Multiple-Locus Variable-Number Tandem-Repeat Analysis of. Acinetobacter baumannii in China: comparison with PFGE typing.

Transcription

1 JCM Accepts, published online ahead of print on 23 January 2013 J. Clin. Microbiol. doi: /jcm Copyright 2013, American Society for Microbiology. All Rights Reserved Multiple-Locus Variable-Number Tandem-Repeat Analysis of Acinetobacter baumannii in China: comparison with PFGE typing. Yuan Hu 1, Boqing Li 2, Dazhi Jin 3, Zhigang Cui 1, Xiaoxia Tao 1, Binghua Zhang 1, Jianzhong Zhang 1* State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention, 2 Binzhou Medical College, Yantai, china, 3 Zhejiang Provincial Center for Disease Control and Prevention, Hangzhou, China. *Correspondence auther: Professor Jian-Zhong Zhang, State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention, 155 Changbai Road Changping District, Beijing, China Phone: zhangjianzhong@icdc.cn Abstract A panel of seven variable-number tandem repeats (VNTRs) markers was selected for Acinetobacter baumannii typing analysis (MLVA-7). Compared with pulsed-field gel electrophoresis (PFGE), MLVA-7 provided greater discrimination. We modified the criteria for MLVA complexes assignment proposed previously, and a remarkable congruence between MLVA-7 and PFGE-based strain clustering was observed. Acinetobacter baumannii is now recognized as a major pathogen involved in nosocomial infections with a documented high mortality rate (1, 2, 3). Major epidemiologic features of this organism include the propensity for clonal spread, long-term persistence in the hospital setting, involvement in hospital outbreaks, as well as resistance to multiple antimicrobial agents (1). The number of

2 multidrug-resistant outbreaks is currently increasing worldwide. Many of the genotypes involved have been found to belong to international clones (lineages) I, II or III (4, 5, 6). Several molecular typing methods have been developed to investigate the molecular epidemiology of A. baumannii infection. Among them, macrorestriction analysis with pulsed-field gel electrophoresis (PFGE) has been used extensively and it can be considered, in some instances, the gold standard for A. baumannii typing (7, 8). However, this method is laborious, time consuming and cannot easily be compared between laboratories. By contrast, multiple-locus variable-number tandem repeat (VNTR) analysis (MLVA), based on the copy numbers of tandem repeat units (TRs) of VNTR loci, is a highly reproducible and portable typing method and can be easily shared. So far, 10 VNTRs that provided sufficient repeat polymorphism for fine typing of clinical isolates have been described in A. baumannii, and three different selections of loci have been evaluated, including a 2 loci assay (9), an MLVA-8 scheme (10), and a partially automated MLVA-10 Orsay scheme (11). A. baumannii isolates. 63 A. baumannii strains, non-outbreak related, isolated from patients in 3 Chinese cities (one hospital per city, one isolates per patient, except for 3 isolates): Haining (n=38), Yantai (n=18) and Wenzhou (n=7), and 59 isolates (54 patients isolates, one isolates per patient, plus three environmental samples and two healthcare workers hands isolates) isolated during an A. baumannii outbreak occurred in one hospital in Beijing were used in the study. All isolates were identified as A. baumannii as described previously (12), and also by the presence of a bla OXA-51-like gene (13). International clones identification. Identification of international clones was done as described previously (14). 41.2% and 90.7% of the non-outbreak and outbreak isolates were identified as belonging to international clonal lineage II (ICL-II) respectively, as shown in Figure 1. Antimicrobial Susceptibility testing. The antimicrobial susceptibilities of the A. baumannii isolates to 11 antimicrobials were examined: ampicillin, piperacillin/tazobactam, ceftriaxone, ceftazidime/clavulanic acid, ciprofloxacin,

3 rifampicin, polymyxin B, doxycycline, chloramphenicol, imipenem, trimethoprim/sulfamethoxazole. The susceptibility tests were performed using E-test on Mueller-Hinton agar and the results were evaluated using the CLSI breakpoints for microdilution methods (15). A strain resistant to at least three classes of antimicrobial agents - all penicillins and cephalosporins (including inhibitor combinations), fluoroquinolones, and aminoglycosides was called multi-drug resistant (MDR) (16). A MDR strain also resistant to imipenem was called extensively-drug resistant (XDR) (16). All the outbreak associated isolates were multidrug resistant (47 XDR and 7 MDR), by contrast, 67.6% of the non-outbreak associated isolates were multidrug resistant (43 XDR and 3 MDR, out of 68 isolates), as shown in Figure 1. PFGE. PFGE was performed and analyzed as described previously (17). Similarity threshold of 95% and 84% were arbitrarily used to define isolates belonging to the same PFGE type and cluster respectively. MLVA. MLVA typing using the MLVA-8 scheme was performed as described previously (10), except that the primers were fluorescently labeled and PCR products were sized on an automated DNA sequencer to accurately determine the number of tandem repeats. The polymorphism index of individual locus or combined VNTR loci was calculated using the Hunter-Gaston diversity index (HGDI) (18). Confidence intervals (CI) were calculated as previously described (19). In order to deduce approximate difference between observed sizes and real sizes, some representative alleles for each VNTR locus was sequenced. The length of repeat, number of repetitions and the deduced size of flanking regions were analyzed using the Tandem Repeats Finder tool ( (20). A total of 4 alleles were obtained at Abaum_1988, but two different repeat units, 78 bp and 59 bp, not 26 bp as previously reported (10), were revealed in Abaum_1988 by sequencing, as shown in Figure 2. Furthermore, no further discriminatory power was gained by Abaum_1988 addition, so a MLVA scheme of 7 loci (MLVA-7) was used in this study, Abaum_1988 was not included. A high proportion (40%, 18/45) of Abaum_0826 amplification failure was revealed in the samples that do not belong to the international clones (non-ic) (Figure 1).

4 Abaum_0826 is entirely located within coding regions of pilus assembly protein (FilE) (10). To determine whether the absence of Abaum_0826 amplification was due to mismatches in the target region of the primers or due to complete deletion of this locus, two primers, FilE-F: CAGTTAAATCCTTCTGTTCAGC and FilE-R: AACATCCACGGCAGAAGA, located in the conserved region of file were designed. With the FilE primers, no PCR products were obtained for all the Abaum_0826 amplification failure cases, while PCR products of the expected length were obtained for the Abaum_0826 PCR positive isolates, suggesting complete deletion of file was the cause for the Abaum_0826 PCR failure. In addition, significant association between drug sensitivity and Abaum_0826 amplification failure was revealed (χ 2 test, p < 0.001). Another recent report obtained similar result (11). So Abaum_0826 PCR failure may be a useful marker to reflect the non-ic identification or sensitive phenotype, thus were considered for the HGDI calculation and clustering analysis. Remarkable differences of flanking region length compared with previous reports were found: 223bp instead of 208bp in Abaum_0826, and 177bp instead of 147bp in Abaum_3468, which resulted in two and five copy deviations, respectively. The 223bp flanking length of Abaum_0826 was mentioned by Pourcel et al in the table S2 (10). Non-outbreak strains analysis. The 68 non-outbreak associated isolates were discriminated into 47 types by MLVA-7 (Figure 3) and into 37 types by PFGE (Figure 1). For MLVA-7, the HGDI value ranged from to per locus with a combined HGDI of % (CI ), as shown in Table 1. The HGDI value for PFGE was 0.95 (95% CI ), lower than MLVA-7. Outbreak strains analysis. A total of 59 strains were isolated not more than 2 months apart during an outbreak period. 54 isolates were found to group together both by PFGE and MLVA-7 (Figure 1 and 3), so were categorized as outbreak associated strains. The remaining 5 isolates, with distinct PFGE profiles, were categorized as non-outbreak strains and were included in the analysis above. Two main PFGE types for the outbreak strains, T09 and T01, including 32 and 13 isolates respectively (Figure 1), were suspected to represent two lineages caused this outbreak. But the T01

5 lineage was refuted by MLVA, for MLVA-7 discriminated the 13 T01 isolates into 10 different types. By contrast, 22 of the 32 T09 isolates were identical in MLVA-7 (type A-1) too (Figure 1), which confirmed this lineage to be responsible for this outbreak. To track the potential transmission, bacterial isolates from medical equipment (n=2), environment sample (n=1) and hands of healthcare workers (n=2) of the outbreak original ward BJ-ICU2 were tested, which were marked in the Figure 1. The hand isolates were found to share identical MLVA and PFGE profiles with the outbreak lineage. Compared with the outbreak lineage (PFGE T09 and MLVA-7 A-1), 6 of the remaining 10 T09 isolates were differed at only one locus (5 at Abaum_0845, one at Abaum_2396, Figure 1). To confirm the S-repeat VNTR loci TRs capable altered in a relative small number of generations, the stability of the 7 loci was examined by 30 in-vitro passages with 13 isolates in this study. Before the 10 th passage, no change in the MLVA profiles was observed. Subsequently, an increase or decrease of one TRs copy number at locus Abaum_0845 occurred in 6 isolates, and a one TRs copy number increase at locus Abaum_0826 occurred in one isolate. No TRs copy number changes were observed in the remaining 5 loci. The tandem repeat region of Abaum_0845 is located downstream of the coding gene of adenosine deaminase (10), which is involved in purine metabolism. The reason for the rapid TRs alteration at this locus is unknown, but it should be noted in the data analysis. Comparison of MLVA-7 with PFGE. All the 122 isolates were resolved into 64 MLVA-7 types and 43 PFGE types, respectively. The dendrogram generated by PFGE analysis was then matched with the results of ICL identification, antimicrobial susceptibilitiy, and MLVA-7 for direct comparison (Figure 1). The discriminatory power was higher with MLVA-7 (0.948; 95% CI ) than with PFGE (0.9; 95% CI ). Taking into account the intrinsic variability of VNTRs, Pourcel et al. proposed an criteria for the assignment of strains to MLVA complexes: (i) identical L-repeat VNTR profiles and (ii) a maximum of three differences of no more than two repeats at S-repeat VNTR loci (10). It is obviously not suitable to determine the clonal

6 relationships of some isolates used in this study, e.g. HN049 and HN051, which shared same PFGE profile, but differed at Abaum_0826 as null and 16 respectively (Figure 1). By analysis the TRs difference between closest MLVA types, with reference to the corresponding PFGE profiles, we modified the criteria: (i) identical L-repeat VNTR profiles and (ii) compared between the closest genotypes, a maximum of two differences of no more than three repeats (excepting Abaum_0826) at S-repeat VNTR loci; and the maximum of three repeats differences occurs at only one locus. TRs differences at S-repeat loci between the closest types in MLVA-7 complex A were showed in Table 2. To evaluate congruence between the typing methods, the Wallace coefficient was calculated (21). Wallace index between MLVA-7 and PFGE was 67.7%, revealing that the chance of isolates with same MLVA profile typed as identical by PFGE was 67.7%, and the Wallace index between PFGE and MLVA-7 was 37.6%. Although concordance between the two methods was not high, the clustering results of them were highly congruent when MLVA results were interpreted according to our modified criteria. MLVA-7 clustered the 122 isolates into two complexes, as shown in Figure 3. All MLVA-7 complex A strains excepting isolate BJ31 typed into PFGE cluster A, and MLVA-7 complex B strains typed into PFGE cluster B too, as shown in Figure 1. A high proportion (82%) of A. baumannii isolates used in this study were multidrug resistant, ICL-II strains accounted for 67% of them, which was agreed with the previously reports that multidrug resistance is often associated with isolates that belong to the international clones (22, 23, 24, 25). Almost all the multidrug resistant strains (96%), even isolated from distant cities, were clustered together by both MLVA-7 and PFGE, whereas the sensitive strains were more heterogeneous (Figure 1). This indicates that antimicrobial resistance may be a major selective advantage that drives the rapid and wide clonal expansion of these strains. Conclusion. In the present investigation, a MLVA scheme of 7 loci was proposed (MLVA-7), which provides sufficiently and greater discriminatory power compared with PFGE, both for the non-outbreak isolates typing and outbreak tracing. And a

7 remarkable congruence between MLVA-7 and PFGE-based strain clustering was observed according to the MLVA complexes assigning criteria we modified in this study. ACKNOWLEDGMENTS This study was supported by the"major infectious diseases such as AIDS and viral hepatitis prevention and control" technology major projects (No. 2008ZX and No. 2011ZX ). REFERENCES 1. Bergogne-Berezin E, Towner KJ Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidemiological features. Clin. Microbiol. Rev. 9: Cisneros JM, Reyes MJ, Pachón J, Becerril B, Caballero FJ, Garcia-Garmendia JL, Ortiz C, Cobacho AR Bacteremia due to Acinetobacter baumannii: epidemiology, clinical findings, and prognostic features. Clin. Infect. Dis. 22: Seifert H, Strate A, Pulverer G Nosocomial bacteremia due to Acinetobacter baumannii. Clinical features, epidemiology, and predictors of mortality. Medicine (Baltimore). 74: Dijkshoorn L, Aucken H, Gerner-Smidt P, Janssen P, Kaufmann ME, Garaizar J, Ursing J, Pitt TL Comparison of outbreak and nonoutbreak Acinetobacter baumannii strains by genotypic and phenotypic methods. J. Clin. Microbiol. 34: van DH, Dijkshoorn L, van der Reijden T, Bakker N, Paauw A, van den Broek P, Verhoef J, Brisse S Identification of a new geographically widespread multiresistant Acinetobacter baumannii clone from European hospitals. Res. Microbiol. 155: Peleg AY, Seifert H, Paterson DL Acinetobacter baumannii: emergence of a successful pathogen. Clin. Microbiol. Rev. 21: Seifert H, Gerner-Smidt P Comparison of ribotyping and pulsed-field gel electrophoresis for molecular typing of Acinetobacter isolates. J. Clin. Microbiol. 33: Wisplinghoff H, Seifert H Molecular Epidemiology of Acinetobacter Species. Acinetobacter Biology and Pathogenesis. pp In Bergogne-Bérézin E, Friedman H, Bendinelli M, Acinetobacter Biology and Pathogenesis, Springer US. 9. Turton JF, Matos J, Kaufmann ME, Pitt TL Variable number tandem repeat loci providing discrimination within widespread genotypes of Acinetobacter baumannii. Eur. J. Clin. Microbiol. Infect. Dis. 5:

8 Pourcel C, Minandri F, Hauck Y, D'Arezzo S, Imperi F, Vergnaud G, Visca P Identification of variable-number tandem-repeat (VNTR) sequences in Acinetobacter baumannii and interlaboratory validation of an optimized multiple-locus VNTR analysis typing scheme. J. Clin. Microbiol. 49: Hauck Y, Soler C, Jault P, Mérens A, Gérome P, Nab CM, Trueba F, Bargues L, Thien HV, Vergnaud G, Pourcel C Diversity of Acinetobacter baumannii in Four French Military Hospitals, as Assessed by Multiple Locus Variable Number of Tandem Repeats Analysis. PLoS One. 7:e Higgins PG, Wisplinghoff H, Krut O, Seifert H A PCR-based method to differentiate between Acinetobacter baumannii and Acinetobacter genomic species 13TU. Clin. Microbiol. Infect. 13: Turton JF, Woodford N, Glover J, Yarde S, Kaufmann ME, Pitt TL Identification of Acinetobacter baumannii by detection of the blaoxa-51-like carbapenemase gene intrinsic to this species. J. Clin. Microbiol. 44: Turton JF, Gabriel SN, Valderrey C, Kaufmann ME, Pitt TL Use of sequence-based typing and multiplex PCR to identify clonal lineages of outbreak strains of Acinetobacter baumannii. Clin. Microbiol. Infect. 13: Clinical and Laboratory Standards Institute Performance Standards for Antimicrobial Susceptibility Testing: Twentieth Informational Supplement, M100-S21. Clinical and Laboratory Standards Institute, Wayne, PA. 16. Manchanda V, Sanchaita S, Singh N Multidrug resistant Acinetobacter. J. Glob. Infect. Dis. 2: Seifert H, Dolzani L, Bressan R, van der Reijden T, van SB, Stefanik D, Heersma H, Dijkshoorn L Standardization and interlaboratory reproducibility assessment of pulsed-field gel electrophoresis-generated fingerprints of Acinetobacter baumannii. J. Clin. Microbiol. 43: Hunter PR, Gaston MA Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26: Grundmann H, Hori S, Tanner G Determining confidence intervals when measuring genetic diversity and the discriminatory abilities of typing methods for microorganisms. J. Clin. Microbiol. 39: Benson G Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res. 27: Carrico JA, Silva-Costa C, Melo-Cristino J, Pinto FR, de LH, Almeida JS, Ramirez M Illustration of a common framework for relating multiple typing methods by application to macrolide-resistant Streptococcus pyogenes. J. Clin. Microbiol. 44: Hujer KM, Hujer AM, Hulten EA, Bajaksouzian S, Adams JM, Donskey CJ, Ecker DJ, Massire C, Eshoo MW, Sampath R, Thomson JM, Rather PN, Craft DW, Fishbain JT, Ewell AJ, Jacobs MR, Paterson DL, Bonomo RA Analysis of antibiotic resistance genes in multidrug-resistant Acinetobacter sp. isolates from military and civilian patients treated at the Walter Reed Army

9 Medical Center. Antimicrob. Agents. Chemother. 50: Nemec A, Dijkshoorn L, van der Reijden TJ Long-term predominance of two pan-european clones among multi-resistant Acinetobacter baumannii strains in the Czech Republic. J. Med. Microbiol. 53: Carretto E, Barbarini D, Dijkshoorn L, van der Reijden TJ, Brisse S, Passet V, Farina C; APSI Acinetobacter Study Group Widespread carbapenem resistant Acinetobacter baumannii clones in Italian hospitals revealed by a multicenter study Infection. Infect Genet Evol. 11: Nemec A, Krízová L, Maixnerová M, Diancourt L, van der Reijden TJ, Brisse S, van den Broek P, Dijkshoorn L Emergence of carbapenem resistance in Acinetobacter baumannii in the Czech Republic is associated with the spread of multidrug-resistant strains of European clone II. J Antimicrob Chemother. 62: Figure 1. Typing concordance between MLVA-7 and PFGE typing. Cluster analysis of 122 A. baumannii strains analyzed by PFGE is shown on the left. The relating results of ICL identification, antimicrobial susceptibility and MLVA-7 allelic profiles are provided for direct comparison. Isolates isolated during the outbreak were marked by BJ in the first two letters of the strain name. The medical equipment isolates, environment sample isolate, healthcare worker s hands isolates and isolates of a single patient were marked by,, and bracket respectively. Isolates belong to MLVA-7 complexes of Figure 3 were marked by A and B respectively in the first letter of the MLVA type names. Isolates clustered out of the MLVA-7 complexes were marked by O. Abbreviation: ICL, international clonal lineage; non-ic, not assigned to the international clones. Figure 2. Alignment of four alleles of VNTR locus Abaum_1988. Sequence analysis and the related proportions of 4 alleles in the 122 isolates are marked at the end of the sequence. Different repeat units of 78 bp or 59 bp were observed. Figure 3. Minimum spanning tree representation of the MLVA-7 clustering of the 122 isolates. Each circle represents a MLVA-7 type (numbers in the circles), and the size is proportional to the number of isolates. Two complexes (A, B) are designated and marked by different colors. A distance of one locus between two MLVA-7 types is indicated by a thick line, a distance of two loci by a thin line, and a distance of three or greater loci by a broken line. The colored pie chart sections inside circles indicate the location of the isolates, the antimicrobial susceptibility of which is indicated besides the corresponding circle.

10 Table 1. HGDI for individual locus or combined VNTR loci in non-outbreak strains Locus Repeat size allele size ranges amplification HGDI CI c K d max(pi) e (bp) (bp [no. of repeats]) failure (%) Abaum_0826 a 9 233(1)~ 406(20) % 18 (26.5%) Abaum_ (2)~315(30) % 2 (2.9%) Abaum_ (8)~267(27) % Abaum_3468 b 6 194(3)~ 267(15) % Abaum_ (4)~ 522(7) % Abaum_ (7)~ 539(8) % 1 (1.5%) Abaum_ (2)~ 826(4) % MLVA %. a 223 bp flanking length was used in this study. b 177 bp flanking length was used in this study. c Precision of the Diversity Index, expressed as 95% upper and lower boundaries d number of different repeats present at this locus in this sample set e Fraction of the most frequent repeat number in this locus Table 2 Tandem repeat copy number differences at S-repeat loci between the closest types in MLVA-7 complex A TRs difference a From same Neighbor pair * hospital A-34 / A Y A-15 / A N A-1 / A N A-1 / A-25 # Y A-1 / A N A-1 / A N A-3 / A N A-12 / A N A-36 / A-11 # Y A-1 / A N A-36 / A-8 # Y A-3 / A N A-1 / A-22 # Y A-6 / A Y A-4 / A N A-4 / A N A-36 / A-10 # Y A-8 / A-38 # Y A-1 / A N

11 A-18 / A N A-8 / A-40 # Y A-1 / A N A-29 / A Y A-1 / A-23 # Y A-1 / A-12 # Y A-10 / A-9 # Y A-8 / A-39 # Y A-1 / A-24 # Y A-2 / A Y A-18 / A Y A-2 / A N A-3 / A N A-4 / A N A-16 / A-17 # Y A-2 / A Y A-3 / A N A-13 / A Y A-2 / A Y A-34 / A N * determined by MLVA-7 clustering (Minimum spanning tree, Figure 3), relating MLVA-7 profiles were shown in Figure 1. a No change at a given locus was indicated with a - symbol. # All isolates typed into the two MLVA-7 types were outbreak associated. Downloaded from on April 26, 2018 by guest

12

13

14