultures for preparation of the antigen of this invention may be harvested at any stage of growth, but late log phase or stationary phase is preferred for recovery of larger amounts of cellular material.

The antigen disclosed herein is generally is not released into the culture medium by logarithmically growing cells, but rather must be extracted from the cellular mass. Extraction is generally into an aqueous buffer, and may be facilitated by freeze-thaw cycling, sonication, refluxing in alcohol, or other known cell lytic procedures. The antigen may be prepared from autoclaved Mtb cells. Usually culture media containing Mtb cells is autoclaved without separation. This source is preferred to attempting extraction of the antigen from viable Mtb; the use of autoclaved material niinimizes the health risk during preparation. Late-log phase Mtb cultures can autoclaved, and the bacilli removed by centrifugation. The desired antigenic activity was not found in medium recovered from viable bacterial cultures, but 40 kDa antigen is released into the culture medium following autoclaving. Additional antigenic material was obtained by reextraction of the cellular pellet, suggesting that the antigen is cell wall associated.

While antigen prepared by extraction from autoclaved M tuberculosis cells will be recognized by sera from TB patients, further purification of the antigen is preferred to increase specificity. In particular, it is preferred to digest the extract with proteinase K, as the antigen of this invention is resistant to such digestion, while nearly all antigenic proteins are degraded by this procedure. Digestion is carried out with sufficient protease to digest most of the contaminating proteins and/or to clarify the lysate. Usually the proteinase is proteinase K, which may be added at lOOμg/ml to provide an excess, and digestion is preferably carried out in the presence of 0.1% SDS. Bulk protein may alternatively be removed from the lysate by digestion with one or more other proteases, so long as substantial protein degradation occurs. Modification of digestion conditions to accommodate protease substitutions is within the skill of the art. Components resistant to digestion can be size-selected by dialysis and concentrated by ultrafiltration. Degraded peptide fragments can be removed by ultrafiltration; the antigen will be retained by a 30 kDa cutoff membrane after digestion with proteinase K. Concentrated preparation of the 40 kDa and 6 kDa antigens can then be filtered and stored at 4°C.

The concentrated antigen may be further purified using ion exchange chromatography. For example, antigen preparation may be applied to a DEAE column equilibrated in 10 mM Tris-HCl (pH 7.4), 0.5% Triton X-100, 1 mM EDTA. The column will then be washed extensively and eluted using a 0-500 mM linear NaCl gradient. The eluted fractions should be collected, tested for immunoreactivity (for example using a dot blot assay as described herein), and pooled. Pooled samples may be dialyzed and concentrated. The antigens may be quantitated based on carbohydrate content of the fractions using the phenol-sulfuric acid colorimetric assay. Additional purification may be achieved by reverse-phase HPLC using, for example, a C-18 HPLC column.

Alternatively, the concentrated material from the Proteinase K digestion may be dialyzed into buffer containing a denaturant, such as 4M urea or l%Triton-X100 and applied to an anion exchange column, such as a DEAE column. The concentrated proteinase K digest typically contains lipid material which may interfere with column flow. This contaminating lipid may be removed by extraction with organic solvents or saponification by NaOH followed by extraction of the unsaponified material into chloroform before applying the antigenic concentrate to the column. Contaminating protein can be removed by washing the column with buffer plus denaturant, the wash preferably having moderate ionic stiength, such as about 300 mM NaCl. The desired antigens are eluted from the washed column by increasing the ionic strength to greater than 500 mM NaCl, preferably at least about 800 mM. In another alternative, the antigen may be purified by fractionating the cellular extract by ion exchange chromatography before Proteinase K digestion.

In yet another alternative, the antigen may be purified by immuno-affinity methods using antibodies specific for the antigen. Preferably, immimo-affinity methods, such as affinity chromatography, will use purified polyclonal or monoclonal antibodies prepared as described herein.

After elution from the column, the mixture of 40 kDa and 6 kDa antigens can be fractionated by selectively precipitating the 40 kDa antigen with a polar organic solvent, such as isopropanol. The precipitated 40 kDa antigen can be redissolved into low ionic strength buffer, such as 10 mM NaCl, 10 mM Tris-Cl, pH=7.4, preferably containing 1 mM EDTA. The 40 kDa antigen may also be separated from the 6 kDa antigen by treating the proteinase K digest with a polar solvent, such as ethanol, isopropyl alcohol, butyl alcohol, acetone, etc. Purification of the 40 kDa antigen by precipitation with a polar solvent is generally less satisfactory if attempted on crude extracts before the bulk protein is removed by Proteinase K digestion and/or column chromatography. The antigenic material which survives proteinase K digestion is surprisingly stable. Little or no loss of antigenic activity is observed at

37 °C for 1 day, and this extrapolates to approximately 3 months of stability when stored at 4°C. Prediluted (ready-to-use form) secondary antibody conjugates are stable for at least one year at 4°C.

Diagnostic Assays

Studies have identified a novel 40 kDa mycobacterial antigen that could be recognized by antibodies contained within the sera of patients with active TB. The sensitivity and specificity of this response was studied using a larger collection of sera from both TB patients and healthy persons. It was also important to determine if co-infection with HIV affected the ability of these patients to raise a humoral response against the 40 kDa antigen. This is particularly important because the diagnostic effectiveness of both skin testing and radiography may be greatly reduced in persons with AIDS.

Serum samples obtained from patients with active TB (either HIV+ or HIV-), and serum samples from patients infected with M avium, which were tested to determine if they contained antibodies which recognized the 40 kDa antigen. Samples from healthy PPD negative, healthy PPD-positive, and BCG-vaccinated persons were also tested. Representative examples of the results obtained using the dot blot assay and normal and TB sera collected from various donors are shown in Example 6. The results indicate that a serodiagnostic assay using the 40 kDa antigen is highly sensitive and specific for active mycobacterial disease state. The sensitivity of this assay appears to be reduced by co-infection with HIV.

The 6 kDa antigen described herein is also associated with the TB disease state, but not limited to sera from patients with active disease. Rather, reactivity of a patient's serum with the 6 kDa antigen reflects a history of active TB, with antibodies to the 6 kDa antigen being found in sera from both active and recovered TB patients. On the other hand, sera from PPD-positive individuals with no histoiy of active TB (such as BCG- vaccinated individuals) do not react with the 6 kDa antigen. Thus, the 6 kDa antigen provides a clinical marker of previous disease. Comparison of serum reactivity with the 40 kDa antigen to reactivity with the 6 kDa antigen can be useful diagnostically in ruling out active TB in an individual with TB-like symptoms, including abnormal chest X-ray, lung scarring, etc.

Detection of antibodies immunoreactive with the antigens of this invention may be accomplished using techniques that are well known to those skilled in the art. These antibodies can detected by a variety of immunometric assay techniques. The assays of the present invention can be directly used to detect antibodies which are specifically immuno-reactive with the 40 kDa and/or the 6 kDa antigens of this invention. One can detect antibody binding by any detection means known in the art. A particularly useful stain employs peroxidase, hydrogen peroxide and a chromogenic substance such as aminoethyl carbazole. The peroxidase (a well known enzyme available from many sources) can be directly coupled to an anti-human IgG or IgM or complexed via one or more antibodies to an antibody which specifically binds the 40 kDa antigen. For example, a goat anti-peroxidase antibody and a goat anti-human antibody can be complexed via an anti-goat IgG. Such techniques are well known in the art. Other chromogenic substances and enzymes may also be used. Radiolabeling of antibodies may also be used to detect antibody binding. Labeled antibodies may be immunoreactive with anti-human IgG or IgM. Again, such techniques are well known.

The precise technique by which antibodies specifically immunoreactive with the 40 kDa antigen are detected in patients is not critical to the invention. Solution assay methods, including calorimetric, chemiluminescent or fluorescent immunoassays such as ELISA, sandwich and competitive immunoassays, immuno-diffusion, radio immunoassay, immunoassay, Western blot and other techniques, may be used to detect and quantitate the antibodies from a patient by assaying any sample from the patient that contains antibodies. Antibodies may be quantitated in a biological fluid, such as whole blood, serum, plasma, effusions, ascites, urine, cerebrospinal fluid, and bronchoalveolar lavage fluid using any immunologic detection means known in the art. Preferred methods employ immunological detection in samples which include serum. Preferred assay techniques include: radioimmunoassay, enzyme linked immunoadsorbent assay, complement fixation, nephelometric assay, immunodiffusion or immunoelectrophoretic assay and the like. Cellular elements and lipid may be removed from fluids, e.g., by centrifugation. For dilute fluids, such as urine or bronchoalveolar lavage fluid, antibodies may be concentrated, e.g., by ultra-filtration or salting-out. Preferred assay formats are described below.

Dot blot analysis.

In a typical embodiment, partially-purified or completely purified 40 kDa and/or 6 kDa Mtb antigens are applied in five serial 2-fold dilutions to strips of nylon membranes (0.7 cm x 5 cm) using high precision ceramic micropumps which are capable of delivering 0.1-1.0μl volume per stroke with a maximum of 5% variability. Purified human IgG (1 μg/dot) is applied on the membrane strip as a positive control for the validity of the test reagents. The membrane dot blot strips are air-dried, blocked with a blocking buffer of PBST (PBS with 0.05% Tween-20) and 5% normal rabbit serum, and then dried at 40 °C for 1 hr. For each assay, control or TB- positive sera (1: 25 dilution) is applied to the strips, incubated for 10 minutes at room temperature, and washed free of non-specifically bound antibodies. A detection antibody (e.g., an alkaline phosphatase-linked goat polyclonal antibody recognizing human IgG and or IgM) is added and incubated at room temperature for 10 minutes. After washing, the appropriate substrate (e.g. BCIP/NBT; Kierkegaard & Perry), is added, and dot blots are scored visually. For analysis of reactivity with individual 40 kDa and 6 kDa antigens, these antigens may be separated as described herein, and individually applied to membrane strips.

Western blotting. In a typical embodiment, the partially-purified Mtb antigen mixture are separated on SDS-PAGE under standard conditions and blotted to nylon membranes. In certain experiments, purified LAM preparations may also separated on gels for comparative purposes. Membranes are cut into several identical strips, blocked with PBST containing 5% rabbit serum, and processed with TB patient sera analogous to the method used for dot blots.

ELISA analysis.

In a typical embodiment, the wells of 96 well microtiter plates are coated with partially purified 40 kDa and 6 kDa antigens diluted in 100 μl of PBS, incubated at room temperature for 60 minutes, and washed with PBST to remove unbound antigen. Wells are then blocked with solution of PBST with 1% BSA for 1 hr at room temperature, washed again, and dried for 1 hr at 40°C. Wells not receiving antigen are used as a negative control, and wells receiving human IgG will serve as positive controls. For each assay, normal human or TB-positive serum (1 :25 dilution) are added, incubated for 60 minutes at room temperature, then washed to remove unbound antibodies. Samples are processed in duplicate. Peroxidase- conjugated goat anti-human Ig (recognizing IgG and/or IgM) is used to detect the bound immune complexes, and an appropriate substrate (e.g., OPD) is added for color development. ELISA plates are read at 450 nm. ELISA assay tests are scored as positive where absorbance reading is greater than a calculated cutoff value (e.g. 0. 10 absorbance units above the mean of several analyzed normal human sera). For subsequent analysis of reactivity with individual 40 kDa and 6 kDa antigens, these antigens may be separated as described herein, and individually applied to the microliter plate wells.

Preliminary studies identified two Mtb antigens that could be detected by antibodies in the sera of TB patients. A larger sample population was studied using both dot blot and ELISA formats to test reactivity against the mixture of 40 kDa and 6 kDa Mtb antigens to allow estimation of the sensitivity and specificity of the assay (see Example 6). These initial results indicate that serodiagnostic assays using the 40 kDa and 6 kDa mycobacterial antigens can be highly sensitive and specific for an active mycobacterial disease state. The diagnostic assay described herein can also be used to monitor the patient during the course of treatment of an active mycobacterial infection. Many individuals who tested TB-positive in the ELISA or dot blot assays were later found to be HIV-positive, suggesting that concurrent infection with HIV did not block the ability of these patients to mount a humoral response to Mtb or avium, although sensitivity of the assays appeared to be lower in these patients. Preferably immunoassays of this invention measure binding of antisera to antigens purified by anion exchange chromatyraphy, or to antigenic material of comparable purity.

Antibody Production

This invention also contemplates specific binding moieties which, as used herein, refer to molecules capable of binding to the 40 kDa antigen with high specificity, as for example an antibody specific for the 40 kDa antigen. Specific binding moieties may include whole immunoglobulin G (IgG) antibodies made up of four immunoglobulin peptide chains, two heavy chains and two light chains and immunoglobulin M (IgM), as well as immunoglobulin fragments, which are protein molecules related to antibodies and which retain the epitopic binding specificity of the original antibody, such as Fab, F(ab)'₂, Fv, etc. Specific binding moieties for the 40 kDa antigen also include single chain antibodies and recombinant peptides constructed to retain the paratope configuration of antibodies specific for the 40 kDa antigen, as well as other molecules constructed to specifically bind the antigen.

Antibodies which are specifically reactive with the antigen of this invention may be obtained in a number of ways which will be readily apparent to those skilled in the art. The antigen obtained as described above can be injected into an animal as an immunogen to elicit polyclonal antibody production. Purification of the antibodies can be accomplished by selective binding of antibodies from the serum of the immunized animal to the 40 kDa antigen purified as described herein. For example, antibodies which specifically bind to this antigen can be isolated (e.g., from serum of humans with active TB) by binding to immobilized 40 kDa antigen and subsequently eluted from the immobilized antigen. This invention also contemplates monoclonal antibodies specifically immunoreactive with the 40 kDa antigen, which may be prepared according to well known methods (See, e.g., Kohler and Milstein, 1976, Eur. J. Immunol, 6:611), using the antigen of this invention as an immunogen, using it for selection or using it for both functions. These and other methods for preparing antibodies that are specifically immunoreactive with the 40 kDa antigen are easily within the skill of the ordinary worker in the art.

Recombinantly produced antibodies, including single chain antibodies of equivalent specificity may be prepared by recombinant DNA methods, including conventional molecular biology, microbiology, and recombinant DNA techniques within the skill of the art. Such techniques are well known to the skilled worker and are explained fully in the literature. See, e.g., Maniatis, Fritsch & Sambrook, "Molecular Cloning: A Laboratory Manual" (1982); "DNA Cloning: A Practical Approach," Volumes I and II (D.N. Glover, ed., 1985); "Oligomicleotide Synthesis" (M.J. Gait, ed., 1984); "Nucleic Acid Hybridization" (B.D. Hames & S.J. Higgins, eds., 1985); "Transcription and Translation" (B.D. Hames & S.J. Higgins, eds., 1984); "Animal Cell Culture" (R.I. Freshney, ed., 1986); "Immobilized Cells and Enzymes" (IRL Press, 1986); B. Perbal, "A Practical Guide to Molecular Cloning" (1984), and Sambrook, et al., "Molecular Cloning: a Laboratory Manual" (1989).

Suitable recombinant antibodies may be obtained using phage display technology, as described in a very comprehensive and well-written article by Burton and Barbas, 1994, Adv. Immunol., 57: 191, and also discussed in Winter et al., 1994, Ann. Rev. Immunol, 12:433, both of which are incorporated herein by reference. The vectors described by Burton and Barbas may be used to make a library from which the sequences of the antibodies of this invention may be selected. DNA segments or oligonucleotides having specific sequences can be synthesized chemically or isolated by one of several approaches. The basic strategies for identifying, amplifying and isolating desired DNA sequences as well as assembling them into larger DNA molecules containing the desired sequence domains in the desired order, are well known to those of ordinary skill in the art. See, e.g., Sambrook, et al., (1989); B. Perbal, (1984). Preferably, DNA segments corresponding to immunoglobulin variable regions may be isolated using the polymerase chain reaction (M. A. Innis, et al., "PCR Protocols: A Guide To Methods and Applications," Academic Press, 1990). Suitable primers may be constructed using sequences selected from the constant region flanking the variable regions, as shown in Kabat, et al. (1991), "Sequences of Proteins of Immunological Interest," Fifth Edition, U.S. Dept. of Health & Human Services, Washington D.C. A complete sequence may be assembled from overlapping oligonucleotides prepared by standard methods and assembled into a complete coding sequence. See, e.g., Edge (1981) Nature, 292:756; Nambair, et al. (1984) Science, 223: 1299; Jay, et al. (1984) J. Biol. Chem., 259:6311.

The assembled sequence can be cloned into any suitable vector or rephcon and maintained there in a composition which is substantially free of vectors that do not contain the assembled sequence. This provides a reservoir of the assembled sequence, and segments or the entire sequence can be extracted from the reservoir by excising from DNA in the reservoir material with restriction enzymes or by PCR amplification. Numerous cloning vectors are known to those of skill in the art, and the selection of an appropriate cloning vector is a matter of choice (see, e.g., Sambrook, et al., incorporated herein by reference). The construction of vectors containing desired DNA segments linked by appropriate DNA sequences is accomplished by techniques similar to those used to construct the segments. These vectors may be constructed to contain additional DNA segments, such as bacterial origins of replication to make shuttle vectors (for shuttling between prokaryotic hosts and mammalian hosts), etc. Antigens purified as described above may be used in standard binding studies to select suitable clones encoding recombinant antibodies specific for the antigens. Antibodies specific for the antigens of this invention, may be used to identify the antigens of this invention. A preparation of affinity- purified antibodies is useful as a standard for the diagnostic assay. The antibodies of this invention may also be used in competitive immunoassays, sandwich immunoassays, etc., for detection of serum antibodies specific for the 40 kDa and 6 kDa antigens in samples of patient serum. Antibodies (e.g., monoclonal antibodies, affinity purified antibodies, antibody fragments or recombinant antibodies) of equivalent specificity may also be injected as immunogens to elicit anti-idiotype antibodies. These anti-idiotype antibodies will bind to serum antibodies specific for the antigens of this invention, and therefore anti-idiotype antibodies may be used in immunoassays, such as those described herein for diagnosis of active TB, in place of the 40 kDa or 6 kDa antigens. Anti-idiotype antibodies may likewise be substituted for the antigens in other applications, such as injection to elicit an immune response.

EXAMPLES

In order to facilitate a more complete understanding of the invention, a number of Examples are provided below. However, the scope of the invention is not limited to specific embodiments disclosed in these Examples, which are for purposes of illustration only.

Example 1. Preparation and purification of the 40 kDa antigen.

Mtb cultures (in 1 liter batches) were grown at 37 °C in Middlebrook 7H9 medium supplemented with Tween 80 and ADC (Sigma). Mtb cultures were maintained in a biosafety level 3 (P-3) laboratory facility. Late-log phase Mtb cultures were autoclaved, and the bacterial debris removed by centrifugation. The 40 kDa antigen and 6 kDa antigen were released into the culture medium following autoclaving. Furthermore, this antigenic activity was not found in medium recovered from viable bacterial cultures.

The autoclave extract was digested with proteinase K, as the antigen is resistant to such digestion, while nearly all antigenic proteins are degraded by this procedure. Supernatant was recovered after autoclaving and digested with Proteinase K (lOOμg/ml) in the presence of 0.1% SDS at 55 °C for 18 hr. Degraded peptide fragments were removed by ultrafiltration; both the 40 kDa and the 6 kDa were retained by a 30 kDa cutoff membrane after digestion with proteinase K. The antigen-containing fraction was then size-selected from the crude digest by dialysis in phosphate buffered saline (PBS) using 30 kDa-cutoff membranes and concentrated 20- 50 fold by ultrafiltration using 30 kDa-cutoff membranes (Centricon-30). Concentrated preparation of the antigens in PBS was then filtered through 0.2 μm filters and stored at 4°C. The antigens which survive proteinase K digestion are surprisingly stable. Little or no loss of antigenic activity was observed in the concentrated preparation in PBS at 37° C for 1 day, and this extrapolates to approximately 3 months of stability when stored at 4°C. Prediluted (ready- to-use form) secondary antibody conjugates are stable for at least one year at 4°C.

Storage of 20-50 fold-concentrated antigen in PBS at 4°C for 2-4 weeks led to the formation of crystals. The antigenic nature of these crystals was determined by showing the loss of antigenic activity from the preparation upon removal of the crystals and by restoration of the antigenic activity by adding back the crystals and resolubilizing at room temperature overnight. When the crystals were isolated and resolubilized in fresh PBS, the resulting solution contained large amounts of antigenic activity. Measurement of this solution with the Bradford Protein assay indicated the presence of some protein. Crystallization may result in substantial purification of the antigen, making this a possible step in a process for preparation of the antigen.

The resolublized antigen preparation was determined not to contain DNA or RNA based on the absence of an absorbance peak at 260 nm and the inability to bind ethidium bromide. The preparation was diluted 1 : 10 in PBS, and the ultraviolet (UV) absorption of this solution was determined in a Beckman DU-65 spectrophotometer. The UV spectrum showed a peak at 280 nm, suggestive of residual protein contamination. DNA contamination may also be expected, if the extraction procedure is particularly severe (such as reextraction of the pellet).

Example 2. Mycobacterial Antigen Preparation

Mycobacteria were cultured in 1 liter of Middlebrook 7H9 medium containing ADC and Tween 80 until the culture reached late-log growth. Then the culture was autoclaved for 30 min., and the autoclaved material was centrifuged to pellet insoluble debris (20,000 rpm, 30 min). The supernatant was transferred to a fresh flask, and the pellet was resuspended in 250 ml Buffer A (10 mM NaCl, 10 mM Tris-HCl (pH 7.4), and 1 mM EDTA) and recentrifuged. The supernatants were pooled and SDS was added to 0.1 %> final concentration. Proteinase K was added to the pooled supernatants to 100 μg/ml final concentration and allowed to digest the mycobacterial lysate at 55 °C for 12-18 hr. The resulting clarified lysate was concentrated 20-fold using an Amicon concentrator unit with a 30 kDa- cutoff membrane, and the concentrate was dialyzed against Buffer A using a 12-14 kDa cutoff membrane. The concentrated lysate was then dialyzed at room temperature against Buffer B (Bμffer A containing 4 M urea). The resultant lysate was loaded onto a DEAE-Sephadex column pre-equilibrated in Buffer B, and the column was washed with 5 column volumes of Buffer B to remove unbound material. Proteins were eluted from the column using 10 column volumes of Buffer B containing 300 mM NaCl. Antigens were eluted from the column using 5 column volumes of Buffer B containing 800 mM NaCl. The eluate containing antigens was dialyzed against Buffer A and concentrated. 40 kDa antigen was isolated from either crude or DEAE- purified material by addition of 2 volumes of isopropanol, followed by incubating for 1 hr at room temperature, and centrifuging to pellet the precipitate which contains the 40 kDa antigen. The pellet was resuspended in Buffer A at 37 °C and dialyzed against Buffer A at room temperature.

Example 3. Comparison of LAM and Mtb Antigens.

The crude Mtb antigens prepared by the procedure of Example 1 were compared with LAM by several criteria. SDS gel fractionated antigens were electrophoretically tiansferred onto nylon membranes. Purified LAM was fractionated in adjacent lanes for comparison. Initially, a binding to the lectin concanavalin A (Con A) was used to detect the presence of carbohydrates. These studies revealed that the crude Mtb antigen preparation contained a rapidly migrating carbohydrate-containing species and a slower migrating carbohydrate-containing species with apparent mobilities of 6 and 40 kDa respectively.

Crude Mtb antigens and purified LAM obtained from Mtb cultures were separated on SDS-PAGE, electrophoretically transferred to nylon membranes, and carbohydrates were detected by incubating the membranes with Con A conjugated to horse radish peroxidase (HRP). Purified LAM was detected as a major carbohydrate-containing band with an apparent mobility on SDS-PAGE of approximately 40 kDa. Soluble Mtb lysates contain carbohydrate containing bands at about 40 kDa and about 6 kDa. Reactivity with the anti-LAM monoclonal antibody CS-35 identified a specific band of approximately 40 kDa in the LAM preparation, which was not observed in the crude Mtb antigen preparation (Figure 1). Detection of LAM was accomplished by incubating the membranes with the anti-LAM monoclonal antibody CS35, followed by goat-anti mouse-HRP. Figure 1 shows SDS gels developed with CS35. Mtb antigen concentrate was apphed to the gel in lane 1, a 1 :5 dilution in lane 2, and a 1:25 dilution in lane 3. Purified LAM was placed in lane 4, and a 1 :5 dilution of purified LAM in lane 5. Furthermore, mild alkaline hydrolysis of the molecules, conditions known to release the lipid moiety from LAM (Hunter, et al., 1990, J. Biol. Chem. 265: 9272), resulted in a loss of serum reactivity against LAM, but not against the 40 kDa and 6 kDa antigens. In Figure 2, lane 1 contains purified Mtb antigens incubated with 0.10 N NaOH at 37 degrees for 2h; lane 2: LAM incubated with 0.10 N NaOH at 37 degrees for 2h; lane 3: Mtb Antigens incubated without 0.10 N NaOH at 37 degrees for 2h; lane 4: Mtb antigens incubated without 0.1 N NaOH at 37 degrees for 2h, lane 5: untreated Mtb antigen, and Lane 6: untreated LAM. The absence of a band in lane 2 shows that immunoreactivity of LAM, but not Mtb antigens, was abolished by mild base treatment.

Treatment with alpha-mannosidase abolished the ability of LAM to be recognized by serum antibodies, but did not affect ability of the 40 kDa and 6 kDa antigens to be recognized by serum antibodies. Lastly, treatment with periodic acid to hydrolyze carbohydrate moieties abolished the antigenicity of 40 kDa antigen, 6 kDa antigen, arid LAM. Together, these data demonstrate that we have identified two carbohydrate Mtb antigens which are chemically and antigenically distinct from LAM. These data are summarized in Table 1. Table 1. Comparison of the heat-resistant (40 kDa and 6 kDa) Mtb antigens with lipoarabinomannan (LAM). Refer to the text for methodology and interpretation.

Characteristic Mtb Antigens LAM

Protease-resistant YES YES

Periodate-Schiff staining YES YES

Con A binding YES YES

LAM MAb CS-35 binding NO YES

Mild NaOH-sensitive NO YES

Mannosidase-sensitive NO YES

Example 4. Reactivity of serum antibodies against crude Mtb antigens and LAM.

The reactivity of serum antibodies present in sera from TB patients with the crude antigens according to this invention and with LAM was evaluated. Crude Mtb antigens (prepared as described in Example 1) and purified LAM were fractionated by SDS-PAGE, transferred to nylon membranes, incubated with TB sera, and developed using goat anti-human antibody conjugated to HRP. The results are shown in Figures 3-5. The 40 kDa antigen extracted from Mtb (lane 2) and purified LAM (lane 3) incubated with TB-positive sera (Figure 3). The relative migration of the 40 kDa and 6 kDa Mtb antigens are indicated. Figures 4 and 5 show similar blots incubated with sera from former TB patients (i.e., previously but not actively infected).

A pattern of immunoreactivity was observed that was virtually identical to that observed using Con A. Antibodies present in TB patient sera bound to the 40 kDa and 6 kDa carbohydrates, and to LAM (Figure 3). In most cases, normal serum did not react with antigens from either of the preparations. However, it should be noted that while both the 40 kDa and 6 kDa antigens were recognized by serum antibodies from patients with active TB, only the 6 kDa antigen and LAM were recognized by serum antibodies from former TB patients (i.e. patients who previously had active TB, but did not have not active disease at the time the serum was obtained). This suggests that the memory B cell response against the 40 kDa antigen declines rapidly after treatment whereas the humoral response against the 6 kDa antigen was long-lived.

Example 5. Antigen extracted from Various Mycobacterial species.

Western blots of antigens extracted from various mycobacterial species were developed with antisera from a normal (TB-free) individual, two different TB-positive sera and antisera from normal and M bovis- infected cows. The antigens used in the Western blots developed with human sera were extracted by the procedure in Example 1 from M bovis, M. kansasii, andM. avium, and the resulting material compared to Mtb antigen prepared by the procedures in Example 1 or Example 2. The antigens used the experiments with cow serum were extiacted by the procedure in Example 1 fromM bovis BCG or a clinical isolate of M bovis.

Sera from human patients with active TB reacted with the 40 kDa and 6 kDa antigens from all Mycobacteria sp. tested. Sera from 13 mycobacterially-infected cows and elk reacted with both antigens, while sera from 2 uninfected animals did not; similar results were found for antigens from Mtb and M kansasii.

Example 6. Dot blot- and ELISA-based serodiagnostics.

Two distinct formats are described below for the analysis of antigenic activity and the detection of serum antibodies which recognize the 40 kDa Mtb antigen: the dot blot and ELISA formats.

Dot blot format. Three dilutions (1: 10, 1:25, 1:50) of the concentrated 40 kDa antigen stock solution from Example 1 are deposited onto a nitrocellulose membrane (0.7 cm x 5 cm) using high precision ceramic micropumps which are capable of delivering 0. l-1.0μl volume per stroke with a maximum of 5% variability. Purified human IgG (1 μg/dot) is applied on the membrane strip as a positive control for the validity of the test reagents. The membrane dot blot strips are air-dried, blocked with a blocking buffer containing 2% BSA, washed with PBS containing 0.05% Tween-20, and then dried at 40 °C for 1 hr. Dried dot blots are packed in static-free, air-tight pouches, heat-sealed, and stored at room temperature. ELISA format. One hundred μl of a 1 :500 dilution (in PBS) of concentrated 40 kDa antigen stock solution prepared as described in Example 1 is dispensed automatically into the wells of 96 well ELISA plates. Plates are then covered with an adhesive film and incubated at room temperature for 1 hr in order to allow antigen binding to the plate. Any unbound antigen in the wells is removed by washing with PBS containing 0.05% Tween-20. The wells are then blocked with a blocking buffer containing 1% BSA for 1 hr at room temperature. Plates are washed with PBS containing Tween-20 and dried for 1 hr at 40 °C. Plates are then sealed in static-free pouches and stored at 4°C. Wells which did not receive antigen are used as a negative control. Purified human IgG and IgM (1 μg/well) are applied to additional wells as a positive control for the validity of the test reagents.

Dot blot and ELISA plates are used for serological testing with panels of clinically confirmed TB-positive and normal human sera (1:25 dilution), and primary antibody binding to the 40 kDa antigen is deteπnined using secondary antibody conjugates (alkaline phosphatase for dot blots and peroxidase for ELISA) with appropriate enzyme substrates for detection. Dot blots are scored visually, and ELISA plates are read at 450 nm. Total time for completion of the dot blot assay is approximately 30 min and less than 2 hr for the ELISA assay. ELISA assay tests are scored as positive where absorbance reading are greater than a calculated cutoff value. Cutoff values are determined by measuring the absorbance reading obtained using normal human sera (typically 0.1-0.3 OD₄₅₀) plus a correction factor to reflect the variability observed between normal serum donors.

Representative examples of the results obtained using the dot blot assay and sera from 11 different donors are shown in Figure 6. Dot blot assays were performed using normal and TB sera collected from various donors. Dilutions of Mtb antigen prepared according to Example 1 were applied to membranes, and the membranes were first incubated with either TB patient or normal sera as indicated. Some TB serum donors were also HIV positive (lanes 1-3). Some normal sera were from healthy PPD positive donors (lanes 9 and 10). Bound human serum antibodies were detected using enzyme-conjugated secondary antibodies specific for both human IgG and IgM. A mixture of human IgG and IgM (hulg) is included as a positive control. Figure 7 shows the results of dot blots using dilutions of similarly prepared Mtb antigen applied to a membrane. Membranes were incubated with either TB patient (TB) or normal (N) sera, and bound human serum antibodies were detected using enzyme-conjugated secondary antibodies (2nd Ab) specific for human IgM (M) or IgG (G). In more than 80% of sera tested, the antigen was recognized by both IgM and IgG antibodies present in the sera of TB patients. In some cases, sera contained only IgM antibodies which recognized the antigen (see Figure 7). The results Figures 6 and 7 indicate that a serodiagnostic assay using the Mtb antigen of this invention is highly sensitive and specific for active mycobacterial disease state. Example 7. Analysis of serum antibody reactivity against the Mtb antigens.

Sera from actively TB-infected or normal individuals were obtained from different sites in the United States, Africa, India, and Estonia. The clinical status of sera was determined by sputum culture. 216 serum samples were tested by ELISA, and 455 serum samples were tested by dot blot analysis, as described in Example 6.

Table 2 summarizes the results of these tests. The ELISA assay showed a sensitivity of 99% and a specificity of 100% for the samples evaluated. The dot blot produced a similarly high estimate of sensitivity (97%) and a somewhat lower estimate of specificity (81%>). The clinical status of these patients, other than sputum culture data, was generally not available. However, many individuals who tested TB-positive in the ELISA or dot blot assays were later found to be HIV-positive, suggesting that concurrent infection with HIV did not block the ability of these patients to mount a humoral response to Mtb, although the sensitivity of the assays appears to be lower in HIV-positive patients. 5 serum samples from patients infected with M avium were tested to determine if antibodies which recognized the 40 kDa antigen and/or 6 kDa antigen could be detected; 4 of 5 sera from patients infected with M avium scored positive in the assay.

Table 2. Preliminary evaluation of the sensitivity and specificity of ELISA and dot blot serodiagnostic assays based on the use of the 40 kDa and 6 kDa Mtb antigens.

ELISA: Dot Blot:

Sample Test posJ Test neg./ Test pos./ Test negJ Source Sputum pos. Sputum neg. Sputum pos. Sputum neg.

USA 103/105 53/53 141/144 40/48

Africa 36/36 ND 44/45 ND

India 22/22 ND 22/22 ND

Estonia ND ND 133/140 44/56

Totals: 161/163 53/53 340/351 84/104

(99%) (100%) (97%) (81%)

ELISA and dot blot assays were performed as described in Example 6. Serum samples were obtained from patients in which a diagnosis of TB was made on the basis of sputum culture. Assays were developed using enzyme- conjugated secondary antibodies which are specific for both human IgG and IgM. ND = not determined.

Example 8. Blinded Study Correlating Antigen with Disease State

A blinded study was performed on 196 serum samples obtained from subjects at Tartu University Lung Hospital in Estonia and Ben Taub General Hospital in Houston, Texas, using the dot blot assay as described in Example 6, except that Mtb antigens prepared as described in Example 1 were applied to each strip in serial 2-fold dilutions. Highest concentration of crude antigen is located at the bottom of the strip. A single application of human IgG at the top of each strip served as a positive control. Four technicians carrying out the testing found the test to be technically straight forward and relatively simple. Table 3 shows the correlation between the TB clinical status and the results of testing using the dot blot test-system. Table 3. Correlation between clinical status and positive dot blot test

Total number of samples tested 196^a Number of clinically positive TB subjects 139

Number of test-system positives of clinically positive subject 119

Number of test-system negatives of clinically positive subjects 20

Number of BCG inoculated but clinically negative subjects 55

Number of BCG inoculated, negative subjects, negative by test-system 46 Number of BCG inoculated, negative subjects, positive by test-system 9

Therefore, the Sensitivity is 119 / 139 = 86%

Specificity is 46 / 55 = 84% ^a There were technical problems with two samples, one from a clinically positive subject and one from a negative subject. These samples were not included in the calculations.

Example 9.

Sera from patients with different diagnoses were tested using the dot blot test-system of Example 8 using proteinase K-digested antigen concentrate. The total number of sera tested for evaluation was 15, out of which 9 were confirmed negative and 10 were positive samples. Among the positive samples, 9 were from actively infected TB patients and one had Sarcoidosis rather than active Tuberculosis. The 9 negative samples consisted of 2 people who had TB in the past, 3 from BCG group who did not have active Tuberculosis, 1 with Silicosis, 2 were from healthy donors, and one from a patient who had active TB. Results are shown in Table 4. Table 4. Results from dot blot screen of patient sera

Sera Type No. of Sera No. (%) of Positive No. (%)of Negative tested reactions reactions

Patients with 10 9 (90%) 1 (10%)

Tuberculosis

Patients without 9 1 (11%) 8 (89%) Tuberculosis

Summary Results for Dot Blot Screen: Sensitivity 90 % Specificity 89 %

Example 10. Testing of a dot blot serological test system.

Tests to determine the anti-TB antibodies in human serum or plasma specimens were performed utilizing the dot blot test-system of Example 8 using proteinase K-digested antigen concentrate. Antibodies were determined in plasma specimens of 108 persons with active TB, 42 persons with unspecified lung diseases and 50 healthy donors. There were the following results among patients infected with the Mycobacterium species complex: 54 with infiltration TB, 23 with focal TB, 9 with fibrous- cavernous TB, 2 with cavernous TB, 5 with disseminated TB, 3 with tubercular TB, 9 with intrathoracic lymph node TB, 3 with TB pleurisy. Among patients without tuberculosis disease of the lungs: 10 with pneumonia, 10 with chronic bronchitis, 8 with sarcoidosis, 8 with bronchial asthma, 6 with different types of alveolitises. In addition, 50 healthy donors were tested. The following test results were determined: a) individuals, who were infected with TB had 71 (or 65.7%) positive results; b) persons without tuberculosis pathology of the lungs had 4 (or 9.5%) positive results; c) healthy donors had 4 (or 8%) positive results. Thus, sensitivity of the test-system for the screening of individuals with TB is 65.7%; specificity is 92% (100 - 8%). Furthermore, the test-system was found to be simple to conduct, no special equipment being necessary for the conducting of tests, and the test could be rapidly completed.

The dot blot test-system for the early diagnosis of Tuberculosis, is especially useful for "field" conditions (primarily in the group of patients with a higher risk of tuberculosis).

Example 11.

224 tests were conducted concerning the clinical testing of the reagent kit dot blot test system described in Example 8 using proteinase K- digested antigen concentrate, which is a rapid test to detect active

Tuberculosis by testing human sera or plasma specimens. For the conducting of the test, the following materials were used: a) 133 healthy donor serum specimens; b) 13 serum specimens of patients with unspecified lung diseases; c) 78 serum specimens of individuals with active TB.

The healthy donor sera were collected at the Moscow city department for blood transfusion. The sera of patients with pulmonary TB and without tuberculosis disease of the lungs were obtained from patients of the Russian Federation Scientific Research Institute of Pulmonology clinic. The patients with active TB were tested, most of them (about 80%) have Mycobacteria in secretions. As a result of the testing, the results in Table 5 were obtained: Table 5. Results of Russian Studies

Group of Testing Number Number (%) of

Number/(%) of of Tests positive reactions negative reactions

Healthy Donors 133 15 (11.3%) 1 1 8

(88.7%)

Without TB Disease 13 2 (15.4%) 1 1

(84.6%) of the Lungs

TB of the Lungs 78 55 (70.5%) 2 3

(29.5%)

Thus, specificity of the test-system for screening patients with TB is 88.7%, with a number of false positive reactions of 11.3%. Sensitivity of the test-system is 70.5%. A positive feature in the use of the test system was reported to be the simplicity of test conducting; in that no special equipment was necessary, and test completion was rapid (30 specimens of the test having been completed in 1.5 hours).

Example 12.

The Russian Scientific Research Institute Pulmonology and Tuberculosis MZMP PF tested 44 specimens of sera from patients with pulmonary TB (adult persons: 30, and children: 14) as well as 51 specimens of sera from healthy donors using the dot blot test-system of Example 8 using proteinase K-digested antigen concentrate. The preliminary test results revealed the following: sensitivity =85%, specificity =94%. Taking into account the simphcity of test conducting, it may be concluded that such tests may be appropriate for screening test purposes even in small laboratories in identifying patients with TB disease. For purposes of clarity of understanding, the foregoing invention has been described in some detail by way of illustration and example in conjunction with specific embodiments, although other aspects, advantages and modifications will be apparent to those skilled in the art to which the invention pertains. The foregoing description and examples are intended to illustrate, but not limit the scope of the invention. Modifications of the above-described modes for carrying out the invention that are apparent to persons of skill in medicine, immunology, and/or related fields are intended to be within the scope of the invention, which is limited only by the appended claims.

All publications and patent applications mentioned in this specification are indicative of the level of skill of those skilled in the art to which this invention pertains. All publications and patent applications are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

Claims

CLAIMS:

1. A mycobacterial antigen of about 40 kDa apparent molecular weight measured by SDS-PAGE, said antigen being specifically bound by antibodies from patients suffering from active tuberculosis and resistant to digestion by proteinase K as determined by retention of antigenicity and mobility on SDS gels, said antigen further being substantially free of 6 kDa mycobacterial antigens said 6 kDa antigen being resistant to proteinase K digestion and specifically immunoreactive with antibodies from patients suffering from active tuberculosis.

2. The mycobacterial antigen of claim 1, wherein said antigen: is retained by a 30 kDa cutoff ultrafiltration membrane after proteinase K digestion, is resistant to alkanolysis by 0.1 N NaOH, as measured by antigenicity, can be stained with periodate-Schiff reagent, but not ethidium bromide, and is not soluble in either phenol or chloroform.

3. The mycobacterial antigen of claim 1, wherein said antigen is extracted from lysed cells of Mycobacterium sp.

4. The mycobacterial antigen of claim 3, wherein the Mycobacterium sp. is selected from M avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M. phei, M. simiae, M. smegmatis, M. tuberculosis, and M ulcerans.

5. The mycobacterial antigen of claim 3, wherein the Mycobacterium sp. is selected fromM avium, M. bovis, M. kansasii, andM. tuberculosis.

6. The mycobacterial antigen of claim 1, wherein said antigen is substantially free of other mycobacterial antigens.

7. A method for obtaining an antigen reactive with antibodies present in patients suffering from active tuberculosis, said method comprising obtaining an extract from lysed cells of Mycobacterium sp., digesting the extract with proteinase K, ultrafiltering the digested extract using an ultrafiltration membrane with size cutoff of 20 kDa or greater, the antigen being retained in the ultrafiltration residue, applying the ultrafiltrate residue to an anion exchange column, washing the column with moderate ionic strength buffer and eluting the antigen with buffer of ionic strength greater than 500 mM, and mixing the eluate with an excess of polar organic solvent and recovering a precipitate containing the antigen.

8. The method of claim 7, wherein the anion exchange column has diethylaminoethyl groups, and the polar organic solvent is isopropanol, added at 2: 1 v/v.

9. The method of claim 7, wherein the Mycobacterium sp. is selected fromM avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M. phei, M. simiae, M. smegmatis, M. tuberculosis, and M ulcerans.

10. The method of claim 7, wherein the Mycobacterium sp. is selected from M avium, M. bovis, M. kansasii, M. leprae, and M tuberculosis.

11. A diagnostic method for determining the presence of active tuberculosis in a patient comprising detecting, in a sample from the patient, antibodies specifically immunoreactive with the antigen according to claim 1.

12. A mycobacterial antigen having an apparent molecular weight of about 6 kDa measured by Western blot developed using antisera from an individual with active tuberculosis (TB), wherein said antigen retaining antigenicity and mobility in SDS PAGE after digestion by proteinase K, said antigen being substantially free of a second mycobacterial antigen of 40 kDa apparent molecular weight measured by Western blot developed using antisera from an individual with active tuberculosis (TB), said second mycobacterial antigen likewise being retained by a 30 kDa cutoff ultrafiltration membrane after digestion by proteinase K.

13. The mycobacterial antigen of claim 12, wherein said antigen: is retained by a 30 kDa cutoff ultrafiltration membrane after digestion by proteinase K, is resistant to alkanolysis by 0.1 N NaOH, as measured by antigenicity, can be stained with periodate-Schiff reagent, but not ethidium bromide, and is not soluble in either phenol or chloroform.

14. The mycobacterial antigen of claim 12, wherein said antigen is extracted from lysed cells of Mycobacterium sp.

15. The mycobacterial antigen of claim 14, wherein the Mycobacterium sp. is selected from M. avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M. phei, M. simiae, M. smegmatis, M. tuberculosis, and M ulcerans.

16. The mycobacterial antigen of claim 14, wherein the Mycobacterium sp. is selected from M avium, M. bovis, M. kansasii, M. leprae, andM. tuberculosis.

17. A method for obtaining an antigen reactive with antibodies present in patients suffering from active tuberculosis, said method comprising obtaining an extract from lysed cells of Mycobacterium sp., digesting the extract with proteinase K, ultrafiltering the digested extract using an ultrafiltration membrane with size cutoff of 20 kDa or greater, the antigen being retained in the ultrafiltration residue, applying the ultrafiltrate residue to an anion exchange column, washing the column with moderate ionic strength buffer and eluting the antigen with buffer of ionic strength greater than 500 mM, and mixing the eluate with an excess of polar organic solvent, and removing precipitate from the antigen containing solution.

18. The method of claim 17, wherein the anion exchange column has ╬▒iemylaminoethyl groups, and the polar organic solvent is isopropanol, added at 2: 1 v/v.

19. The method of claim 17, wherein the Mycobacterium sp. is selected fromM avium, M. bovis, M. chelonei, M. fortuitum, M. gastris, M. intracellulare, M. kansasii, M. leprae, M. nonchromogenicum, M phei, M. simiae, M. smegmatis, M. tuberculosis, and M ulcerans.

20. The method of claim 17, wherein the Mycobacterium sp. is selected from M avium, M. bovis, M. kansasii, M. leprae, and M tuberculosis.

21. A diagnostic method to aid in diagnosis of tuberculosis in a patient comprising detecting, in a sample from the patient, antibodies specifically immunoreactive with the antigen according to claim 12.

22. A diagnostic method to aid in diagnosis of tuberculosis in a patient comprising deteπnining the presence of absence of antibodies which specifically bind a mycobacterial antigen having an apparent molecular weight of about

6 kDa measured by Western blot developed using antisera from an individual with active tuberculosis (TB), said antigen being resistant to digestion by proteinase K; and determining the presence of absence of antibodies which specifically bind a mycobacterial antigen having an apparent molecular weight of about

40 kDa measured by Western blot developed using antisera from an individual with active tuberculosis (TB), said antigen being resistant to digestion by proteinase K; wherein the presence of antibodies which bind the 6 kDa antigen and the absence of antibodies that bind the 40 kDa antigen is indicative of past tuberculosis disease in the patient.

23. A diagnostic kit for detecting active tuberculosis, comprising a first mycobacterial antigen of about 40 kDa apparent molecular weight measured by SDS-PAGE, said first antigen being specifically bound by antibodies from patients suffering from active tuberculosis and resistant to digestion by proteinase K as determined by retention of the antigen by a 30 kDa cutoff membrane; and a second mycobacterial antigen having an apparent molecular weight of about 6 kDa measured by Western blot developed using antisera from an individual with active tuberculosis (TB), said second antigen being retained by a 30 kDa cutoff ultrafiltration membrane after digestion by proteinase K; said first antigen being substantially free of said second antigen, and said second antigen being substantially free of said first antigen.

24. A diagnostic kit according to claim 23, wherein said first antigen is immobilized on a solid support, and said second antigen is immobilized on a solid support.