|
|
FEMS Microbiol Lett. 2007 October; 275(1): 16–23. Published online 2007 August 10. doi: 10.1111/j.1574-6968.2007.00875.x. | PMCID: PMC2040188 |
Copyright © 2007 Michigan State University Journal compilation © 2007 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. Multiple-locus variable-number tandem repeat analysis of Salmonella Enteritidis isolates from human and non-human sources using a single multiplex PCR Seongbeom Cho,1 David J Boxrud,2 Joanne M Bartkus,2 Thomas S Whittam,1 and Mahdi Saeed1,2 1Department of Large Animal Sciences and Epidemiology, and National Food Safety and Toxicology Center, Michigan State University, East Lansing, MI, USA 2Department of Large Animal Sciences and Epidemiology, and National Food Safety and Toxicology Center, Minnesota Department of Health, St Paul, MN, USA Section Editor: Stefan Schwarz Received April 4, 2007; Revised June 5, 2007; Accepted June 8, 2007. Simplified multiple-locus variable-number tandem repeat analysis (MLVA) was developed using one-shot multiplex PCR for seven variable-number tandem repeats (VNTR) markers with high diversity capacity. MLVA, phage typing, and PFGE methods were applied on 34 diverse Salmonella Enteritidis isolates from human and non-human sources. MLVA detected allelic variations that helped to classify the S. Enteritidis isolates into more evenly distributed subtypes than other methods. MLVA-based S. Enteritidis clonal groups were largely associated with sources of the isolates. Nei's diversity indices for polymorphism ranged from 0.25 to 0.70 for seven VNTR loci markers. Based on Simpson's and Shannon's diversity indices, MLVA had a higher discriminatory power than pulsed field gel electrophoresis (PFGE), phage typing, or multilocus enzyme electrophoresis. Therefore, MLVA may be used along with PFGE to enhance the effectiveness of the molecular epidemiologic investigation of S. Enteritidis infections. Keywords: Salmonella Enteritidis, MLVA, VNTR, PFGE, molecular epidemiology Salmonella infections are estimated to cause 1.4 million cases and more than 500 deaths a year in the United States ( Mead et al., 1999) and economic losses caused by salmonellosis are projected to be from $0.5 to $2.3 billion per year. Currently, Salmonella enterica serovar Enteritidis is the most common Salmonella serotype encountered in foodborne salmonellosis in Europe and is one of the most common Salmonella serotypes recovered from human infections in North America. Phage typing is the most common method used for subtyping of S. Enteritidis but is available only at a small number of laboratories. However, phage types may be unstable and several S. Enteritidis strains do not conform to any of the specific phage types in current use ( Rankin & Platt, 1995). Furthermore, pulsed-field gel electrophoresis (PFGE), which is currently the gold-standard technique for subtyping bacterial isolates associated with common foodborne infections, is laborious, requires precise standardization, and is also of limited effectiveness in subtyping S. Enteritidis isolates ( Hudson et al., 2001; Liebana et al., 2001; Swaminathan et al., 2001; Lukinmaa et al., 2004; Hyytia-Trees et al., 2006). The changing epidemiology of S. Enteritidis infections requires typing methods of high discriminatory power, that are reproducible, stable, easy to interpret, and compare between laboratories. Recently, DNA sequence-based methods such as multilocus sequence typing (MLST) have been commonly used in subtyping of bacteria with high conserved sequences, as genome sequence data are available at GenBank ( Lukinmaa et al., 2004; Maiden, 2006). However, MLST is laborious and time-consuming and had limited ability to resolve genetic diversity between S. enterica serovars compared with PFGE ( Fakhr et al., 2005; Torpdahl et al., 2005). Recently, multiple-locus variable-number tandem repeat analysis (MLVA) has been proposed as an alternative to PFGE for subtyping of a number of pathogenic bacteria, especially highly clonal groups of microorganisms including Francisella tularensis, Bacillus anthracis, Escherichia coli O157 : H7, Mycobacterium paratuberculosis, and Coxiella burnetii ( Keim et al., 2000; Farlow et al., 2001; Klevytska et al., 2001; Lindstedt et al., 2003b, 2004a, b; Overduin et al., 2004; Ramisse et al., 2004; Svraka et al., 2006). MLVA is based on the detection of short sequence repeats that vary in copy number (i.e. variable number of tandem repeats or VNTR) in the microbial genome at various regions. The VNTR are often highly polymorphic with variation in both the number of repeat units and by sequence heterogeneity among individual units ( van Belkum, 1999). MLVA typing of S. Enteritidis has been described previously ( Boxrud et al., 2007). The objectives of this report were to describe the optimization of this MLVA to enable performing a one-shot multiplex PCR for seven VNTR markers with high diversity and to compare its discriminatory power with PFGE and phage typing on a diverse group of S. Enteritidis isolates. Bacterial strains Thirty-four S. Enteritidis strains originally isolated from a variety of sources from 1978 to 2004 were used in this study ( Table 1). The strains represent the major phage types identified in the United States and include a variety of PFGE profiles, antimicrobial resistance profiles, and distinct multilocus enzyme electrophoresis (MLEE) types. The strains were selected to represent various host and tissue sources including humans, chickens, eggs, mice, cows, wild animals, and the environment (chicken farms). The S. Enteritidis strains were phage typed at the National Veterinary Service and CDC laboratories ( Ward et al., 1987). These strains were also analyzed for attachment and invasiveness to Hep-2 tissue culture cells, antimicrobial resistance profiling using the NARMS panel of 12 antimicrobials in a ‘Sensititer’ system, and MLEE to screen the isolates for 18 housekeeping enzymes were performed as described earlier ( Saeed et al., 2006). PFGE for the isolates was performed at the Michigan Department of Community Health Bureau of Laboratories following the PulseNet standardized protocol ( Ribot et al., 2002). | Table 1Phenotypic and Genotypic Characterization of Salmonella Enteritidis isolates (n = 34) from multiple sources |
DNA isolation Strains were cultured overnight at 37°C on tryptic soy agar plates. A generous loopful of each culture was suspended in 200 μL distilled water. For DNA extraction, cell suspensions were heated to 98°C on a heat block for 5 min and then immediately cooled on ice for 5 min. Cell suspensions were centrifuged (20 000 g, 5 min), and supernatants were used as the DNA template for PCR amplification. DNA quantification was measured with a NanoDrop© ND-1000 spectrophotometer (NanoDrop Technologies Inc., Rockland, DE), after which diluted samples (1 : 100–1 : 300 sterile DI water) were prepared and the remainder was frozen at −20°C in 30% glycerol for later use. VNTR loci and multiplex PCR Tandem repeat loci and their primers were based on the genome sequences of S. Enteritidis LK5, S. Enteritidis PT4, and S. Typhimurium LT2 strains as described previously ( Boxrud et al., 2007). A primer set for SE7 locus was redesigned in this study to yield a PCR amplicon that can be analyzed with the Beckman 600-bp DNA size marker and that can be separated by capillary electrophoresis under the same condition from amplicons of the other loci. Seven VNTR loci selected for MLVA were amplified in a single multiplex PCR with seven primer sets (SE1, SE2, SE3, SE5, SE7, SE8, and SE9) using forward primers labeled with a WellRED dye (Sigma-Proligo, Boulder, CO) at the 5′ end and nonlabeled reverse primers (Integrated DNA Technologies, Coralville, IA) described in Table 2. A master mix was made for 20 μL reactions containing the following components: 10 μL of 2 × Qiagen multiplex PCR Master mix, 2 μL of MgCl 2 (25 mM), 2 μL of 10 × primer mix (0.3–1.5 μM per each primer), 3 μL of the dilute DNA template (0.1–1.0 ng μL −1), and RNase-free water to a volume of 20 μL. Samples were loaded into a GeneAmp PCR system 9700 (Applied Biosystems). PCR was performed with predenature of 95°C for 15 min, then 35 cycles of 94°C for 30 s, 58°C for 90 s, 72°C for 90 s followed by a final elongation of 60°C for 30 min. | Table 2Tandem repeats and primers for one-shot multiplex PCR amplification of seven VNTR loci for MLVA |
Capillary electrophoresis The PCR products were purified using MultiScreen™ 96-well plates (Millipore, Bedford, MA) with hydrated Sephadex and loaded into a Beckman CEQ 8000 automated DNA sequencer (Beckman-Coulter) according to the manufacturer's suggested protocol. The fragment size for each locus was determined by combining 1.0 μL of PCR product, 38.5 μL sample loading solution, and 0.5 μL of CEQ 600-bp DNA size standard (Beckman-Coulter) to each sample well. The conditions for capillary separation of fragments for all loci included a capillary temperature of 35°C, denature temperature of 90°C, duration of 120 s, injection voltage of 2.0 kV, duration of 30 s, separation voltage of 7.5 kV, and separation duration of 50 min. DNA sequencing To verify the genetic basis of the results from MLVA, the copy number variations of tandem repeats for distinct alleles at all seven VNTR loci were analyzed by direct sequencing. After VNTR amplification using nonfluorescently labeled primers, the PCR products were purified using QIAquick PCR purification kits (QIAGen Inc., Valencia, CA). The forward and reverse strands of the purified PCR amplicons were sequenced using a Beckman CEQ 8000 DNA sequencer (Beckman Coulter, Fullerton, CA) and GenomeLab DTCS Quick Start kit (Beckman Coulter, Fullerton, CA) according to the manufacturer's suggested protocol. After sequences were aligned to create contigs using SeqMan (DNASTAR Inc., Madison, WI), the numbers of tandem repeats were counted using a tandem repeats finder software (accessible at http://tandem.bu.edu/trf/trf.html) ( Benson, 1999). The copy numbers were rounded to the nearest integer and entered into the MLVA profiles. Diversity and discriminatory power For the evaluation of the discriminatory capacity of the allelic variation at each VNTR marker, Nei's diversity index was calculated as 1–Σ (allele frequency)2. The probability that another S. Enteritidis isolate in the population has the same MLVA type was measured by multiplying the frequency of each allele for seven loci, which assumes that the alleles occur at random in genotypes. Simpson's diversity (1-D) and Shannon's diversity (H′) indices were calculated to evaluate the discriminatory power of subtyping methods as described previously ( Boxrud et al., 2007). Cluster analysis After sequencing verification process, the allele scores based on the fragment size were converted into repeat numbers of the seven loci and entered into bionumerics software (Applied-Maths, St-Martens-Latem, Belgium) as character data for cluster analysis. A dendrogram was generated using the categorical coefficient and unweighted pair group method with arithmetic means (UPGMA) of the bionumerics software (version 4.5). This categorical parameter implies that the same weight is given to any multistate character at each locus, whatever the repeat number is ( Ramisse et al., 2004). A minimum-spanning tree (MST) was also generated using the categorical coefficient of the software to calculate the distance matrix. In case of equivalent solutions in terms of calculated distance, the highest number of single locus variants (SLVs; in case two types have an equal distance to a linkage position in the tree, the type that has the highest number of SLVs is linked first) associated was used as the priority rule for linking types in the tree. Creation of hypothetical types (missing links) was allowed to introduce hypothetical types as branches of the MST, causing the total spanning of the tree to decrease significantly. The MLVA method has been optimized for the typing of a diverse group of 34 S. Enteritidis isolates. Diversity of 34 S. Enteritidis isolates is demonstrated by 10 different phage types, several patterns of attachment and invasiveness, and antimicrobial resistance expressed by six isolates, including four isolates with multi-drug resistance ( Table 1). MLEE type (ET) 3 was the most common among the 34 S. Enteritidis isolates. Ten PFGE patterns were identified using XbaI enzyme while 11 PFGE patterns were identified using BlnI enzyme. Combined patterns of the two enzymes resulted in 13 distinct PFGE genotypes. Tandem repeat sequences and diversity for VNTR loci are shown in Table 2. The genetic diversity estimates for seven loci ranged from 0.25 for SE9 to 0.70 for SE5. MLVA distinguished 13 different types among the 34 S. Enteritidis isolates from diverse sources (). These were classified into four main clusters by similarity matrices based on the categorical coefficient and UPGMA (). In the dendrogram, cluster A is composed mostly of S. Enteritidis strains recently isolated from humans. The strains are SLVs at locus SE5. However, based on categorical parameters, these isolates may be considered to belong to the same cluster regardless of copy numbers in this dendrogram (). Cluster B consists of a single strain isolated from mule deer (resistant to Sulfisoxazole) and is distinct from the lineages with allele differences at loci 2 and 5. Cluster C is composed of nine human clinical isolates, one bovine isolate, four egg isolates, and one chicken isolate. MLVA types Ca and Cb are SLVs at locus SE7 and are very closely related. A major type Ca is also related to double locus variants (DLVs) type Cc that includes isolates from eggs and chicken. Type Cd includes an isolate from human source. Cluster D comprises five types: Da (chicken farm), Db (chicken and egg), Dc (human), Dd (mice), and De (chicken). There are SLVs between types Da and Db and between types Dd and De. Defined S. Enteritidis MLVA types appeared to cluster largely by source and year of isolation (). The probability that another S. Enteritidis isolate with the same MLVA profile based on seven VNTR markers occurs in the population ranged from 0.038 to 0.00001 (). It was found that MLVA types Ca, Cc, Da, Db, Dd, and De appeared to be clonal groups, which can be more frequently found than other types based on allele frequency at seven VNTR loci among S. Enteritidis populations (). The MLVA type with the highest frequency alleles was Db, which included chicken and egg isolates. An MST analysis was conducted to develop a model for the evolutionary steps in the divergence of the S. Enteritidis MLVA genotypes (). The MST posits an intermediate clone that is hypothetical and not part of the real data set. MLVA type Db and a hypothetical intermediate type are located in the center of the MST, indicating they may be the evolutionary origin of other types but more S. Enteritidis isolates from diversified sources are needed for inclusion into this model for a more valid comparison. Clusters C and D could have branched off from a hypothetical type even though they were clustered together into a single clonal complex in the MST. There are two branches (Da, Dc, and Dd, De) off from type Db that constitute cluster D. Cluster A (Aa, Ab, and Ac) has neighbor distance of changes at four loci from a major central cluster (clusters C and D). Three recent isolates of antimicrobial-resistant S. Enteritidis from humans (Aa, Ab, and Ac) and one multidrug-resistant S. Enteritidis isolate from mink (Aa) belonged to this cluster. Cluster B is distant in changes at three loci from the major central cluster. A mule deer strain in cluster B is distinct in its unique PFGE, phage type, and MLEE type, suggesting its overall level of genetic divergence. The fact that these isolates were recently obtained from the listed sources may suggest the occurrence of an evolutionary process that affected the profiles of the isolates. MLVA has been reported to produce larger numbers of subtypes of E. coli O157 ( Noller et al., 2003; Keys et al., 2005; Hyytia-Trees et al., 2006) and S. enterica serovar Typhimurium ( Lindstedt et al., 2003a) than does PFGE. A previous study on S. Enteritidis ( Boxrud et al., 2007) demonstrated that MLVA yielded higher discriminatory power than did PFGE when tested using a panel of human clinical isolates of S. Enteritidis recently isolated from Minnesota. In the present study, the higher discriminatory power of MLVA for S. Enteritidis enabled the resolution of 18 S. Enteritidis isolates from diverse sources, but with identical PFGE patterns, into seven MLVA types. These S. Enteritidis MLVA types were largely clustered by source of isolation irrespective of their multiple phage types, which can serve as a useful epidemiologic tool in the investigation of S. Enteritidis infections (). For example, S. Enteritidis isolates from chickens and eggs were clustered together in each of two types (Cc and Db), which may support the widely perceived epidemiological linkage between these two sources. MLVA type Ca, Cb, Cd, and Dc were associated with isolates from human sources and type Dd was associated with mouse S. Enteritidis isolates. Simpson's diversity index and the 95% confidence intervals (95% CI) of subtyping methods were calculated in Table 3. Shannon's diversity index (H′) for MLVA was compared with other subtyping methods. Diversity indices of both Simpson and Shannon's were significantly higher in MLVA than other subtyping methods ( Table 3). Although PFGE and MLVA grouped the 34 S. Enteritidis isolates into 13 types (), MLVA had higher diversity indices than PFGE since MLVA types were more-evenly distributed than PFGE types. The number of S. Enteritidis isolates for the maximum, third quartile, and median were 8, 4, and 2 for MLVA types and 18, 2, and 1 for PFGE types, respectively. | Table 3Diversity indices of subtyping methods |
Molecular subtyping methods that can be used to relate disease-causing pathogens to their probable sources are needed. PFGE has been the standardized method used in PulseNet, the national molecular subtyping network for the epidemiological investigation of several foodborne pathogens. However, a combination of PFGE and phage typing has been applied for the characterization of S. Enteritidis isolates from different sources ( Liebana et al., 2001, 2002). PFGE subtyping of S. Enteritidis did not result in diverse profiles because the genomes of S. Enteritidis are highly similar ( Hudson et al., 2001; Liebana et al., 2001). In this study, the previously described MLVA method for S. Enteritidis subtyping has been optimized, so that it may be performed as a single multiplex PCR followed by DNA fragment analysis. In epidemiologic investigations, the sensitivity and effectiveness of a laboratory method, along with other information, to identify the source of infectious agents, are vital if early control and preventive measures against the disease. If the host adaptation leaves a print of particular variation in even one locus of the VNTR, MLVA is more likely to help in tracing back the S. Enteritidis to its source than PFGE and phage tying. Because the identification of sources of foodborne pathogens is important for the control of foodborne diseases, the high discriminatory power of MLVA can be of significant value in the epidemiologic investigation of outbreak and sporadic cases of S. Enteritidis infection ( Boxrud et al., 2007). Further study may be needed to compare allele distribution or genetic diversity for each locus of S. Enteritidis isolates from different sources for tracing purposes. Among the limitations of this study is the relatively small number of S. Enteritidis isolates; for example, the small numbers of samples used in an MST-based cluster may not be adequate to construct a valid population modeling. In conclusion, the MLVA method has been optimized based on a single multiplex PCR using seven VNTR markers with high diversity capacity. Tandem repeat numbers for each locus were verified after the sequencing step. This method can be used along with PFGE to enhance the effectiveness of the molecular epidemiologic investigation of S. Enteritidis infections. This study was funded, in part, with federal funds from the National Institute of Allergy and Infectious Disease (NIAID) National Institute of Health (NIH), US Department of Health and Human Services under contract N01-AI-30058. Dr Thomas Besser and the Washington State University ZRU team are thanked for providing part of the S. Enteritidis isolates for this project. Reuse of this article is permitted in accordance with the Creative Commons Deed, Attribution 2.5, which does not permit commercial exploitation. - Benson G. Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res. 1999;27:573–580. [PubMed]
- Boxrud D. Comparison of multiple-locus variable-number tandem repeat analysis, pulsed-field gel electrophoresis, and phage typing for subtype analysis of Salmonella enterica serotype Enteritidis. J Clin Microbiol. 2007;45:536–543. [PubMed]
- Fakhr MK. Multilocus sequence typing lacks the discriminatory ability of pulsed-field gel electrophoresis for typing Salmonella enterica serovar Typhimurium. J Clin Microbiol. 2005;43:2215–2219. [PubMed]
- Farlow J. Francisella tularensis strain typing using multiple-locus, variable-number tandem repeat analysis. J Clin Microbiol. 2001;39:3186–3192. [PubMed]
- Hudson CR. Determination of close genetic relatedness of the major Salmonella enteritidis phage types by pulsed-field gel electrophoresis and DNA sequence analysis of several Salmonella virulence genes. Avian Dis. 2001;45:875–886. [PubMed]
- Hyytia-Trees E. Second generation subtyping: a proposed PulseNet protocol for multiple-locus variable-number tandem repeat analysis of Shiga toxin-producing Escherichia coli O157 (STEC O157). Foodborne Pathog Dis. 2006;3:118–131. [PubMed]
- Keim P. Multiple-locus variable-number tandem repeat analysis reveals genetic relationships within Bacillus anthracis. J Bacteriol. 2000;182:2928–2936. [PubMed]
- Keys C. Highly diverse variable number tandem repeat loci in the E. coli O157:H7 and O55:H7 genomes for high-resolution molecular typing. J Appl Microbiol. 2005;98:928–940. [PubMed]
- Klevytska AM. Identification and characterization of variable-number tandem repeats in the Yersinia pestis genome. J Clin Microbiol. 2001;39:3179–3185. [PubMed]
- Liebana E. Diversity of strains of Salmonella enterica serotype enteritidis from English poultry farms assessed by multiple genetic fingerprinting. J Clin Microbiol. 2001;39:154–161. [PubMed]
- Liebana E. Salmonella enterica serotype Enteritidis phage types 4, 7, 6, 8, 13a, 29 and 34: a comparative analysis of genomic fingerprints from geographically distant isolates. J Appl Microbiol. 2002;92:196–209. [PubMed]
- Lindstedt BA. DNA fingerprinting of Salmonella enterica subsp. enterica serovar typhimurium with emphasis on phage type DT104 based on variable number of tandem repeat loci. J Clin Microbiol. 2003a;41:1469–1479. [PubMed]
- Lindstedt BA. DNA fingerprinting of Shiga-toxin producing Escherichia coli O157 based on Multiple-locus variable-number tandem-repeats analysis (MLVA). Ann Clin Microbiol Antimicrob. 2003b;2:12. [PubMed]
- Lindstedt BA. Multiple-locus variable-number tandem-repeats analysis of Salmonella enterica subsp. enterica serovar Typhimurium using PCR multiplexing and multicolor capillary electrophoresis. J Microbiol Methods. 2004a;59:163–172. [PubMed]
- Lindstedt BA. Multiple-locus variable-number tandem-repeats analysis of Escherichia coli O157 using PCR multiplexing and multi-colored capillary electrophoresis. J Microbiol Methods. 2004b;58:213–222. [PubMed]
- Lukinmaa S. Application of molecular genetic methods in diagnostics and epidemiology of food-borne bacterial pathogens. Apmis. 2004;112:908–929. [PubMed]
- Maiden MC. Multilocus sequence typing of bacteria. Annu Rev Microbiol. 2006;60:561–588. [PubMed]
- Mead PS. Food-related illness and death in the United States. Emerg Infect Dis. 1999;5:607–625. [PubMed]
- Noller AC. Multilocus variable-number tandem repeat analysis distinguishes outbreak and sporadic Escherichia coli O157: H7 isolates. J Clin Microbiol. 2003;41:5389–5397. [PubMed]
- Overduin P. Use of multilocus variable-number tandem-repeat analysis for typing Mycobacterium avium subsp. paratuberculosis. J Clin Microbiol. 2004;42:5022–5028. [PubMed]
- Ramisse V. Variable number of tandem repeats in Salmonella enterica subsp. enterica for typing purposes. J Clin Microbiol. 2004;42:5722–5730. [PubMed]
- Rankin S. Phage conversion in Salmonella enterica serotype enteritidis: implications for epidemiology. Epidemiol Infect. 1995;114:227–236. [PubMed]
- Ribot EM. Salmonella enterica serotype Typhimurium DT104 isolated from humans, United States, 1985, 1990, and 1995. Emerg Infect Dis. 2002;8:387–391. [PubMed]
- Saeed AM. Clonal structure and variation in virulence of Salmonella Enteritidis isolated from mice, chickens, and humans. J AOAC Int. 2006;89:504–511. [PubMed]
- Svraka S. Establishment of a genotyping scheme for Coxiella burnetii. FEMS Microbiol Lett. 2006;254:268–274. [PubMed]
- Swaminathan B. PulseNet: the molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg Infect Dis. 2001;7:382–389. [PubMed]
- Torpdahl M. Genotypic characterization of Salmonella by multilocus sequence typing, pulsed-field gel electrophoresis and amplified fragment length polymorphism. J Microbiol Methods. 2005;63:173–184. [PubMed]
- van Belkum A. Short sequence repeats in microbial pathogenesis and evolution. Cell Mol Life Sci. 1999;56:729–734. [PubMed]
- Ward LR. A phage-typing scheme for Salmonella enteritidis. Epidemiol Infect. 1987;99:291–294. [PubMed]
|
This article has been 
