Infectious Diseases, 2015; Early Online: 1–7
Original article
Simultaneous detection of IgG and IgM antibodies against a recombinant polyprotein PstS1-LEP for tuberculosis diagnosis
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
Jun-Li Li1,2*, Xiang-Yu Huang1*, Hong-Bing Chen3, Xin-Jing Wang3, Chuan-Zhi Zhu1, Ming Zhao4, Qing-De Song1,2, Hai-Yan Huang1, Li Xiao1 & Xiu-Yun He1 From the 1Central Laboratory, Beijing Key Laboratory of Organ Transplantation and Immunology Regulation, 309th Hospital of the Chinese People’s Liberation Army, Beijing, 2School of Life Sciences, Southwest University, Chongqing, 3Tuberculosis Institute, and 4Department of Osteology, 309th Hospital of Chinese People’s Liberation Army, Beijing, China
Abstract Background: Commercial serological tests for the diagnosis of tuberculosis (TB) show poor sensitivity and specificity, and a new approach to antigen screening is required to improve the accuracy of serodiagnosis. Methods: Using an indirect enzyme-linked immunosorbent assay (ELISA), we evaluated the responses of IgG and IgM antibodies to the recombinant PstS1-LEP protein expressed in Escherichia coli, a polyprotein of PstS1 and line multi-epitopes polypeptide (LEP). Results: The mixture of anti-human IgG and IgM added to a well [Ig(G M)], which was different from the combination of IgG and IgM (IgG IgM), had a stronger immunoreactivity to PstS1-LEP than the single antibody. IgG and Ig(G M), but not IgM against the PstS1-LEP protein effectively distinguished TB patients from patients with nontuberculous pulmonary disease (NTBPD) and healthy controls (HCs). Compared with IgG, the sensitivities of Ig(G M) and IgG IgM varied from 71.4% to 77.6% and 72.7% in pulmonary TB (PTB) patients and from 42.1% to 64.0% and 55.8% in extrapulmonary TB (EPTB) patients, respectively. The specificity of Ig(G M) did not decrease, and was higher than that provided by IgG IgM in HCs with positive tuberculin skin test. Conclusion: These findings suggest that PstS1-LEP can act as a candidate for detecting Ig(G M) in serum from PTB and EPTB patients.
Keywords: Line multi-epitope polypeptide, ELISA, tuberculosis, Ig(G M)
Introduction Despite impressive advances in tuberculosis (TB) control over the last decades [1], missed diagnoses continue to fuel the global epidemic, leading to more severe illness for patients and enabling further transmission of Mycobacterium tuberculosis (MTB) [2]. The conventional laboratory tests for the diagnosis of TB include bacterial culture (gold standard) and sputum smear microscopy. However, these tests have their own limitations because of low sensitivity or because they are time-consuming. The Xpert® MTB/RIF, the first rapid molecular test endorsed by the World Health Organization (WHO), can simultaneously detect TB and rifampicin resistance [3]. However, some limitations (e.g. high cost) of Xpert® MTB/RIF may restrain its global application, especially in undeveloped areas where the TB epidemic remains very severe [2]. Therefore, the
development of reliable, quick, and low-cost laboratory tests is vital for TB diagnosis [4]. For many years, serological tests have been used successfully for the rapid diagnosis of many infectious diseases (e.g. HIV infection, syphilis, and viral hepatitis). Serological TB tests can potentially enable rapid diagnosis as they have the advantages of speed, technological simplicity, and modest training requirements [5]. Dozens of commercial serological tests for TB diagnosis are offered for sale in many parts of the world [6]. A recent meta-analysis showed that commercial serological tests produce inconsistent and imprecise estimates of sensitivity and specificity [6]. The WHO Strategic and Technical Advisory Group for TB has issued negative recommendations for the use of currently available commercial serological tests for active TB diagnosis [7]. According to a recent survey,
*These authors contributed equally to this work. Correspondence: Xiu-Yun He, No. 17 Heishanhu Road, Haidian District, Beijing 100091, China. Tel: 86 10 66775520. Fax: 86 10 66767722. E-mail: [email protected] (Received 5 December 2014; accepted 16 April 2015) ISSN 2374-4235 print/ISSN 2374-4243 online © 2015 Informa Healthcare DOI: 10.3109/23744235.2015.1043941
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
2
J.-L. Li et al.
serological tests are still widely used in countries with a high TB burden [5,8]. Meanwhile, WHO guidelines do not limit future research for new antigen screening methods [9]. Thus, serological tests based on novel antigens are necessary to improve the accuracy of TB serodiagnosis. Many tests have been performed to increase the sensitivity of TB serodiagnosis, i.e. cocktails of multiple antigens, fusion proteins, combinations of two or three antibody isotypes [6,10]. The antigens of MTB are related to a number of proteins from nontuberculous mycobacterial pathogens or commensal nonpathogenic mycobacterial species, which may reduce the specificity of TB serodiagnosis [11]. However, immunodominant species-specific T- and B-cell epitopes can be found in antigens with highly conserved amino acid sequences [12]. Due to the rapid development of bioinformatics and immune information networks, the accuracy of epitope prediction has been improved dramatically [13]. B- and T-cell epitopes have been used for serodiagnosis and vaccine candidates. Commercial interferon-g release assays based on peptide cocktails have been widely used for the specific detection of latent MTB [14]. IgG responses to three immunodominant peptides show higher rates of positivity with TB serum (positives from 66% to 93%) than serum from healthy subjects (positives from 10% to 28%) [15]. Thus T- and B-cell epitopes can diagnose TB. Twelve immunodominant Th epitopes covering eight antigens of MTB form a line epitope polypeptide (LEP), which has five B-cell epitopes with a score 0.80 (www.imtech.res.in/raghava/abcpred) /ABC_submission.html. Our ongoing study shows that LEP protein as inclusion bodies in Escherichia coli is hardly dissolved in 8 M urea buffer. Previous studies have shown that PstS1 (also called as 38kDa or Rv0934) is overexpressed in E. coli and aggregated into bacterial inclusion bodies that are easily dissolved in 8 M urea buffer [16,17]. The PstS1 protein is an antigenic composition of several commercial serological diagnostic kits [6,18]. Unfortunately, the recognition frequency reported for the PstS1 antigen is highly variable in sensitivity (40.0– 89.0%) and specificity (44.0–100.0%) [19,20]. Due to the heterogeneity of the antibody responses in TB patients [21,22], the detection of antibodies against a single antigen usually offers a low sensitivity in TB diagnosis [23]. Thus, the fusion protein PstS1-LEP expressed in E. coli can induce specific cellular responses in mice [24]. Here, we addressed the potential values of PstS1-LEP in the serodiagnosis of TB.
The recombinant plasmid PET-28a-PstS1 was constructed and preserved in our laboratory. Anti-PstS1 rat monoclonal antibody (mAb) and horseradish peroxidase (HRP)-conjugated secondary antibodies were products of Jackson Immuno Research Laboratories Inc., West Grove, PA, USA. Subjects A total of 442 TB patients and 75 patients with nontuberculous pulmonary disease (pneumonia and lung cancer, NTBPD) were recruited from the 309th Hospital of the People’s Liberation Army (PLA). Patients with pulmonary TB (PTB; n 245) were identified according to the guidelines for PTB diagnosis and therapy authorized by the Tuberculosis Branch Association of the Chinese Medical Association. Patients with extrapulmonary TB (EPTB; n 197) were identified according to clinical presentation, Ziehl-Neelesenstain or positive mycobacterial culture, and X-ray inspection. PTB and EPTP patients showed appropriate responses to antituberculosis chemotherapy. NTBPD patients were diagnosed via clinical presentation, chest X-ray or pathological examination. Healthy controls (HCs) were recruited from draftees. In all, 102 HCs showed positive responses to purified protein derivative of BCG (BCG-PPD, Chengdu Institute of Biological Products Co.. Ltd., Chengdu, China) [HC(PPD)], and 48 HCs showed the opposite reaction [HC(PPD–)]. All the participants were HIV negative. Approval was granted by the Medicine Ethics Committee of the 309th Hospital, PLA. All participants provided written and informed consent before enrollment. Cloning, expression, and purification The DNA sequence encoding LEP (LEP), including BamHI restriction sites of N terminal and termination code (TAA) and HindIII restriction sites of C terminal, was commercially synthesized by Sangon Biotech (Shanghai) Co. Ltd, Shanghai, China. LEP was cloned into the recombinant plasmid pET28a-PstS1. The recombinant protein PstS1-LEP was expressed in E. coli BL21 (DE3) induced by IPTG (isopropyl-betaD-thiogalactopyranoside) and purified with ionexchange chromatography (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). The concentration of the renatured PstS1-LEP protein was determined by the bicinchoninic acid method (BCA Protein Assay Kit,Thermo Fisher Scientific Inc., Waltham, MA, USA).
Materials and Methods Bacterial strains, plasmid, and antibodies
Western blot analysis
Competent E. coli DH5a and BL21 (DE3) were purchased from Tiangen Biotech Co. Ltd (Beijing, China).
PstS1-LEP protein was resolved by 12% sodium dodecyl sulfate-polyacrylamide minigel electrophoresis
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
(SDS-PAGE) and transferred to a polyvinylidene fluoride membrane. After blocking for 2 h at room temperature (RT) with Tris-NaCl buffer (TBS; pH 7.5) containing 0.1% Tween-20 (TBST) and 5% fat-free milk (TBSTM), the membrane was rinsed twice with 15 ml TBST for 2 min each with gentle agitation. The membrane was incubated with primary antibodies diluted in 15 ml TBSTM (1:100 for serum and 1:10 000 for anti-PstS1 rat mAb) for 2 h at RT with gentle agitation. After six washes with TBST for 5 min each with gentle agitation, the membrane was incubated with corresponding HRP-conjugated secondary antibodies diluted (1:10 000) in 20 ml TBSTM for 1 h at RT with gentle agitation. After six washes with TBST for 5 min each with gentle agitation, the membrane was incubated with chemiluminescent substrate (Pierce ECL Western Blotting Substrate; Thermo Fisher Scientific Inc.) for 1 min and then exposed to X-ray film for 10–60 s as required. Indirect ELISA ELISA plates were coated with 1 mg/ml PstS1-LEP protein at 4°C overnight. After washing three times with PBS containing 0.05% Tween-20 (PBST) for 3 min each, 200 ml of PBST containing 5% fat-free milk (PBSTM) was added to each well. The plates were sealed, incubated at 37°C for 1 h, and washed as described above. Subsequently, 100 ml of serum diluted in PBSTM (1:100) was added to a well.The plates were sealed, incubated at 37°C for 2 h, and then washed as above. Then 100 ml of HRP-conjugated goat antihuman IgG or IgM diluted in PBSTM (1:30 000) was added to each well. The plates were sealed, incubated at 37°C for 2 h, and then washed as above. Freshly prepared tetramethyl benzidine (TMB) substrate solution (100 ml) was added, and the plates were protected from light for 20–30 min. The optical density (OD) was measured at 450 nm after stopping the reaction by
Simultaneous detection of IgG and IgM
3
adding 50 ml of 2 N H2SO4. ELISA for in-well simultaneous detection of IgG and IgM antibodies (Ig(G M)) was performed as above except that 100 ml of mixture of HRP-conjugated goat anti-human IgG and IgM dilution was added to each well. Statistical analysis Scatterplot was performed using GraphPad Prism (version 5.0). Receiver operating characteristic (ROC) curves were constructed by plotting the true-positive rate (sensitivity) and false-positive rate (1 – specificity) with each unique OD450 of antibodies against the PstS1-LEP protein for TB patients and NTBPD patients. The differences in OD450 values were determined using the Student’s t test or one-way analysis of variance with least significant difference comparisons. Pearson correlation analysis was performed with bivariate correlation. The sensitivity and specificity were analysed by the chi-squared test. All analyses were performed with SPSS version 16.0 (SPSS Inc., Chicago, IL, USA). A p value 0.05 was considered statistically significant. Results PstS1-LEP protein was expressed as inclusion bodies in E. coli BL21 (DE3), and its purity was about 90.0% (Figure 1a). PstS1-LEP protein showed positive responses to IgG, IgM, and Ig(G M) in serum from TB patients and retained a positive immunoreactivity to anti-PstS1 rat mAb (Figure 1b). The levels of anti-PstS1-LEP Ig(G M) were positively correlated with the levels of anti-PstS1-LEP IgG (r 0.91, p 0.001) and anti-PstS1-LEP IgM (r 0.45, p 0.001), and the levels of anti-PstS1-LEP Ig(G M) and IgG were higher than those of antiPstS1-LEP IgG and IgM in PTB, EPTB and NTBPD patients, HC(PPD–), and HC(PPD), respectively
Figure 1. (a) SDS-PAGE. Lane M1, protein marker; lane 1, total proteins of pET28a-PstS1-LEP/BL21 (DE3) without IPTG induction; lane 2, total proteins of pET28a-PstS1-LEP/BL21 (DE3) with IPTG induction; lane 3, the supernatant of supersonic lysates; lane 4, inclusion bodies; lane 5, purified PstS1-LEP; lane 6, renatured PstS1-LEP. (b) Western blot analysis. Lane M2, prestained protein marker; lanes 1–4, anti-PstS1 monoclonal antibody, IgG, Ig(G M), and IgM responses to PstS1-LEP, respectively.
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
4
J.-L. Li et al.
Figure 2. Anti-PstS1-LEP IgG, IgM, and Ig(G M) antibodies. (a) The relation of the levels of anti-PstS1-LEP Ig(G M) with the levels of anti-PstS1-LEP IgG and IgM. (b–d) Anti-PstS1-LEP IgG, IgM, and Ig(G M) antibodies in sera from PTB patients, EPTB patients, NTBPD patients, HC(PPD–), and HC(PPD), respectively. ** p 0.01. Each spot represents the OD450 value obtained with an individual serum. The solid lines indicate the median values of OD450. PTB, pulmonary tuberculosis; EPTB, extrapulmonary tuberculosis; NTBPD, nontuberculous pulmonary disease; HC(PPD–), healthy controls with negative reaction to purified protein derivative of BCG; HC(PPD), healthy controls with positive reaction to purified protein derivative of BCG.
(all p 0.001, Figure 2a). The levels of anti-PstS1-LEP Ig(G M) roughly equalled the sum of the levels of anti-PstS1-LEP IgG and IgM (Figure 2a). Area under the curve (AUC) of PstS1-LEP used to detect IgG, IgM, and Ig(G M) was 0.78 (95% confidence interval (CI) 0.73–0.83), 0.65 (95% CI 0.58–0.72), and 0.86 (95% CI 0.82–0.90), respectively. The cut-off values of IgG, IgM, and Ig(G M), indicated by OD450 nm, were 0.49, 0.26, and 0.60, respectively. Thus, the ROC analysis indicated that PstS1-LEP had greater diagnostic values
for IgG and Ig(G M) than for IgM. The levels of anti-PstS1-LEP IgG and Ig(G M) were higher in PTB and EPTB patients than in NTBPD patients, HC(PPD–), and HC(PPD); as well as in NTBPD patients and in HC(PPD) than in HC(PPD–) (all p 0.001, Figure 2b and d). The levels of anti-PstS1LEP IgM were higher in EPTB patients than in PTB and NTBPD patients, HC(PPD–), and HC(PPD) (all p 0.001, Figure 2c). The positivity was relatively low for IgM antibodies in TB patients, that is, 18.0% for PTB patients and 33.5% for EPTB patients (Table
Table I. IgG and IgM antibodies to PstS1-LEP.
Group
PTB (n 245)
EPTB (n 197)
IgG
IgG
IgM
IgM
NTBPD (n 75) IgG
IgM
HC(PPD–) (n 48) IgG
IgM
HC(PPD) (n 102) IgG
IgM
#Positive
no. (%)a 175 (71.4) 44 (18.0) 83 (42.1) 66 (33.5) 9 (12.0) 9 (12.0) 1 (2.1) 3 (6.3) 21 (20.6) 14 (13.7) Positive no. for one isotype of 134 3 44 27 6 6 1 3 18 11 antibody Double positive no. b 41 39 3 0 3 Combination positive no. (%)c 178 (72.7) 110 (55.8) 15 (20.0) 4 (8.3) 32 (31.4)
PTB, pulmonary tuberculosis; EPTB, extrapulmonary tuberculosis; NTBPD, nontuberculous pulmonary disease; HC(PPD–), healthy controls with negative reaction to purified protein derivative of BCG; HC(PPD), healthy controls with positive reaction to purified protein derivative of BCG; LEP, line multi-epitopes polypeptide. aPositive no. (%), the number of positive sera (percentage of positive sera). bDouble positive no., the number positive for IgG and IgM. cCombination positive no. (%), the number of IgG- and/or IgM-positive sera (percentage of positive sera).
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
I). We found that some of the sera, which were negative for IgG antibody, were positive for IgM and/or Ig(G M) antibodies (Tables I and II). By combination of IgG and IgM (IgG IgM), the sensitivity increased from 71.4% to 72.7% in PTB and from 42.1% to 55.8% in EPTB patients (Table I). The sensitivity of the tests improved to 77.6% for Ig(G M) and 82.0% for IgG Ig(G M) in PTB patients and 64.0% for Ig(G M) and 67.5% for IgG Ig(G M) in EPTB patients (Table II). The sensitivities of the tests based on IgG, Ig(G M), IgG IgM, and IgG Ig(G M) were higher in PTB patients than in EPTB patients (c2 38.6, c2 9.9, c2 11.6, and c2 12.5, respectively; all p 0.001). Compared with the tests based on IgG, the sensitivities of Ig(G M), IgG IgM, and IgG Ig(G M) were higher in EPTB patients (c2 18.8, c2 7.4, and c2 20.9, all p 0.001), and only the sensitivity of IgG Ig(G M) was higher in PTB patients (c2 18.8, p 0.001, Tables I and II). However, PTB patients showed a lower sensitivity provided by IgM responses to PstS1LEP than EPTB patients (c2 14.1, p 0.001). The specificity of the tests based on IgG, Ig(G M), IgG IgM, and IgG Ig(G M) was lower in HC(PPD) than in HC(PPD–) (c2 8.9, c2 5.1, c2 12.2, and c2 9.2, respectively; all p 0.05, Tables I and II). In addition, NTBPD patients had a lower specificity for IgG IgM than HC(PPD–) (c2 4.4, p 0.035, Table I). The specificity of IgG and Ig (G M) was marginally or significantly higher than that of IgG IgM in HC(PPD), respectively (c2 3.6, p 0.075 and c2 5.9, p 0.015). The levels of anti-PstS1-LEP IgG and Ig(G M) were higher in HCs with BCG scar than in HCs without BCG scar (Figure 3a, t 2.4, p 0.017; Figure 3c, t 3.1, p 0.003). However, HCs with BCG scar and without BCG scar had similar levels of antiPstS1-LEP IgM (t 1.4, p 0.16, Figure 3b). There were no differences in specificity of the tests based on IgM, Ig(G M), and IgG IgM between HCs with BCG scar and HCs without BCG scar (all p 0.05), but the specificity of anti-PstS1-LEP IgG was higher in HCs without BCG scar than in HCs with BCG scar (c2 4.6, p 0.030, Table III). HCs without BCG scar showed no differences in specificity of IgG, IgM, Ig(G M), and IgG IgM (all p 0.05), and HCs with BCG scar showed a higher specificity of IgM and Ig(G M) than that of IgG IgM (c2 3.9, p 0.050 and c2 8.4, p 0.004, Table III). Discussion Due to the high prevalence of TB and the lack of accurate serological tests, MTB proteins and synthetic peptides corresponding to MTB proteins have been used for serodiagnosis of TB [15,25]. The
Simultaneous detection of IgG and IgM
5
Figure 3. The levels of IgG, IgM, and Ig(G M) against the PstS1-LEP protein in healthy controls (HC). (a) The levels of IgG against the PstS1-LEP protein. (b) The levels of IgM against the PstS1-LEP protein. (c) The levels of Ig(G M) against the PstS1LEP protein. *p 0.05, **p 0.01. Each spot represents the OD450 value obtained with an individual serum. The solid lines indicate the median values of OD450.
immunogenetic backgrounds of the infected hosts result in heterogeneous recognition of antigens by serum antibodies in TB patients [21,26]. Thus, the TB serodiagnostic assay with high sensitivity and specificity is most likely to require a panel of several recombinant antigens or a fusion polyprotein constructed by the critical epitopes [27]. Although peptides overcome the problem in obtaining full-length pure recombinant proteins, only 3 peptides of 775 synthetic peptides covering 39 proteins have been identified to be reactive with TB serum [15]. LEP constructed by tandem of 12 line Th epitopes
6
J.-L. Li et al.
Table II. IgG and Ig(G M) antibodies to PstS1-LEP. PTB (n 245) Group
IgG
Ig(G M)
EPTB (n 197)
NTBPD (n 75)
IgG
IgG
Ig(G M)
Ig(G M)
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
Positive no. (%)a 175 (71.4) 190 (77.6) 83 (42.1) 126 (64.0) 9 (12.0) 9 (12.0) Positive no. for one 11 26 7 50 2 2 isotype of antibody Double positive no.b 164 76 7 Combination positive 201 (82.0) 133 (67.5) 11 (14.7) no. (%)c
HC(PPD–) (n 48) IgG
Ig(G M)
1 (2.1) 2 (4.2) 0 1 1 2 (4.2)
HC(PPD) (n 102) IgG
Ig(G M)
21 (20.6) 18 (17.6) 7 4 14 25 (24.5)
PTB, extrapulmonary tuberculosis; HC (PPD–), healthy controls with negative reaction to purified protein derivative of BCG; HC E (PPD), healthy controls with positive reaction to purified protein derivative of BCG; LEP, line multi-epitopes polypeptide; NTBPD, nontuberculous pulmonary disease; PTB, pulmonary tuberculosis. aPositive no. (%), the number of positive sera (percentage of positive sera). bDouble positive no., the number positive for IgG and IgM. cCombination positive no. (%), the number of IgG- and/or IgM-positive sera (percentage of positive sera).
contains B-cell epitopes. The recombinant PstS1LEP protein showed positive responses to IgG, IgM, and Ig(G M) antibodies in serum from TB patients and retained immunoreactivity of PstS1 because it displayed positive responses to anti-PstS1 mAb. This evidence suggests that PstS1-LEP may be a candidate for serodiagnosis of TB. IgM against PstS1-LEP may be used to diagnose EPTB but not PTB. The values of AUCs usually indicate the diagnostic accuracy, which are too low to suggest meaningful use in clinical practice [28,29]. Our findings suggest that PstS1-LEP protein used to detect IgM antibodies has a very poor discriminatory ability with AUCs of 0.65. The levels of IgM antibody were higher in EPTB patients than in PTB patients, NTBPD patients, and HCs. Meanwhile, the sensitivity of anti-PstS1-LEP IgM was higher in EPTB patients than in PTB patients. The sensitivity of anti-PstS1-LEP IgM in TB patients, in accord with fusion protein 38kD-ESAT6-CFP10 detecting active TB [30], was lower than that of the combination of lipoarabinomannan antigen and PstS1 detecting PTB (48.7%) [31]. Our results suggest that IgM against PstS1-LEP may be able to detect EPTB. Combining IgG with other antibody isotypes may increase sensitivity but reduce specificity. Combinations of IgG, IgA, and IgM (IgG IgA IgM) can increase the sensitivity because some sera, that are negative for IgG, are still positive for IgA and/or IgM Table III. Effects of BCG vaccination on the specificity of PstS1-LEP protein. Specificity (%) BCG scar
IgG
IgM
Yes (n 108) No (n 42)
81.5 95.2
87.9 90.5
LEP, line multi-epitopes polypeptide.
Ig(G M) 83.3 95.2
antibodies [19,30]. We found that the sensitivity of IgG was lower than that of IgG IgM in EPTB patients, but the specificity of IgG was higher than that of IgG IgM in HC(PPD). Other researchers reported that IgG IgA IgM against 38kD-ESAT6-CFP10 increased sensitivity from 65.7% to 75.5% and decreased specificity slightly from 94.7% to 91.5% compared with IgG [30]. Our previous study showed that the sensitivity of IgG IgM provided by 38 kDa increased in EPTB patients [32]. Ig(G M) displayed a stronger immunoreactivity to PstS1-LEP than a single antibody. We found that the serological test based on Ig(G M) against PstS1-LEP increased its sensitivity but did not decrease its specificity. The specificity of IgG IgM was lower than that of Ig(G M) in HC(PPD). Thus, Ig(G M) against PstS1-LEP has advantages over IgG IgM responses to PstS1-LEP. BCG vaccination reduced the specificity of PstS1LEP. The specificities of IgG and Ig(G M) responses to PstS1-LEP were higher in HC(PPD–) than in HC(PPD), but such differences were not found in NTBPD patients in comparison with HC(PPD). The specificity of anti-PstS1-LEP Ig(G M) was higher in HCs without BCG scar than in HCs with BCG scar. Our findings suggest that HC(PPD–) may provide a high specificity of PstS1-LEP due to ignoring some practical problems, such as BCG vaccination, environmental mycobacteria, and other respiratory diseases. Thus, many studies on TB serodiagnosis use BCG-vaccinated HCs as controls [15,33]. In summary, PstS1-LEP has the potential to detect IgG and Ig(G M) antibodies in PTB patients and Ig(G M) antibody against PstS1-LEP exhibits high diagnostic values to detect EPTB.
IgG IgM 72.2 85.7
Acknowledgments This study was supported by funding from the National Mega Projects on Key Infectious Disease
Control of Ministry of Science and Technology, China (2013ZX10003006-003-001). Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
References [1] Lonnroth K, Castro KG, Chakaya JM, Chauhan LS, Floyd K, Glaziou P, et al. Tuberculosis control and elimination 2010-50: cure, care, and social development. Lancet 2010;375: 1814–29. [2] Small PM, Pai M. Tuberculosis diagnosis – time for a game change. N Engl J Med 2010;363:1070–1. [3] World Health Organization. Global tuberculosis report 2013. WHO; 2013. [4] Rasanathan K, Sivasankara Kurup A, Jaramillo E, Lonnroth K. The social determinants of health: key to global tuberculosis control. Int J Tuberc Lung Dis 2011;15(Suppl 2):S30–6. [5] Steingart KR, Ramsay A, Dowdy DW, Pai M. Serological tests for the diagnosis of active tuberculosis: relevance for India. Indian J Med Res 2012;135:695–702. [6] Steingart KR, Flores LL, Dendukuri N, Schiller I, Laal S, Ramsay A, et al. Commercial serological tests for the diagnosis of active pulmonary and extrapulmonary tuberculosis: an updated systematic review and meta-analysis. PLoS Med 2011;8:e1001062. [7] World Health Organization. Strategic and Technical Advisory Group for Tuberculosis (STAG-TB). Report of the Tenth Meeting. WHO; 2010. [8] Grenier J, Pinto L, Nair D, Steingart K, Dowdy D, Ramsay A, et al. Widespread use of serological tests for tuberculosis: data from 22 high-burden countries. Eur Respir J 2012;39:502–5. [9] Morris K. WHO recommends against inaccurate tuberculosis tests. Lancet 2011;377:113–14. [10] Raja A, Ranganathan UD, Bethunaickan R. Improved diagnosis of pulmonary tuberculosis by detection of antibodies against multiple Mycobacterium tuberculosis antigens. Diagn Microbiol Infect Dis 2008;60:361–8. [11] Ivanyi J. Function and potentials of M. tuberculosis epitopes. Front Immunol 2014;5:107. [12] Jackett PS, Bothamley GH, Batra HV, Mistry A, Young DB, Ivanyi J. Specificity of antibodies to immunodominant mycobacterial antigens in pulmonary tuberculosis. J Clin Microbiol 1988;26:2313–18. [13] Reche PA, Glutting JP, Reinherz EL. Prediction of MHC class I binding peptides using profile motifs. Hum Immunol 2002;63:701–9. [14] Pai M, Riley LW, Colford JM Jr. Interferon-gamma assays in the immunodiagnosis of tuberculosis: a systematic review. Lancet Infect Dis 2004;4:761–76. [15] Al-Khodari NY, Al-Attiyah R, Mustafa AS. Identification, diagnostic potential, and natural expression of immunodominant seroreactive peptides encoded by five Mycobacterium tuberculosis-specific genomic regions. Clin Vaccine Immunol 2011;18:477–82. [16] Singh M, Andersen AB, McCarthy JE, Rohde M, Schutte H, Sanders E, et al. The Mycobacterium tuberculosis 38-kDa antigen: overproduction in Escherichia coli, purification and characterization. Gene 1992;117:53–60.
Simultaneous detection of IgG and IgM
7
[17] He X, Zhuang Y, Zhang X, Xiong Z, Chang Z. The 38,000 protein of Mycobacterium tuberculosis is overexpressed in Escherichia coli. Zhonghua Jie He He Hu Xi Za Zhi 1999;22:138–41. [18] Anderson BL, Welch RJ, Litwin CM. Assessment of three commercially available serologic assays for detection of antibodies to Mycobacterium tuberculosis and identification of active tuberculosis. Clin Vaccine Immunol 2008;15:1644–9. [19] Uma Devi KR, Ramalingam B, Brennan PJ, Narayanan PR, Raja A. Specific and early detection of IgG, IgA and IgM antibodies to Mycobacterium tuberculosis 38kDa antigen in pulmonary tuberculosis. Tuberculosis (Edinb) 2001;81:249–53. [20] Senol G, Ecevit C, Ozturk A. Humoral immune response against 38- and 16-kDa mycobacterial antigens in childhood tuberculosis. Pediatr Pulmonol 2009;44:839–44. [21] Lyashchenko K, Colangeli R, Houde M, Al Jahdali H, Menzies D, Gennaro ML. Heterogeneous antibody responses in tuberculosis. Infect Immun 1998;66:3936–40. [22] Kunnath-Velayudhan S, Salamon H, Wang HY, Davidow AL, Molina DM, Huynh VT, et al. Dynamic antibody responses to the Mycobacterium tuberculosis proteome. Proc Natl Acad Sci U S A 2010;107:14703–8. [23] Lange C, Mori T. Advances in the diagnosis of tuberculosis. Respirology 2010;15:220–40. [24] Zhang CQ, Huang XY, Shao JS, Li JL, Song QD, Zhuang YH, et al. Effects of MF59 in combination with heat-killed BCG as adjuvant on the immunogenicity of Mycobacterium tuberculosis fusion protein PstS1-LEP. Zhong Hua Wei Sheng Wu Xue Yu Mian Yi Xue Za Zhi 2014;34:241–6. [25] Goyal B, Kumar K, Gupta D, Agarwal R, Latawa R, Sheikh JA, et al. Utility of B-cell epitopes based peptides of RD1 and RD2 antigens for immunodiagnosis of pulmonary tuberculosis. Diagn Microbiol Infect Dis 2014;78:391–7. [26] Demkow U, Filewska M, Michalowska-Mitczuk D, Kus J, Jagodzinski J, Zielonka T, et al. Heterogeneity of antibody response to myobacterial antigens in different clinical manifestations of pulmonary tuberculosis. J Physiol Pharmacol 2007;58(Suppl 5;Pt 1):117–27. [27] Lodes MJ, Dillon DC, Mohamath R, Day CH, Benson DR, Reynolds LD, et al. Serological expression cloning and immunological evaluation of MTB48, a novel Mycobacterium tuberculosis antigen. J Clin Microbiol 2001;39:2485–93. [28] Everson RM, Fieldsend JE. Multi-class ROC analysis from a multi-objective optimisation perspective. Pattern Recog Lett 2006;27:918–27. [29] Van der Schouw YT, Verbeek AL, Ruijs JH. ROC curves for the initial assessment of new diagnostic tests. Fam Pract 1992;9:506–11. [30] Feng X, Yang X, Xiu B, Qie S, Dai Z, Chen K, et al. IgG, IgM and IgA antibodies against the novel polyprotein in active tuberculosis. BMC Infect Dis 2014;14:336. [31] Singh S, Singh J, Kumar S, Gopinath K, Balooni V, Singh N, et al. Poor performance of serological tests in the diagnosis of pulmonary tuberculosis: evidence from a contact tracing field study. PLoS One 2012;7:e40213. [32] He XY, Li J, Hao J, Chen HB, Zhao YZ, Huang XY, et al. Assessment of five antigens from Mycobacterium tuberculosis for serodiagnosis of tuberculosis. Clin Vaccine Immunol 2011;18:565–70. [33] Khan N, Alam K, Nair S, Valluri VL, Murthy KJ, Mukhopadhyay S. Association of strong immune responses to PPE protein Rv1168c with active tuberculosis. Clin Vaccine Immunol 2008;15:974–80.
Original article
Simultaneous detection of IgG and IgM antibodies against a recombinant polyprotein PstS1-LEP for tuberculosis diagnosis
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
Jun-Li Li1,2*, Xiang-Yu Huang1*, Hong-Bing Chen3, Xin-Jing Wang3, Chuan-Zhi Zhu1, Ming Zhao4, Qing-De Song1,2, Hai-Yan Huang1, Li Xiao1 & Xiu-Yun He1 From the 1Central Laboratory, Beijing Key Laboratory of Organ Transplantation and Immunology Regulation, 309th Hospital of the Chinese People’s Liberation Army, Beijing, 2School of Life Sciences, Southwest University, Chongqing, 3Tuberculosis Institute, and 4Department of Osteology, 309th Hospital of Chinese People’s Liberation Army, Beijing, China
Abstract Background: Commercial serological tests for the diagnosis of tuberculosis (TB) show poor sensitivity and specificity, and a new approach to antigen screening is required to improve the accuracy of serodiagnosis. Methods: Using an indirect enzyme-linked immunosorbent assay (ELISA), we evaluated the responses of IgG and IgM antibodies to the recombinant PstS1-LEP protein expressed in Escherichia coli, a polyprotein of PstS1 and line multi-epitopes polypeptide (LEP). Results: The mixture of anti-human IgG and IgM added to a well [Ig(G M)], which was different from the combination of IgG and IgM (IgG IgM), had a stronger immunoreactivity to PstS1-LEP than the single antibody. IgG and Ig(G M), but not IgM against the PstS1-LEP protein effectively distinguished TB patients from patients with nontuberculous pulmonary disease (NTBPD) and healthy controls (HCs). Compared with IgG, the sensitivities of Ig(G M) and IgG IgM varied from 71.4% to 77.6% and 72.7% in pulmonary TB (PTB) patients and from 42.1% to 64.0% and 55.8% in extrapulmonary TB (EPTB) patients, respectively. The specificity of Ig(G M) did not decrease, and was higher than that provided by IgG IgM in HCs with positive tuberculin skin test. Conclusion: These findings suggest that PstS1-LEP can act as a candidate for detecting Ig(G M) in serum from PTB and EPTB patients.
Keywords: Line multi-epitope polypeptide, ELISA, tuberculosis, Ig(G M)
Introduction Despite impressive advances in tuberculosis (TB) control over the last decades [1], missed diagnoses continue to fuel the global epidemic, leading to more severe illness for patients and enabling further transmission of Mycobacterium tuberculosis (MTB) [2]. The conventional laboratory tests for the diagnosis of TB include bacterial culture (gold standard) and sputum smear microscopy. However, these tests have their own limitations because of low sensitivity or because they are time-consuming. The Xpert® MTB/RIF, the first rapid molecular test endorsed by the World Health Organization (WHO), can simultaneously detect TB and rifampicin resistance [3]. However, some limitations (e.g. high cost) of Xpert® MTB/RIF may restrain its global application, especially in undeveloped areas where the TB epidemic remains very severe [2]. Therefore, the
development of reliable, quick, and low-cost laboratory tests is vital for TB diagnosis [4]. For many years, serological tests have been used successfully for the rapid diagnosis of many infectious diseases (e.g. HIV infection, syphilis, and viral hepatitis). Serological TB tests can potentially enable rapid diagnosis as they have the advantages of speed, technological simplicity, and modest training requirements [5]. Dozens of commercial serological tests for TB diagnosis are offered for sale in many parts of the world [6]. A recent meta-analysis showed that commercial serological tests produce inconsistent and imprecise estimates of sensitivity and specificity [6]. The WHO Strategic and Technical Advisory Group for TB has issued negative recommendations for the use of currently available commercial serological tests for active TB diagnosis [7]. According to a recent survey,
*These authors contributed equally to this work. Correspondence: Xiu-Yun He, No. 17 Heishanhu Road, Haidian District, Beijing 100091, China. Tel: 86 10 66775520. Fax: 86 10 66767722. E-mail: [email protected] (Received 5 December 2014; accepted 16 April 2015) ISSN 2374-4235 print/ISSN 2374-4243 online © 2015 Informa Healthcare DOI: 10.3109/23744235.2015.1043941
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
2
J.-L. Li et al.
serological tests are still widely used in countries with a high TB burden [5,8]. Meanwhile, WHO guidelines do not limit future research for new antigen screening methods [9]. Thus, serological tests based on novel antigens are necessary to improve the accuracy of TB serodiagnosis. Many tests have been performed to increase the sensitivity of TB serodiagnosis, i.e. cocktails of multiple antigens, fusion proteins, combinations of two or three antibody isotypes [6,10]. The antigens of MTB are related to a number of proteins from nontuberculous mycobacterial pathogens or commensal nonpathogenic mycobacterial species, which may reduce the specificity of TB serodiagnosis [11]. However, immunodominant species-specific T- and B-cell epitopes can be found in antigens with highly conserved amino acid sequences [12]. Due to the rapid development of bioinformatics and immune information networks, the accuracy of epitope prediction has been improved dramatically [13]. B- and T-cell epitopes have been used for serodiagnosis and vaccine candidates. Commercial interferon-g release assays based on peptide cocktails have been widely used for the specific detection of latent MTB [14]. IgG responses to three immunodominant peptides show higher rates of positivity with TB serum (positives from 66% to 93%) than serum from healthy subjects (positives from 10% to 28%) [15]. Thus T- and B-cell epitopes can diagnose TB. Twelve immunodominant Th epitopes covering eight antigens of MTB form a line epitope polypeptide (LEP), which has five B-cell epitopes with a score 0.80 (www.imtech.res.in/raghava/abcpred) /ABC_submission.html. Our ongoing study shows that LEP protein as inclusion bodies in Escherichia coli is hardly dissolved in 8 M urea buffer. Previous studies have shown that PstS1 (also called as 38kDa or Rv0934) is overexpressed in E. coli and aggregated into bacterial inclusion bodies that are easily dissolved in 8 M urea buffer [16,17]. The PstS1 protein is an antigenic composition of several commercial serological diagnostic kits [6,18]. Unfortunately, the recognition frequency reported for the PstS1 antigen is highly variable in sensitivity (40.0– 89.0%) and specificity (44.0–100.0%) [19,20]. Due to the heterogeneity of the antibody responses in TB patients [21,22], the detection of antibodies against a single antigen usually offers a low sensitivity in TB diagnosis [23]. Thus, the fusion protein PstS1-LEP expressed in E. coli can induce specific cellular responses in mice [24]. Here, we addressed the potential values of PstS1-LEP in the serodiagnosis of TB.
The recombinant plasmid PET-28a-PstS1 was constructed and preserved in our laboratory. Anti-PstS1 rat monoclonal antibody (mAb) and horseradish peroxidase (HRP)-conjugated secondary antibodies were products of Jackson Immuno Research Laboratories Inc., West Grove, PA, USA. Subjects A total of 442 TB patients and 75 patients with nontuberculous pulmonary disease (pneumonia and lung cancer, NTBPD) were recruited from the 309th Hospital of the People’s Liberation Army (PLA). Patients with pulmonary TB (PTB; n 245) were identified according to the guidelines for PTB diagnosis and therapy authorized by the Tuberculosis Branch Association of the Chinese Medical Association. Patients with extrapulmonary TB (EPTB; n 197) were identified according to clinical presentation, Ziehl-Neelesenstain or positive mycobacterial culture, and X-ray inspection. PTB and EPTP patients showed appropriate responses to antituberculosis chemotherapy. NTBPD patients were diagnosed via clinical presentation, chest X-ray or pathological examination. Healthy controls (HCs) were recruited from draftees. In all, 102 HCs showed positive responses to purified protein derivative of BCG (BCG-PPD, Chengdu Institute of Biological Products Co.. Ltd., Chengdu, China) [HC(PPD)], and 48 HCs showed the opposite reaction [HC(PPD–)]. All the participants were HIV negative. Approval was granted by the Medicine Ethics Committee of the 309th Hospital, PLA. All participants provided written and informed consent before enrollment. Cloning, expression, and purification The DNA sequence encoding LEP (LEP), including BamHI restriction sites of N terminal and termination code (TAA) and HindIII restriction sites of C terminal, was commercially synthesized by Sangon Biotech (Shanghai) Co. Ltd, Shanghai, China. LEP was cloned into the recombinant plasmid pET28a-PstS1. The recombinant protein PstS1-LEP was expressed in E. coli BL21 (DE3) induced by IPTG (isopropyl-betaD-thiogalactopyranoside) and purified with ionexchange chromatography (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). The concentration of the renatured PstS1-LEP protein was determined by the bicinchoninic acid method (BCA Protein Assay Kit,Thermo Fisher Scientific Inc., Waltham, MA, USA).
Materials and Methods Bacterial strains, plasmid, and antibodies
Western blot analysis
Competent E. coli DH5a and BL21 (DE3) were purchased from Tiangen Biotech Co. Ltd (Beijing, China).
PstS1-LEP protein was resolved by 12% sodium dodecyl sulfate-polyacrylamide minigel electrophoresis
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
(SDS-PAGE) and transferred to a polyvinylidene fluoride membrane. After blocking for 2 h at room temperature (RT) with Tris-NaCl buffer (TBS; pH 7.5) containing 0.1% Tween-20 (TBST) and 5% fat-free milk (TBSTM), the membrane was rinsed twice with 15 ml TBST for 2 min each with gentle agitation. The membrane was incubated with primary antibodies diluted in 15 ml TBSTM (1:100 for serum and 1:10 000 for anti-PstS1 rat mAb) for 2 h at RT with gentle agitation. After six washes with TBST for 5 min each with gentle agitation, the membrane was incubated with corresponding HRP-conjugated secondary antibodies diluted (1:10 000) in 20 ml TBSTM for 1 h at RT with gentle agitation. After six washes with TBST for 5 min each with gentle agitation, the membrane was incubated with chemiluminescent substrate (Pierce ECL Western Blotting Substrate; Thermo Fisher Scientific Inc.) for 1 min and then exposed to X-ray film for 10–60 s as required. Indirect ELISA ELISA plates were coated with 1 mg/ml PstS1-LEP protein at 4°C overnight. After washing three times with PBS containing 0.05% Tween-20 (PBST) for 3 min each, 200 ml of PBST containing 5% fat-free milk (PBSTM) was added to each well. The plates were sealed, incubated at 37°C for 1 h, and washed as described above. Subsequently, 100 ml of serum diluted in PBSTM (1:100) was added to a well.The plates were sealed, incubated at 37°C for 2 h, and then washed as above. Then 100 ml of HRP-conjugated goat antihuman IgG or IgM diluted in PBSTM (1:30 000) was added to each well. The plates were sealed, incubated at 37°C for 2 h, and then washed as above. Freshly prepared tetramethyl benzidine (TMB) substrate solution (100 ml) was added, and the plates were protected from light for 20–30 min. The optical density (OD) was measured at 450 nm after stopping the reaction by
Simultaneous detection of IgG and IgM
3
adding 50 ml of 2 N H2SO4. ELISA for in-well simultaneous detection of IgG and IgM antibodies (Ig(G M)) was performed as above except that 100 ml of mixture of HRP-conjugated goat anti-human IgG and IgM dilution was added to each well. Statistical analysis Scatterplot was performed using GraphPad Prism (version 5.0). Receiver operating characteristic (ROC) curves were constructed by plotting the true-positive rate (sensitivity) and false-positive rate (1 – specificity) with each unique OD450 of antibodies against the PstS1-LEP protein for TB patients and NTBPD patients. The differences in OD450 values were determined using the Student’s t test or one-way analysis of variance with least significant difference comparisons. Pearson correlation analysis was performed with bivariate correlation. The sensitivity and specificity were analysed by the chi-squared test. All analyses were performed with SPSS version 16.0 (SPSS Inc., Chicago, IL, USA). A p value 0.05 was considered statistically significant. Results PstS1-LEP protein was expressed as inclusion bodies in E. coli BL21 (DE3), and its purity was about 90.0% (Figure 1a). PstS1-LEP protein showed positive responses to IgG, IgM, and Ig(G M) in serum from TB patients and retained a positive immunoreactivity to anti-PstS1 rat mAb (Figure 1b). The levels of anti-PstS1-LEP Ig(G M) were positively correlated with the levels of anti-PstS1-LEP IgG (r 0.91, p 0.001) and anti-PstS1-LEP IgM (r 0.45, p 0.001), and the levels of anti-PstS1-LEP Ig(G M) and IgG were higher than those of antiPstS1-LEP IgG and IgM in PTB, EPTB and NTBPD patients, HC(PPD–), and HC(PPD), respectively
Figure 1. (a) SDS-PAGE. Lane M1, protein marker; lane 1, total proteins of pET28a-PstS1-LEP/BL21 (DE3) without IPTG induction; lane 2, total proteins of pET28a-PstS1-LEP/BL21 (DE3) with IPTG induction; lane 3, the supernatant of supersonic lysates; lane 4, inclusion bodies; lane 5, purified PstS1-LEP; lane 6, renatured PstS1-LEP. (b) Western blot analysis. Lane M2, prestained protein marker; lanes 1–4, anti-PstS1 monoclonal antibody, IgG, Ig(G M), and IgM responses to PstS1-LEP, respectively.
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
4
J.-L. Li et al.
Figure 2. Anti-PstS1-LEP IgG, IgM, and Ig(G M) antibodies. (a) The relation of the levels of anti-PstS1-LEP Ig(G M) with the levels of anti-PstS1-LEP IgG and IgM. (b–d) Anti-PstS1-LEP IgG, IgM, and Ig(G M) antibodies in sera from PTB patients, EPTB patients, NTBPD patients, HC(PPD–), and HC(PPD), respectively. ** p 0.01. Each spot represents the OD450 value obtained with an individual serum. The solid lines indicate the median values of OD450. PTB, pulmonary tuberculosis; EPTB, extrapulmonary tuberculosis; NTBPD, nontuberculous pulmonary disease; HC(PPD–), healthy controls with negative reaction to purified protein derivative of BCG; HC(PPD), healthy controls with positive reaction to purified protein derivative of BCG.
(all p 0.001, Figure 2a). The levels of anti-PstS1-LEP Ig(G M) roughly equalled the sum of the levels of anti-PstS1-LEP IgG and IgM (Figure 2a). Area under the curve (AUC) of PstS1-LEP used to detect IgG, IgM, and Ig(G M) was 0.78 (95% confidence interval (CI) 0.73–0.83), 0.65 (95% CI 0.58–0.72), and 0.86 (95% CI 0.82–0.90), respectively. The cut-off values of IgG, IgM, and Ig(G M), indicated by OD450 nm, were 0.49, 0.26, and 0.60, respectively. Thus, the ROC analysis indicated that PstS1-LEP had greater diagnostic values
for IgG and Ig(G M) than for IgM. The levels of anti-PstS1-LEP IgG and Ig(G M) were higher in PTB and EPTB patients than in NTBPD patients, HC(PPD–), and HC(PPD); as well as in NTBPD patients and in HC(PPD) than in HC(PPD–) (all p 0.001, Figure 2b and d). The levels of anti-PstS1LEP IgM were higher in EPTB patients than in PTB and NTBPD patients, HC(PPD–), and HC(PPD) (all p 0.001, Figure 2c). The positivity was relatively low for IgM antibodies in TB patients, that is, 18.0% for PTB patients and 33.5% for EPTB patients (Table
Table I. IgG and IgM antibodies to PstS1-LEP.
Group
PTB (n 245)
EPTB (n 197)
IgG
IgG
IgM
IgM
NTBPD (n 75) IgG
IgM
HC(PPD–) (n 48) IgG
IgM
HC(PPD) (n 102) IgG
IgM
#Positive
no. (%)a 175 (71.4) 44 (18.0) 83 (42.1) 66 (33.5) 9 (12.0) 9 (12.0) 1 (2.1) 3 (6.3) 21 (20.6) 14 (13.7) Positive no. for one isotype of 134 3 44 27 6 6 1 3 18 11 antibody Double positive no. b 41 39 3 0 3 Combination positive no. (%)c 178 (72.7) 110 (55.8) 15 (20.0) 4 (8.3) 32 (31.4)
PTB, pulmonary tuberculosis; EPTB, extrapulmonary tuberculosis; NTBPD, nontuberculous pulmonary disease; HC(PPD–), healthy controls with negative reaction to purified protein derivative of BCG; HC(PPD), healthy controls with positive reaction to purified protein derivative of BCG; LEP, line multi-epitopes polypeptide. aPositive no. (%), the number of positive sera (percentage of positive sera). bDouble positive no., the number positive for IgG and IgM. cCombination positive no. (%), the number of IgG- and/or IgM-positive sera (percentage of positive sera).
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
I). We found that some of the sera, which were negative for IgG antibody, were positive for IgM and/or Ig(G M) antibodies (Tables I and II). By combination of IgG and IgM (IgG IgM), the sensitivity increased from 71.4% to 72.7% in PTB and from 42.1% to 55.8% in EPTB patients (Table I). The sensitivity of the tests improved to 77.6% for Ig(G M) and 82.0% for IgG Ig(G M) in PTB patients and 64.0% for Ig(G M) and 67.5% for IgG Ig(G M) in EPTB patients (Table II). The sensitivities of the tests based on IgG, Ig(G M), IgG IgM, and IgG Ig(G M) were higher in PTB patients than in EPTB patients (c2 38.6, c2 9.9, c2 11.6, and c2 12.5, respectively; all p 0.001). Compared with the tests based on IgG, the sensitivities of Ig(G M), IgG IgM, and IgG Ig(G M) were higher in EPTB patients (c2 18.8, c2 7.4, and c2 20.9, all p 0.001), and only the sensitivity of IgG Ig(G M) was higher in PTB patients (c2 18.8, p 0.001, Tables I and II). However, PTB patients showed a lower sensitivity provided by IgM responses to PstS1LEP than EPTB patients (c2 14.1, p 0.001). The specificity of the tests based on IgG, Ig(G M), IgG IgM, and IgG Ig(G M) was lower in HC(PPD) than in HC(PPD–) (c2 8.9, c2 5.1, c2 12.2, and c2 9.2, respectively; all p 0.05, Tables I and II). In addition, NTBPD patients had a lower specificity for IgG IgM than HC(PPD–) (c2 4.4, p 0.035, Table I). The specificity of IgG and Ig (G M) was marginally or significantly higher than that of IgG IgM in HC(PPD), respectively (c2 3.6, p 0.075 and c2 5.9, p 0.015). The levels of anti-PstS1-LEP IgG and Ig(G M) were higher in HCs with BCG scar than in HCs without BCG scar (Figure 3a, t 2.4, p 0.017; Figure 3c, t 3.1, p 0.003). However, HCs with BCG scar and without BCG scar had similar levels of antiPstS1-LEP IgM (t 1.4, p 0.16, Figure 3b). There were no differences in specificity of the tests based on IgM, Ig(G M), and IgG IgM between HCs with BCG scar and HCs without BCG scar (all p 0.05), but the specificity of anti-PstS1-LEP IgG was higher in HCs without BCG scar than in HCs with BCG scar (c2 4.6, p 0.030, Table III). HCs without BCG scar showed no differences in specificity of IgG, IgM, Ig(G M), and IgG IgM (all p 0.05), and HCs with BCG scar showed a higher specificity of IgM and Ig(G M) than that of IgG IgM (c2 3.9, p 0.050 and c2 8.4, p 0.004, Table III). Discussion Due to the high prevalence of TB and the lack of accurate serological tests, MTB proteins and synthetic peptides corresponding to MTB proteins have been used for serodiagnosis of TB [15,25]. The
Simultaneous detection of IgG and IgM
5
Figure 3. The levels of IgG, IgM, and Ig(G M) against the PstS1-LEP protein in healthy controls (HC). (a) The levels of IgG against the PstS1-LEP protein. (b) The levels of IgM against the PstS1-LEP protein. (c) The levels of Ig(G M) against the PstS1LEP protein. *p 0.05, **p 0.01. Each spot represents the OD450 value obtained with an individual serum. The solid lines indicate the median values of OD450.
immunogenetic backgrounds of the infected hosts result in heterogeneous recognition of antigens by serum antibodies in TB patients [21,26]. Thus, the TB serodiagnostic assay with high sensitivity and specificity is most likely to require a panel of several recombinant antigens or a fusion polyprotein constructed by the critical epitopes [27]. Although peptides overcome the problem in obtaining full-length pure recombinant proteins, only 3 peptides of 775 synthetic peptides covering 39 proteins have been identified to be reactive with TB serum [15]. LEP constructed by tandem of 12 line Th epitopes
6
J.-L. Li et al.
Table II. IgG and Ig(G M) antibodies to PstS1-LEP. PTB (n 245) Group
IgG
Ig(G M)
EPTB (n 197)
NTBPD (n 75)
IgG
IgG
Ig(G M)
Ig(G M)
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
Positive no. (%)a 175 (71.4) 190 (77.6) 83 (42.1) 126 (64.0) 9 (12.0) 9 (12.0) Positive no. for one 11 26 7 50 2 2 isotype of antibody Double positive no.b 164 76 7 Combination positive 201 (82.0) 133 (67.5) 11 (14.7) no. (%)c
HC(PPD–) (n 48) IgG
Ig(G M)
1 (2.1) 2 (4.2) 0 1 1 2 (4.2)
HC(PPD) (n 102) IgG
Ig(G M)
21 (20.6) 18 (17.6) 7 4 14 25 (24.5)
PTB, extrapulmonary tuberculosis; HC (PPD–), healthy controls with negative reaction to purified protein derivative of BCG; HC E (PPD), healthy controls with positive reaction to purified protein derivative of BCG; LEP, line multi-epitopes polypeptide; NTBPD, nontuberculous pulmonary disease; PTB, pulmonary tuberculosis. aPositive no. (%), the number of positive sera (percentage of positive sera). bDouble positive no., the number positive for IgG and IgM. cCombination positive no. (%), the number of IgG- and/or IgM-positive sera (percentage of positive sera).
contains B-cell epitopes. The recombinant PstS1LEP protein showed positive responses to IgG, IgM, and Ig(G M) antibodies in serum from TB patients and retained immunoreactivity of PstS1 because it displayed positive responses to anti-PstS1 mAb. This evidence suggests that PstS1-LEP may be a candidate for serodiagnosis of TB. IgM against PstS1-LEP may be used to diagnose EPTB but not PTB. The values of AUCs usually indicate the diagnostic accuracy, which are too low to suggest meaningful use in clinical practice [28,29]. Our findings suggest that PstS1-LEP protein used to detect IgM antibodies has a very poor discriminatory ability with AUCs of 0.65. The levels of IgM antibody were higher in EPTB patients than in PTB patients, NTBPD patients, and HCs. Meanwhile, the sensitivity of anti-PstS1-LEP IgM was higher in EPTB patients than in PTB patients. The sensitivity of anti-PstS1-LEP IgM in TB patients, in accord with fusion protein 38kD-ESAT6-CFP10 detecting active TB [30], was lower than that of the combination of lipoarabinomannan antigen and PstS1 detecting PTB (48.7%) [31]. Our results suggest that IgM against PstS1-LEP may be able to detect EPTB. Combining IgG with other antibody isotypes may increase sensitivity but reduce specificity. Combinations of IgG, IgA, and IgM (IgG IgA IgM) can increase the sensitivity because some sera, that are negative for IgG, are still positive for IgA and/or IgM Table III. Effects of BCG vaccination on the specificity of PstS1-LEP protein. Specificity (%) BCG scar
IgG
IgM
Yes (n 108) No (n 42)
81.5 95.2
87.9 90.5
LEP, line multi-epitopes polypeptide.
Ig(G M) 83.3 95.2
antibodies [19,30]. We found that the sensitivity of IgG was lower than that of IgG IgM in EPTB patients, but the specificity of IgG was higher than that of IgG IgM in HC(PPD). Other researchers reported that IgG IgA IgM against 38kD-ESAT6-CFP10 increased sensitivity from 65.7% to 75.5% and decreased specificity slightly from 94.7% to 91.5% compared with IgG [30]. Our previous study showed that the sensitivity of IgG IgM provided by 38 kDa increased in EPTB patients [32]. Ig(G M) displayed a stronger immunoreactivity to PstS1-LEP than a single antibody. We found that the serological test based on Ig(G M) against PstS1-LEP increased its sensitivity but did not decrease its specificity. The specificity of IgG IgM was lower than that of Ig(G M) in HC(PPD). Thus, Ig(G M) against PstS1-LEP has advantages over IgG IgM responses to PstS1-LEP. BCG vaccination reduced the specificity of PstS1LEP. The specificities of IgG and Ig(G M) responses to PstS1-LEP were higher in HC(PPD–) than in HC(PPD), but such differences were not found in NTBPD patients in comparison with HC(PPD). The specificity of anti-PstS1-LEP Ig(G M) was higher in HCs without BCG scar than in HCs with BCG scar. Our findings suggest that HC(PPD–) may provide a high specificity of PstS1-LEP due to ignoring some practical problems, such as BCG vaccination, environmental mycobacteria, and other respiratory diseases. Thus, many studies on TB serodiagnosis use BCG-vaccinated HCs as controls [15,33]. In summary, PstS1-LEP has the potential to detect IgG and Ig(G M) antibodies in PTB patients and Ig(G M) antibody against PstS1-LEP exhibits high diagnostic values to detect EPTB.
IgG IgM 72.2 85.7
Acknowledgments This study was supported by funding from the National Mega Projects on Key Infectious Disease
Control of Ministry of Science and Technology, China (2013ZX10003006-003-001). Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Infect Dis Downloaded from informahealthcare.com by Nyu Medical Center on 05/13/15 For personal use only.
References [1] Lonnroth K, Castro KG, Chakaya JM, Chauhan LS, Floyd K, Glaziou P, et al. Tuberculosis control and elimination 2010-50: cure, care, and social development. Lancet 2010;375: 1814–29. [2] Small PM, Pai M. Tuberculosis diagnosis – time for a game change. N Engl J Med 2010;363:1070–1. [3] World Health Organization. Global tuberculosis report 2013. WHO; 2013. [4] Rasanathan K, Sivasankara Kurup A, Jaramillo E, Lonnroth K. The social determinants of health: key to global tuberculosis control. Int J Tuberc Lung Dis 2011;15(Suppl 2):S30–6. [5] Steingart KR, Ramsay A, Dowdy DW, Pai M. Serological tests for the diagnosis of active tuberculosis: relevance for India. Indian J Med Res 2012;135:695–702. [6] Steingart KR, Flores LL, Dendukuri N, Schiller I, Laal S, Ramsay A, et al. Commercial serological tests for the diagnosis of active pulmonary and extrapulmonary tuberculosis: an updated systematic review and meta-analysis. PLoS Med 2011;8:e1001062. [7] World Health Organization. Strategic and Technical Advisory Group for Tuberculosis (STAG-TB). Report of the Tenth Meeting. WHO; 2010. [8] Grenier J, Pinto L, Nair D, Steingart K, Dowdy D, Ramsay A, et al. Widespread use of serological tests for tuberculosis: data from 22 high-burden countries. Eur Respir J 2012;39:502–5. [9] Morris K. WHO recommends against inaccurate tuberculosis tests. Lancet 2011;377:113–14. [10] Raja A, Ranganathan UD, Bethunaickan R. Improved diagnosis of pulmonary tuberculosis by detection of antibodies against multiple Mycobacterium tuberculosis antigens. Diagn Microbiol Infect Dis 2008;60:361–8. [11] Ivanyi J. Function and potentials of M. tuberculosis epitopes. Front Immunol 2014;5:107. [12] Jackett PS, Bothamley GH, Batra HV, Mistry A, Young DB, Ivanyi J. Specificity of antibodies to immunodominant mycobacterial antigens in pulmonary tuberculosis. J Clin Microbiol 1988;26:2313–18. [13] Reche PA, Glutting JP, Reinherz EL. Prediction of MHC class I binding peptides using profile motifs. Hum Immunol 2002;63:701–9. [14] Pai M, Riley LW, Colford JM Jr. Interferon-gamma assays in the immunodiagnosis of tuberculosis: a systematic review. Lancet Infect Dis 2004;4:761–76. [15] Al-Khodari NY, Al-Attiyah R, Mustafa AS. Identification, diagnostic potential, and natural expression of immunodominant seroreactive peptides encoded by five Mycobacterium tuberculosis-specific genomic regions. Clin Vaccine Immunol 2011;18:477–82. [16] Singh M, Andersen AB, McCarthy JE, Rohde M, Schutte H, Sanders E, et al. The Mycobacterium tuberculosis 38-kDa antigen: overproduction in Escherichia coli, purification and characterization. Gene 1992;117:53–60.
Simultaneous detection of IgG and IgM
7
[17] He X, Zhuang Y, Zhang X, Xiong Z, Chang Z. The 38,000 protein of Mycobacterium tuberculosis is overexpressed in Escherichia coli. Zhonghua Jie He He Hu Xi Za Zhi 1999;22:138–41. [18] Anderson BL, Welch RJ, Litwin CM. Assessment of three commercially available serologic assays for detection of antibodies to Mycobacterium tuberculosis and identification of active tuberculosis. Clin Vaccine Immunol 2008;15:1644–9. [19] Uma Devi KR, Ramalingam B, Brennan PJ, Narayanan PR, Raja A. Specific and early detection of IgG, IgA and IgM antibodies to Mycobacterium tuberculosis 38kDa antigen in pulmonary tuberculosis. Tuberculosis (Edinb) 2001;81:249–53. [20] Senol G, Ecevit C, Ozturk A. Humoral immune response against 38- and 16-kDa mycobacterial antigens in childhood tuberculosis. Pediatr Pulmonol 2009;44:839–44. [21] Lyashchenko K, Colangeli R, Houde M, Al Jahdali H, Menzies D, Gennaro ML. Heterogeneous antibody responses in tuberculosis. Infect Immun 1998;66:3936–40. [22] Kunnath-Velayudhan S, Salamon H, Wang HY, Davidow AL, Molina DM, Huynh VT, et al. Dynamic antibody responses to the Mycobacterium tuberculosis proteome. Proc Natl Acad Sci U S A 2010;107:14703–8. [23] Lange C, Mori T. Advances in the diagnosis of tuberculosis. Respirology 2010;15:220–40. [24] Zhang CQ, Huang XY, Shao JS, Li JL, Song QD, Zhuang YH, et al. Effects of MF59 in combination with heat-killed BCG as adjuvant on the immunogenicity of Mycobacterium tuberculosis fusion protein PstS1-LEP. Zhong Hua Wei Sheng Wu Xue Yu Mian Yi Xue Za Zhi 2014;34:241–6. [25] Goyal B, Kumar K, Gupta D, Agarwal R, Latawa R, Sheikh JA, et al. Utility of B-cell epitopes based peptides of RD1 and RD2 antigens for immunodiagnosis of pulmonary tuberculosis. Diagn Microbiol Infect Dis 2014;78:391–7. [26] Demkow U, Filewska M, Michalowska-Mitczuk D, Kus J, Jagodzinski J, Zielonka T, et al. Heterogeneity of antibody response to myobacterial antigens in different clinical manifestations of pulmonary tuberculosis. J Physiol Pharmacol 2007;58(Suppl 5;Pt 1):117–27. [27] Lodes MJ, Dillon DC, Mohamath R, Day CH, Benson DR, Reynolds LD, et al. Serological expression cloning and immunological evaluation of MTB48, a novel Mycobacterium tuberculosis antigen. J Clin Microbiol 2001;39:2485–93. [28] Everson RM, Fieldsend JE. Multi-class ROC analysis from a multi-objective optimisation perspective. Pattern Recog Lett 2006;27:918–27. [29] Van der Schouw YT, Verbeek AL, Ruijs JH. ROC curves for the initial assessment of new diagnostic tests. Fam Pract 1992;9:506–11. [30] Feng X, Yang X, Xiu B, Qie S, Dai Z, Chen K, et al. IgG, IgM and IgA antibodies against the novel polyprotein in active tuberculosis. BMC Infect Dis 2014;14:336. [31] Singh S, Singh J, Kumar S, Gopinath K, Balooni V, Singh N, et al. Poor performance of serological tests in the diagnosis of pulmonary tuberculosis: evidence from a contact tracing field study. PLoS One 2012;7:e40213. [32] He XY, Li J, Hao J, Chen HB, Zhao YZ, Huang XY, et al. Assessment of five antigens from Mycobacterium tuberculosis for serodiagnosis of tuberculosis. Clin Vaccine Immunol 2011;18:565–70. [33] Khan N, Alam K, Nair S, Valluri VL, Murthy KJ, Mukhopadhyay S. Association of strong immune responses to PPE protein Rv1168c with active tuberculosis. Clin Vaccine Immunol 2008;15:974–80.
