Comparison between Salmonella enterica Serotype ...Comparison between Salmonella enterica Serotype Enteritidis Genotyping Methods and Phage Type Alessandra De Cesare,a Keshav Krishnamani,b
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Comparison between Salmonella enterica Serotype EnteritidisGenotyping Methods and Phage Type
Alessandra De Cesare,a Keshav Krishnamani,b Antonio Parisi,c Antonia Ricci,d Ida Luzzi,e Lisa Barco,d Alex Lucchi,a
Angela Miccolupo,c Gerardo Manfredaa
Department of Agricultural and Food Sciences, Ozzano dell’Emilia, Italya; DuPont Nutrition & Health, Geneva, Switzerlandb; Istituto Zooprofilattico Sperimentale dellaPuglia e della Basilicata, Putignano, Bari, Italyc; OIE/National Reference Laboratory for Salmonella-Istituto Zooprofilattico Sperimentale delle Venezie, Legnaro, Italyd;Istituto Superiore di Sanità, Rome, Italye
A quantitative comparison between discriminatory indexes and concordance among multilocus variable-number tandem-repeatanalysis (MLVA), pulsed-field gel electrophoresis (PFGE), automated ribotyping, and phage typing has been performed, testing238 Salmonella enterica serotype Enteritidis isolates not epidemiologically correlated. The results show that MLVA is the bestchoice, but each typing method provides a piece of information for establishing clonal relationships between the isolates.
Salmonella enterica serotype Enteritidis is the most frequentlyreported Salmonella serovar in humans in the European Union
(1). This study evaluated the discriminatory power and congru-ence among typing results collected by applying multilocus vari-able-number tandem-repeat analysis (MLVA), pulsed-field gelelectrophoresis (PFGE), and automated ribotyping, with a newcombination of restriction enzymes, to a collection of 238 S. En-teritidis isolates not correlated from an epidemiological point ofview. They were isolated in 19 regions in Italy from humans, ani-mals, foods, and environments between 2008 and 2012 (Table 1).For all isolates, except those isolated from humans, the phage type(PT) was kindly provided by the Italian Reference Laboratory forSalmonella and included in the analysis of typing results (Table 1).
MLVA was performed according to the protocol published byHopkins et al. (2) based on the loci SE3, SENTR4, SENTR5,SENTR6, and SENTR7. The five loci were amplified in one mul-tiplex PCR (25-�l volume) containing 5 pmol of each primer,using a multiplex PCR kit (Qiagen). The amplification productswere diluted 1:50 in sterile distilled water, and 1 �l of each dilutionwas mixed with 10 �l of Hi-Di formamide (Applied Biosystems)and 0.4 �l GeneScan LIZ 600 size standard (Applied Biosystems)before being subjected to capillary electrophoresis using POP-7polymer on a 3130 genetic analyzer (Applied Biosystems). AnMLVA type, labeled as a number, was assigned to each isolatebased on the difference in the variable-number tandem repeat(VNTR) profile in at least one locus. Minimum spanning trees(MST) based on the MLVA profiles were built in BioNumerics 7.5(Applied Maths) using the categorical coefficient. Distances be-tween MLVA profiles were calculated based on the numbers ofdifferent loci between profiles, irrespective of their within-locusdifferences in the number of repeats.
PFGE was performed according to the PulseNet protocol (3),digesting the plug with 50 U of XbaI (Fermentas). The SalmonellaBraenderup strain H9812 PulseNet standard was used as a molec-ular weight marker. The fingerprinting profiles were analyzed us-ing BioNumerics 7.1 and compared by cluster analysis using theDice coefficient and the unweighted pair group method witharithmetic means (UPGMA), with a position tolerance limit andoptimization of 1%. Isolates showing a PFGE similarity levelof �95% were assigned to the same pulsotype. The pulsotypeswere labeled as numbers.
Automated ribotyping was performed with the RiboPrinter ac-cording to the manufacturer’s instructions (4), using a mixture of1,250 U of PvuII (Qualicon) and 1.250 U of PstI (New EnglandBioLabs). The restriction digestion was performed at 37°C for 20min. The characterization consisted of combining profiles withina similarity range (as calculated using the RiboPrinter’s proprie-tary algorithm) �0.93 to form a dynamic ribogroup (RIBO) la-beled with an alphanumeric code (4).
The Comparing Partitions website (http://darwin.phyloviz.net/ComparingPartitions/index.php?link�Toll) was used to calcu-late the discriminatory index (DI) of each typing method throughSimpson’s diversity index (5, 6), the unidirectional concordancebetween methods by applying the Wallace (W) and adjustedWallace (AW) coefficients (7), and the bidirectional concor-dance using the adjusted Rand (AR) coefficient (8).
Overall, 21 MLVA types were each associated to a single isolate,whereas 23 comprised between 2 and 67 isolates. The two largestMLVA types grouped isolates collected from all sampling years(Fig. 1). However, small clusters of isolates collected only in spe-cific years were also identified (Fig. 1). One of the two main MLVAtypes clustered together human, poultry, and egg isolates. Fur-thermore, the same cluster included isolates from other sources,confirming that S. Enteritidis can be transmitted between hu-mans, wildlife, livestock, and pets (Fig. 2). In relation to the PT, anMST was built, including the MLVA types of human isolates andthe reaction does not conform (RDNC) isolate (in red). Humanisolates clustered in the two main groups with isolates belonging
Received 29 April 2015 Returned for modification 12 June 2015Accepted 21 June 2015
Accepted manuscript posted online 1 July 2015
Citation De Cesare A, Krishnamani K, Parisi A, Ricci A, Luzzi I, Barco L, Lucchi A,Miccolupo A, Manfreda G. 2015. Comparison between Salmonella entericaserotype Enteritidis genotyping methods and phage type. J Clin Microbiol53:3021–3031. doi:10.1128/JCM.01122-15.
to the PT identified in S. Enteritidis isolates collected in Italy fromconfirmed cases of human salmonellosis (Fig. 3).
The S. Enteritidis isolates were characterized by 23 pulsotypes,with bands between 20 and 900 kbp, showing a similarity levelranging between 36 and 92% (Fig. 4), and seven ribotyping pro-files, characterized by 7 to 10 bands, with molecular weight rang-ing between 2 and 15 kbp (Fig. 5). The ribotyping profiles 153-494-S-1 and 153-507-S-2 were identified in 57.1 and 36.5% of theisolates, respectively. PFGE is the current gold standard to assessrelatedness among Salmonella isolates from different sources (9,10) and for outbreak investigations (11, 12). However, accordingto the literature, PFGE exhibits limited discriminatory power forS. Enteritidis (13). This aspect was confirmed in a multicountryoutbreak of S. Enteritidis recently reported in Europe (14, 15).Therefore, the European Centre for Disease Prevention and Con-trol (ECDC) decided to promote the use of MLVA to subtype S.Enteritidis isolates (14). Since the European Food Safety Author-ity (EFSA) stated that there is currently no comprehensive collec-tion of comparable background on MLVA typing data for S. En-teritidis available at the European Union level, this paper shouldhelp to start filling this gap (14).
For each typing method applied alone, the DI ranged between0.54 for automated ribotyping and 0.88 for MLVA (Table 2),whereas combining two or three methods resulted in a DI rangingbetween 0.88 and 0.97 (Table 2). Since S. Enteritidis is one of the
most genetically homogeneous serotypes of Salmonella (16) it isnot surprising that in this study no typing method applied alonereached a DI of �0.90. The combination MLVA-PT has beenidentified as the best option by Cho et al. (17) and Dewaele et al.(18), even though they applied a different MLVA scheme.
All methods tested, including PT, showed a weak directionalconcordance (i.e., �0.5) (Table 3). However, MLVA showed anAW of �0.16 with PFGE and PT. Moreover, PT exhibited an AWof �0.18 with MLVA and PFGE. The bidirectional concordancebetween methods was also very low and was �0.19 for MLVA-PTand PFGE-PT only (Table 4). The same combinations of typingmethods reached the highest AW values. The low congruence be-tween the applied typing methods demonstrated that they all pro-vide a piece of information for establishing possible clonal rela-tionships among the isolates tested. Such relationships concerned42.4% of the isolates forming 32 clusters containing 2 to 16 iso-lates sharing the same type strain, as defined according to theMLVA, PFGE, RIBO, and PT profiles (Table 5). Overall, 65.6% ofthese clusters grouped isolates of common origin (e.g., poultry),whereas 34.4% isolates were classically not correlated (e.g., poul-try and cheese) (Table 5).
In conclusion, the results of this study allow a quantitativecomparison between different typing methods for S. Enteritidis interms of discriminatory power and concordance. These results,
TABLE 1 (Continued)
Source (no. of isolates) Isolation year, phage type (no. of isolates if �1) [region(s)]a
FIG 2 Minimum spanning tree calculated for MLVA profiles of 238 S. Enteritidis isolates collected from environmental (n � 33), animal (n � 112), food/feed(n � 63), and human (n � 30) sources.
Salmonella Enteritidis Genotyping and Phage Typing
September 2015 Volume 53 Number 9 jcm.asm.org 3025Journal of Clinical Microbiology
along with execution time, cost effectiveness, and the level ofcomplexity of data interpretation and sharing, should help inmaking a critical choice on the most appropriate method toapply for typing S. Enteritidis during both outbreak investiga-tions and longtime surveillance. Even if the most promisingroutine epidemiological typing tool for Salmonella enterica
seems to be whole-genome sequencing (WGS) (19), when an al-ternative method must be selected, MLVA is confirmed to be thebest choice. In fact, it is highly discriminatory, reproducible, fast,and easy to perform. Moreover, it produces results easy to inter-pret and analyze, which can be shared using international data-bases.
TABLE 2 Discriminatory power of each typing method applied alone or in combination with other methods on the S. Enteritidis isolates tested
REFERENCES1. European Food Safety Authority and European Centre for Disease
Prevention and Control. 2015. The European Union summary report ontrends and sources of zoonoses, zoonotic agents and food-borne out-breaks in 2013. EFSA J 13(1):3991.
2. Hopkins KL, Peters TM, de Pinna E, Wain J. 2011. Standardisationof multilocus variable-number tandem repeat analysis (MLVA) forsubtyping of Salmonella enterica serovar Enteritidis. Euro Surveill16(32):pii�19942. http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId�19942.
3. Ribot EM, Fair MA, Gautom R, Cameron DN, Hunter SB, Swamina-than B, Barrett TJ. 2006. Standardization of pulsed-field gel electropho-resis protocols for the subtyping of Escherichia coli O157:H7, Salmonella,and Shigella for PulseNet. Foodborne Pathog Dis 3:59 – 67. http://dx.doi.org/10.1089/fpd.2006.3.59.
4. Bruce J. 1996. Automated system rapidly identifies and characterizes mi-cro-organisms in foods. Food Technol 50:77–78.
5. Hunter PR. 1990. Reproducibility and indices of discriminatory of dis-criminatory power of microbial typing methods. J Clin Microbiol 28:1903–1905.
6. Hunter PR, Gaston MA. 1988. Numerical index of the discriminatoryability of typing systems: an application of Simpson’s index of diversity. JClin Microbiol 26:2465–2466.
7. Severiano A, Pinto FR, Ramirez M, Carriço JA. 2011. Adjusted Wallacecoefficient as a measure of congruence between typing methods. J ClinMicrobiol 49:3997– 4000. http://dx.doi.org/10.1128/JCM.00624-11.
8. Carriço JA, Silva-Costa C, Melo-Cristino J, Pinto FR, de Lencastre H,Almeida JS, Ramirez M. 2006. Illustration of a common framework forrelating multiple typing methods by application to macrolide-resistantStreptococcus pyogenes. J Clin Microbiol 44:2524 –2532. http://dx.doi.org/10.1128/JCM.02536-05.
9. Lynne AM, Kaldhone P, David D, White DG, Foley SL. 2009. Charac-terization of antimicrobial resistance in Salmonella enterica serotypeHeidelberg isolated from food animals. Foodborne Pathog Dis 6:207–215.http://dx.doi.org/10.1089/fpd.2008.0172.
10. Ponce E, Khan AA, Cheng CM, Summage-West C, Cerniglia CE. 2008.Prevalence and characterization of Salmonella enterica serovar Weltevre-den from imported seafood. Food Microbiol 25:29 –35. http://dx.doi.org/10.1016/j.fm.2007.09.001.
11. Centers for Disease Control and Prevention. 2010. Investigation update:multistate outbreak of human Salmonella enteritidis infections associatedwith shell eggs. Centers for Disease Control and Prevention, Atlanta, GA.http://www.cdc.gov/salmonella/Enteritidis.
12. Jacobs W, Kuiling S, van der Zwaluw K. 2014. Molecular typing ofSalmonella strains isolated from food, feed and animals: state of playand standard operating procedures for pulsed field gel electrophoresis(PFGE) and multiple-locus variable number tandem repeat Analysis(MLVA) typing, profiles interpretation and curation. EFSA supportingpublication 2014:EN-703. European Food Safety Authority, Parma,Italy.
13. Boxrud D, Pederson-Gulrud K, Wotton J, Medus C, Lyszkowicz E,Besser J, Bartkus JM. 2007. Comparison of multiple-locus variable-number tandem repeat analysis, pulsed-field gel electrophoresis, andphage typing for subtype analysis of Salmonella enterica serotype Enter-itidis. J Clin Microbiol 45:536 –543. http://dx.doi.org/10.1128/JCM.01595-06.
14. European Food Safety Authority and European Centre for DiseasePrevention and Control. 2014. Multi-country outbreak of SalmonellaEnteritidis infections associated with consumption of eggs from Germany.EFSA supporting publication 2014:EN-646. http://www.efsa.europa.eu/it/supporting/doc/646e.pdf
15. Peters TM, Berghold C, Brown D, Coia J, Dionisi AM, Echeita A, FisherIS, Gatto AJ, Gill N, Green J, Gerner-Smidt P, Heck M, Lederer I,Lukinmaa S, Luzzi I, Maguire C, Prager R, Usera M, Siitonen A,Threlfall EJ, Torpdahl M, Tschäpe H, Wannet W, Zwaluw WK. 2007.Relationship of pulsed-field profiles with key phage types of Salmonellaenterica serotype Enteritidis in Europe: results of an international multi-center study. Epidemiol Infect 135:1274 –1281.
16. Saeed AM, Walk ST, Arshad M, Whittam TS. 2006. Clonal structure andvariation in virulence of Salmonella Enteritidis isolated from mice, chick-ens, and humans. J AOAC Int 9:504 –511.
17. Cho S, Whittam TS, Boxrud DJ, Bartkus JM, Rankin SC, WilkinsMJ, Somsel P, Downes FP, Musser KA, Root TP, Warnick LD,Wiedmann M, Saeed AM. 2010. Use of multiple-locus variable num-ber tandem repeat analysis and phage typing for subtyping of Salmo-nella Enteritidis from sporadic human cases in the United States. J Appl
TABLE 5 (Continued)
Cluster label Type strain MLVA-PFGE(RIBO)PT (no. of isolates) Source (no. of isolates if �1) Isolation year (no. of isolates if �1) [region(s)]a
a Region: A, Abruzzo; CA, Campania; ER, Emilia-Romagna; FR, Friuli Venezia Giulia; LA, Lazio; LI, Liguria; LO, Lombardia; MA, Marche; P, Piemonte; T, Tuscany; TA, TrentinoAlto Adige; U, Umbria; V, Veneto.
De Cesare et al.
3030 jcm.asm.org September 2015 Volume 53 Number 9Journal of Clinical Microbiology
18. Dewaele I, Rasschaert G, Bertrand S, Wildemauwe C, Wattiau P,Imberechts H, Herman L, Ducatelle R, De Reu K, Heyndrickx M.2012. Molecular characterization of Salmonella Enteritidis: comparisonof an optimized multi-locus variable-number of tandem repeat analysis
(MLVA) and pulsed-field gel electrophoresis. Foodborne Pathog Dis9:885– 895. http://dx.doi.org/10.1089/fpd.2012.1199.
19. Leekitcharoenphon P, Nielsen EM, Kaas RS, Lund O, Aarestrup FM.2014. Evaluation of whole genome sequencing for outbreak detectionof Salmonella enterica. PLoS One 9:e87991. http://dx.doi.org/10.1371/journal.pone.0087991.
Salmonella Enteritidis Genotyping and Phage Typing
September 2015 Volume 53 Number 9 jcm.asm.org 3031Journal of Clinical Microbiology