Multiplex PCR-Based Reverse Line Blot Hybridization Assay To Identify 23 Streptococcus pneumoniae Polysaccharide Vaccine Serotypes
http://www.100md.com
微生物临床杂志 2006年第5期
Centre for Infectious Diseases and Microbiology, Institute of Clinical Pathology and Medical Research, Westmead, New South Wales, Australia
Department of Cell and Molecular Biology, Faculty of Science, University of Technology, Sydney, New South Wales, Australia
School of Biotechnology and Biomolecular Sciences, The University of New South Wales, New South Wales, Australia
Wuhan First Hospital, Hubei Province, Wuhan 430022, People's Republic of China
ABSTRACT
We developed a multiplex PCR-based reverse line blot assay to identify 23 pneumococcal serotypes represented in the polysaccharide vaccine, using 334 well-characterized isolates, representing all 90 serotypes, and 268 "unknowns." The assay identified all target serotypes, but 11, which cross-react with 1 to 4 nonvaccine serotypes, could be distinguished using serotype-specific antisera.
TEXT
Previously, we developed a molecular capsule type prediction system for 90 Streptococcus pneumoniae serotypes, based on a combination of partial cpsA-cpsB sequencing and serotype- and/or serogroup-specific PCR (3, 6). While this system is useful, it is too slow, expensive, and labor-intensive for routine use.
Multiplex PCR-based reverse line blot hybridization (mPCR/RLB) is a promising method for simultaneous detection and genotyping of microorganisms, which we have used previously in several applications (5, 9, 10). In the present study we applied it to identification of the 23 S. pneumoniae serotypes represented in the polysaccharide vaccine (Pneumovax 23; Merck & Co, Inc.).
The design of primers and probes was based on recently published full cps gene cluster sequences of all 90 pneumococcal serotypes (http://www.sanger.ac.uk/Projects/S_pneumoniae/CPS/) and others available in GenBank (3, 6; F. Kong, G. L. Gilbert, L. Wang, D. Liu, and J. Tao, 13 April 2004, Australian Patent Office). For the mPCR, we modified the primers we had used previously for serotype-specific PCR for the 23 vaccine serotypes (3, 6; Kong et al., Australian Patent Office) and a published S. pneumoniae-specific primer (7). For the RLB, we designed two probes for each of the 23 serotypes, as well as two S. pneumoniae-specific control probes (Table 1). Probes and primers were designed to have similar physical characteristics so as to allow simultaneous amplification and hybridization in a multiplex reaction system (5); they were synthesized by Sigma-Aldrich (Sydney, Australia). Primers were biotinylated at the 5' end and probes had a 5'-amine group (5).
The DNA extraction method (4), the mPCR system, and the thermal profiles used were as previously described (5), except that we included 24 (rather than 10) primer pairs (Table 1) in a 25-μl mPCR system and used 1 U (rather than 0.5) of QIAGEN Hotstart Taq polymerase. For serotype 23F, 25 pM of each primer were used, whereas 12.5 pM of each primer was used for all other serotypes.
RLB hybridization was based on previously published methods (http://www.nioo.knaw.nl/cl/me/) (8, 11) with the following modifications: the hybridization temperature was 60°C, and 1.25 pM of each probe, in 150 μl of 500 mM NaHCO3 (pH 8.4), was used in each slot to label the membrane (Fig. 1).
A collection of 334 reference strains and well-characterized clinical isolates were used to develop the assay, of which 244 had been used in our previous studies (3, 6) and 90 were serotype reference strains, newly purchased from Statens Serum Institut of Denmark (1, 2). Except for serotypes 10C, 11F, 12B, 25A, 33D, and 44, all serotypes were represented by two or more strains.
All of the putative serotype-specific primers or probes yielded mPCR products and RLB signals from isolates with the corresponding serotypes. However, as demonstrated previously, serotype discrimination based solely on the wzy gene (or even the whole cps gene cluster) is not straightforward (3, 6). Primers and probes designed to identify serotypes 6B, 7F, 9N, 9V, 10A, 11A, 12F, 15B, 18C, 22F, and 33F could not distinguish between the target and one or more closely related serotypes, usually but not always in the corresponding serogroup (Table 1 and Fig. 1) because they share virtually identical wzy sequences. Thus, 17 serotypes, in addition to those in the 23-valent polysaccharide vaccine, were amplified and hybridized by the mPCR/RLB system. These cross-reactions are predictable, and individual serotypes can be identified using a limited number of factor antisera. The remaining 12 primer pairs and probe pairs were truly serotype specific, and there was excellent agreement between paired probes for the same serotypes, indicating that single probes would be adequate for most serotypes.
The method was further evaluated using 268 clinical isolates, the serotypes of which were unknown at the time of mPCR/RLB testing. These isolates included 135 consecutive invasive isolates referred to the NSW Pneumococcal Reference Laboratory for serotyping and 133 colonizing isolates from patients with respiratory infections at the Children's Hospital, Westmead, New South Wales, Australia. Conventional serotyping was performed using the Quellung reaction, as previously described (3). Two isolates were not amplified by mPCR/RLB, and phenotypic retesting showed that they were not S. pneumoniae. Of 266 pneumococcus strains, 12 (4.5%) were nontypeable by mPCR/RLB (amplified by mPCR but hybridized only with the pneumococcal control probe); of these, 10 belonged to serotypes not represented in the current mPCR/RLB assay, namely, 16F (one isolate), 23A (three isolates), 35F (four isolates), 35B (one isolate), and 38 (one isolate), and 2 were not serotypeable. Another 4 of the 266 pneumococcus isolates were not serotypeable but were identified by mPCR/RLB as serotype(s) 4, 11A/11D, 14, and 33F/33A/37. The predicted serotype(s) of the other 250 isolates were as follows: 3 (5 isolates), 4 (13 isolates), 6B/6A (45 isolates), 7F/7A (1 isolate), 9V/9A (15 isolates), 10A/10B (2 isolates), 11A/11D (2 isolates), 14 (51 isolates), 15B/15C (4 isolates), 18 (14 isolates), 19F (63 isolates), 19A (12 isolates), 22F/22A (2 isolates), and 23F (21 isolates). Thus, 181 of 254 (71%) isolates that were typeable by mPCR/RLB were identified exactly, and another 73 (29%) were identified to within one to five serotypes, for which a limited number of antisera were needed to distinguish individual serotypes.
Three isolates identified by mPCR/RLB as serotypes 19A, 9V/9A, and 23F had been serotyped, initially, as 19F, 19A, and 6B, respectively. The discrepancies were resolved by repeating the serotyping, which confirmed that the mPCR/RLB results were correct.
Our molecular serotype prediction mPCR/RLB assay for 23 pneumococcal "vaccine" serotypes is clearly less discriminatory than conventional serotyping (1-3, 6; Kong et al., Australian Patent Office) because individual serotype-specific targets are not available within cps gene clusters of some serotypes. However, there are obvious advantages of the mPCR/RLB. First, it uses reagents and techniques that are available in many microbiology laboratories; an uncomplicated, rapid DNA preparation method (4); and a single mPCR/RLB reaction (5). Second, it provides more consistent and objective results than immunological methods such as the Quellung reaction; cross-reactions are predictable and can be resolved with a small number of antisera. Serotypes that are nontypeable by the mPCR/RLB are isolated uncommonly from clinical specimens (5% of invasive isolates in NSW in the past 3 years [data not shown]). Third, the method, potentially, could to be used directly to test clinical specimens and so rapidly identify possible vaccine failure.
Currently, there is no single, ideal technique for pneumococcal serotyping, but this mPCR/RLB format is convenient, rapid, objective, reproducible, and discriminatory when used in conjunction with a limited number of antisera.
ACKNOWLEDGMENTS
We thank Denise Murphy, Queensland Health Scientific Services, Brisbane, Australia; Diana Martin, Institute of Environmental Science Research, Porirua, New Zealand for providing isolates; and Ping Zhu for technical assistance.
This study was funded, in part, by the National Centre for Immunization Research and Surveillance of Vaccine Preventable Disease, Children's Hospital at Westmead, New South Wales, Australia.
REFERENCES
Henrichsen, J. 1995. Six newly recognized types of Streptococcus pneumoniae. J. Clin. Microbiol. 33:2759-2762.
Henrichsen, J. 1999. Typing of Streptococcus pneumoniae: past, present, and future. Am. J. Med. 107:50S-54S.
Kong, F., and G. L. Gilbert. 2003. Using cpsA-cpsB sequence polymorphisms and serotype-/group-specific PCR to predict 51 Streptococcus pneumoniae capsular serotypes. J. Med. Microbiol. 52:1047-1058.
Kong, F., S. Gowan, D. Martin, G. James, and G. L. Gilbert. 2002. Serotype identification of group B streptococci by PCR and sequencing. J. Clin. Microbiol. 40:216-226.
Kong, F., L. Ma, and G. L. Gilbert. 2005. Simultaneous detection and serotype identification of Streptococcus agalactiae using multiplex PCR and reverse line blot hybridization. J. Med. Microbiol. 54:1133-1138.
Kong, F., W. Wang, J. Tao, L. Wang, Q. Wang, A. Sabananthan, and G. L. Gilbert. 2005. A molecular-capsular-type prediction system for 90 Streptococcus pneumoniae serotypes using partial cpsA-cpsB sequencing and wzy- or wzx-specific PCR. J. Med. Microbiol. 54:351-356.
Salo, P., A. Ortqvist, and M. Leinonen. 1995. Diagnosis of bacteremic pneumococcal pneumonia by amplification of pneumolysin gene fragment in serum. J. Infect. Dis. 171:479-482.
van den Brule, A. J., R. Pol, N. Fransen-Daalmeijer, L. M. Schouls, C. J. Meijer, and P. J. Snijders. 2002. GP5+/6+ PCR followed by reverse line blot analysis enables rapid and high-throughput identification of human papillomavirus genotypes. J. Clin. Microbiol. 40:779-787.
Zeng, X., F. Kong, H. Wang, A. Darbar, and G. L. Gilbert. 2006. Simultaneous detection of nine antibiotic resistance-related genes in Streptococcus agalactiae using multiplex PCR and reverse line blot hybridization assay. Antimicrob. Agents Chemother. 50:204-209.
Zhao, Z., F. Kong, and G. L. Gilbert. 2006. Reverse line blot assay for direct identification of seven Streptococcus agalactiae major surface protein antigen genes. Clin. Vaccine Immunol. 13:145-149.
Zwart, G., E. J. van Hannen, M. P. Kamst-van Agterveld, G. K. Van der, E. S. Lindstrom, J. Van Wichelen, T. Lauridsen, B. C. Crump, S. K. Han, and S. Declerck. 2003. Rapid screening for freshwater bacterial groups by using reverse line blot hybridization. Appl. Environ. Microbiol. 69:5875-5883.(Fanrong Kong, Mitchell Br)
Department of Cell and Molecular Biology, Faculty of Science, University of Technology, Sydney, New South Wales, Australia
School of Biotechnology and Biomolecular Sciences, The University of New South Wales, New South Wales, Australia
Wuhan First Hospital, Hubei Province, Wuhan 430022, People's Republic of China
ABSTRACT
We developed a multiplex PCR-based reverse line blot assay to identify 23 pneumococcal serotypes represented in the polysaccharide vaccine, using 334 well-characterized isolates, representing all 90 serotypes, and 268 "unknowns." The assay identified all target serotypes, but 11, which cross-react with 1 to 4 nonvaccine serotypes, could be distinguished using serotype-specific antisera.
TEXT
Previously, we developed a molecular capsule type prediction system for 90 Streptococcus pneumoniae serotypes, based on a combination of partial cpsA-cpsB sequencing and serotype- and/or serogroup-specific PCR (3, 6). While this system is useful, it is too slow, expensive, and labor-intensive for routine use.
Multiplex PCR-based reverse line blot hybridization (mPCR/RLB) is a promising method for simultaneous detection and genotyping of microorganisms, which we have used previously in several applications (5, 9, 10). In the present study we applied it to identification of the 23 S. pneumoniae serotypes represented in the polysaccharide vaccine (Pneumovax 23; Merck & Co, Inc.).
The design of primers and probes was based on recently published full cps gene cluster sequences of all 90 pneumococcal serotypes (http://www.sanger.ac.uk/Projects/S_pneumoniae/CPS/) and others available in GenBank (3, 6; F. Kong, G. L. Gilbert, L. Wang, D. Liu, and J. Tao, 13 April 2004, Australian Patent Office). For the mPCR, we modified the primers we had used previously for serotype-specific PCR for the 23 vaccine serotypes (3, 6; Kong et al., Australian Patent Office) and a published S. pneumoniae-specific primer (7). For the RLB, we designed two probes for each of the 23 serotypes, as well as two S. pneumoniae-specific control probes (Table 1). Probes and primers were designed to have similar physical characteristics so as to allow simultaneous amplification and hybridization in a multiplex reaction system (5); they were synthesized by Sigma-Aldrich (Sydney, Australia). Primers were biotinylated at the 5' end and probes had a 5'-amine group (5).
The DNA extraction method (4), the mPCR system, and the thermal profiles used were as previously described (5), except that we included 24 (rather than 10) primer pairs (Table 1) in a 25-μl mPCR system and used 1 U (rather than 0.5) of QIAGEN Hotstart Taq polymerase. For serotype 23F, 25 pM of each primer were used, whereas 12.5 pM of each primer was used for all other serotypes.
RLB hybridization was based on previously published methods (http://www.nioo.knaw.nl/cl/me/) (8, 11) with the following modifications: the hybridization temperature was 60°C, and 1.25 pM of each probe, in 150 μl of 500 mM NaHCO3 (pH 8.4), was used in each slot to label the membrane (Fig. 1).
A collection of 334 reference strains and well-characterized clinical isolates were used to develop the assay, of which 244 had been used in our previous studies (3, 6) and 90 were serotype reference strains, newly purchased from Statens Serum Institut of Denmark (1, 2). Except for serotypes 10C, 11F, 12B, 25A, 33D, and 44, all serotypes were represented by two or more strains.
All of the putative serotype-specific primers or probes yielded mPCR products and RLB signals from isolates with the corresponding serotypes. However, as demonstrated previously, serotype discrimination based solely on the wzy gene (or even the whole cps gene cluster) is not straightforward (3, 6). Primers and probes designed to identify serotypes 6B, 7F, 9N, 9V, 10A, 11A, 12F, 15B, 18C, 22F, and 33F could not distinguish between the target and one or more closely related serotypes, usually but not always in the corresponding serogroup (Table 1 and Fig. 1) because they share virtually identical wzy sequences. Thus, 17 serotypes, in addition to those in the 23-valent polysaccharide vaccine, were amplified and hybridized by the mPCR/RLB system. These cross-reactions are predictable, and individual serotypes can be identified using a limited number of factor antisera. The remaining 12 primer pairs and probe pairs were truly serotype specific, and there was excellent agreement between paired probes for the same serotypes, indicating that single probes would be adequate for most serotypes.
The method was further evaluated using 268 clinical isolates, the serotypes of which were unknown at the time of mPCR/RLB testing. These isolates included 135 consecutive invasive isolates referred to the NSW Pneumococcal Reference Laboratory for serotyping and 133 colonizing isolates from patients with respiratory infections at the Children's Hospital, Westmead, New South Wales, Australia. Conventional serotyping was performed using the Quellung reaction, as previously described (3). Two isolates were not amplified by mPCR/RLB, and phenotypic retesting showed that they were not S. pneumoniae. Of 266 pneumococcus strains, 12 (4.5%) were nontypeable by mPCR/RLB (amplified by mPCR but hybridized only with the pneumococcal control probe); of these, 10 belonged to serotypes not represented in the current mPCR/RLB assay, namely, 16F (one isolate), 23A (three isolates), 35F (four isolates), 35B (one isolate), and 38 (one isolate), and 2 were not serotypeable. Another 4 of the 266 pneumococcus isolates were not serotypeable but were identified by mPCR/RLB as serotype(s) 4, 11A/11D, 14, and 33F/33A/37. The predicted serotype(s) of the other 250 isolates were as follows: 3 (5 isolates), 4 (13 isolates), 6B/6A (45 isolates), 7F/7A (1 isolate), 9V/9A (15 isolates), 10A/10B (2 isolates), 11A/11D (2 isolates), 14 (51 isolates), 15B/15C (4 isolates), 18 (14 isolates), 19F (63 isolates), 19A (12 isolates), 22F/22A (2 isolates), and 23F (21 isolates). Thus, 181 of 254 (71%) isolates that were typeable by mPCR/RLB were identified exactly, and another 73 (29%) were identified to within one to five serotypes, for which a limited number of antisera were needed to distinguish individual serotypes.
Three isolates identified by mPCR/RLB as serotypes 19A, 9V/9A, and 23F had been serotyped, initially, as 19F, 19A, and 6B, respectively. The discrepancies were resolved by repeating the serotyping, which confirmed that the mPCR/RLB results were correct.
Our molecular serotype prediction mPCR/RLB assay for 23 pneumococcal "vaccine" serotypes is clearly less discriminatory than conventional serotyping (1-3, 6; Kong et al., Australian Patent Office) because individual serotype-specific targets are not available within cps gene clusters of some serotypes. However, there are obvious advantages of the mPCR/RLB. First, it uses reagents and techniques that are available in many microbiology laboratories; an uncomplicated, rapid DNA preparation method (4); and a single mPCR/RLB reaction (5). Second, it provides more consistent and objective results than immunological methods such as the Quellung reaction; cross-reactions are predictable and can be resolved with a small number of antisera. Serotypes that are nontypeable by the mPCR/RLB are isolated uncommonly from clinical specimens (5% of invasive isolates in NSW in the past 3 years [data not shown]). Third, the method, potentially, could to be used directly to test clinical specimens and so rapidly identify possible vaccine failure.
Currently, there is no single, ideal technique for pneumococcal serotyping, but this mPCR/RLB format is convenient, rapid, objective, reproducible, and discriminatory when used in conjunction with a limited number of antisera.
ACKNOWLEDGMENTS
We thank Denise Murphy, Queensland Health Scientific Services, Brisbane, Australia; Diana Martin, Institute of Environmental Science Research, Porirua, New Zealand for providing isolates; and Ping Zhu for technical assistance.
This study was funded, in part, by the National Centre for Immunization Research and Surveillance of Vaccine Preventable Disease, Children's Hospital at Westmead, New South Wales, Australia.
REFERENCES
Henrichsen, J. 1995. Six newly recognized types of Streptococcus pneumoniae. J. Clin. Microbiol. 33:2759-2762.
Henrichsen, J. 1999. Typing of Streptococcus pneumoniae: past, present, and future. Am. J. Med. 107:50S-54S.
Kong, F., and G. L. Gilbert. 2003. Using cpsA-cpsB sequence polymorphisms and serotype-/group-specific PCR to predict 51 Streptococcus pneumoniae capsular serotypes. J. Med. Microbiol. 52:1047-1058.
Kong, F., S. Gowan, D. Martin, G. James, and G. L. Gilbert. 2002. Serotype identification of group B streptococci by PCR and sequencing. J. Clin. Microbiol. 40:216-226.
Kong, F., L. Ma, and G. L. Gilbert. 2005. Simultaneous detection and serotype identification of Streptococcus agalactiae using multiplex PCR and reverse line blot hybridization. J. Med. Microbiol. 54:1133-1138.
Kong, F., W. Wang, J. Tao, L. Wang, Q. Wang, A. Sabananthan, and G. L. Gilbert. 2005. A molecular-capsular-type prediction system for 90 Streptococcus pneumoniae serotypes using partial cpsA-cpsB sequencing and wzy- or wzx-specific PCR. J. Med. Microbiol. 54:351-356.
Salo, P., A. Ortqvist, and M. Leinonen. 1995. Diagnosis of bacteremic pneumococcal pneumonia by amplification of pneumolysin gene fragment in serum. J. Infect. Dis. 171:479-482.
van den Brule, A. J., R. Pol, N. Fransen-Daalmeijer, L. M. Schouls, C. J. Meijer, and P. J. Snijders. 2002. GP5+/6+ PCR followed by reverse line blot analysis enables rapid and high-throughput identification of human papillomavirus genotypes. J. Clin. Microbiol. 40:779-787.
Zeng, X., F. Kong, H. Wang, A. Darbar, and G. L. Gilbert. 2006. Simultaneous detection of nine antibiotic resistance-related genes in Streptococcus agalactiae using multiplex PCR and reverse line blot hybridization assay. Antimicrob. Agents Chemother. 50:204-209.
Zhao, Z., F. Kong, and G. L. Gilbert. 2006. Reverse line blot assay for direct identification of seven Streptococcus agalactiae major surface protein antigen genes. Clin. Vaccine Immunol. 13:145-149.
Zwart, G., E. J. van Hannen, M. P. Kamst-van Agterveld, G. K. Van der, E. S. Lindstrom, J. Van Wichelen, T. Lauridsen, B. C. Crump, S. K. Han, and S. Declerck. 2003. Rapid screening for freshwater bacterial groups by using reverse line blot hybridization. Appl. Environ. Microbiol. 69:5875-5883.(Fanrong Kong, Mitchell Br)