- Email: [email protected]
Reverse line probe assay for cheap detection of Single Nucleotide Polymorphisms in Mycobacterium tuberculosis
Tuberculosis 110 (2018) 52–55
Contents lists available at ScienceDirect
Tuberculosis journal homepage: www.elsevier.com/locate/tube
Diagnostics
Reverse line probe assay for cheap detection of Single Nucleotide Polymorphisms in Mycobacterium tuberculosis
T
Memona Yasmina,b, Guislaine Refregierc, Rubina Tabassum Siddiquia,d,∗, Rizwan Iqbale, Shahid Ahmad Abbasif, Sabira Tahseeng a
Health Biotechnology Division, National Institute for Biotechnology and Genetic Engineering (NIBGE), P.O.Box No. 577, Jhang Road, Faisalabad, Pakistan Pakistan Institute of Engineering and Applied Sciences (PIEAS), Nilore, Islamabad, Pakistan c Institut de Génétique et Microbiologie, UMR8621, CNRS – Univ. Paris-Sud, Universud, Campus d’Orsay, F-91405 Orsay-Cedex, France d Punjab Institute of Nuclear Medicine (PINUM), Faisalabad, Pakistan e Pakistan Medical Research Council, Lahore, Pakistan f Armed Forces Institute of Pathology, Rawalpindi, Pakistan g National TB Reference Laboratory, Islamabad, Pakistan b
A R T I C LE I N FO
A B S T R A C T
Keywords: SNPs detection MDR-TB Molecular detection of RIF resistance Line probe assay Global health
More and more Single Nucleotide Polymosrphisms of interest among pathogenic organisms are described with the advent of Whole Genome Sequencing but WGS approach is still too expensive, time consuming, and relying on bioinformatical means that are not available in many developing countries. This study presents a low-cost reverse hybridization line probe technique for detecting SNPs in Mycobacterium tuberculosis. The proposed test is able to detect mutations in the RRDR of rpoB gene in M. tuberculosis with specificity and sensitivity of 98% and 100%, respectively and for an average cost of less than €3 per sample. The technique proved efficient not only on pure DNA samples extracted from culture isolates but also on crude extracts from clinical samples. The flexibility of the platform allows to get it transformed to any kind of test detection, hence, building a bridge between rich countries performing SNP discovery and countries with high burden that can target these SNPs on the collected samples.
1. Introduction Molecular epidemiology of bacterial pathogens enters a new era: Whole Genome Sequencing becomes a common technique in rich countries [1] that enables to gather maximized information to infer transmission chains and assessing drug resistance when SNPs accounting for resistance are almost or all known. For instance, Mycobacterium tuberculosis reads can be analyzed using Phyresse interface [2, 3]. Methods specifically designed to detect resistance are also beginning to be developed (Deeplex by Genoscreen, P. Supply, personal communication, and ongoing clinical trial NCT03303963). However, many infectious diseases concern low-income and middle-income countries that have massive amounts of patients and cannot benefit from WGS. Alternative methods are thus still needed. The rising issue for tuberculosis is the continuous increase of multiple drug resistance (MDR) i.e. the resistance to most potent first-line drugs; rifampicin (RIF) and isoniazid. In 2014, there were 480,000 (range: 360,000–600,000) estimated new cases of MDR-TB worldwide with approximately 190,000 deaths [4]. The spread of MDR could be
∗
controlled by rapid drug sensitivity testing as it plays an important role in timely and accurate initiation of effective therapy and surveillance of drug resistance [5, 6]. RIF resistance is found to be coupled with isoniazid resistance in more than 90% cases [7, 8] and hence RIF resistance can be used as a surrogate marker for the detection of MDR-TB. Mutations that confer resistance to this drug are frequently found in “RIF Resistance Determining Region (RRDR)”, an 81 bp “hot-spot” region consisting of 27 amino acids encompassing codon 426 to 452 of M. tuberculosis [9] which corresponds to 507 to 533 in Escherichia coli [10]. We propose “Reverse Hybridization Line Probe Assay”, using tailing of the oligonucleotides with deoxyribose thymidine (dT) and Nylon membrane as cheap option to immobilize probes. The present study aimed to identify key parameters to build up in-house reverse hybridization assays to screen mutations in bacterial pathogens such as those leading to RIF resistance in M. tuberculosis.
Corresponding author. Punjab Institute of Nuclear Medicine (PINUM), Faisalabad, Pakistan. E-mail address: [email protected] (R.T. Siddiqui).
https://doi.org/10.1016/j.tube.2018.03.007 Received 15 December 2017; Received in revised form 21 March 2018; Accepted 23 March 2018 1472-9792/ © 2018 Elsevier Ltd. All rights reserved.
Tuberculosis 110 (2018) 52–55
M. Yasmin et al.
Table 1 –Effect of hybridization and washing conditions on the signal intensity. Combination
Hybridization Temp. (oC)
Washing Temp. (oC)
Pre-hybridization solution
Probes with efficient detection
1 2 3 4 5 6 7 8
42 45 50 55 42 45 50 50
50 53 58 63 50 53 58 58
Yes Yes Yes Yes No No No No
None All Only W430, W434, W445, W450, W452 Only W430, W434, W450, W452 High background All High background High background
and biotin labelled 5′ end of reverse primer) and 1 μl of PCR product from the first round of PCR cycle. The cycling parameters were; 95 °C for 5 min followed by 30 cycles of 95 °C for 60 s, 65 °C for 30 s and 72 °C for 30 s.
2. Materials and methods 2.1. Oligonucleotides used to screen mutations A total of 40 oligonucleotides derived from already published literature [11] were used in this study. These correspond to 5 wild type probes spanning the mutation hotspot region of rpoB gene of M. tuberculosis and 35 mutant probes complementary to different mutations (Table S1). All these oligonucleotides were poly(dT) tailed.
2.6. Sequencing of RRDR region of rpoB gene To assess the accuracy of in house developed reverse hybridization assay, DNA sequencing was used as “Gold Standard”. The 257 bp PCR amplified fragments of hotspot region of rpoB gene for 81 isolates were commercially sequenced (Macrogen, Korea). Sequence information was stored, assembled, and analyzed using the Lasergene sequence analysis package (DNAStar Inc., Madison, WI, USA) and SeqScape (Version 2.6; Applied Biosystems) software.
2.2. Immobilization of oligonucleotides on filters All the oligonucleotides were applied on Hybond N + Nylon membrane (Amersham RPN 303N) using Miniblotter 45 (Immunetics, Inc. USA). They were immobilized by exposing membrane twice to UV radiation using UV cross linker at auto-crosslink mode (1200 μJ). The unbound nucleotides were removed by washing with 5XSSPE/0.5%SDS for 20 min at room temperature followed by washing with double distilled water and stored at room temperature till further use.
2.7. Reverse hybridization For hybridization, varying volumes of PCR product were mixed with a hybridization solution (5XSSPE /0.5% SDS) of 150 μL final volume and heat denatured for 2 min. This solution was applied to Hybond N + nylon membrane at right angle to the oligonucleotides probes, using Miniblotter 45. Hybridization conditions were optimized for quantity of PCR product, hybridization temperature, hybridization time and washing temperature (Table 1). Detection of hybridization signals was carried out using Biotin Chromogenic Detection kit (Fermentas Cat. #K0662) according to the suppliers' instructions. Color development on membrane indicated the positive hybridization signal.
2.3. M. tuberculosis samples A total of 125 bacterial samples enriched in MDR were included in this study. Eighty-one (81) bacterial cultures on Lowenstein Jensen (LJ) medium came from different regions of Pakistan (Peshawar = 20, Rawalpindi = 13, Quetta = 1, Faisalabad = 2, Lahore = 25, Karachi = 8, Islamabad = 12), collected during 2009–2010. The rest (44) were clinical specimens (blood; n = 35 and sputum; n = 6). All the cultures and clinical specimens were confirmed as M. tuberculosis complex by PCR amplification using primers specific for IS6110.
3. Results and discussion Sequencing analysis of 81 culture isolates revealed that 46 had 48 mutations of 10 different kinds in the RRDR of rpoB gene, while 35 isolates did not carry any mutation. Codon 450 was found to be most frequently mutated (65%) followed by codons 445 (14%), 435 (12%), 437 (2%), 430 (2%) and 439 (2%). Mutations in codon 450 were TCG to TTG[Ser→Leu] or TGG[Ser→Trp] while codon 445 exhibited four different types of SNPs: CAC to TAC/GAC/CTG/AAC[His→Tyr/Asp/Leu/ Phe]. Two isolates showed double mutation: 450(TCG→TTG) [Ser→ Leu] along with codon 445(CAC→AAC) [His→Asn]. Using these reference samples, we set up and measured the quality of a test aiming at detecting rpoB mutations potentially linked to RIF resistance. Five wild type probes and 30 probes corresponding to the most frequent mutations were coupled to a membrane and used to detect mutations in a subset of samples. Some rare mutations for which resistance was not proven, such as 445(CAC→AAC)[His→Asn] and 445(CAC→CTG)[His→Leu] mutation were not included. H37Rv was used to identify and to assess the sensitivity and specificity of wild-type probes while sequenced isolates allowed assessing 17 of 30 mutant probes. To find out the common parameters to get high sensitivity and specificity of hybridization signals, several conditions were tested where hybridization using 20 μL of PCR product, at 45 °C for 20 min with washing at 53 °C gave better results (Table 1). Worthy of note is that varying conditions provided varying levels of background or
2.4. Extraction of genomic DNA DNA from M. tuberculosis cultures on LJ slants was extracted by CTAB method (van Soolingen et al., 1991) while the extraction of DNA from clinical samples was performed using SDS/proteinase K method (Goldenberger et al., 1995). Sputum samples were decontaminated by NaOH/N-acetyl-L-cystein solution before extraction of DNA. About 40 ng of DNA from culture isolates, while 5 μL of DNA from clinical specimens, was used for subsequent PCRs. 2.5. Amplification of RRDR of rpoB gene Amplification of RRDR of rpoB gene of M. tuberculosis was achieved by nested PCR using primers reported in literature [12]. The outer and inner pair of primers generated a fragment of 395 bp and 257 bp, respectively. The PCR was performed in a final volume of 25 μL containing 1X PCR buffer, 1.5 mM MgCl2, 0.2 mM each dNTPs (Fermentas Cat. #R0181) and 1U of Taq DNA polymerase (Fermentas Cat. # EP0402). Outer pair of primers (10 pmol) was used in the first round of PCR cycle with amplification program consisted of: 5 min at 95 °C for 5 min; 30 cycles at 95 °C for 60 s, 58 °C for 30 s and 72 °C for 30 s. The second round of PCR was performed using inner set of primers (20 pmol 53
Tuberculosis 110 (2018) 52–55
M. Yasmin et al.
Fig. 1. Final display of the in-house line probe assay applied on a panel of pure extracts and clinical isolates. lanes indicate the distribution of wild • Horizontal type and corresponding mutant probes (W corre-
•
•
sponds to Wild type probe and M corresponds to Mutant probes) Vertical lanes indicate the samples (Lane 31 indicates control H37Rv; Lanes 1, 2, 3, 4 6, 7, 10, 11, 12, 14, 15, 16, 19, 20, 22, 24, 25, 28 and 30 indicate susceptible isolates; rectangles at lane 5, 8, 9, 13, 18, 21, 23, 26, 27 and 29 underline resistant isolates possessing one of the specific mutations indicated in Table 1.) Sample 19 to 21 are clinical samples
testing arena and research. The cost of commercially available assays makes them difficult to be used in routine especially in the resource poor settings with high prevalence and incidence of diseases. The assay we propose is particularly cheap thanks to the use of cheap components: nylon membrane with the poly(dT) tailed oligonucleotides instead of Biodyn C membrane with amino linkage resulted in approximately 25% reduction of costs at this step. Reagent costs correspond to a price of only €2.1 per sample as per pricing in Pakistan when handling 40 samples simultaneously, and €2.55 for batch of 11 samples while €3.54 for single sample. The difference in price between handling large batch and single sample is due to the necessity of including controls in every experiment, at least water and a sample with known rpoB sequence such as H37Rv, as a quality check (See Supplementary information S4). Costs may vary in other countries and depending on contracts with the suppliers. In comparison, the RIFO assay including only 11 oligonucleotides costs €4.18 per sample for 40 samples batches in 2002 [14]. Our assay exhibits high specificity and sensitivity (98% and 100%, respectively) as detected using reference culture samples, and we obtained 100% sensitivity on clinical samples as already observed on other pathogens when using nested PCR [15, 16]. The very high specificity we obtained on mutations included in our sample may rely on the inclusion of internal controls: a sample is declared mutant only when both mutant probe signal is on, and wild-type probe signal is off. We could not confirm good sensitivity and specificity of probes targeting mutations absent from our sample, however, presence of wild and mutant probes makes it possible to check whether they give concordant information. Quality control is also ensured by the inclusion of reference samples in each run (for instance, H37Rv and water). Since its set-up, this test has been efficiently implemented on 150 additional clinical samples.
sensitivity but specificity was never affected (Table 1). Reference samples produced 100% reproducible patterns and served as a test to confirm the quality of the experiment. Increased UV exposure to crosslink the probes resulted in the decrease of signal intensity, probably because of DNA nicking of the probe (data not shown). Nine out of ten rpoB mutations for which we had a positive control were successfully detected by in-house reverse hybridization line probe assay. The only undetected mutation was a rare silent mutation at codon 439(CCG→CCA)[Pro→Pro] in one isolate. Double mutations at codon 450 and codon 445 in two isolates were successfully detected. The mutations 445(CAC→AAC)[His→Asn] and 445(CAC→CTG)[His→ Leu] that were not included in our test and concerning a total of three isolates in our sample were indirectly detected by the absence of signal for both with wild and mutant probes. With one single discrepant result, the overall sensitivity and specificity of the assay was found to be 98% and 100%, respectively. Final display of membrane is exemplified in Fig. 1. Exhaustive results are presented in Supplementary Table 2. We then explored the ability of the test to provide results on clinical samples. Out of 44 specimens, PCR products from two samples hybridized with two mutant probes: carrying 430(CTG→CCG)[Leu→Pro] and 450(TCG→TTG)[Ser→Leu] mutation while all other samples gave signals only with wild type probes. The specimen that hybridized to mutant 450 probe also gave positive signal with 450 wild type probe, suggesting the presence of both wild type and mutant bacteria. Exhaustive results are given in Supplementary Table 3. Single nucleotide polymorphisms are distributed along the entire genome of organisms and their potential utility is varying from detection of disease causing genes, detection of drug resistance of pathogens, and pharmacogenesis to phylogenetic analysis [13]. Reliable and cost effective SNPs detection technologies are an important part of medical 54
Tuberculosis 110 (2018) 52–55
M. Yasmin et al.
Appendix A. Supplementary data
However, due to lack of means to enter approval procedures, it is not used routinely. The test has a turnaround time of approximately 14 (not full) working hours including DNA extraction, PCRs and reverse hybridization (See Supplementary information S4). As we obtained unambiguous results on clinical samples of both sputum and blood, the global turnaround time and costs are further reduced as compared to tests that require culture. The fact that clinical samples could be easily typed suggested that raw DNA extraction from cultured samples may allow for good test performances. We later occasionally used our membranes with 15 boiled cultures and indeed obtained unambiguous patterns. A limitation of the test we propose is that we do not target mutations outside the RRDR region. This would require the amplification of other parts of rpoB gene which may reduce the ease of its implementation. Future developments should investigate multiplex amplifications. With optimization steps taking one week to fix protocol, the flexibility of the blotting device allows to set up any diagnostic test with up to 43 targets where multiplexing could be used to amplify different target genes, hence, reducing work load and time to detection. To ensure good quality, two to three reference samples should be included in each run. This approach can be used to screen mutations in organisms other than M. tuberculosis. As it is relatively cheap and rapid, this technique is particularly relevant in developing countries where the cost is the major factor to be shifted. Only two key factors must be met for implementing such technique: the presence of technicians trained in molecular biology and the availability of standard molecular tools such as UV-linker and PCR machine, i.e. a real molecular biology laboratory. Of note is that most recent and possibly less robust machines such as Real-Time PCR are not needed for this test. Basic molecular laboratory remains a true lever for versatility of the service and seems a relevant investment for any regional hospital.
Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.tube.2018.03.007. References [1] Robinson ER, Walker TM, Pallen MJ. Genomics and outbreak investigation: from sequence to consequence. Genome Med 2013;5(4):36. [2] Feuerriegel S, et al. PhyResSE: a web tool delineating Mycobacterium tuberculosis antibiotic Resistance and Lineage from whole-genome sequencing data. J Clin Microbiol 2015;53(6):1908–14. [3] Pankhurst LJ, et al. Rapid, comprehensive, and affordable mycobacterial diagnosis with whole-genome sequencing: a prospective study. Lancet Respir Med 2016;4(1):49–58. [4] WHO, Global Tuberculosis Report 2015. World Health Organization; 2015. [5] Coban AY, et al. Drug susceptibility testing of Mycobacterium tuberculosis by the broth microdilution method with 7H9 broth. Mem Inst Oswaldo Cruz 2004;99(1):111–3. [6] van Klingeren B, et al. Drug susceptibility testing of Mycobacterium tuberculosis complex by use of a high-throughput, reproducible, absolute concentration method. J Clin Microbiol 2007;45(8):2662–8. [7] Drobniewski FA, Wilson SM. The rapid diagnosis of isoniazid and rifampicin resistance in Mycobacterium tuberculosis–a molecular story. J Med Microbiol 1998;47(3):189–96. [8] Ormerod LP. Multidrug-resistant tuberculosis (MDR-TB): epidemiology, prevention and treatment. Br Med Bull 2005;73-74:17–24. [9] Andre E, et al. Consensus numbering system for the rifampicin resistance-associated rpoB gene mutations in pathogenic mycobacteria. Clin Microbiol Infect 2017;23(3):167–72. [10] Telenti A, et al. Direct, automated detection of rifampin-resistant Mycobacterium tuberculosis by polymerase chain reaction and single-strand conformation polymorphism analysis. Antimicrob Agents Chemother 1993;37(10):2054–8. [11] Mengsu Yang, S.W, Mutation detection on RNA polmerase beta subunit gene having rifampin resistance, U.S. Patent, Editor. 2005, Genetel Pharmaceuticals Ltd., Hong Kong (CN). [12] Cavusoglu C, et al. Mycobacterium tuberculosis infection and laboratory diagnosis in solid-organ transplant recipients. Clin Transplant 2002;16(4):257–61. [13] Harvey JJ, et al. SNP analysis using CataCleave probes. J Clin Lab Anal 2008;22(3):192–203. [14] Morcillo N, et al. A low cost, home-made, reverse-line blot hybridisation assay for rapid detection of rifampicin resistance in Mycobacterium tuberculosis. Int J Tuberc Lung Dis 2002;6(11):959–65. [15] Hu S, et al. Detection of Sporothrix schenckii in clinical samples by a nested PCR assay. J Clin Microbiol 2003;41(4):1414–8. [16] Luo XF, et al. Establishment of a nested-ASP-PCR method to determine the clarithromycin resistance of Helicobacter pylori. World J Gastroenterol 2016;22(25):5822–30.
Conflicts of interest Authors have no conflict of interest. Acknowledgments We are thankful to Higher Education Commission for providing part of the funding for this work.
55
Contents lists available at ScienceDirect
Tuberculosis journal homepage: www.elsevier.com/locate/tube
Diagnostics
Reverse line probe assay for cheap detection of Single Nucleotide Polymorphisms in Mycobacterium tuberculosis
T
Memona Yasmina,b, Guislaine Refregierc, Rubina Tabassum Siddiquia,d,∗, Rizwan Iqbale, Shahid Ahmad Abbasif, Sabira Tahseeng a
Health Biotechnology Division, National Institute for Biotechnology and Genetic Engineering (NIBGE), P.O.Box No. 577, Jhang Road, Faisalabad, Pakistan Pakistan Institute of Engineering and Applied Sciences (PIEAS), Nilore, Islamabad, Pakistan c Institut de Génétique et Microbiologie, UMR8621, CNRS – Univ. Paris-Sud, Universud, Campus d’Orsay, F-91405 Orsay-Cedex, France d Punjab Institute of Nuclear Medicine (PINUM), Faisalabad, Pakistan e Pakistan Medical Research Council, Lahore, Pakistan f Armed Forces Institute of Pathology, Rawalpindi, Pakistan g National TB Reference Laboratory, Islamabad, Pakistan b
A R T I C LE I N FO
A B S T R A C T
Keywords: SNPs detection MDR-TB Molecular detection of RIF resistance Line probe assay Global health
More and more Single Nucleotide Polymosrphisms of interest among pathogenic organisms are described with the advent of Whole Genome Sequencing but WGS approach is still too expensive, time consuming, and relying on bioinformatical means that are not available in many developing countries. This study presents a low-cost reverse hybridization line probe technique for detecting SNPs in Mycobacterium tuberculosis. The proposed test is able to detect mutations in the RRDR of rpoB gene in M. tuberculosis with specificity and sensitivity of 98% and 100%, respectively and for an average cost of less than €3 per sample. The technique proved efficient not only on pure DNA samples extracted from culture isolates but also on crude extracts from clinical samples. The flexibility of the platform allows to get it transformed to any kind of test detection, hence, building a bridge between rich countries performing SNP discovery and countries with high burden that can target these SNPs on the collected samples.
1. Introduction Molecular epidemiology of bacterial pathogens enters a new era: Whole Genome Sequencing becomes a common technique in rich countries [1] that enables to gather maximized information to infer transmission chains and assessing drug resistance when SNPs accounting for resistance are almost or all known. For instance, Mycobacterium tuberculosis reads can be analyzed using Phyresse interface [2, 3]. Methods specifically designed to detect resistance are also beginning to be developed (Deeplex by Genoscreen, P. Supply, personal communication, and ongoing clinical trial NCT03303963). However, many infectious diseases concern low-income and middle-income countries that have massive amounts of patients and cannot benefit from WGS. Alternative methods are thus still needed. The rising issue for tuberculosis is the continuous increase of multiple drug resistance (MDR) i.e. the resistance to most potent first-line drugs; rifampicin (RIF) and isoniazid. In 2014, there were 480,000 (range: 360,000–600,000) estimated new cases of MDR-TB worldwide with approximately 190,000 deaths [4]. The spread of MDR could be
∗
controlled by rapid drug sensitivity testing as it plays an important role in timely and accurate initiation of effective therapy and surveillance of drug resistance [5, 6]. RIF resistance is found to be coupled with isoniazid resistance in more than 90% cases [7, 8] and hence RIF resistance can be used as a surrogate marker for the detection of MDR-TB. Mutations that confer resistance to this drug are frequently found in “RIF Resistance Determining Region (RRDR)”, an 81 bp “hot-spot” region consisting of 27 amino acids encompassing codon 426 to 452 of M. tuberculosis [9] which corresponds to 507 to 533 in Escherichia coli [10]. We propose “Reverse Hybridization Line Probe Assay”, using tailing of the oligonucleotides with deoxyribose thymidine (dT) and Nylon membrane as cheap option to immobilize probes. The present study aimed to identify key parameters to build up in-house reverse hybridization assays to screen mutations in bacterial pathogens such as those leading to RIF resistance in M. tuberculosis.
Corresponding author. Punjab Institute of Nuclear Medicine (PINUM), Faisalabad, Pakistan. E-mail address: [email protected] (R.T. Siddiqui).
https://doi.org/10.1016/j.tube.2018.03.007 Received 15 December 2017; Received in revised form 21 March 2018; Accepted 23 March 2018 1472-9792/ © 2018 Elsevier Ltd. All rights reserved.
Tuberculosis 110 (2018) 52–55
M. Yasmin et al.
Table 1 –Effect of hybridization and washing conditions on the signal intensity. Combination
Hybridization Temp. (oC)
Washing Temp. (oC)
Pre-hybridization solution
Probes with efficient detection
1 2 3 4 5 6 7 8
42 45 50 55 42 45 50 50
50 53 58 63 50 53 58 58
Yes Yes Yes Yes No No No No
None All Only W430, W434, W445, W450, W452 Only W430, W434, W450, W452 High background All High background High background
and biotin labelled 5′ end of reverse primer) and 1 μl of PCR product from the first round of PCR cycle. The cycling parameters were; 95 °C for 5 min followed by 30 cycles of 95 °C for 60 s, 65 °C for 30 s and 72 °C for 30 s.
2. Materials and methods 2.1. Oligonucleotides used to screen mutations A total of 40 oligonucleotides derived from already published literature [11] were used in this study. These correspond to 5 wild type probes spanning the mutation hotspot region of rpoB gene of M. tuberculosis and 35 mutant probes complementary to different mutations (Table S1). All these oligonucleotides were poly(dT) tailed.
2.6. Sequencing of RRDR region of rpoB gene To assess the accuracy of in house developed reverse hybridization assay, DNA sequencing was used as “Gold Standard”. The 257 bp PCR amplified fragments of hotspot region of rpoB gene for 81 isolates were commercially sequenced (Macrogen, Korea). Sequence information was stored, assembled, and analyzed using the Lasergene sequence analysis package (DNAStar Inc., Madison, WI, USA) and SeqScape (Version 2.6; Applied Biosystems) software.
2.2. Immobilization of oligonucleotides on filters All the oligonucleotides were applied on Hybond N + Nylon membrane (Amersham RPN 303N) using Miniblotter 45 (Immunetics, Inc. USA). They were immobilized by exposing membrane twice to UV radiation using UV cross linker at auto-crosslink mode (1200 μJ). The unbound nucleotides were removed by washing with 5XSSPE/0.5%SDS for 20 min at room temperature followed by washing with double distilled water and stored at room temperature till further use.
2.7. Reverse hybridization For hybridization, varying volumes of PCR product were mixed with a hybridization solution (5XSSPE /0.5% SDS) of 150 μL final volume and heat denatured for 2 min. This solution was applied to Hybond N + nylon membrane at right angle to the oligonucleotides probes, using Miniblotter 45. Hybridization conditions were optimized for quantity of PCR product, hybridization temperature, hybridization time and washing temperature (Table 1). Detection of hybridization signals was carried out using Biotin Chromogenic Detection kit (Fermentas Cat. #K0662) according to the suppliers' instructions. Color development on membrane indicated the positive hybridization signal.
2.3. M. tuberculosis samples A total of 125 bacterial samples enriched in MDR were included in this study. Eighty-one (81) bacterial cultures on Lowenstein Jensen (LJ) medium came from different regions of Pakistan (Peshawar = 20, Rawalpindi = 13, Quetta = 1, Faisalabad = 2, Lahore = 25, Karachi = 8, Islamabad = 12), collected during 2009–2010. The rest (44) were clinical specimens (blood; n = 35 and sputum; n = 6). All the cultures and clinical specimens were confirmed as M. tuberculosis complex by PCR amplification using primers specific for IS6110.
3. Results and discussion Sequencing analysis of 81 culture isolates revealed that 46 had 48 mutations of 10 different kinds in the RRDR of rpoB gene, while 35 isolates did not carry any mutation. Codon 450 was found to be most frequently mutated (65%) followed by codons 445 (14%), 435 (12%), 437 (2%), 430 (2%) and 439 (2%). Mutations in codon 450 were TCG to TTG[Ser→Leu] or TGG[Ser→Trp] while codon 445 exhibited four different types of SNPs: CAC to TAC/GAC/CTG/AAC[His→Tyr/Asp/Leu/ Phe]. Two isolates showed double mutation: 450(TCG→TTG) [Ser→ Leu] along with codon 445(CAC→AAC) [His→Asn]. Using these reference samples, we set up and measured the quality of a test aiming at detecting rpoB mutations potentially linked to RIF resistance. Five wild type probes and 30 probes corresponding to the most frequent mutations were coupled to a membrane and used to detect mutations in a subset of samples. Some rare mutations for which resistance was not proven, such as 445(CAC→AAC)[His→Asn] and 445(CAC→CTG)[His→Leu] mutation were not included. H37Rv was used to identify and to assess the sensitivity and specificity of wild-type probes while sequenced isolates allowed assessing 17 of 30 mutant probes. To find out the common parameters to get high sensitivity and specificity of hybridization signals, several conditions were tested where hybridization using 20 μL of PCR product, at 45 °C for 20 min with washing at 53 °C gave better results (Table 1). Worthy of note is that varying conditions provided varying levels of background or
2.4. Extraction of genomic DNA DNA from M. tuberculosis cultures on LJ slants was extracted by CTAB method (van Soolingen et al., 1991) while the extraction of DNA from clinical samples was performed using SDS/proteinase K method (Goldenberger et al., 1995). Sputum samples were decontaminated by NaOH/N-acetyl-L-cystein solution before extraction of DNA. About 40 ng of DNA from culture isolates, while 5 μL of DNA from clinical specimens, was used for subsequent PCRs. 2.5. Amplification of RRDR of rpoB gene Amplification of RRDR of rpoB gene of M. tuberculosis was achieved by nested PCR using primers reported in literature [12]. The outer and inner pair of primers generated a fragment of 395 bp and 257 bp, respectively. The PCR was performed in a final volume of 25 μL containing 1X PCR buffer, 1.5 mM MgCl2, 0.2 mM each dNTPs (Fermentas Cat. #R0181) and 1U of Taq DNA polymerase (Fermentas Cat. # EP0402). Outer pair of primers (10 pmol) was used in the first round of PCR cycle with amplification program consisted of: 5 min at 95 °C for 5 min; 30 cycles at 95 °C for 60 s, 58 °C for 30 s and 72 °C for 30 s. The second round of PCR was performed using inner set of primers (20 pmol 53
Tuberculosis 110 (2018) 52–55
M. Yasmin et al.
Fig. 1. Final display of the in-house line probe assay applied on a panel of pure extracts and clinical isolates. lanes indicate the distribution of wild • Horizontal type and corresponding mutant probes (W corre-
•
•
sponds to Wild type probe and M corresponds to Mutant probes) Vertical lanes indicate the samples (Lane 31 indicates control H37Rv; Lanes 1, 2, 3, 4 6, 7, 10, 11, 12, 14, 15, 16, 19, 20, 22, 24, 25, 28 and 30 indicate susceptible isolates; rectangles at lane 5, 8, 9, 13, 18, 21, 23, 26, 27 and 29 underline resistant isolates possessing one of the specific mutations indicated in Table 1.) Sample 19 to 21 are clinical samples
testing arena and research. The cost of commercially available assays makes them difficult to be used in routine especially in the resource poor settings with high prevalence and incidence of diseases. The assay we propose is particularly cheap thanks to the use of cheap components: nylon membrane with the poly(dT) tailed oligonucleotides instead of Biodyn C membrane with amino linkage resulted in approximately 25% reduction of costs at this step. Reagent costs correspond to a price of only €2.1 per sample as per pricing in Pakistan when handling 40 samples simultaneously, and €2.55 for batch of 11 samples while €3.54 for single sample. The difference in price between handling large batch and single sample is due to the necessity of including controls in every experiment, at least water and a sample with known rpoB sequence such as H37Rv, as a quality check (See Supplementary information S4). Costs may vary in other countries and depending on contracts with the suppliers. In comparison, the RIFO assay including only 11 oligonucleotides costs €4.18 per sample for 40 samples batches in 2002 [14]. Our assay exhibits high specificity and sensitivity (98% and 100%, respectively) as detected using reference culture samples, and we obtained 100% sensitivity on clinical samples as already observed on other pathogens when using nested PCR [15, 16]. The very high specificity we obtained on mutations included in our sample may rely on the inclusion of internal controls: a sample is declared mutant only when both mutant probe signal is on, and wild-type probe signal is off. We could not confirm good sensitivity and specificity of probes targeting mutations absent from our sample, however, presence of wild and mutant probes makes it possible to check whether they give concordant information. Quality control is also ensured by the inclusion of reference samples in each run (for instance, H37Rv and water). Since its set-up, this test has been efficiently implemented on 150 additional clinical samples.
sensitivity but specificity was never affected (Table 1). Reference samples produced 100% reproducible patterns and served as a test to confirm the quality of the experiment. Increased UV exposure to crosslink the probes resulted in the decrease of signal intensity, probably because of DNA nicking of the probe (data not shown). Nine out of ten rpoB mutations for which we had a positive control were successfully detected by in-house reverse hybridization line probe assay. The only undetected mutation was a rare silent mutation at codon 439(CCG→CCA)[Pro→Pro] in one isolate. Double mutations at codon 450 and codon 445 in two isolates were successfully detected. The mutations 445(CAC→AAC)[His→Asn] and 445(CAC→CTG)[His→ Leu] that were not included in our test and concerning a total of three isolates in our sample were indirectly detected by the absence of signal for both with wild and mutant probes. With one single discrepant result, the overall sensitivity and specificity of the assay was found to be 98% and 100%, respectively. Final display of membrane is exemplified in Fig. 1. Exhaustive results are presented in Supplementary Table 2. We then explored the ability of the test to provide results on clinical samples. Out of 44 specimens, PCR products from two samples hybridized with two mutant probes: carrying 430(CTG→CCG)[Leu→Pro] and 450(TCG→TTG)[Ser→Leu] mutation while all other samples gave signals only with wild type probes. The specimen that hybridized to mutant 450 probe also gave positive signal with 450 wild type probe, suggesting the presence of both wild type and mutant bacteria. Exhaustive results are given in Supplementary Table 3. Single nucleotide polymorphisms are distributed along the entire genome of organisms and their potential utility is varying from detection of disease causing genes, detection of drug resistance of pathogens, and pharmacogenesis to phylogenetic analysis [13]. Reliable and cost effective SNPs detection technologies are an important part of medical 54
Tuberculosis 110 (2018) 52–55
M. Yasmin et al.
Appendix A. Supplementary data
However, due to lack of means to enter approval procedures, it is not used routinely. The test has a turnaround time of approximately 14 (not full) working hours including DNA extraction, PCRs and reverse hybridization (See Supplementary information S4). As we obtained unambiguous results on clinical samples of both sputum and blood, the global turnaround time and costs are further reduced as compared to tests that require culture. The fact that clinical samples could be easily typed suggested that raw DNA extraction from cultured samples may allow for good test performances. We later occasionally used our membranes with 15 boiled cultures and indeed obtained unambiguous patterns. A limitation of the test we propose is that we do not target mutations outside the RRDR region. This would require the amplification of other parts of rpoB gene which may reduce the ease of its implementation. Future developments should investigate multiplex amplifications. With optimization steps taking one week to fix protocol, the flexibility of the blotting device allows to set up any diagnostic test with up to 43 targets where multiplexing could be used to amplify different target genes, hence, reducing work load and time to detection. To ensure good quality, two to three reference samples should be included in each run. This approach can be used to screen mutations in organisms other than M. tuberculosis. As it is relatively cheap and rapid, this technique is particularly relevant in developing countries where the cost is the major factor to be shifted. Only two key factors must be met for implementing such technique: the presence of technicians trained in molecular biology and the availability of standard molecular tools such as UV-linker and PCR machine, i.e. a real molecular biology laboratory. Of note is that most recent and possibly less robust machines such as Real-Time PCR are not needed for this test. Basic molecular laboratory remains a true lever for versatility of the service and seems a relevant investment for any regional hospital.
Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.tube.2018.03.007. References [1] Robinson ER, Walker TM, Pallen MJ. Genomics and outbreak investigation: from sequence to consequence. Genome Med 2013;5(4):36. [2] Feuerriegel S, et al. PhyResSE: a web tool delineating Mycobacterium tuberculosis antibiotic Resistance and Lineage from whole-genome sequencing data. J Clin Microbiol 2015;53(6):1908–14. [3] Pankhurst LJ, et al. Rapid, comprehensive, and affordable mycobacterial diagnosis with whole-genome sequencing: a prospective study. Lancet Respir Med 2016;4(1):49–58. [4] WHO, Global Tuberculosis Report 2015. World Health Organization; 2015. [5] Coban AY, et al. Drug susceptibility testing of Mycobacterium tuberculosis by the broth microdilution method with 7H9 broth. Mem Inst Oswaldo Cruz 2004;99(1):111–3. [6] van Klingeren B, et al. Drug susceptibility testing of Mycobacterium tuberculosis complex by use of a high-throughput, reproducible, absolute concentration method. J Clin Microbiol 2007;45(8):2662–8. [7] Drobniewski FA, Wilson SM. The rapid diagnosis of isoniazid and rifampicin resistance in Mycobacterium tuberculosis–a molecular story. J Med Microbiol 1998;47(3):189–96. [8] Ormerod LP. Multidrug-resistant tuberculosis (MDR-TB): epidemiology, prevention and treatment. Br Med Bull 2005;73-74:17–24. [9] Andre E, et al. Consensus numbering system for the rifampicin resistance-associated rpoB gene mutations in pathogenic mycobacteria. Clin Microbiol Infect 2017;23(3):167–72. [10] Telenti A, et al. Direct, automated detection of rifampin-resistant Mycobacterium tuberculosis by polymerase chain reaction and single-strand conformation polymorphism analysis. Antimicrob Agents Chemother 1993;37(10):2054–8. [11] Mengsu Yang, S.W, Mutation detection on RNA polmerase beta subunit gene having rifampin resistance, U.S. Patent, Editor. 2005, Genetel Pharmaceuticals Ltd., Hong Kong (CN). [12] Cavusoglu C, et al. Mycobacterium tuberculosis infection and laboratory diagnosis in solid-organ transplant recipients. Clin Transplant 2002;16(4):257–61. [13] Harvey JJ, et al. SNP analysis using CataCleave probes. J Clin Lab Anal 2008;22(3):192–203. [14] Morcillo N, et al. A low cost, home-made, reverse-line blot hybridisation assay for rapid detection of rifampicin resistance in Mycobacterium tuberculosis. Int J Tuberc Lung Dis 2002;6(11):959–65. [15] Hu S, et al. Detection of Sporothrix schenckii in clinical samples by a nested PCR assay. J Clin Microbiol 2003;41(4):1414–8. [16] Luo XF, et al. Establishment of a nested-ASP-PCR method to determine the clarithromycin resistance of Helicobacter pylori. World J Gastroenterol 2016;22(25):5822–30.
Conflicts of interest Authors have no conflict of interest. Acknowledgments We are thankful to Higher Education Commission for providing part of the funding for this work.
55