- Email: [email protected]
Evaluation of multiple genetic markers for typing drug-resistant Mycobacterium tuberculosis strains from Poland
Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64 www.elsevier.com/locate/diagmicrobio
Mycobacteriology
Evaluation of multiple genetic markers for typing drug-resistant Mycobacterium tuberculosis strains from PolandB Anna Sajdudaa, Jarosyaw Dziadekb, Roman Kotyowskic, Franc¸oise Portaelsd,4 a
Department of Genetics of Microorganisms, University of xo´dz´, xo´dz´ 90-237, Poland b Center for Medical Biology, Polish Academy of Sciences, xo´dz´ 93-232, Poland c Chemical Faculty, Gdan´sk University of Technology, Gdan´sk 80-952, Poland d Mycobacteriology Unit, Department of Microbiology, Institute of Tropical Medicine, 2000 Antwerp, Belgium Received 5 August 2005; revised 7 December 2005; accepted 7 December 2005
Abstract In the present study, 77 drug-resistant Mycobacterium tuberculosis strains isolated in Poland in 2000 were characterized by the mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) typing and our novel method based on PCR amplification of DNA regions between IS6110 and 16-bp GC-rich frequent repeats (designated IS6110-Mtb1/Mtb2 PCR). The results were compared with previous data of the more commonly used methods, IS6110 restriction fragment length polymorphism (RFLP) and spoligotyping. The discriminatory power of IS6110-Mtb1/Mtb2 method was only slightly lower than that of IS6110 RFLP, whereas MIRU-VNTR typing was the least discriminative among the 4 methods used. Clustering of strains by using results of IS6110-Mtb1/Mtb2 PCR correlated well with RFLP-defined clusters, further confirming epidemiologic relationships among patients. These results indicate that the novel genotyping method could be an attractive alternative for other PCR-based typing procedures, such as spoligotyping and MIRU-VNTR typing. Also, it seems to be a valuable adjunct to the reference IS6110 RFLP method for studying the genetic diversity of drugresistant M. tuberculosis strains in Poland. D 2006 Elsevier Inc. All rights reserved. Keywords: Genetic markers; Genotyping; Drug resistance; Mycobacterium tuberculosis; Poland
1. Introduction Tuberculosis (TB) remains a major infectious disease and causes high morbidity and mortality worldwide. The situation is made even worse by the emergence of drugresistant strains of Mycobacterium tuberculosis (Dahle et al., 2003). Poland, such as many other countries in Europe, has witnessed a dramatic fall in the incidence of TB during the last century. However, in 2000, the prevalence of primary drug resistance (including multidrug resistance) increased 2-fold in comparison with that of 1997 (6.1% and 1.2%, respectively) (Augustynowicz-Kopec et al., 2003). To better
B This work was partly supported by the Fund for Scientific Research of Flanders (Brussels, Belgium, grant G.0471.03 N). A.S. was supported by a NATO Science Fellowship at the Institute of Tropical Medicine, Antwerp, Belgium. 4 Corresponding author. Tel.: +32-3-2476317; fax: +32-3-2476333. E-mail address: [email protected] (F. Portaels).
0732-8893/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.diagmicrobio.2005.12.004
understand the epidemiology of TB, we have developed in recent years a large number of DNA fingerprinting methods based on various genetic markers (Mostro¨m et al., 2002). Because a single genotyping method cannot define all unique isolates, the current studies undertaken require various typing strategies to increase the power of strain differentiation (Cowan et al., 2005; Sun et al., 2004a, 2004b). The most widely used and generally accepted international bgold standard Q method has been IS6110 restriction fragment length polymorphism (RFLP), based on the variability in number of copies and chromosomal locations of the IS6110 insertion element between strains (van Embden et al., 1993). The IS6110 RFLP is the most discriminatory method at the population level (van Embden et al., 2000). Unfortunately, it is laborious and expensive, and requires uniform data processing systems, complicating the comparison of results from different laboratories due to subjectivity in the analysis (Cowan et al., 2005). To overcome the technical difficulties encountered by IS6110
60
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64
Table 1 Genotyping results for 77 drug-resistant M. tuberculosis strains from Poland by using multiple genetic markers IS6110 RFLP pattern (no. of strains) 1 2 3 4 5 6 7 8 9
(2) (2) (2) (2) (2) (2) (7) (2) (8)
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
(2)b (2) (3) (2) (2) (4) (5) (2) (2) (7) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1)b (1) (1) (1) a b
Spoligotype
MIRU-VNTR
Pattern no. in IS6110-Mtb1/Mtb2 PCR using primers
STa
Octal
No.
Pattern
Mtb1-IS1-IS2
Mtb2-IS1-IS2
53 90 1555 52 42 53 1558 462 1051
777777777760771 677777776000371 700001403760371 777777777760731 777777607760771 777777777760771 777777775660771 777777777560771 777777437760771
1 53 237 463 1559 264 891 2 47 1557 54 Orphan 280 180 280 237 1557 1564 39 50 52 Orphan 1746 1 53 46 1558
000000000003771 777777777760771 777777777700000 777777777720571 777777400000171 777740003760771 777777607660771 000000004020771 777777774020771 777677774020771 777777777763771 777777776000740 770000777760771 677777777720771 770000777760771 777777777700000 777677774020771 777776737760571 777777347760471 777777777720771 777777777760731 777777776760571 476377737760771 000000000003771 777777777760771 777777770000000 777777775660771
18 13 10 13 11 6 6 9 10 16 19 20 14 21 4 15 2 7 1 24 11 25 3 26 3 23 24 30 22 29 12 8 28 17 7 27 5
223326133323 223325153325 223126153324 223325153325 223226143321 223115153324 223115153324 223125153324 223126153324 223325163533 223326133523 224125153324 223325153533 224325113322 222226151323 223325154322 124325152222 223125143324 124225163224 225323153225 223226143321 225323153323 222125153324 225325153223 222125153324 224325153322 225323153225 23V4315143311 224325143424 235325153323 223325153324 223125153224 233325153225 223325173533 223125143324 225325153323 222325153224
1 2 3 2 4 5 6 7 3 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 21 23 24 25 26 27 28 29 30 31 32 33 34
1 2 3 2 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 10 20 21 22 21 23 24 25 26 27 28 29 30 31 32 33 34
ST = shared type in spoligotyping database SpolDB3 (www.pasteur-guadeloupe.fr/tb/spoldb3). Beijing family strains.
RFLP analysis, we have developed easier PCR-based methods, such as spoligotyping (Kamerbeek et al., 1997) and mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) typing (Mazars et al., 2001). Spoligotyping, based on polymorphism in the chromosomal direct repeat locus, has been the most commonly used PCR-based typing procedure. This method is fast and reproducible; however, it is less discriminatory than the IS6110 RFLP (Kremer et al., 2005; Sun et al., 2004a, 2004b). A new method, MIRU-VNTR typing, is based on the determination of the number of copies of repeated units at 12 independent loci scattered throughout the genome. It allows rapid analysis of the results (numbers of repeats) as digital patterns, with subsequent comparison with worldwide databases. The high resolution and the
possibility of high-throughput analysis make MIRU-VNTR an attractive method for analysis of the global genetic diversity of M. tuberculosis strains (Blackwood et al., 2004; Supply et al., 2000, 2001). Recently, we developed a genotypic method based on PCR amplification of DNA regions between IS6110 and frequently repeated 16-bp GCrich sequences Mtb1 or Mtb2. Preliminary results indicate the usefulness of our method, designated IS6110-Mtb1/ Mtb2 PCR, in differentiation of M. tuberculosis strains for epidemiologic purposes (Kotlowski et al., 2004). In the present study, we applied IS6110-Mtb1/Mtb2 genotyping method and MIRU-VNTR analysis to further characterize drug-resistant M. tuberculosis strains from Poland, previously typed by the reference IS6110 RFLP and spoligotyping (Sajduda et al., 2004). We evaluated a
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64 Table 2 Comparison of the discriminatory power of genotyping methods Set and methoda
No. of distinct patterns
No. of unique isolates
No. of clusters (ranges)
No. of clustered isolates (%)
HGDI
Set 1 IS6110 RFLP Spoligotyping MIRU-VNTR Mtb1-IS1-IS2 Mtb2-IS1-IS2
36 27 30 34 34
17 9 14 17 15
19 18 16 17 19
(2– 8) (2– 8) (2– 10) (2–10) (2– 8)
60 68 63 60 62
(78) (88) (82) (78) (81)
0.965 0.951 0.950 0.958 0.963
Set 2 IS6110 RFLP Spoligotyping MIRU-VNTR Mtb1-IS1-IS2 Mtb2-IS1-IS2
19 17 17 18 19
19 17 15 16 17
(2– 8) (2– 8) (2–10) (2–10) (2– 8)
60 60 58 58 58
(100) (100) (97) (97) (97)
0.942 0.936 0.924 0.932 0.941
0 0 2b 2b 2b
61
we prepared bacterial DNA either by boiling of the mycobacterial cells for 10 min or by the internationally standardized protocol (van Embden et al., 1993). 2.2. Genotyping methods
a Set 1 includes all the strains tested (n = 77), whereas set 2 comprises strains clustered by IS6110 RFLP (n = 60). b Beijing family strains.
multistep typing strategy to attain maximum specificity and compared the discriminatory ability of the 4 methods used. 2. Materials and methods 2.1. Bacterial strains and DNA preparation The 72 M. tuberculosis strains (of 251 drug-resistant strains studied) isolated in Poland in 2000 during the second national survey of drug resistance have been previously shown to cluster by the IS6110 RFLP and spoligotyping (Sajduda et al., 2004). Sixty of these clustered strains and, in addition, 17 strains as representative of the 179 strains with unique IS6110 RFLP patterns were included in further analysis in the present study. M. tuberculosis H37Rv reference strain served as a control. To genotype the 77 drug-resistant strains by the use of PCR-based methods,
MIRU-VNTR analyses were performed using primers for the 12 MIRU-VNTR loci as described by Supply et al. (2001). Three different concentrations of MgCl2 were used depending on MIRU-VNTR locus: 1.5 mmol/L for MIRU loci 20, 24, 26, and 27; 2 mmol/L for MIRU loci 2, 4, 10, 16, 31, and 40; and 2.5 mmol/L for MIRU loci 23 and 39. Each MIRU-VNTR locus was individually amplified in a 25-AL reaction volume containing 100 Amol/L of each deoxynucleoside triphosphate, 0.1 Amol/L of each primer, 0.5 U of HotStarTaq DNA polymerase (Qiagen, Hilden, Germany), 1 Q solution, and 2.5 AL of template DNA. The cycling parameters were 95 8C for 15 min to activate the HotStarTaq DNA polymerase, followed by 40 amplification cycles at 94 8C for 1 min, 59 8C for 1 min, and 72 8C for 1 min 30 s, with a final extension at 72 8C for 10 min in a PTC 100 thermocycler (MJ Research, Waltham, MA). PCR products were electrophoresed on a 3% NuSieve agarose gel and sized with a 50- and 100-bp ladder (Promega, Leiden, The Netherlands), and/or by using the Agilent 2100 bioanalyzer (Agilent Technologies, Waldbronn, Germany) for more accurate determination of amplicon sizes. Strain typing with IS6110-Mtb1/Mtb2 PCR that uses combination of primers IS1 and IS2, directed at inverted repeats flanking IS6110, and Mtb1 or Mtb2, targeting the repeated GC-rich motif, was performed essentially as we described previously (Kotlowski et al., 2004). In a PCR, 2.5 AL template DNA (purified or crude extract) was amplified in a 25-AL reaction volume. An aliquot of PCR products (1 AL) was electrophoresed and analyzed with the bioanalyzer system.
Table 3 MIRU-VNTR allelic diversity in 77 drug-resistant M. tuberculosis strains from Poland Locus
2 4b 10 16 20 23 24 26 27 31 39 40
No. of strains with the specified MIRU copy number 1
2
3
4
5
6
7
7 0 0 28 10 0 77 2 2 0 1 4
70 74 5 8 67 0 0 0 5 19 71 12
0 2 46 41 0 9 0 3 66 51 5 14
0 0 14 0 0 0 0 8 4 1 0 34
0 0 12 0 0 50 0 60 0 6 0 13
0 0 0 0 0 18 0 3 0 0 0 0
0 0 0 0 0 0 0 1 0 0 0 0
Allelic diversitya
Discriminatory power
0.154 0.038 0.576 0.568 0.216 0.504 0.000 0.369 0.252 0.488 0.134 0.712
Poor Poor Moderate Moderate Poor Moderate Poor Moderate Poor Moderate Poor High
a Allelic diversity (h) at a locus was calculated as follows: h ¼ 1 Rx2i ½n=ðn 1Þ, where x i is the frequency of the ith allele at the locus and n is the number of isolates (Mazars et al., 2001). b In 1 strain, the 53-bp unit at the 3V terminus of locus 4 was absent.
62
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64
2.3. Calculation of discriminatory power The Hunter–Gaston Discriminatory Index (HGDI) (Hunter and Gaston, 1988) was used as a numerical index for the discriminatory power of each typing method. The HGDI was calculated using the following formu la: HGDI ¼ 1 f1=½ N ð N 1ÞgfRð j¼1S Þ nj nj 1 g , where N is the total number of strains in the typing scheme, s is the total number of different patterns, and n j is the number of strains belonging to the jth pattern. The calculation was applied to both the entire study population and a discrete sample set including strains originally clustered by the IS6110 RFLP method.
3. Results and discussion The 77 drug-resistant M. tuberculosis strains were characterized in the present study by using MIRU-VNTR typing and IS6110-Mtb1/Mtb2 method (Kotlowski et al., 2004) for maximum specificity, and by comparison of the discriminatory power of the 2 new genetic markers with the most widely used reference IS6110 RFLP method and spoligotyping. Table 1 summarizes the strains sorted by IS6110 RFLP pattern. Seventeen strains had unique IS6110 RFLP patterns, and 60 strains clustered in 19 groups consisting of 2–8 isolates with identical IS6110 RFLP patterns (Tables 1 and 2). All strains were high copy numbers (contained z 6 copies of IS6110).
Spoligotyping identified 27 distinct patterns in the 77 strains split into 9 unique isolates and 68 isolates clustered in 18 groups containing 2 – 8 isolates having the same spoligotype (Tables 1 and 2). Of the 18 spoligotype clusters, 7 included strains with multiple IS6110 RFLP patterns. Three strains had the Beijing spoligotype (van Soolingen et al., 1995), of which 2 strains shared the same IS6110 RFLP pattern, whereas the third isolate was unique (Table 1). The new MIRU-VNTR method (Mazars et al., 2001) was applied here for the first time as an additional molecular tool to differentiate M. tuberculosis isolates in Poland. The number of repeats for each of the 12 MIRU-VNTR loci (Supply et al., 2000) ranged from 1 to 7 copies. One strain contained three 77-bp repeats without the 3V 53-bp unit in locus 4, resulting in aberrant copy number (Table 1, pattern 30) characteristic for M. tuberculosis strains H37Rv and H37Ra, Mycobacterium bovis BCG, and less than 1% of the recent clinical isolates (Cowan et al., 2002; Supply et al., 2000). Table 3 shows allelic diversities of the 12 MIRUVNTR loci in our study set. According to the classification proposed by Sola et al. (2003), only locus 40 provided a high discrimination level ( z 0.6), whereas loci 10, 16, 23, 26, and 31 were moderately discriminative (z 0.3), and loci 2, 4, 20, 24, 27, and 39 were poorly discriminative (b 0.3). In general, the allelic diversity was comparable to some previous studies (Cowan et al., 2002; Supply et al., 2000), but it was lower than in others (Sola et al., 2003; Sun et al.,
Fig. 1. Fingerprinting results for 2 Beijing family strains by various methods: IS6110 RFLP (A), spoligotyping (B), MIRU-VNTR typing (C), and IS6110Mtb1/Mtb2 PCR with primers Mtb1-IS1-IS2 (D) or Mtb2-IS1-IS2 (E). Lanes: 1, M. tuberculosis DS1520; 2, M. tuberculosis PD629; H, M. tuberculosis H37Rv; M, MIRU locus; L, DNA ladder. Differences in MIRU copy number in loci 23, 26, and 39 are underlined. The 15- and 1500-bp bands in all lanes in panels D and E represent lower and upper internal markers.
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64
2004a, 2004b), most likely due to the lower diversity and smaller number of isolates examined here. The MIRU-VNTR typing revealed 30 distinct patterns among the 77 strains tested (Tables 1 and 2). Fourteen strains were unique, and 63 strains were grouped in 16 clusters including 2–10 isolates. Of the 16 MIRU clusters, only 9 (56%) contained strains with identical IS6110 RFLP patterns and spoligotypes. The remaining 7 MIRU clusters could be subdivided by spoligotyping and/ or IS6110 RFLP. On the other hand, 2 members of 1 RFLP cluster found in the Beijing family were further differentiated by MIRU-VNTR (Fig. 1C), thus, confirming lack of epidemiologic link between those patients based on contact tracing. The third Beijing strain, isolated from a patient of Armenian origin, also had a distinct MIRU-VNTR pattern (Table 1, pattern 17) that was found to be predominant among members of this genotype family in different geographic areas (Cowan et al., 2005; Kam et al., 2005; Mokrousov et al., 2004; Sun et al., 2004a, 2004b). Mokrousov et al. (2004) hypothesize that this type could have resulted from a convergent evolution due to a possible biologic role of some MIRUs or could present a stable conserved combination achieved long ago and unchanged since evolutionary distant time (Mokrousov et al., 2004). Overall, in contrast to other studies (Cowan et al., 2005; Kam et al., 2005; Mokrousov et al., 2004), MIRU-VNTR typing performed better for the Beijing strains than IS6110 RFLP, most likely due to the limited number of isolates tested here. As shown in Tables 1 and 2, the IS6110-Mtb1/Mtb2 PCR (Kotlowski et al., 2004) yielded results similar to those obtained by the reference method. Using either primer sets (Mtb1-IS1-IS2 or Mtb2-IS1-IS2) in a simple PCR, we were able to detect only 3 clusters (of 17 or 19, respectively) that could be subtyped by spoligotyping and/or IS6110 RFLP, thus, proving close epidemiologic relationships between isolates in most of the clusters. Similar to MIRU-VNTR typing, IS6110-Mtb1/Mtb2 method discriminated between 2 Beijing strains clustered by IS6110 RFLP (Fig. 1D and E). However, in that case, the primer combination Mtb1-IS1-IS2 performed slightly better, producing distinct banding pattern for each strain, whereas PCR with primers Mtb2IS1-IS2 subdivided the aforementioned 2 strains but clustered 1 of them with a unique non-Beijing isolate (Table 1, pattern 10). In general, the IS6110-Mtb1/Mtb2 method seems a promising tool in differentiation of the increasingly important Beijing family of M. tuberculosis strains, but it would have to be evaluated on a larger collection of isolates. Clustering results from each method were compared with each other to determine the discriminatory power of the typing methods applied in the present study. As shown in Table 2, IS6110 RFLP was the most discriminative when applied both to the entire population (HGDI = 0.965) and to a set of 60 strains previously clustered by this method (HGDI = 0.942). Interestingly, in both study sets, IS6110-
63
Mtb1/Mtb2 PCR based on primers Mtb2-IS1-IS2 gave resolving power very close to that of the reference method (HGDI = 0.963 and 0.941, respectively). Although the degree of discrimination obtained with each method did not differ significantly, MIRU-VNTR typing was the least discriminative. A low sensitivity of MIRU-VNTR method for detecting IS6110 RFLP-based clusters and its poor specificity for providing distinct patterns among isolates unmatched by IS6110 RFLP have been reported recently by Scott et al. (2005) in IS6110 high–copy-number M. tuberculosis isolates from Montreal. Our results concur with their observations but appear to contrast with previous findings, demonstrating close discriminatory power of the MIRU-VNTR typing to that of the RFLP method (Blackwood et al., 2004; Cowan et al., 2002; Supply et al., 2001). It is most likely due to the sample studied that included smaller number of less heterogeneous, IS6110 high–copy-number M. tuberculosis strains than the previous studies. On the other hand, in a similar study, Hawkey et al. (2003) recently found a very close correlation between the RFLP and MIRU typing of apparently clustered cases of TB. However, their study sample included 53 M. tuberculosis isolates grouped in only 4 clusters. It can be expected that with increasing number of different RFLP clusters, probability of clustering epidemiologically nonrelated strains by the MIRU-VNTR method would be also higher. In this study, the PCR products were analyzed on NuSieve agarose gels for better resolution, and they were also labeled with fluorescent dye for increased throughput automated analysis on a microfluidic labchip instrument, the Agilent 2100 bioanalyzer. This new technology offers several important advantages over the traditional agarose electrophoresis that could be especially useful in MIRUVNTR analyses. It is fast (12 samples on each disposable labchip are analyzed within 30 min), easy (a minimum of technical skill is required), accurate, reproducible, and it generates low waste. Also, it is cost-effective (the volumes of reagents may be reduced because only 1-AL sample is evaluated) and versatile (it is complemented with kits for dsDNA, RNA, protein, and flow cytometric analyses). Despite the high cost of equipment, which might limit its usefulness in developing countries and/or areas with high prevalence of TB, the bioanalyzer system appears a suitable format for performance of PCR-based typing analyses. Although various typing strategies have been proposed depending on the population structure of M. tuberculosis, the present tendency is that MIRU-VNTR typing will likely predominate as a prescreening method regardless of the population structure. However, there is still a need for alternative simple, fast, and reliable genotyping procedures. Based on our previous and present results, the IS6110Mtb1/Mtb2 method seems a valuable adjunct to the reference IS6110 RFLP typing. For us to further confirm its utility in typing M. tuberculosis, its evaluation on a larger
64
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64
number of strains from various geographic areas would, however, be necessary.
Acknowledgments We acknowledge the Department of Microbiology, National Research Institute of Tuberculosis and Lung Diseases, Warsaw, Poland, for providing M. tuberculosis strains. References Augustynowicz-Kopec E, Zwolska Z, Jaworski A, Kostrzewa E, Klatt M (2003) Drug-resistant tuberculosis in Poland in 2000: Second national survey and comparison with the 1997 survey. Int J Tuberc Lung Dis 7:645 – 651. Blackwood KS, Wolfe JN, Kabani AM (2004) Application of mycobacterial interspersed repetitive unit typing to Manitoba tuberculosis cases: Can restriction fragment length polymorphism be forgotten? J Clin Microbiol 42:5001 – 5006. Cowan LS, Mosher L, Diem L, Massey JP, Crawford JT (2002) Variablenumber tandem repeat typing of Mycobacterium tuberculosis isolates with low copy numbers of IS6110 by using mycobacterial interspersed repetitive units. J Clin Microbiol 40:1592 – 1602. Cowan LS, Diem L, Monson T, Wand P, Temporado D, Oemig TV, Crawford JT (2005) Evaluation of a two-step approach for large-scale, prospective genotyping of Mycobacterium tuberculosis isolates in the United States. J Clin Microbiol 43:688 – 695. Dahle UR, Sandven P, Heldal E, Mannsaaker T, Caugant DA (2003) Deciphering an outbreak of drug-resistant Mycobacterium tuberculosis. J Clin Microbiol 41:67 – 72. Hawkey PM, Smith EG, Evans JT, Monk P, Bryan G, Mohamed HH, Bardhan M, Pugh RN (2003) Mycobacterial interspersed repetitive unit typing of Mycobacterium tuberculosis compared to IS6110-based restriction fragment length polymorphism analysis for investigation of apparently clustered cases of tuberculosis. J Clin Microbiol 41: 3514 – 3520. Hunter PR, Gaston MA (1988) Numerical index of the discriminatory ability of typing systems: An application of Simpson’s index of diversity. J Clin Microbiol 26:2465 – 2466. Kam KM, Yip CW, Tse LW, Wong KL, Lam TK, Kremer K, Au BKY, van Soolingen D (2005) Utility of mycobacterial interspersed repetitive unit typing for differentiating multidrug-resistant Mycobacterium tuberculosis isolates of the Beijing family. J Clin Microbiol 43:306 – 313. Kamerbeek J, Schouls L, Kolk A, van Agterveld M, van Soolingen D, Kuijper S, Bunschoten A, Molhuizen H, Shaw R, Goyal M, van Embden JDA (1997) Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin Microbiol 35:907 – 914. Kotyowski R, Shamputa IC, El Aila NA, Sajduda A, Rigouts L, van Deun A, Portaels F (2004) PCR-based genotyping of Mycobacterium tuberculosis with new GC-rich repeated sequences and IS6110 inverted repeats used as primers. J Clin Microbiol 42:372 – 377. Kremer K, Au BKY, Yip PCW, Skuce R, Supply P, Kam KM, van Soolingen D (2005) Use of variable-number tandem-repeat typing to differentiate Mycobacterium tuberculosis Beijing family isolates
form Hong Kong and comparison with IS6110 restriction fragment length polymorphism typing and spoligotyping. J Clin Microbiol 43: 314 – 320. Mazars E, Lesjean S, Banuls A-L, Gilbert M, Vincent V, Gicquel B, Tibayrenc M, Locht C, Supply P (2001) High-resolution minisatellitebased typing as a portable approach to global analysis of Mycobacterium tuberculosis molecular epidemiology. Proc Natl Acad Sci U S A 98:1901 – 1906. Mokrousov B, Narvskaya O, Limeschenko E, Vyazovaya A, Otten T, Vyshnevskiy B (2004) Analysis of the allelic diversity of the mycobacterial interspersed repetitive units in Mycobacterium tuberculosis strains of the Beijing family: Practical implications and evolutionary considerations. J Clin Microbiol 42:2438 – 2444. Mostrfm P, Gordon M, Sola C, Ridell M, Rastogi N (2002) Methods used in the molecular epidemiology of tuberculosis. Clin Microbiol Infect 8:694 – 704. Sajduda A, Brzostek A, Popyawska M, Rastogi N, Sola C, AugustynowiczKopec´ E, Zwolska Z, Dziadek J, Portaels F (2004) Molecular epidemiology of drug-resistant Mycobacterium tuberculosis strains isolated from patients with pulmonary tuberculosis in Poland: A 1year study. Int J Tuberc Lung Dis 8:1448 – 1457. Scott AN, Menzies D, Tannenbaum T-N, Thibert L, Kozak R, Joseph L, Schwartzman K, Behr MA (2005) Sensitivities and specificities of spoligotyping and mycobacterial interspersed repetitive unit-variablenumber tandem repeat typing methods for studying molecular epidemiology of tuberculosis. J Clin Microbiol 43:89 – 94. Sola C, Filliol I, Legrand E, Lesjean S, Locht C, Supply P, Rastogi N (2003) Genotyping of the Mycobacterium tuberculosis complex using MIRUs: Association with VNTR and spoligotyping for molecular epidemiology and evolutionary genetics. Infect Genet Evol 3:125 – 133. Sun Y-J, Bellamy R, Lee ASG, Ng ST, Ravindran S, Wong S-Y, Locht C, Supply P, Paton NI (2004a) Use of mycobacterial interspersed repetitive unit-variable-number tandem repeat typing to examine genetic diversity of Mycobacterium tuberculosis in Singapore. J Clin Microbiol 42: 1986 – 1993. Sun Y-J, Lee ASG, Ng ST, Ravindran S, Kremer K, Bellamy R, Wong S-Y, van Soolingen D, Supply P, Paton NI (2004b) Characterization of ancestral Mycobacterium tuberculosis by multiple genetic markers and proposal of genotyping strategy. J Clin Microbiol 42:5058 – 5064. Supply P, Mazars E, Lesjean S, Vincent V, Gicquel B, Locht C (2000) Variable human minisatellite-like regions in the Mycobacterium tuberculosis genome. Mol Microbiol 36:762 – 771. Supply P, Lesjean S, Savine E, Kremer K, van Soolingen D, Locht C (2001) Automated high-throughput genotyping for study of global epidemiology of Mycobacterium tuberculosis based on mycobacterial interspersed repetitive units. J Clin Microbiol 39:3563 – 3571. van Embden JD, Cave MD, Crawford JT, Dale JW, Eisenach KD, Gicquel B, Hermans P, Martin C, McAdam R, Shinnick TM (1993) Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: Recommendations for a standardized methodology. J Clin Microbiol 31:406 – 409. van Embden JD, van Gorkom T, Kremer K, Jansen R, van Der Zeijst BA, Schouls LM (2000) Genetic variation and evolutionary origin of the direct repeat locus of Mycobacterium tuberculosis complex bacteria. J Bacteriol 182:2393 – 2401. van Soolingen D, Qian L, de Haas PEW, Douglas JT, Traore H, Portaels F, Qing HZ, Enkhasaikan D, Nymadawa P, van Embden JDA (1995) Predominance of a single genotype of Mycobacterium tuberculosis in countries of East Asia. J Clin Microbiol 33:3234 – 3238.
Mycobacteriology
Evaluation of multiple genetic markers for typing drug-resistant Mycobacterium tuberculosis strains from PolandB Anna Sajdudaa, Jarosyaw Dziadekb, Roman Kotyowskic, Franc¸oise Portaelsd,4 a
Department of Genetics of Microorganisms, University of xo´dz´, xo´dz´ 90-237, Poland b Center for Medical Biology, Polish Academy of Sciences, xo´dz´ 93-232, Poland c Chemical Faculty, Gdan´sk University of Technology, Gdan´sk 80-952, Poland d Mycobacteriology Unit, Department of Microbiology, Institute of Tropical Medicine, 2000 Antwerp, Belgium Received 5 August 2005; revised 7 December 2005; accepted 7 December 2005
Abstract In the present study, 77 drug-resistant Mycobacterium tuberculosis strains isolated in Poland in 2000 were characterized by the mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) typing and our novel method based on PCR amplification of DNA regions between IS6110 and 16-bp GC-rich frequent repeats (designated IS6110-Mtb1/Mtb2 PCR). The results were compared with previous data of the more commonly used methods, IS6110 restriction fragment length polymorphism (RFLP) and spoligotyping. The discriminatory power of IS6110-Mtb1/Mtb2 method was only slightly lower than that of IS6110 RFLP, whereas MIRU-VNTR typing was the least discriminative among the 4 methods used. Clustering of strains by using results of IS6110-Mtb1/Mtb2 PCR correlated well with RFLP-defined clusters, further confirming epidemiologic relationships among patients. These results indicate that the novel genotyping method could be an attractive alternative for other PCR-based typing procedures, such as spoligotyping and MIRU-VNTR typing. Also, it seems to be a valuable adjunct to the reference IS6110 RFLP method for studying the genetic diversity of drugresistant M. tuberculosis strains in Poland. D 2006 Elsevier Inc. All rights reserved. Keywords: Genetic markers; Genotyping; Drug resistance; Mycobacterium tuberculosis; Poland
1. Introduction Tuberculosis (TB) remains a major infectious disease and causes high morbidity and mortality worldwide. The situation is made even worse by the emergence of drugresistant strains of Mycobacterium tuberculosis (Dahle et al., 2003). Poland, such as many other countries in Europe, has witnessed a dramatic fall in the incidence of TB during the last century. However, in 2000, the prevalence of primary drug resistance (including multidrug resistance) increased 2-fold in comparison with that of 1997 (6.1% and 1.2%, respectively) (Augustynowicz-Kopec et al., 2003). To better
B This work was partly supported by the Fund for Scientific Research of Flanders (Brussels, Belgium, grant G.0471.03 N). A.S. was supported by a NATO Science Fellowship at the Institute of Tropical Medicine, Antwerp, Belgium. 4 Corresponding author. Tel.: +32-3-2476317; fax: +32-3-2476333. E-mail address: [email protected] (F. Portaels).
0732-8893/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.diagmicrobio.2005.12.004
understand the epidemiology of TB, we have developed in recent years a large number of DNA fingerprinting methods based on various genetic markers (Mostro¨m et al., 2002). Because a single genotyping method cannot define all unique isolates, the current studies undertaken require various typing strategies to increase the power of strain differentiation (Cowan et al., 2005; Sun et al., 2004a, 2004b). The most widely used and generally accepted international bgold standard Q method has been IS6110 restriction fragment length polymorphism (RFLP), based on the variability in number of copies and chromosomal locations of the IS6110 insertion element between strains (van Embden et al., 1993). The IS6110 RFLP is the most discriminatory method at the population level (van Embden et al., 2000). Unfortunately, it is laborious and expensive, and requires uniform data processing systems, complicating the comparison of results from different laboratories due to subjectivity in the analysis (Cowan et al., 2005). To overcome the technical difficulties encountered by IS6110
60
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64
Table 1 Genotyping results for 77 drug-resistant M. tuberculosis strains from Poland by using multiple genetic markers IS6110 RFLP pattern (no. of strains) 1 2 3 4 5 6 7 8 9
(2) (2) (2) (2) (2) (2) (7) (2) (8)
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
(2)b (2) (3) (2) (2) (4) (5) (2) (2) (7) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1)b (1) (1) (1) a b
Spoligotype
MIRU-VNTR
Pattern no. in IS6110-Mtb1/Mtb2 PCR using primers
STa
Octal
No.
Pattern
Mtb1-IS1-IS2
Mtb2-IS1-IS2
53 90 1555 52 42 53 1558 462 1051
777777777760771 677777776000371 700001403760371 777777777760731 777777607760771 777777777760771 777777775660771 777777777560771 777777437760771
1 53 237 463 1559 264 891 2 47 1557 54 Orphan 280 180 280 237 1557 1564 39 50 52 Orphan 1746 1 53 46 1558
000000000003771 777777777760771 777777777700000 777777777720571 777777400000171 777740003760771 777777607660771 000000004020771 777777774020771 777677774020771 777777777763771 777777776000740 770000777760771 677777777720771 770000777760771 777777777700000 777677774020771 777776737760571 777777347760471 777777777720771 777777777760731 777777776760571 476377737760771 000000000003771 777777777760771 777777770000000 777777775660771
18 13 10 13 11 6 6 9 10 16 19 20 14 21 4 15 2 7 1 24 11 25 3 26 3 23 24 30 22 29 12 8 28 17 7 27 5
223326133323 223325153325 223126153324 223325153325 223226143321 223115153324 223115153324 223125153324 223126153324 223325163533 223326133523 224125153324 223325153533 224325113322 222226151323 223325154322 124325152222 223125143324 124225163224 225323153225 223226143321 225323153323 222125153324 225325153223 222125153324 224325153322 225323153225 23V4315143311 224325143424 235325153323 223325153324 223125153224 233325153225 223325173533 223125143324 225325153323 222325153224
1 2 3 2 4 5 6 7 3 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 21 23 24 25 26 27 28 29 30 31 32 33 34
1 2 3 2 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 10 20 21 22 21 23 24 25 26 27 28 29 30 31 32 33 34
ST = shared type in spoligotyping database SpolDB3 (www.pasteur-guadeloupe.fr/tb/spoldb3). Beijing family strains.
RFLP analysis, we have developed easier PCR-based methods, such as spoligotyping (Kamerbeek et al., 1997) and mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) typing (Mazars et al., 2001). Spoligotyping, based on polymorphism in the chromosomal direct repeat locus, has been the most commonly used PCR-based typing procedure. This method is fast and reproducible; however, it is less discriminatory than the IS6110 RFLP (Kremer et al., 2005; Sun et al., 2004a, 2004b). A new method, MIRU-VNTR typing, is based on the determination of the number of copies of repeated units at 12 independent loci scattered throughout the genome. It allows rapid analysis of the results (numbers of repeats) as digital patterns, with subsequent comparison with worldwide databases. The high resolution and the
possibility of high-throughput analysis make MIRU-VNTR an attractive method for analysis of the global genetic diversity of M. tuberculosis strains (Blackwood et al., 2004; Supply et al., 2000, 2001). Recently, we developed a genotypic method based on PCR amplification of DNA regions between IS6110 and frequently repeated 16-bp GCrich sequences Mtb1 or Mtb2. Preliminary results indicate the usefulness of our method, designated IS6110-Mtb1/ Mtb2 PCR, in differentiation of M. tuberculosis strains for epidemiologic purposes (Kotlowski et al., 2004). In the present study, we applied IS6110-Mtb1/Mtb2 genotyping method and MIRU-VNTR analysis to further characterize drug-resistant M. tuberculosis strains from Poland, previously typed by the reference IS6110 RFLP and spoligotyping (Sajduda et al., 2004). We evaluated a
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64 Table 2 Comparison of the discriminatory power of genotyping methods Set and methoda
No. of distinct patterns
No. of unique isolates
No. of clusters (ranges)
No. of clustered isolates (%)
HGDI
Set 1 IS6110 RFLP Spoligotyping MIRU-VNTR Mtb1-IS1-IS2 Mtb2-IS1-IS2
36 27 30 34 34
17 9 14 17 15
19 18 16 17 19
(2– 8) (2– 8) (2– 10) (2–10) (2– 8)
60 68 63 60 62
(78) (88) (82) (78) (81)
0.965 0.951 0.950 0.958 0.963
Set 2 IS6110 RFLP Spoligotyping MIRU-VNTR Mtb1-IS1-IS2 Mtb2-IS1-IS2
19 17 17 18 19
19 17 15 16 17
(2– 8) (2– 8) (2–10) (2–10) (2– 8)
60 60 58 58 58
(100) (100) (97) (97) (97)
0.942 0.936 0.924 0.932 0.941
0 0 2b 2b 2b
61
we prepared bacterial DNA either by boiling of the mycobacterial cells for 10 min or by the internationally standardized protocol (van Embden et al., 1993). 2.2. Genotyping methods
a Set 1 includes all the strains tested (n = 77), whereas set 2 comprises strains clustered by IS6110 RFLP (n = 60). b Beijing family strains.
multistep typing strategy to attain maximum specificity and compared the discriminatory ability of the 4 methods used. 2. Materials and methods 2.1. Bacterial strains and DNA preparation The 72 M. tuberculosis strains (of 251 drug-resistant strains studied) isolated in Poland in 2000 during the second national survey of drug resistance have been previously shown to cluster by the IS6110 RFLP and spoligotyping (Sajduda et al., 2004). Sixty of these clustered strains and, in addition, 17 strains as representative of the 179 strains with unique IS6110 RFLP patterns were included in further analysis in the present study. M. tuberculosis H37Rv reference strain served as a control. To genotype the 77 drug-resistant strains by the use of PCR-based methods,
MIRU-VNTR analyses were performed using primers for the 12 MIRU-VNTR loci as described by Supply et al. (2001). Three different concentrations of MgCl2 were used depending on MIRU-VNTR locus: 1.5 mmol/L for MIRU loci 20, 24, 26, and 27; 2 mmol/L for MIRU loci 2, 4, 10, 16, 31, and 40; and 2.5 mmol/L for MIRU loci 23 and 39. Each MIRU-VNTR locus was individually amplified in a 25-AL reaction volume containing 100 Amol/L of each deoxynucleoside triphosphate, 0.1 Amol/L of each primer, 0.5 U of HotStarTaq DNA polymerase (Qiagen, Hilden, Germany), 1 Q solution, and 2.5 AL of template DNA. The cycling parameters were 95 8C for 15 min to activate the HotStarTaq DNA polymerase, followed by 40 amplification cycles at 94 8C for 1 min, 59 8C for 1 min, and 72 8C for 1 min 30 s, with a final extension at 72 8C for 10 min in a PTC 100 thermocycler (MJ Research, Waltham, MA). PCR products were electrophoresed on a 3% NuSieve agarose gel and sized with a 50- and 100-bp ladder (Promega, Leiden, The Netherlands), and/or by using the Agilent 2100 bioanalyzer (Agilent Technologies, Waldbronn, Germany) for more accurate determination of amplicon sizes. Strain typing with IS6110-Mtb1/Mtb2 PCR that uses combination of primers IS1 and IS2, directed at inverted repeats flanking IS6110, and Mtb1 or Mtb2, targeting the repeated GC-rich motif, was performed essentially as we described previously (Kotlowski et al., 2004). In a PCR, 2.5 AL template DNA (purified or crude extract) was amplified in a 25-AL reaction volume. An aliquot of PCR products (1 AL) was electrophoresed and analyzed with the bioanalyzer system.
Table 3 MIRU-VNTR allelic diversity in 77 drug-resistant M. tuberculosis strains from Poland Locus
2 4b 10 16 20 23 24 26 27 31 39 40
No. of strains with the specified MIRU copy number 1
2
3
4
5
6
7
7 0 0 28 10 0 77 2 2 0 1 4
70 74 5 8 67 0 0 0 5 19 71 12
0 2 46 41 0 9 0 3 66 51 5 14
0 0 14 0 0 0 0 8 4 1 0 34
0 0 12 0 0 50 0 60 0 6 0 13
0 0 0 0 0 18 0 3 0 0 0 0
0 0 0 0 0 0 0 1 0 0 0 0
Allelic diversitya
Discriminatory power
0.154 0.038 0.576 0.568 0.216 0.504 0.000 0.369 0.252 0.488 0.134 0.712
Poor Poor Moderate Moderate Poor Moderate Poor Moderate Poor Moderate Poor High
a Allelic diversity (h) at a locus was calculated as follows: h ¼ 1 Rx2i ½n=ðn 1Þ, where x i is the frequency of the ith allele at the locus and n is the number of isolates (Mazars et al., 2001). b In 1 strain, the 53-bp unit at the 3V terminus of locus 4 was absent.
62
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64
2.3. Calculation of discriminatory power The Hunter–Gaston Discriminatory Index (HGDI) (Hunter and Gaston, 1988) was used as a numerical index for the discriminatory power of each typing method. The HGDI was calculated using the following formu la: HGDI ¼ 1 f1=½ N ð N 1ÞgfRð j¼1S Þ nj nj 1 g , where N is the total number of strains in the typing scheme, s is the total number of different patterns, and n j is the number of strains belonging to the jth pattern. The calculation was applied to both the entire study population and a discrete sample set including strains originally clustered by the IS6110 RFLP method.
3. Results and discussion The 77 drug-resistant M. tuberculosis strains were characterized in the present study by using MIRU-VNTR typing and IS6110-Mtb1/Mtb2 method (Kotlowski et al., 2004) for maximum specificity, and by comparison of the discriminatory power of the 2 new genetic markers with the most widely used reference IS6110 RFLP method and spoligotyping. Table 1 summarizes the strains sorted by IS6110 RFLP pattern. Seventeen strains had unique IS6110 RFLP patterns, and 60 strains clustered in 19 groups consisting of 2–8 isolates with identical IS6110 RFLP patterns (Tables 1 and 2). All strains were high copy numbers (contained z 6 copies of IS6110).
Spoligotyping identified 27 distinct patterns in the 77 strains split into 9 unique isolates and 68 isolates clustered in 18 groups containing 2 – 8 isolates having the same spoligotype (Tables 1 and 2). Of the 18 spoligotype clusters, 7 included strains with multiple IS6110 RFLP patterns. Three strains had the Beijing spoligotype (van Soolingen et al., 1995), of which 2 strains shared the same IS6110 RFLP pattern, whereas the third isolate was unique (Table 1). The new MIRU-VNTR method (Mazars et al., 2001) was applied here for the first time as an additional molecular tool to differentiate M. tuberculosis isolates in Poland. The number of repeats for each of the 12 MIRU-VNTR loci (Supply et al., 2000) ranged from 1 to 7 copies. One strain contained three 77-bp repeats without the 3V 53-bp unit in locus 4, resulting in aberrant copy number (Table 1, pattern 30) characteristic for M. tuberculosis strains H37Rv and H37Ra, Mycobacterium bovis BCG, and less than 1% of the recent clinical isolates (Cowan et al., 2002; Supply et al., 2000). Table 3 shows allelic diversities of the 12 MIRUVNTR loci in our study set. According to the classification proposed by Sola et al. (2003), only locus 40 provided a high discrimination level ( z 0.6), whereas loci 10, 16, 23, 26, and 31 were moderately discriminative (z 0.3), and loci 2, 4, 20, 24, 27, and 39 were poorly discriminative (b 0.3). In general, the allelic diversity was comparable to some previous studies (Cowan et al., 2002; Supply et al., 2000), but it was lower than in others (Sola et al., 2003; Sun et al.,
Fig. 1. Fingerprinting results for 2 Beijing family strains by various methods: IS6110 RFLP (A), spoligotyping (B), MIRU-VNTR typing (C), and IS6110Mtb1/Mtb2 PCR with primers Mtb1-IS1-IS2 (D) or Mtb2-IS1-IS2 (E). Lanes: 1, M. tuberculosis DS1520; 2, M. tuberculosis PD629; H, M. tuberculosis H37Rv; M, MIRU locus; L, DNA ladder. Differences in MIRU copy number in loci 23, 26, and 39 are underlined. The 15- and 1500-bp bands in all lanes in panels D and E represent lower and upper internal markers.
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64
2004a, 2004b), most likely due to the lower diversity and smaller number of isolates examined here. The MIRU-VNTR typing revealed 30 distinct patterns among the 77 strains tested (Tables 1 and 2). Fourteen strains were unique, and 63 strains were grouped in 16 clusters including 2–10 isolates. Of the 16 MIRU clusters, only 9 (56%) contained strains with identical IS6110 RFLP patterns and spoligotypes. The remaining 7 MIRU clusters could be subdivided by spoligotyping and/ or IS6110 RFLP. On the other hand, 2 members of 1 RFLP cluster found in the Beijing family were further differentiated by MIRU-VNTR (Fig. 1C), thus, confirming lack of epidemiologic link between those patients based on contact tracing. The third Beijing strain, isolated from a patient of Armenian origin, also had a distinct MIRU-VNTR pattern (Table 1, pattern 17) that was found to be predominant among members of this genotype family in different geographic areas (Cowan et al., 2005; Kam et al., 2005; Mokrousov et al., 2004; Sun et al., 2004a, 2004b). Mokrousov et al. (2004) hypothesize that this type could have resulted from a convergent evolution due to a possible biologic role of some MIRUs or could present a stable conserved combination achieved long ago and unchanged since evolutionary distant time (Mokrousov et al., 2004). Overall, in contrast to other studies (Cowan et al., 2005; Kam et al., 2005; Mokrousov et al., 2004), MIRU-VNTR typing performed better for the Beijing strains than IS6110 RFLP, most likely due to the limited number of isolates tested here. As shown in Tables 1 and 2, the IS6110-Mtb1/Mtb2 PCR (Kotlowski et al., 2004) yielded results similar to those obtained by the reference method. Using either primer sets (Mtb1-IS1-IS2 or Mtb2-IS1-IS2) in a simple PCR, we were able to detect only 3 clusters (of 17 or 19, respectively) that could be subtyped by spoligotyping and/or IS6110 RFLP, thus, proving close epidemiologic relationships between isolates in most of the clusters. Similar to MIRU-VNTR typing, IS6110-Mtb1/Mtb2 method discriminated between 2 Beijing strains clustered by IS6110 RFLP (Fig. 1D and E). However, in that case, the primer combination Mtb1-IS1-IS2 performed slightly better, producing distinct banding pattern for each strain, whereas PCR with primers Mtb2IS1-IS2 subdivided the aforementioned 2 strains but clustered 1 of them with a unique non-Beijing isolate (Table 1, pattern 10). In general, the IS6110-Mtb1/Mtb2 method seems a promising tool in differentiation of the increasingly important Beijing family of M. tuberculosis strains, but it would have to be evaluated on a larger collection of isolates. Clustering results from each method were compared with each other to determine the discriminatory power of the typing methods applied in the present study. As shown in Table 2, IS6110 RFLP was the most discriminative when applied both to the entire population (HGDI = 0.965) and to a set of 60 strains previously clustered by this method (HGDI = 0.942). Interestingly, in both study sets, IS6110-
63
Mtb1/Mtb2 PCR based on primers Mtb2-IS1-IS2 gave resolving power very close to that of the reference method (HGDI = 0.963 and 0.941, respectively). Although the degree of discrimination obtained with each method did not differ significantly, MIRU-VNTR typing was the least discriminative. A low sensitivity of MIRU-VNTR method for detecting IS6110 RFLP-based clusters and its poor specificity for providing distinct patterns among isolates unmatched by IS6110 RFLP have been reported recently by Scott et al. (2005) in IS6110 high–copy-number M. tuberculosis isolates from Montreal. Our results concur with their observations but appear to contrast with previous findings, demonstrating close discriminatory power of the MIRU-VNTR typing to that of the RFLP method (Blackwood et al., 2004; Cowan et al., 2002; Supply et al., 2001). It is most likely due to the sample studied that included smaller number of less heterogeneous, IS6110 high–copy-number M. tuberculosis strains than the previous studies. On the other hand, in a similar study, Hawkey et al. (2003) recently found a very close correlation between the RFLP and MIRU typing of apparently clustered cases of TB. However, their study sample included 53 M. tuberculosis isolates grouped in only 4 clusters. It can be expected that with increasing number of different RFLP clusters, probability of clustering epidemiologically nonrelated strains by the MIRU-VNTR method would be also higher. In this study, the PCR products were analyzed on NuSieve agarose gels for better resolution, and they were also labeled with fluorescent dye for increased throughput automated analysis on a microfluidic labchip instrument, the Agilent 2100 bioanalyzer. This new technology offers several important advantages over the traditional agarose electrophoresis that could be especially useful in MIRUVNTR analyses. It is fast (12 samples on each disposable labchip are analyzed within 30 min), easy (a minimum of technical skill is required), accurate, reproducible, and it generates low waste. Also, it is cost-effective (the volumes of reagents may be reduced because only 1-AL sample is evaluated) and versatile (it is complemented with kits for dsDNA, RNA, protein, and flow cytometric analyses). Despite the high cost of equipment, which might limit its usefulness in developing countries and/or areas with high prevalence of TB, the bioanalyzer system appears a suitable format for performance of PCR-based typing analyses. Although various typing strategies have been proposed depending on the population structure of M. tuberculosis, the present tendency is that MIRU-VNTR typing will likely predominate as a prescreening method regardless of the population structure. However, there is still a need for alternative simple, fast, and reliable genotyping procedures. Based on our previous and present results, the IS6110Mtb1/Mtb2 method seems a valuable adjunct to the reference IS6110 RFLP typing. For us to further confirm its utility in typing M. tuberculosis, its evaluation on a larger
64
A. Sajduda et al. / Diagnostic Microbiology and Infectious Disease 55 (2006) 59 – 64
number of strains from various geographic areas would, however, be necessary.
Acknowledgments We acknowledge the Department of Microbiology, National Research Institute of Tuberculosis and Lung Diseases, Warsaw, Poland, for providing M. tuberculosis strains. References Augustynowicz-Kopec E, Zwolska Z, Jaworski A, Kostrzewa E, Klatt M (2003) Drug-resistant tuberculosis in Poland in 2000: Second national survey and comparison with the 1997 survey. Int J Tuberc Lung Dis 7:645 – 651. Blackwood KS, Wolfe JN, Kabani AM (2004) Application of mycobacterial interspersed repetitive unit typing to Manitoba tuberculosis cases: Can restriction fragment length polymorphism be forgotten? J Clin Microbiol 42:5001 – 5006. Cowan LS, Mosher L, Diem L, Massey JP, Crawford JT (2002) Variablenumber tandem repeat typing of Mycobacterium tuberculosis isolates with low copy numbers of IS6110 by using mycobacterial interspersed repetitive units. J Clin Microbiol 40:1592 – 1602. Cowan LS, Diem L, Monson T, Wand P, Temporado D, Oemig TV, Crawford JT (2005) Evaluation of a two-step approach for large-scale, prospective genotyping of Mycobacterium tuberculosis isolates in the United States. J Clin Microbiol 43:688 – 695. Dahle UR, Sandven P, Heldal E, Mannsaaker T, Caugant DA (2003) Deciphering an outbreak of drug-resistant Mycobacterium tuberculosis. J Clin Microbiol 41:67 – 72. Hawkey PM, Smith EG, Evans JT, Monk P, Bryan G, Mohamed HH, Bardhan M, Pugh RN (2003) Mycobacterial interspersed repetitive unit typing of Mycobacterium tuberculosis compared to IS6110-based restriction fragment length polymorphism analysis for investigation of apparently clustered cases of tuberculosis. J Clin Microbiol 41: 3514 – 3520. Hunter PR, Gaston MA (1988) Numerical index of the discriminatory ability of typing systems: An application of Simpson’s index of diversity. J Clin Microbiol 26:2465 – 2466. Kam KM, Yip CW, Tse LW, Wong KL, Lam TK, Kremer K, Au BKY, van Soolingen D (2005) Utility of mycobacterial interspersed repetitive unit typing for differentiating multidrug-resistant Mycobacterium tuberculosis isolates of the Beijing family. J Clin Microbiol 43:306 – 313. Kamerbeek J, Schouls L, Kolk A, van Agterveld M, van Soolingen D, Kuijper S, Bunschoten A, Molhuizen H, Shaw R, Goyal M, van Embden JDA (1997) Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin Microbiol 35:907 – 914. Kotyowski R, Shamputa IC, El Aila NA, Sajduda A, Rigouts L, van Deun A, Portaels F (2004) PCR-based genotyping of Mycobacterium tuberculosis with new GC-rich repeated sequences and IS6110 inverted repeats used as primers. J Clin Microbiol 42:372 – 377. Kremer K, Au BKY, Yip PCW, Skuce R, Supply P, Kam KM, van Soolingen D (2005) Use of variable-number tandem-repeat typing to differentiate Mycobacterium tuberculosis Beijing family isolates
form Hong Kong and comparison with IS6110 restriction fragment length polymorphism typing and spoligotyping. J Clin Microbiol 43: 314 – 320. Mazars E, Lesjean S, Banuls A-L, Gilbert M, Vincent V, Gicquel B, Tibayrenc M, Locht C, Supply P (2001) High-resolution minisatellitebased typing as a portable approach to global analysis of Mycobacterium tuberculosis molecular epidemiology. Proc Natl Acad Sci U S A 98:1901 – 1906. Mokrousov B, Narvskaya O, Limeschenko E, Vyazovaya A, Otten T, Vyshnevskiy B (2004) Analysis of the allelic diversity of the mycobacterial interspersed repetitive units in Mycobacterium tuberculosis strains of the Beijing family: Practical implications and evolutionary considerations. J Clin Microbiol 42:2438 – 2444. Mostrfm P, Gordon M, Sola C, Ridell M, Rastogi N (2002) Methods used in the molecular epidemiology of tuberculosis. Clin Microbiol Infect 8:694 – 704. Sajduda A, Brzostek A, Popyawska M, Rastogi N, Sola C, AugustynowiczKopec´ E, Zwolska Z, Dziadek J, Portaels F (2004) Molecular epidemiology of drug-resistant Mycobacterium tuberculosis strains isolated from patients with pulmonary tuberculosis in Poland: A 1year study. Int J Tuberc Lung Dis 8:1448 – 1457. Scott AN, Menzies D, Tannenbaum T-N, Thibert L, Kozak R, Joseph L, Schwartzman K, Behr MA (2005) Sensitivities and specificities of spoligotyping and mycobacterial interspersed repetitive unit-variablenumber tandem repeat typing methods for studying molecular epidemiology of tuberculosis. J Clin Microbiol 43:89 – 94. Sola C, Filliol I, Legrand E, Lesjean S, Locht C, Supply P, Rastogi N (2003) Genotyping of the Mycobacterium tuberculosis complex using MIRUs: Association with VNTR and spoligotyping for molecular epidemiology and evolutionary genetics. Infect Genet Evol 3:125 – 133. Sun Y-J, Bellamy R, Lee ASG, Ng ST, Ravindran S, Wong S-Y, Locht C, Supply P, Paton NI (2004a) Use of mycobacterial interspersed repetitive unit-variable-number tandem repeat typing to examine genetic diversity of Mycobacterium tuberculosis in Singapore. J Clin Microbiol 42: 1986 – 1993. Sun Y-J, Lee ASG, Ng ST, Ravindran S, Kremer K, Bellamy R, Wong S-Y, van Soolingen D, Supply P, Paton NI (2004b) Characterization of ancestral Mycobacterium tuberculosis by multiple genetic markers and proposal of genotyping strategy. J Clin Microbiol 42:5058 – 5064. Supply P, Mazars E, Lesjean S, Vincent V, Gicquel B, Locht C (2000) Variable human minisatellite-like regions in the Mycobacterium tuberculosis genome. Mol Microbiol 36:762 – 771. Supply P, Lesjean S, Savine E, Kremer K, van Soolingen D, Locht C (2001) Automated high-throughput genotyping for study of global epidemiology of Mycobacterium tuberculosis based on mycobacterial interspersed repetitive units. J Clin Microbiol 39:3563 – 3571. van Embden JD, Cave MD, Crawford JT, Dale JW, Eisenach KD, Gicquel B, Hermans P, Martin C, McAdam R, Shinnick TM (1993) Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: Recommendations for a standardized methodology. J Clin Microbiol 31:406 – 409. van Embden JD, van Gorkom T, Kremer K, Jansen R, van Der Zeijst BA, Schouls LM (2000) Genetic variation and evolutionary origin of the direct repeat locus of Mycobacterium tuberculosis complex bacteria. J Bacteriol 182:2393 – 2401. van Soolingen D, Qian L, de Haas PEW, Douglas JT, Traore H, Portaels F, Qing HZ, Enkhasaikan D, Nymadawa P, van Embden JDA (1995) Predominance of a single genotype of Mycobacterium tuberculosis in countries of East Asia. J Clin Microbiol 33:3234 – 3238.