Dynamic IgG antibody response to immunodominant ...

1 downloads 0 Views 674KB Size Report
[4] C.C. Boehme, P. Nabeta, D. Hillemann, M.P. Nicol, S. Shenai, F. Krapp, J. Allen, R. .... [47] K.R. Steingart, N. Dendukuri, M. Henry, I. Schiller, P. Nahid, P.C. ...
Clinica Chimica Acta 461 (2016) 25–33

Contents lists available at ScienceDirect

Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim

Dynamic IgG antibody response to immunodominant antigens of M. tuberculosis for active TB diagnosis in high endemic settings Balaji Pathakumari, Maddineni Prabhavathi, Deenadayalan Anbarasu, Pukazhvanthen Paramanandhan, Alamelu Raja ⁎ Department of Immunology, National Institute for Research in Tuberculosis (ICMR), No. 1, Mayor Sathyamoorthy Road, Chetput, Chennai 600 031, India

a r t i c l e

i n f o

Article history: Received 28 February 2016 Received in revised form 27 June 2016 Accepted 27 June 2016 Available online 28 June 2016 Keywords: M. tuberculosis Pulmonary tuberculosis Latent tuberculosis Recombinant proteins Sero diagnosis Immunoglobulin G ELISA

a b s t r a c t Background: Even though various techniques have been developed for rapid diagnosis of tuberculosis (TB), still there is an immense need for a simple, cost effective, highly sensitive and specific test. Hence, one of the possibilities is identification of Mycobacterium tuberculosis specific antibodies in infected serum by using specific antigens. Methods: We tested 10 recombinant M. tuberculosis antigens to evaluate IgG levels among Healthy control subjects (HCS), Healthy household contacts (HHC) and pulmonary TB patients (PTB) by ELISA. Results: The median IgG levels specific to all the antigens are higher in PTB than HHC and HCS. Amongst single antigens, 38-kDa antigen has showed maximum sensitivity of 50% than any other antigens at 95.5% specificity. Among the two antigen combination, 38-kDa + Rv1860 has showed maximum sensitivity of 66.6% with specificity of 92.2%. The same antigen combination (38-kDa and Rv1860) predominantly identifies smear negative and culture positive TB patients with 68% sensitivity and 92.2% specificity. Most of the antigens have exhibited higher antibody titre in cavitary TB than non cavitary. With regard to latent TB infection (LTBI) identification, Rv1860 has exhibited maximum sensitivity of 53.3% with 95% specificity. Conclusions: IgG response to combination of recombinant mycobacterial antigens (38-kDa, Rv1860, Rv2204c and Rv0753c) presents good specificity with acceptable level of sensitivity for TB diagnosis. © 2016 Published by Elsevier B.V.

1. Introduction According to World Health Organization (WHO) global tuberculosis (TB) 2015 report, one-third of the world's population is infected with Mycobacterium tuberculosis (M. tuberculosis), the causative agent of human TB. It is estimated that ~ 9.6 million people developed TB and 1.5 million died from TB in 2014 [1]. India alone accounted for 24% of total cases among the 95% of TB deaths that occurred in low- and middle-income countries [1]. Targets for the post-2015 global TB strategy include a 95% reduction in TB deaths and a 90% reduction in TB incidence by 2035. To achieve these targets, implementation of new diagnostic tools is utmost important. There are two major gaps in the existing diagnostic tools: lack of simple and accurate point of care Abbreviations: M. tb, Mycobacterium tuberculosis; LTBI, latent tuberculosis infection; WHO, World Health Organization; DOTS, Directly Observed Treatment Short Course; TST, tuberculin skin test; IGRA, interferon-gamma release assay; PPD, purified protein derivative; HCS, healthy control subjects; HHC, healthy house-hold contacts; PTB, pulmonary tuberculosis; QFT-GIT, Quantiferon TB Gold In Tube assay; TMB, tetramethylbenzidine. ⁎ Corresponding author at: National Institute for Research in Tuberculosis (ICMR), No. 1, Mayor Sathyamoorthy Road, Chetput, Chennai 600 031, India. E-mail address: [email protected] (A. Raja).

http://dx.doi.org/10.1016/j.cca.2016.06.033 0009-8981/© 2016 Published by Elsevier B.V.

(POC) test for TB that can be used for rapid diagnosis at the primary care level; and lack of a biomarker (or combination of biomarkers) that can be used to identify latently infected individuals who will benefit from preventive therapy. Although acid-fast staining of bacilli in sputum smear is a simple and relatively fast test, it has reduced sensitivity and requires N 104 bacilli per ml of sputum for the reliable detection of active TB [2]. Culture test is considered as the gold standard diagnostic test for active TB, however results take weeks to obtain and it is expensive and need a well-equipped laboratory, trained staff, and an efficient transport system to ensure viable specimens. Moreover about 5–10% active TB cases often gives false negative results [3]. Although chest X-ray can be useful for the diagnosis of active TB, it is not specific. To make the diagnosis of TB as more accurate, rapid and convenient, new diagnostic techniques such as molecular methods and immune reactions based on cell-mediated-immune (CMI) or humoral immune response have been investigated. Polymerase chain reaction (PCR) based molecular methods for detecting M. tuberculosis-specific nucleic acids, especially WHO endorsed GeneXpert MTB/RIF have revolutionized the diagnosis of active TB and rifampicin resistant TB. However, they are costly and require technological investment [4]. Recently introduced T cell immune based assay, Interferon Gamma Release Assays (IGRA) have been successful for detection of M. tuberculosis infection, by virtue

26

B. Pathakumari et al. / Clinica Chimica Acta 461 (2016) 25–33

of their higher sensitivity and specificity than TST [5]. Although, IGRAs show great sensitivity in detecting latent TB infection (LTBI) and active TB cases, their performance is affected in immunosupressed individuals [6]. In addition, IGRA cannot differentiate between LTBI and active TB disease [5,7], which forbid their usage in high endemic settings. Lack of effective diagnostic tests is responsible for the delay in TB diagnosis, which leads to progression of the disease and ultimately becomes the potential source for M. tuberculosis transmission. Humoral immune-based tests confer advantages over the conventional methods due to their speed (results may be available within hours) and simplicity [8]. It has been demonstrated that identification of M. tuberculosis specific antibodies in the infected serum by enzyme linked immune sorbent assay (ELISA) is highly sensitive and reproducible technique [9]. Serodiagnosis is characterized by convenient sampling, low costs, easy operation, and rapid determination which can be implemented in clinical laboratories in resource-constrained settings, where access to diagnostic instruments is limited and cost efficiency has high priority [10]. 1.1. Rationale and hypothesis of the study Selection of antigen and its antibody isotype are the key elements for any successful serological assays. Most of the serological assays for active TB diagnosis use a single antigen (38-kDa protein), which may not be recognized uniquely by the host immune system. Since, there is significant heterogeneity in antigen recognition and response, the diagnostic efficacy of single antigen is hampered. However, it can be improved by identification and inclusion of cocktail antigens which will react with the sera of most of the infected individuals [11,12]. Currently, number of M. tuberculosis antigens has been characterized for serodiagnosis of active TB. Unfortunately, none of the antigens showed as the promising candidate for active TB diagnosis. Identification of additional immunodominant antigens to enhance the performance of sero-diagnostic tests is of great interest in the field of TB diagnosis. Therefore, it is necessary to test immuno-dominant antigens for improving the accuracy of TB identification by serodiagnosis. Many immuno-dominant antigens have been identified by various stringent techniques for serodiagnosis of TB disease [13–15]. In this context, our laboratory has identified various secreted M. tuberculosis antigens based on their ability to provoke cellular immune response [16]. Most of these antigens act as both T cell and B cell antigens. Amongst, we have selected 10 antigens (Rv2204c, Rv3716c, Rv0753c, Rv0009, Rv1860, Rv2626c, Rv3914, Rv1908, Ag85a and 38-kDa) for this study, since there is no serodiagnostic data on most of the selected antigens. In this study, we have cloned, over expressed and purified these 10 antigens and evaluated IgG response among Healthy control subjects (HCS), Healthy house-hold contacts (HHC) and pulmonary TB patients (PTB) by ELISA. We checked the usefulness of these antigens in the identification of active TB and differential diagnosis of latent and active TB. In addition, we also analysed dynamic IgG response to the selected antigens for the identification of different stages of active TB based on smear and cavity status. 2. Materials and methods 2.1. Antigen preparation The details of recombinant proteins, expression vectors and expression strains of Escherichia coli, used in this study were showed in Table 1. As described previously, the expression and purification of recombinant proteins were done by standard protocols followed in our laboratory [17–19]. 2.2. Study population This study was approved by institutional ethical committee established by National institute for Research in Tuberculosis (NIRT),

Table 1 Recombinant antigens of M. tuberculosis used in this study. Protein name

Gene number

Molecular mass (kDa)

Expression vector

Expression strain of E. coli

38-kDa Ag85a mpt32 Rv2204c (hp) mmsA PpiA Trxc Rv2626c (hp) KatG Rv3716c (hp)

Rv0934c Rv3804c Rv1860 Rv2204c Rv0753c Rv0009 Rv3914 Rv2626c Rv1908 Rv3716c

38 36 32.7 12.5 55 16 12.5 15.5 80.5 13.3

pET23a pET23a pET23a pRSET-A pRSET-A pRSET-A pET23a pET23a pET23a pRSET-A

BL21(DE3)pLysS BL21(DE3) BL21(DE3) BL21(DE3) BL21(DE3) BL21(DE3)pLysS BL21(DE3)pLysS BL21(DE3)pLysS BL21(DE3)pLysS BL21(DE3)pLysS

hp - hypothetical protein.

Chennai. All the subjects provided written informed consent before drawing the blood. Subjects with previous history of TB, those under immunosuppressive therapy were excluded from this study. All the subjects were confirmed as human immunodeficiency virus (HIV) negative by routine acquired immunodeficiency syndrome (AIDS) tests. HIV positive individuals, patients with diabetes, cancer, autoimmune diseases or other conditions that may affect the immune system of the individual, pregnant women, children, and individuals with indeterminate Quantiferon-TB Gold In-Tube test (QFT-GIT) results were also excluded from the study. A total of 270 subjects were included in three groups. 2.2.1. Healthy control subjects (HCS) Ninety serum samples were collected from control subjects. All the enrolled HCS were from families where there was no history of TB and know exposure to TB; and they were recruited from offices, schools, colleges and slum areas in Chennai city and nearby villages. Active TB among HCS were ruled out by negative sputum smear microscopy, had normal chest X-ray and without clinical symptoms. Tuberculin skin test (TST) and QFT-GIT were performed in healthy controls to rule out LTBI in controls. 2.2.2. Healthy house hold contacts (HHC) Ninety study subjects were recruited from families where there was one sputum positive case (index case) staying in the same household for at least 3 months immediately before the start of the anti TB treatment of the index case. These study subjects were identified by visiting the households of adult smear positive pulmonary TB patients who were enrolled for treatment in Revised National Tuberculosis Control Program (RNTCP) centres, Chennai. Disease free status was ruled out by negative sputum smear, normal X-ray and without clinical symptoms. Latent infection was confirmed by both TST and QFT-GIT. Among the HHCs, 58 were showed as positive by both TST and QFTGIT. In the present study both TST and QFT-GIT positive HHCs subjects only considered as latently infected individuals. The subjects of this group were followed up for possible breakdown to TB and at the end of 6 months found to be healthy. 2.2.3. Pulmonary TB patients (PTB) For this study, 90 Pulmonary tuberculosis patients (PTB) were recruited from RNTCP centres, Chennai, India. Two spot and one overnight sputum specimens were collected from each patient. The presence of active TB was defined as Ziehl-Neelsen-stained sputum smear positive or positive for mycobacterial culture or positive for chest X-ray. Among the 90 PTB patients, 65 patients were positive for both smear and culture; and 25 were negative for smear and positive for culture. Out of 90 patients, cavities were found in the lungs of 13 patients and 40 did not show cavities in the lungs. The cavity status of majority of the patients (37) was unknown. TB patients with multidrug-resistant tuberculosis (MDR-TB) infection, active TB patients with more than

B. Pathakumari et al. / Clinica Chimica Acta 461 (2016) 25–33

seven consecutive days of anti -TB treatment were excluded from the study. 2.3. Tuberculin skin (TST) Individual HCS and HHC were injected intradermally with 2TU purified protein derivative (PPD) (Statens Serum Institute) in the left forearm. After 48–72 h post-injection, the diameters of both axes of skin induration were measured and recorded by a certified doctor. The cutoff point for TST positivity was considered as N10 mm. 2.4. Quantiferon-TB Gold In-Tube assay (QFT-GIT) IGRA was performed using QFT-GIT test (Cellestis, Qiagen, Venlo, Netherlands) [20,21]. Briefly, 1 ml of blood was taken in each of the three tubes which were precoated with M. tuberculosis-specific antigens (ESAT-6 & CFP-10) for test sample tube, phytohemagglutinin (PHA) for positive control tube and no antigen or saline coated tube for the negative control. The tubes were incubated for 16–24 h at 37 °C and plasma was collected after centrifugation. The collected supernatants were stored at − 20 °C until assayed. The cytokine IFN-γ was measured in plasma by ELISA as per manufacturer's instructions. The test results were interpreted using software supplied by the manufacturer (Cellestis, Qiagen, Venlo, Netherlands). 2.5. ELISA As described elsewhere, indirect ELISA was performed to estimate IgG levels in sera against all the antigens [22]. Briefly, 96-well polystyrene plates (Nunc-Immuno Maxisorp polystyrene plates, Nunc A/S) were coated with 100 μl of antigen (concentration 1, 2, or 5 μg/ml) of our interest and incubated overnight at 4 °C. After washing 3 times with 1 × PBST (300 μl per well), plates were blocked with 200 μL of blocking buffer (10% goat serum in 1% casein) for 1 h at 37 °C and then washed with 1 × PBST. Then, serum was added in 1:1000 dilutions to the plates and then incubated for 1 h at 37 °C. After washing 5 times with 1 × PBST, 100 μl of goat antihuman IgG peroxidase conjugate (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) was added at optimal dilution (1:20,000) and then incubated for 1 h at 37 °C. The plates were washed for 7 times. Then 100 μl of 3.3′,5,5′ tetramethylbenzidine (TMB; Sigma Aldrich) was added to the plates and incubated for 20–30 min until the colour was seen. Finally colour development was arrested with 50 μl of 2 mol/l H2SO4 and the plates were read at 450 nm with 650 nm as the correction wavelength using an ELISA reader (Molecular Devices, CA, USA). 2.6. Statistical analysis All analyses were performed with GraphPad Prism ver 5.00 for Windows. Median antibody levels among three groups were compared by Kruskal-Wallis (KW) test. Dunn's multiple compare test was used as post hoc test. Receiver operator characteristic (ROC) curves were used to determine the cut-off values for each antigen. Differences were accepted as significant when the p-value was b0.05. 3. Results Demographic characteristics of recruited subjects were depicted in Table 2. IgG antibody titre in all the three groups against all the antigens were showed in Fig. 1. 3.1. Serological antibody response to recombinant antigens in HCS and PTB Diagnostic utility of M. tuberculosis antigens in HCS and PTB were evaluated by ROC. p value b 0.05 was considered as statistically significant. The AUC of M. tuberculosis antigens were above 0.6, suggesting

27

Table 2 Demographic profiles of study subjects. Group

No of subjects

Healthy control subject (HCS) TST+ TST− TST status unknown QFT-GIT+ QFT-GIT− TST− QFT-GIT− Healthy household contact (HHC) TST+ TST− TST status unknown QFT-GIT+ QFT-GIT− TST+ -QFT-GIT+ Pulmonary tuberculosis (PTB) S+ C+ S− C+ Cavity Non-cavity Unknown status

90 10 70 10 26 64 60 90 75 10 5 65 25 58 90 65 25 13 40 37

TST - tuberculin skin test; QFT-GIT - Quantiferon Gold In Tube.

that these antigens may have discriminative capacity between HCS and PTB (Fig. 2). The lowest cut-off value with N 95% specificity and maximum sensitivity was considered as optimal. Individuals were scored as positive for the specific antibody response if the absorbing OD value was above the cut-off value. Based on the cut-off value, we calculated specificity in HCS and sensitivity in PTB group (Table 3). The median IgG levels of all the antigens were higher in PTB compared to HCS. Except Rv3914, remaining 9 antigens showed significant difference between HCS and PTB (p b 0.05). In this study, the recombinant antigens demonstrated various sensitivities ranging from 14% to 39%, at constant 95% specificity. As we expected, the antigen 38-kDa appeared to recognize a large proportion of the TB sera with low levels of background binding (specificity). This antigen showed maximum sensitivity of 50% than any other antigens studied. Rv1860 showed 39% sensitivity, followed by Ag85a which showed 33% sensitivity. Two of the hypothetical proteins, Rv2204c and Rv3716c showed 29% sensitivity and another hypothetical protein Rv2626c showed 15.5% sensitivity. The other antigens such as Rv0009, Rv0753c and Rv1908 presented the lowest reactivity to PTB patients and showed 14.4%, 23%, 16.6% sensitivity in PTB respectively (Table 3). 3.2. IgG response against combination of antigens In order to increase the sensitivity and specificity of antigens, we combined the antibody response to 2 antigens and 3 antigen results. For combination analysis, sensitivity was calculated as number of subjects positive for either of antigens among PTB patients and specificity was calculated as number of subjects negative for both the antigens among HCS subjects. As we expected after combination of antigens, sensitivity was increased but specificity was also slightly reduced (Table 3). Among the two antigen combination, 38-kDa + Rv1860 showed maximum sensitivity of 66.6% (60/90) with specificity of 92.2% (7/90). Another antigen combination Rv2204c + 38-kDa exhibited 64.4% (58/90) sensitivity with 91.1% (8/90) specificity. Among the three antigen combination Rv2204c + Rv0753c + 38-kDa showed 72.2% (65/90) sensitivity and 86.6% (12/90) specificity. The combination, 38-kDa + Ag85a + Rv1860 yielded maximum specificity of 88.8% (10/90) than any other three antigen combination and also showed 71.1% (64/90) sensitivity. Combination of four antigens did not further improve the sensitivity and moreover showed reduced specificity. Therefore, compared to single antigen, two (38-kDa + Rv1860) or three antigen (38-kDa + Ag85a + Rv1860) combinations were effective for the diagnosis of active TB (Table 3). Thus, antibody detection against

28

B. Pathakumari et al. / Clinica Chimica Acta 461 (2016) 25–33

Fig. 1. Scatter plot of IgG levels against nine antigens in HCS, HHC and PTB. Each dot represents optical density (OD450) of individual serum sample. The solid horizontal line in each group indicates median. HCS, healthy control subject; HHC, healthy house hold contacts; PTB, pulmonary tuberculosis.

multiple M. tuberculosis antigens is crucial for improving the detection accuracy of TB.

66.1% (43/65) sensitivity and 91.1% (8/90) specificity. Amongst the three antigen combinations, Rv2204c + Rv0753c + 38-kDa gave 72.3% (47/65) sensitivity and 86.6% (12/90) specificity (Table 4).

3.3. Serological reactivity of each patient group The results were further analyzed based on smear, culture and cavity nature of TB patients. Thus, based on smear and culture status, the patients were sub divided into two groups: sputum and culture positive TB patients (S+ C+) and sputum-negative and culture positive TB patients (S− C+). Based on nature of cavity, TB patients were sub divided into two groups: patients with cavities and without cavities. 3.3.1. IgG response based on smear and culture status of PTB Out of 90 PTB patients, 65 were positive for both smear and culture tests. Remaining 25 patients were smear negative and culture positive. 3.3.1.1. Serum antibody response in S+ C+ category. At a fixed 95% specificity as determined by cut-off value, individual antigens showed sensitivity ranged from 17% to 52% in smear and culture positive patients. Patients who were positive for both sputum smear and culture, were more likely to react to 38-kDa (53.2%) followed by Rv1860 (40%) and Ag85a (38%). Amongst the two antigen combinations, Rv2204c + 38-kDa showed

3.3.1.2. Serum antibody response in S− C+ category. The individual antigens yielded 8 to 48% sensitivity (at a prefixed 95% specificity) in smear negative and culture positive cases. In S− C+ PTB cases, 38-kDa identified 12 patients with 36% sensitivity at 95% specificity. Amongst combination analysis, the combination of 38-kDa and Rv1860 identified 19 patients with sensitivity of 68% at 92.2% specificity. In three antigen combination 38-kDa + Rv1860 + Rv3716c exhibited a maximum sensitivity of 76% (19/25) and a maximum specificity of 90% (9/90) (Table 4). 3.3.2. IgG response based on nature of cavity We further analyzed the sensitivity of antigens based on severity of TB disease i.e., patients surviving with and without cavity. Out of 90 patients, cavities were found in the lungs of 13 patients and 40 patients did not show cavities in the lungs. The cavity status of 37 patients was unknown. In the current study, we found PTB patients with radiographic cavities and/or infiltrates were more likely to react to 38-kDa antigen. This antigen identified 6 TB cases with the sensitivity of 38.6% at 95%

B. Pathakumari et al. / Clinica Chimica Acta 461 (2016) 25–33

29

Fig. 2. Receiver operating characteristics curve analysis of IgG antibody response against nine antigens in healthy control subjects and pulmonary TB. ROC plots showing the accuracy of IgG in discriminating between HCS and PTB.

Table 3 Diagnostic potential of IgG response in HCS and PTB against single antigens or combinations of antigens. Antigen

Sensitivity

Specificity

Cut-off

p value

AUC

Rv2204c Rv0753c Rv0009 Ag85a 38-kDa Rv1860 Rv2626c Rv1908 Rv3716c 38-kDa + Rv1860 Rv2204c + 38-kDa 38-kDa + Rv3716c Ag85a + Rv1860 Ag85a + 38-kDa + Rv1860 Rv2204c + Rv0753c + 38-kDa

28.89 (26/90) 23.33 (21/90) 14.44 (13/90) 33.33 (30/90) 50 (45/90) 38.89 (35/90) 15.56 (14/90) 16.67 (15/90) 28.89 (26/90) 66.6 (60/90) 64.4 (58/90) 60 (54/90) 55.5 (50/90) 71.1 (64/90)

95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 92.2 (7/90) 91.1(8/90) 93.3 (6/90) 92.2 (7/90) 88.8 (10/90)

1.875 2.575 2.206 2.732 1.449 1.458 2.091 1.795 0.692 NA NA NA NA NA

0.0001 0.0001 0.0002 0.0001 0.0001 0.0001 0.0001 0.0292 0.0001 NA NA NA NA NA

0.7307 0.6928 0.6606 0.773 0.7202 0.7471 0.8009 0.5941 0.6725 NA NA NA NA NA

72.2 (65/90)

86.6 (12/90)

NA

NA

NA

NA-not applicable.

specificity. Among the two antigen combination, 38-kDa + Rv1860 displayed 61.5% (8/31) sensitivity with 92.2% (7/90) specificity. Within the three antigen combinations, Rv0009 + Ag85a + Rv1860 identified maximum of 9 cases with the sensitivity of 69.2% at 88.8% (10/90) specificity (Table 4). The reference antigen 38-kDa had higher sensitivity than the rest of the antigens in TB patients without cavity. Out of 40 non-cavity TB cases, 38-kDa antigen identified half of the cases (20), followed by Rv1860 which recognized 14 cases with 95% specificity. Among the two antigen combinations, maximum sensitivity of 50% (20/40) was attained by 38-kDa and Rv1860 combination, with the specificity of 92.2% (7/90). Among the three antigen combination, 38-kDa + Rv1860 + Rv2626c displayed maximum sensitivity of 72.5% (29/40) with the specificity of 88.8% (10/90) (Table 4). 3.4. IgG response against M. tuberculosis antigens in HHC Among the 10 antigens Rv2204c and Rv0753c showed significant difference between HCS and HHC. At predefined 95% specificity, Rv2204c showed 27.7% (25/90) sensitivity whereas Rv0753c showed 30% (27/90) sensitivity. In the current study, both TST/QFT positive

30

B. Pathakumari et al. / Clinica Chimica Acta 461 (2016) 25–33

Table 4 Sensitivity and specificity based on smear and cavitary status in PTB group. Antigen

Specificity in HCS (n = 90)

% sensitivity Cavitary PTB (n = 13)

Non-cavitary PTB (n = 40)

PTB [S+ C+] (n = 65)

PTB [S− C+] (n = 25)

Rv2204c Rv0753c Rv0009 Ag85a 38-kDa Rv1860 Rv2626c Rv1908 Rv3716c Rv2204c + 38-kDa Rv0753c + 38-kDa 38-kDa + Rv1860 Rv2204c + Rv0753c + 38-kDa Ag85a + 38-kDa + Rv1860 Rv0009 + Ag85a + 1860 38-kDa + Rv1860 + Rv2626

95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 95.56 (4/90) 91.1(8/90) 91.1(8/90) 92.2 (7/90) 86.6 (12/90) 88.8 (10/90) 88.8 (10/90) 88.8 (10/90)

23 (3/13) 15.3 (2/13) 23 (3/13) 38.4 (5/13) 46.1 (6/13) 30.7 (4/13) 15.3 (2/13) 0 7.6 (1/13) 61.5 (8/13) 61.5 (8/13) 61.5 (8/13) 69.2 (9/13) 69.2(9/13) 69.2(9/13) 61.5 (8/13)

27.5 (11/40) 22.5 (9/40) 15 (6/40) 27.5 (11/40) 50 (20/40) 35 (14/40) 17.5 (7/40) 12.5 (5/40) 22.5 (9/40) 67.5 (27/40) 55 (22/40) 67.5 (27/40) 72.5 (29/40) 72.5 (29/40) 57.5 (23/40) 72.5 (29/40)

33.8 (22/65) 23 (15/65) 15.3 (10/65) 38.4 (25/65) 53.2 (33/65) 40 (26/65) 16.9 (11/65) 20 (13/65) 27.6 (18/65) 66.1 (43/65) 60 (39/65) 64.6 (42/65) 72.3 (47/65) 67.6 (44/65) 61.5 (40/65) 67.6 (44/65)

16 (4/25) 24 (6/25) 12 (3/25) 20 (5/25) 48 (12/25) 36 (9/25) 12 (3/25) 8 (2/25) 32 (8/25) 60 (15/25) 56 (14/25) 68 (17/25) 68 (17/25) 76 (19/25) 64 (16/25) 72 (18/25)

and negative subjects were included in HHC group. So, we regrouped the HHC subjects by taking, HHC subjects who were positive for both TST and QFT and considered them as LTBI. In similar way HCS subjects, negative for both TST and IGRA were considered as uninfected healthy controls. Out of 90 HHC, 58 subjects were positive for both TST and IGRA. On the other hand out of 90 HCS, 60 showed negative for both TST and IGRA. After stratifying the groups, we assessed sero-diagnostic performance of selected antigens for discriminating HCS and HHC. Among the 10 antigens, Rv0753c and Rv2626c antigens showed significant difference between HCS and HHC. At a prefixed specificity of 95%, Rv0753c antigen showed 39.6% (23/58) sensitivity whereas Rv2626c showed 10.3% (6/58) sensitivity.

3.5. Diagnostic potential of IgG response in discriminating HHC and PTB Although the median values against all the antigens were higher in PTB than HHC, only 6 antigens showed significant difference between these two groups. These include Rv2204c, Rv0009, Ag85a, 38-kDa, Rv1860 and Rv3716c. the optimal cut-off values (where the maximum specificity of 95%) were chosen based on ROC (Table 5). Among the 6 antigens, Rv1860 was more associated to latent TB than any other antigen. This antigen showed maximum sensitivity of 53.3 (48/90) and maximum specificity of 95% (3/58). The sensitivities of Rv2204c, Rv0009, Ag85a, 38-kDa and Rv3716c exhibited 8.8%, 10%, 6.6%, 22.2%, and 7.7% respectively at a fixed 95% specificity. To improve the efficiency of the sero-diagnostic test for discriminating latent and active TB, we approached combined antigen analysis. Among the two antigen combination, 38-kDa + Rv1860 showed maximum sensitivity of 58.8% (53/ 90) with maximum specificity of 91.3% (5/58). Among the three antigen Table 5 Diagnostic evaluations of serum IgG levels in LTBI and PTB group against single antigens or combinations 2 or 3 antigens. Antigen

Sensitivity

Specificity

Cut-off

p value

AUC

Rv2204c Rv0009 Ag85a 38-kDa Rv1860 Rv3716c Rv2204c + Rv1860 Rv38-kDa + Rv1860 Rv2204c + Rv0009 + Rv1860 Rv0009 + Ag85a + Rv1860

8.889 (8/90) 10 (9/90) 6.667 (6/90) 22.22 (20/90) 53.33 (48/90) 7.778 (7/90) 57.5 (52/90) 58.8 (53/90) 63.3 (57/90)

94.83 (3/58) 94.83 (3/58) 94.83 (3/58) 94.83 (3/58) 94.83 (3/58) 94.83 (3/58) 91.3 (5/58) 91.3 (5/58) 87.9 (7/58)

2.660 2.310 3.413 1.67 1.353 1.127 NA NA NA

0.0101 0.0008 0.004 0.0001 0.0001 0.0055 NA NA NA

0.6254 0.6635 0.6404 0.7033 0.7761 0.6353 NA NA NA

62.2 (56/90)

87.9 (7/58)

NA

NA

NA

combination, Rv2204c + Rv0009 + Rv1860 identified 57 cases with 63.3% sensitivity and specificity of 88% (7/58) (Table 5). 4. Discussion Though efficient treatment is available for active TB, thousands of TB deaths are reported every day worldwide. Unfortunately to date, there is no simple, rapid and efficient diagnostic test available for identifying TB, leads to continued transmission of infection in the community. Therefore, early detection and successful treatment of patients with TB is the cornerstone of TB control programmes. Patients with active TB disease have elevated level of anti-M. tuberculosis IgG antibodies [23,24]. Although the role of these antibodies in protective immunity is still under investigation, it has been proposed that they can be utilized as a diagnostic marker for TB disease [25–27]. However, WHO issued a negative policy statement, concluding that M. tuberculosis antibody tests should not be used for the diagnosis of pulmonary and extrapulmonary TB [28]. However, the Expert Group strongly recommending further research to identify and validate alternative serological tests with improved accuracy [29]. Therefore, it is necessary to test novel antigens for improving the accuracy of active TB identification by serodiagnosis. In this context, our laboratory has identified various immunodominant antigens by screening entire secretory proteome of M. tuberculosis [16]. Amongst, we have selected 10 antigens for this study and there is no serodiagnostic data on most of the selected antigens. In this study, we have cloned, over expressed and purified in these proteins in E. coli by recombinant technology, since the specificity of purified recombinant antigens is highly specific for M. tuberculosis complex [30,31]. Although recombinant antigens may have a risk if losing some immunogenic epitopes, yet majority of the epitopes can be recognized by antibodies from infected individual. For the past decade, antibodies against several highly purified M. tuberculosis recombinant antigens have been evaluated as potential serological markers for the identification of active TB disease with relatively high sensitivity and specificity [22,32]. These findings have renovated the concept that serological tests using recombinant M. tuberculosis antigens are workable for the diagnosis of active TB. In this study, we measured IgG response against 10 recombinant proteins by ELISA and evaluate the diagnostic usefulness of different antigen combination for identifying active TB. Five antigens (Rv2204c, Rv3716c, Rv0753c, Rv0009 and Rv1860) are known to be novel B cell antigens among the selected antigens. To our knowledge, this is the first study to evaluate sero-diagnostic potential of these antigens in high endemic country. Among these antigens, Rv2204c and Rv3716c are noteworthy. Being hypothetical proteins, Rv2204c and Rv3716c both have showed a sensitivity of 28.8% and specificity of 95.6%. When the positivity of these two antigens is summed up,

B. Pathakumari et al. / Clinica Chimica Acta 461 (2016) 25–33

sensitivity has augmented to 42.2% with 92.2% specificity. The additive response of Rv2204c and 38-kDa antigen has showed a higher sensitivity of 64.4% with 91% specificity. In the similar manner, the result of both Rv3716c and 38-kDa has displayed 60% sensitivity at 93% specificity. Upon combination of Rv1860 with Ag85a, the sensitivity is increased to 55.5% in PTB with 92% specificity. Therefore, refinement of antigen cocktails (Rv2204c + 38-kDa, Rv3716c + 38-kDa, Rv1860 + Ag85a) may lead to improved sensitivity and specificity for active TB diagnosis. In line with our study, Wu and co-workers achieved 69.5% sensitivity and 91.1% specificity by using five antigens combination [33]. Therefore, the combination of potential antigens not only enhances sensitivity but also hold higher specificity, which may be valuable for the clinical diagnosis of active TB. The 38-kDa antigen has been the most studied antigen for the serodiagnosis of TB [34]. In line with other serological studies, we also used 38-kDa antigen as a reference antigen. However, the recognition frequency reported for the 38-kDa antigen varies greatly (16–94%), depending on the smear status and stage of the disease [33,35]. Though 38-kDa antigen showed high specificity, it lacks sensitivity due to heterogeneous antibody response to single antigen. However, 38-kDa antigen's sensitivity can be improved by combining with other antigens. In this study, we observed 38-kDa antigen showed significantly higher IgG levels in PTB than HCS. This antigen exhibited 50% sensitivity in PTB and 95% specificity in HCS. After combining the positivity of Rv1860 with positivity of 38-kDa, the duo sensitivity is increased to 66.6% and specificity is reduced to 92%. Thus, 38-kDa antigen and Rv1860 has shown low IgG response in HCS suggesting that these antigens can be implemented for diagnosis of active TB. The results of three antigen combination (Rv2204c, Rv0753c and 38-kDa) have offered the better sensitivity (72%) in PTB and 87% specificity in HCS. Since, the other tested antigens showed redundant pattern of reactivity, the overall sensitivity has not been increased. Even though, some of the antigens predicted to have potential immunogenicity, there is no significant difference between HCS and active TB. This is due to the fact that, some antigens can highly elicit T cell response instead of humoral response. Among the 10 antigens, 9 have shown significant difference between HCS and TB patients in terms of IgG response, however the diagnostic performance of these antigens is poor compared to other studies. First, the antigen from different sources has different sensitivities in various study population. Moreover the different post-translational modifications between recombinant proteins and native antigens extracted from a M. tuberculosis complex may affect the humoral immune response. Besides, the immune response also varies with the great heterogeneity from individual to individual, smear status and disease stages [11,36]. Further, smear status, stages of disease, geographical regions may also be the reasons resulting in the different sensitivities of the same antigen. Since antibody levels associate with the bacterial load, serological test can able to detect sputum smear-positive TB cases with higher sensitivity [37]. Sputum smear positive cases of pulmonary TB patients are the main source of transmission of infection. They are responsible for almost 95% of the transmission of infection in the community. An untreated sputum smear-positive patient may infect on average N 10 contacts annually and over 20 contacts during the natural history of the disease until death [38]. In this study, we have scrutinized the positivity obtained by individual antigens and their combinations based on smear and culture positivity. Among the individual antigens, 38-kDa antigen showed maximum of 53.3% sensitivity. Among the two antigen combinations, Rv2204c + Rv1860 showed 66% (43/65) sensitivity and 91% specificity (8/90). The three antigen combination Rv2204c + Rv0753c + 38-kDa has resulted in 72.3% (47/65) sensitivity and 86.6% (12/90) specificity. Although the obtained sensitivity is comparable with other studies, still it is lower than other studies [22,33]. A previous study on Rv1860 reported that, recombinant Rv1860 showed reactivity in 11 out of 39 (28%) smear-positive patients and none from smear negative patients [15]. However, in the present study, Rv1860

31

has showed comparable sensitivity in both smear positive (40%) and smear negative TB patients (36%). Previous studies have reported that recombinant 38-kDa antigen showed the sensitivity ranged 16–36% for smear-negative patients and 36–67% for smear-positive patients [10,39]. In contrast to above studies, in the current study we did not find much difference in the sensitivity for both smear positive (53%) and smear negative cases (48%). Diagnosis of smear-negative TB cases is also crucial because these patients are capable of transmitting M. tuberculosis. Because, studies showed that smear negative patients is reporting more often with HIV co-infection [40–42]. Averagely 30–40% untreated smear negative patient will develop active TB within first two year period. Accumulating literature highlighted that antibody response is more common in smear positive active TB patients than smear negative TB patients [14, 42]. Importantly, the levels of antibodies against 38-kDa, Ag85a and Rv1860 antigens in smear-negative TB patients are significantly higher than those in healthy controls indicating their value in the diagnosis of smear negative patients. In the current study, there are 25 smear negative and culture positive TB patients. The 38-kDa antigen has identified 12 out of 25 smear negative and culture positive TB patients at 95% specificity. However, we have attained maximum sensitivity of 76% (19/25) and maximum specificity of 90% (9/90) by three antigen combination (Rv0009 + Ag85a + Rv1860). Thus, compared to individual antigens, using combination of antigens will be helpful in the diagnosis of smear negative and culture positive cases. Earlier in our laboratory, by measuring IgG and IgA isotypes against three antigen combinations attained 70% sensitivity in smear negative TB cases [35], whereas in the present study by measuring single isotype (IgG), we achieved 76% sensitivity and 90% specificity. Thus, characterization of IgG against these M. tuberculosis antigens (Rv0009 + Ag85a + Rv1860) can effectively distinguish between smear negative TB patients and healthy controls. Since we evaluated diagnostic potential of these antigens in less number of smear negative cases, further evaluation has to be done in large number of subjects. In general, cavities will form in the lungs of advanced stage of TB patients than early stage of disease Compared to non cavitary, cavitary TB patients are highly contagious [8]. So, early identification of cavitary TB patients is foremost important. In the present study, among 90 TB patients, 13 patients are known to be cavitary patients. Among the individual antigens, 38-kDa antigen has showed 46% sensitivity at 95% specificity. By adding Rv1860 antigen to 38-kDa, the sensitivity is increased from 46% to 61.5% with the specificity of 92.1%. The three antigen combination, Rv0009 + Ag85a + Rv1860 has identified maximum of 9 cases with the sensitivity of 69.2% with 88.8% (10/90) specificity. In non-cavitary TB, 38-kDa antigen has identified maximum of 20 cases followed by Rv1860, identifies 14 cases with 95% specificity. By adding Rv1860 to 38-kDa antigen, sensitivity is increased from 50% to 67.5% and the specificity is reduced from 95% to 92.2% (7/90). The antigen Ag85a showed 27.5% sensitivity in non-cavitary group compared to 38.4% sensitivity in cavitary TB. Thus, as a single antigen Ag85a can able to predominantly recognizes the advanced stages of TB such as cavitary TB compared to early stage of TB such as non cavitary TB. The WHO estimated that about one-third of the total world population is harbouring LTBI [1]. Five to 10% of the LTBI individuals will develop active TB disease during their life span [43,44]. This makes individuals with LTBI as the largest reservoir of potential future source of active TB. Treatment of high risk individuals in LTBI population effectively reduces the risk of progression to active TB. In addition, early treatment of LTBI may decrease the emergence of multidrug resistance strains [45]. Moreover, those LTBI individuals showing the highest IgA or IgG antibody signals might have the highest risk of progression into active TB [46]. Therefore detection and management of LTBI in persons who are at risk is very important. Current diagnostic methods TST and IGRA are insufficient for differentiating LTBI and active TB. Studies in humans and animals demonstrated that the levels of antibody titre depend on the stages of M. tuberculosis infection [47]. In

32

B. Pathakumari et al. / Clinica Chimica Acta 461 (2016) 25–33

the present study, we also investigated the IgG responses to the selected antigens, and evaluated their ability to discriminate between LTBI and active TB. Earlier studies have focused on serodiagnostic potential of M. tuberculosis antigens that can distinguish active TB from LTBI [48]. In our study 90 HHC subjects are included. Since our setting is endemic country, the exact infection status of HHC is unknown. So, those HHC who are positive for both the tests such as TST and IGRA, are considered as latently infected individuals. Among the 90 HHC, 58 HHC subjects are positive for both TST and IGRA. We detected higher levels of anti IgG titre in the group with active TB compared to LTBI group. This finding is supported by other studies, reporting significant difference in IgG titres when compared active TB cases with LTBI and control cases [11,22]. Among the individual antigens, Rv1860 showed highest sensitivity than any other antigens. It showed 53.3% sensitivity in PTB with specificity of 94.5% in HHC. So, the positivity rate is significantly high in active TB compared to LTBI, which indicates 38-kDa is the optimal antigen for discriminating LTBI and active TB. After combination analysis, we didn't observe much difference between HHC and PTB compared to single antigen. Most of the patients who had circulating antibodies to Rv1860 also had antibodies to other antigens. This leads to, no increment in sensitivity by taking latter antigens into account. Thus, compare to combination of antigens, individual antigen showed significant difference between LTBI and active TB. Our results also revealed that higher IgG titres in active TB patients reflect the high antigen load. On the other hand, we also assessed the sero-diagnostic potential of selected antigens to differentiate the TST and QFT negative HCS and TST and QFT positive HHC. None of the antigens showed significant difference between these two groups. Thus, the selected antigens used in this study, may not be useful for discriminating healthy controls and LTBI. 5. Conclusion From this study we concluded that the IgG response to the selected recombinant mycobacterial antigens (38-kDa, Rv1860, Rv2204c and Rv0753c) present very good specificity and acceptable level of sensitivity for the diagnosis of TB in concordance with the published literature. These antigens are highly reactive and are potentially valuable candidates for inclusion in a serodiagnostic test. Further, our findings also indicated that IgG response to Rv1860 antigen effectively distinguished patients with active TB from LTBI. Acknowledgements The authors thank all the study subjects who participated in this study. Mr. Balaji Pathakumari expresses his gratitude to the Indian Council of Medical Research (ICMR), New Delhi, India, for providing the Senior Research Fellowship. Ms. Maddineni Prabhavathi expresses her gratitude to Council of Scientific and Industrial Research (CSIR), New Delhi, India for providing Senior Research Fellowship. The help rendered by Mr. Madhavan Dhanapal in performing ELISA is acknowledged. References [1] World Health Organization (WHO), Global Tuberculosis Control - Surveillance, Planning, Financing. Geneva: WHO Report, 2014 (WHO/HTM/TB/2014.08). [2] P. Mathew, Y.H. Kuo, B. Vazirani, R.H. Eng, M.P. Weinstein, Are three sputum acidfast bacillus smears necessary for discontinuing tuberculosis isolation? J. Clin. Microbiol. 40 (9) (2002) 3482–3484. [3] P. Andersen, M.E. Munk, J.M. Pollock, T.M. Doherty, Specific immune-based diagnosis of tuberculosis, Lancet 356 (9235) (2000) 1099–1104. [4] C.C. Boehme, P. Nabeta, D. Hillemann, M.P. Nicol, S. Shenai, F. Krapp, J. Allen, R. Tahirli, R. Blakemore, R. Rustomjee, et al., Rapid molecular detection of tuberculosis and rifampin resistance, N. Engl. J. Med. 363 (11) (2010) 1005–1015. [5] M. Pai, Spectrum of latent tuberculosis - existing tests cannot resolve the underlying phenotypes, Nat. Rev. Microbiol. 8 (3) (2010) 242 (author reply 242).

[6] B. Syed Ahamed Kabeer, R. Sikhamani, S. Swaminathan, V. Perumal, P. Paramasivam, A. Raja, Role of interferon gamma release assay in active TB diagnosis among HIV infected individuals, PLoS One 4 (5) (2009) e5718. [7] K. Dheda, S.R. van Zyl, M. Badri, M. Pai, T-cell interferon-gamma release assays for the rapid immunodiagnosis of tuberculosis: clinical utility in high-burden vs. lowburden settings, Curr. Opin. Pulm. Med. 15 (3) (2009) 188–200. [8] Y. Okuda, R. Maekura, A. Hirotani, S. Kitada, K. Yoshimura, T. Hiraga, Y. Yamamoto, M. Itou, T. Ogura, T. Ogihara, Rapid serodiagnosis of active pulmonary Mycobacterium tuberculosis by analysis of results from multiple antigen-specific tests, J. Clin. Microbiol. 42 (3) (2004) 1136–1141. [9] T.M. Daniel, S.M. Debanne, The serodiagnosis of tuberculosis and other mycobacterial diseases by enzyme-linked immunosorbent assay, Am. Rev. Respir. Dis. 135 (5) (1987) 1137–1151. [10] V.M. Silva, G. Kanaujia, M.L. Gennaro, D. Menzies, Factors associated with humoral response to ESAT-6, 38 kDa and 14 kDa in patients with a spectrum of tuberculosis, Int. J. Tuberc. Lung Dis. 7 (5) (2003) 478–484. [11] A. Davidow, G.V. Kanaujia, L. Shi, J. Kaviar, X. Guo, N. Sung, G. Kaplan, D. Menzies, M.L. Gennaro, Antibody profiles characteristic of Mycobacterium tuberculosis infection state, Infect. Immun. 73 (10) (2005) 6846–6851. [12] D. Tiwari, R.P. Tiwari, R. Chandra, et al., Efficient ELISA for diagnosis of active tuberculosis employing a cocktail of secretory proteins of Mycobacterium tuberculosis, Folia Biol. (Praha) 60 (2014) 10–20. [13] K. Weldingh, I. Rosenkrands, L.M. Okkels, T.M. Doherty, P. Andersen, Assessing the serodiagnostic potential of 35 Mycobacterium tuberculosis proteins and identification of four novel serological antigens, J. Clin. Microbiol. 43 (1) (2005) 57–65. [14] M.J. Sartain, R.A. Slayden, K.K. Singh, S. Laal, J.T. Belisle, Disease state differentiation and identification of tuberculosis biomarkers via native antigen array profiling, Mol. Cell. Proteomics 5 (11) (2006) 2102–2113. [15] K.M. Samanich, M.A. Keen, V.D. Vissa, J.D. Harder, J.S. Spencer, J.T. Belisle, S. ZollaPazner, S. Laal, Serodiagnostic potential of culture filtrate antigens of Mycobacterium tuberculosis, Clin. Diagn. Lab. Immunol. 7 (4) (2000) 662–668. [16] A. Deenadayalan, D. Heaslip, A.A. Rajendiran, B.V. Velayudham, S. Frederick, H.L. Yang, K. Dobos, J.T. Belisle, A. Raja, Immunoproteomic identification of human T cell antigens of Mycobacterium tuberculosis that differentiate healthy contacts from tuberculosis patients, Mol. Cell. Proteomics 9 (3) (2010) 538–549. [17] M. Prabhavathi, B. Pathakumari, A. Raja, IFN-gamma/TNF-alpha ratio in response to immuno proteomically identified human T-cell antigens of Mycobacterium tuberculosis - the most suitable surrogate biomarker for latent TB infection, J. Infect. 71 (2) (2015) 238–249. [18] B. Pathakumari, D. Anbarasu, R.T. Parthasarathy, A. Raja, PpiA antigen specific immune response is a potential biomarker for latent tuberculosis infection, Tuberculosis (Edinb) 95 (6) (2015) 736–743. [19] B. Pathakumari, M. Prabhavathi, A. Raja, Evaluation of cytokine and chemokine response elicited by Rv2204c and Rv0753c to detect latent tuberculosis infection, Cytokine 76 (2) (2015) 496–504. [20] M. Prabhavathi, B.S. Kabeer, A. Deenadayalan, A. Raja, Role of QuantiFERON-TB Gold antigen-specific IL-1beta in diagnosis of active tuberculosis, Med. Microbiol. Immunol. 204 (5) (2015) 567–574. [21] M. Prabhavathi, B.S. Ahamed Kabeer, A. Deenadayalan, A. Raja, In vitro QuantiFERON-TB gold antigen specific interleukin-1beta to diagnose TB among HIV-positive subjects, Tuberculosis (Edinb) 96 (2016) 27–30. [22] P. Pukazhvanthen, D. Anbarasu, S.A. Basirudeen, A. Raja, M. Singh, Assessing humoral immune response of 4 recombinant antigens for serodiagnosis of tuberculosis, Tuberculosis (Edinb) 94 (6) (2014) 622–633. [23] S. Gupta, N. Shende, A.S. Bhatia, S. Kumar, B.C. Harinath, IgG subclass antibody response to mycobacterial serine protease at different stages of pulmonary tuberculosis, Med. Sci. Monit. 11 (12) (2005) CR585–CR588. [24] M. Mizusawa, M. Kawamura, M. Takamori, T. Kashiyama, A. Fujita, M. Usuzawa, H. Saitoh, Y. Ashino, I. Yano, T. Hattori, Increased synthesis of anti-tuberculous glycolipid immunoglobulin G (IgG) and IgA with cavity formation in patients with pulmonary tuberculosis, Clin. Vaccine Immunol. 15 (3) (2008) 544–548. [25] B.L. Anderson, R.J. Welch, C.M. Litwin, Assessment of three commercially available serologic assays for detection of antibodies to Mycobacterium tuberculosis and identification of active tuberculosis, Clin. Vaccine Immunol. 15 (11) (2008) 1644–1649. [26] F. Abebe, G. Bjune, The protective role of antibody responses during Mycobacterium tuberculosis infection, Clin. Exp. Immunol. 157 (2) (2009) 235–243. [27] J. Grenier, L. Pinto, D. Nair, K. Steingart, D. Dowdy, A. Ramsay, et al., Widespread use of serological tests for tuberculosis: data from 22 high-burden countries, Eur. Respir. J. 39 (2012) 502–505. [28] World Health Organization, Commercial Serodiagnostic Tests for Diagnosis of Tuberculosis. Policy Statement, World Health Organization, Geneva, 2011. [29] K. Morris, WHO recommends against inaccurate tuberculosis tests, Lancet 377 (9760) (2011) 113–114. [30] U. Demkow, J. Ziolkowski, M. Filewska, B. Bialas-Chromiec, T. Zielonka, D. Michalowska-Mitczuk, J. Kus, E. Augustynowicz, Z. Zwolska, E. SkopinskaRozewska, et al., Diagnostic value of different serological tests for tuberculosis in Poland, J. Physiol. Pharmacol. 55 (Suppl. 3) (2004) 57–66. [31] L. Wu, M. Zhang, M. Sun, et al., Humoural immune responses to a recombinant 16kDa-38-kDa-ESAT-6 mycobacterial antigen in tuberculosis, J. Int. Med. Res. 39 (2011) 514–521. [32] A. Raja, U.D. Ranganathan, R. Bethunaickan, Improved diagnosis of pulmonary tuberculosis by detection of antibodies against multiple Mycobacterium tuberculosis antigens, Diagn. Microbiol. Infect. Dis. 60 (4) (2008) 361 e 8. [33] X. Wu, Y. Yang, J. Zhang, B. Li, Y. Liang, C. Zhang, M. Dong, Comparison of antibody responses to seventeen antigens from Mycobacterium tuberculosis, Clin. Chim. Acta 411 (19–20) (2010) 1520–1528.

B. Pathakumari et al. / Clinica Chimica Acta 461 (2016) 25–33 [34] W.H. Hwang, W.K. Lee, et al., Expression, purification and improved antigenicity of the Mycobacterium tuberculosis PstS1 antigen for serodiagnosis, Protein Expr. Purif. 95 (2013) 77e83. [35] K.R. Uma Devi, B. Ramalingam, P.J. Brennan, P.R. Narayanan, A. Raja, Specific and early detection of IgG, IgA and IgM antibodies to Mycobacterium tuberculosis 38kDa antigen in pulmonary tuberculosis, Tuberculosis (Edinb) 81 (3) (2001) 249–253. [36] M.S. Imaz, E. Zerbini, Antibody response to culture filtrate antigens of Mycobacterium tuberculosis during and after treatment of tuberculosis patients, Int. J. Tuberc. Lung. Dis. 4 (6) (2000) 562–569. [37] S. Kunnath-Velayudhan, A.L. Davidow, H.-Y. Wang, et al., Proteome-scale antibody responses and outcome of Mycobacterium tuberculosis infection in nonhuman primates and in tuberculosis patients, J. Infect. Dis. 206 (5) (2012) 697–705. [38] S.D. Lawn, B. Afful, J.W. Acheampong, Pulmonary tuberculosis: diagnostic delay in Ghanaian adults, Int. J. Tuberc. Lung Dis. 2 (8) (1998) 635–640. [39] S.T. Beck, O.M. Leite, R.S. Arruda, A.W. Ferreira, Combined use of Western blot/ELISA to improve the serological diagnosis of human tuberculosis, Braz. J. Infect. Dis. 9 (1) (2005) 35–43. [40] N.J. Hargreaves, O. Kadzakumanja, C.J. Whitty, F.M. Salaniponi, A.D. Harries, Squire SB: ‘smear-negative’ pulmonary tuberculosis in a DOTS programme: poor outcomes in an area of high HIV seroprevalence, Int. J. Tuberc. Lung Dis. 5 (9) (2001) 847–854. [41] G. Senol, Erer OF, Y.A. Yalcin, M. Coskun, A.T. Gunduz, C. Bicmen, M. Ertas, S.A. Ozkan, Humoral immune response against 38-kDa and 16-kDa mycobacterial antigens in tuberculosis, Eur. Respir. J. 29 (1) (2007) 143–148.

33

[42] B.L. Wang, Y. Xu, Z.M. Li, Y.M. Xu, X.H. Weng, H.H. Wang, Antibody response to four secretory proteins from Mycobacterium tuberculosis and their complex antigen in TB patients, Int. J. Tuberc. Lung Dis. 9 (12) (2005) 1327–1334. [43] H. Milburn, Key issues in the diagnosis and management of tuberculosis, J. R. Soc. Med. 100 (3) (2007) 134–141. [44] S. Ahmad, Pathogenesis, immunology, and diagnosis of latent Mycobacterium tuberculosis infection, Clin. Dev. Immunol. 814943 (2011). [45] M. Klopper, R.M. Warren, C. Hayes, N.C. Gey van Pittius, E.M. Streicher, B. Muller, F.A. Sirgel, M. Chabula-Nxiweni, E. Hoosain, G. Coetzee, et al., Emergence and spread of extensively and totally drug-resistant tuberculosis, South Africa, Emerg. Infect. Dis. 19 (3) (2013) 449–455. [46] R. Baumann, S. Kaempfer, N.N. Chegou, W. Oehlmann, A.G. Loxton, S.H. Kaufmann, P.D. van Helden, G.F. Black, M. Singh, G. Walzl, Serologic diagnosis of tuberculosis by combining Ig classes against selected mycobacterial targets, J. Infect. 69 (6) (2014) 581–589. [47] K.R. Steingart, N. Dendukuri, M. Henry, I. Schiller, P. Nahid, P.C. Hopewell, A. Ramsay, M. Pai, S. Laal, Performance of purified antigens for serodiagnosis of pulmonary tuberculosis: a meta-analysis, Clin. Vaccine Immunol. 16 (2) (2009) 260–276. [48] G.C. Ireton, R. Greenwald, H. Liang, J. Esfandiari, K.P. Lyashchenko, S.G. Reed, Identification of Mycobacterium tuberculosis antigens of high serodiagnostic value, Clin. Vaccine Immunol. 17 (10) (2010) 1539–1547.