Repeat-based Sequence Typing of Carnobacterium ...

International Journal of Food Microbiology 226 (2016) 1–4

Contents lists available at ScienceDirect

International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro

Repeat-based Sequence Typing of Carnobacterium maltaromaticum Abdur Rahman a,b, Sara M. El Kheir a, Alexandre Back a, Cécile Mangavel a, Anne-Marie Revol-Junelles a, Frédéric Borges a,⁎ a b

Université de Lorraine, Laboratoire d'Ingénierie des Biomolécules (LIBio), ENSAIA, 2 avenue de la Forêt de Haye, TSA, 40602 54518 Vandœuvre-lès-Nancy, France National University of Sciences and Technology, Atta ur Rahman School of Applied Biosciences, Department of Industrial Biotechnology, 44000, H-12, Islamabad, Pakistan

a r t i c l e

i n f o

Article history: Received 16 July 2015 Received in revised form 24 November 2015 Accepted 4 March 2016 Available online 4 March 2016

a b s t r a c t Carnobacterium maltaromaticum is a Lactic Acid Bacterium (LAB) of technological interest for the food industry, especially the dairy as bioprotection and ripening flora. The industrial use of this LAB requires accurate and resolutive typing tools. A new typing method for C. maltaromaticum inspired from MLVA analysis and called Repeat-based Sequence Typing (RST) is described. Rather than electrophoresis analysis, our RST method is based on sequence analysis of multiple loci containing Variable-Number Tandem-Repeats (VNTRs). The method described here for C. maltaromaticum relies on the analysis of three VNTR loci, and was applied to a collection of 24 strains. For each strain, a PCR product corresponding to the amplification of each VNTR loci was sequenced. Sequence analysis allowed delineating 11, 11, and 12 alleles for loci VNTR-A, VNTR-B, and VNTR-C, respectively. Considering the allele combination exhibited by each strain allowed defining 15 genotypes, ending in a discriminatory index of 0.94. Comparison with MLST revealed that both methods were complementary for strain typing in C. maltaromaticum. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Carnobacterium maltaromaticum is a non-starter Lactic Acid Bacterium (LAB) with technological interest in the field of food bioprotection and dairy ripening (Afzal et al., 2010). This bacterium is able to inhibit the pathogen Listeria monocytogenes and thus can be used to extend the shelf-life of food including lightly preserved food (Azuma et al., 2007; Brillet et al., 2004; Duffes et al., 1999; Mathieu et al., 1994; Schobitz et al., 2003; Wan et al., 1997). In soft cheese, C. maltaromaticum was shown to be associated with the development of malty and butyric flavors (Afzal et al., 2013). The genetic diversity of C. maltaromaticum dairy strains is high compared to starter LAB such as Streptococcus thermophilus and Lactococcus lactis (Delorme et al., 2010; Passerini et al., 2010; Rahman et al., 2014a), suggesting that dairy strains of the species C. maltaromaticum hold a high potential for technological innovation in the dairy industry. Moreover, it can survive the acidification process with no apparent interference with commercial starters (Edima et al., 2008). The use of this bacterium as adjunct cultures is presumptively safe since (i) it frequently colonizes cheeses (10% of soft cheeses) where it can grow up to 109 CFU/g indicating that this bacterium is massively ingested by consumers, (ii) it does not produce detectable amounts of tyramine and histamine in soft cheeses (Afzal et al., 2010), (iii) shows neutral to slightly anti-inflammatory properties when compared with other probiotic LAB (Rahman et al., 2014b) and (iv) it has been very scarcely ⁎ Corresponding author. E-mail address: [email protected] (F. Borges).

http://dx.doi.org/10.1016/j.ijfoodmicro.2016.03.003 0168-1605/© 2016 Elsevier B.V. All rights reserved.

associated with bacterial infections (Chmelar et al., 2002). It is Generally Recognized As Safe (GRAS) in the USA (FDA, 2005) and is considered as technological beneficial microorganisms in EU (Bourdichon et al., 2012). The industrial use of C. maltaromaticum strains necessitates the availability of tools that enable distinguishing the members within this species for strain tracking, collection management, traceability, and strain isolation. Lately, a method based on MultiLocus Sequence Typing allowed to distinguish strains from this species, including from food origin, with high discriminatory index (0.98) (Rahman et al., 2014a). However, several strains remained undistinguishable, as exemplified by strains exhibiting the genotypes ST4 and ST3; these genotypes are mainly represented by dairy strains, each genotype being represented by 5 strains (Rahman et al., 2014a). In the field of medical microbiology, one alternative powerful method for typing of closely related isolates is Multiple-Locus Variable-number tandem-repeat Analysis (MLVA) (Van Belkum, 2007). This method is based on repetitive DNA elements arrayed in tandem. Tandem repeats with short repeat units are genetically unstable due to DNA replication errors and are therefore characterized by fast evolution. By targeting several different tandem repeat loci in the chromosome, it is possible to distinguish very closely related strains by analyzing the length of PCR products bearing VNTR and thereby the evaluation of repeat number by using electrophoresis methods (Van Belkum, 2007). MLVA is successfully used for foodborne pathogens including Salmonella enterica serotypes Typhimurium and Enteritidis for which strain typing is a great challenge (Nadon et al., 2013). Lately, MLVA was applied to the LAB Oenococcus oeni and revealed a higher discriminatory power compared to Pulsed Field Gel

2

A. Rahman et al. / International Journal of Food Microbiology 226 (2016) 1–4

Table 1 Strains used in this study and corresponding VNTR allele numbers and RSTs. Strain name

Origin

MLST ST

VNTR-A

VNTR-B

VNTR-C

RST

F1 F4 F84 F87 CP7 CP14 F7 F44 LMA28 F86 L1 DSM20342 N15 F42 CP23 F48 F73 L11 G97 3BO4 9.4 F43 CIP100481 DSM20590

Camembert Le montagnard Kernhemmer Tynjethaler Brie Brie Halloum Mont d'or Brie Cabrales Milk Raw milk Halibut sugar salted Rocamadour Brie Gormas Selles-sur-Cher Raw milk Raw milk Sphagnum Beef MAP St. Marcellin Human blood Vaccum packaged beef

1 3 3 3 3 3 4 4 4 11 13 4 7 7 16 9 10 14 33 31 30 8 18 22

1 2 2 2 2 2 3 3 3 3 3 4 5 2 3 6 6 7 8 9 10 10 10 11

1 2 2 2 2 2 3 3 3 3 3 4 2 2 5 6 6 7 8 9 10 10 10 11

1 2 3 3 4 3 5 6 6 6 6 7 4 3 6 8 8 9 10 11 12 12 12 9

1 2 3 3 4 3 5 6 6 6 6 7 8 3 9 10 10 11 12 13 14 14 14 15

Electrophoresis (PFGE) based method and MLST (Claisse and LonvaudFunel, 2012). The aim of this study was to establish a typing method based on the analysis of VNTR loci for C. maltaromaticum and to determine its discriminatory power. 2. Material and methods 2.1. Strains and media The strains of C. maltaromaticum used in this study are described in Table 1 and were cultivated in Tryptic Soy Broth supplemented with Yeast Extract (TSBYE-medium).

Types) was investigated by using eBURST (Feil et al., 2004). RSTs differing at one VNTR locus were considered to belong to the same RST-clonal complex. The discriminatory index was calculated as the Simpson's index of diversity (Hunter and Gaston, 1988). 2.4. Nucleotide sequence accession numbers The sequence data are hosted at GenBank under the following accession numbers: KR820018 to KR820041 for VNTR-A, KR820042 to KR820065 for VNTR-B, and KR820066 to KR820089 for VNTR-C. 3. Results 3.1. Identification of potential VNTR loci by genome mining In order to identify potential VNTR loci, tandem repeats were searched in the chromosome sequence of C. maltaromaticum LMA28 (Cailliez-Grimal et al., 2013) with tandem repeat finder. One hundred and twenty tandem repeats were identified in the chromosome sequence of this strain. The results were subsequently manually filtered based on the criteria of high repeat number, variations between repeat units, and the size of the repeat units. Three loci, presenting the highest scores for those criteria were retained for further analysis and were subsequently named VNTR-A, VNTR-B, and VNTR-C. The VNTR-A sequence is localized within the coding sequence BN424_278, at the 3′ extremity, which encodes a hypothetical protein. The repeated nucleic acid sequence results in a repeated amino acid sequence enriched with aspartate residues at the C-terminus of the putative protein. VNTR-B is localized in an intergenic region, between the coding sequences BN424_521 and BN424_522 which encode a putative hydroxyacid dehydrogenase and a putative ABC transporter extracellular-binding protein, respectively. VNTR-C completely overlaps the coding sequences BN424_902 and partially overlaps the 5′ extremity of BN424_903. BN424_902 and BN424_903 encode a hypothetical protein and a putative rifampin ADP-ribosyl transferase, respectively. The VNTR loci A, B, and C each consist of a unique repeated tandem motif of 36 bp, 82 bp, and 121 bp, respectively. The repeat units are imperfectly repeated in three selected VNTR loci. Repeats indeed contain indels of one or several nucleotides, and point mutations. In addition, VNTR-A and VNTR-C both contain a partial repeat unit at one extremity.

2.2. DNA preparation, polymerase chain reaction (PCR) and sequencing 3.2. Variability of the VNTR loci DNA extraction was performed as previously described (Rahman et al., 2014a). PCR reactions were performed in a final volume of 25 μL containing TaKaRa mix (Takara Bio Incorporated, Otsu, Japan), 5– 10 ng of DNA template and 0.4 ng of each primer. The primer sequences were as follows: VNTR-A-f gcgattaatagtggaacgac, VNTR-A-r taatgttgatgaggagatggcacag, VNTR-B-f tgccccggaatgtcattgat, VNTR-B-r gcattacgcttcacgaatcca, VNTR-C-f taaagttgaaattacactggataagg, and VNTR-C-r tctcctatcgctagatttgcctttg. The BioRad T100™ thermocycler (BioRad, Hercules, USA) was used for amplification using the following program: 95 °C for 5 min, followed by 35 cycles of 95 °C for 30 s, 54 °C for 45 s, and 72 °C for 1 min, followed by a final extension at 72 °C for 5 min. PCR products were analyzed by 16 h migration at 40 V and 40 mA on a 2.5% w/v agarose gel made in a Tris acetate EDTA buffer. PCR products were sequenced according to the Sanger method (Eurofins).

The three loci were targeted for PCR amplification by using the genomic DNA of 24 strains of C. maltaromaticum. Agarose gel electrophoresis

2.3. Data analysis Tandem repeats were searched with tandem repeat finder (Benson, 1999) by using the following parameters: 2 for match, 7 for mismatch, 7 for indels, 50 for minimum alignment core to report repeat, and 500 for maximum period size. The genome localization of the VNTR loci was determined by using MicroScope (Vallenet et al., 2013). Allele determination was performed by sequentially building a local database with BioEdit (Hall, 1999). Relatedness between RSTs (Repeat based Sequence

Fig. 1. Allele size variation of the three VNTR loci.

A. Rahman et al. / International Journal of Food Microbiology 226 (2016) 1–4

3

size variability and VNTR-A the lowest (Fig. 1). Several alleles exhibited the same size i.e. 341 bp for VNTRA-6 and -8; 347 bp for VNTRA-7, -9, and -10; 242 bp for VNTRB-2 and VNTRB-11; 570 bp for VNTRB-3 and -4; 104 bp for VNTRB-7 and -10; 272 for VNTRC-2, -3, and -8; 496 for VNTRC-6 and -7; 416 for VNTRC-10 and -12. Although they exhibited the same size they were different thanks to point mutations, except for VNTRA-10, − 7 and − 9: VNTRA-A-10 contains an insertion of 3 base pairs after position 24 and a deletion of 3 other base pairs between positions 205 and 206, compared to VNTRA-7 and VNTRA-9. Point mutations were also observable between alleles of different sizes. 3.3. Strain discrimination through RST analysis

Fig. 2. Scheme representing the link between MLST and RST genotypes. STs and RSTs are represented by circles. Two genotypes connected by a solid line indicate they were carried by the same strain. Thick gray bars embrace genotypes that belong to the same clonal complex. Clonal complexes based on MLST analysis are CC1, CC2, CC3, CC3, and CC4. Clonal complexes based on RST analysis are RST-CC1 and RST-CC2.

revealed that one band was obtained for each strain and that all VNTR loci were variable in size between strains. The analysis of the resulting profiles allowed identifying 6, 6, and 5 different alleles based on size determination for VNTR-A, -B, and -C, respectively, and consequently ended in a low discrimination between strains. In order to increase the discrimination power, each PCR product was sequenced and the sequences were analyzed to delineate alleles. Sequence comparison revealed 11, 11, and 12 alleles for VNTR-A, VNTR-B, and VNTR-C, respectively, showing that DNA sequencing allow delineating more efficiently the different alleles than size comparison by agarose gel electrophoresis. The loci VNTR-A, VNTR-B, and VNTR-C display a size variability ranging from 236 to 380, 104 to 570, and 181 to 496, respectively. The locus VNTR-B exhibited the highest

Fig. 3. Graphical representation of eBURST analysis of the C. maltaromaticum strains forming a clonal complex. Each RST is represented by a circle, the size of which is proportional to the number of strains exhibiting the subjected RST genotype. The number of strains is also indicated in brackets when there is more than one strain. Solid lines connect single locus variants, i.e. strains differing by one locus. The population is characterized by two RST-clonal complexes: RST-CC1 and RST-CC2. Singletons are not represented.

By combining the alleles from each VNTR locus carried by each strain, Repeat based Sequence Type (RST) was delineated. Combining VNTR loci allowed reaching a discrimination index value of 0.94. Fifteen RSTs were defined on the basis of the 24 analyzed strains, whereas the MLST scheme allowed identifying 16 STs with the same set of strains. This result suggests that the MLST scheme has a slightly higher discriminatory power than the RST scheme used in this study. Nevertheless, several strains exhibiting the same ST were found to exhibit different RSTs. This is exemplified with genotype ST7, which is represented by strains N15 and F42 that are differentiated with RST as RST8, and RST3 respectively (Table 1, Fig. 2). Similarly ST3, which is shared by 5 different strains is divided into RST-2, -3 and -4 respectively. Conversely, several strains that were not differentiated by RST could be differentiated by MLST. Thus, strains with genotype RST6 were described to exhibit ST4, ST11, and ST13 (Fig. 2). The relatedness between RST was estimated by using allelic profile comparison with eBURST as previously described (Dimovski et al., 2014). This allowed identifying two RST-clonal complexes (RST-CC, Fig. 3). RST-CC1 and RST-CC2 are exhibited by strains showing MLST ST from clonal complexes CC1, and CC2, respectively (Fig. 2), which are the major clonal complexes identified in C. maltaromaticum. 4. Discussion In this paper, a new method for strain typing in the species C. maltaromaticum based on the analysis of 3 VNTR loci is described. Analyzing the PCR products by agarose gel electrophoresis ended in low resolution. However, comparing the PCR products at the sequence level allowed significantly increasing the discriminatory power. This is likely due to (i) the small size of the repeat units in the case of VNTRA, and (ii) the difference between alleles that can be due to point mutations. The discriminatory power of this RST method is comparable to that of MLST. The main result is that RST and MLST schemes for C. maltaromaticum are complementary. Indeed, several strains exhibiting the same MLST genotype could be differentiated with RST and reciprocally. This is contrasting with other bacteria such as O. oeni for which a higher resolution was obtained with the MLVA scheme (Claisse and Lonvaud-Funel, 2012). The comparison of RSTs by using eBURST revealed closely related genotypes forming two RST-clonal complexes. These clonal complexes mirror the clonal complexes previously identified by MLST (Rahman et al., 2014a). The strains populating these clonal complexes are mainly strains originating from dairy products. It can be speculated that these strains are particularly well adapted to the dairy environment which would result in the clonal expansion of these lineages. Acknowledgment We thank Angélique Behr, Myriam Michel, Sylvie Wolff, and Arnaud Khemisti for technical assistance. We thank the reviewers for criticisms that significantly improved the manuscript. We are also grateful to the National Infrastructure “France Génomique”.

4

A. Rahman et al. / International Journal of Food Microbiology 226 (2016) 1–4

References Afzal, M.I., Jacquet, T., Delaunay, S., Borges, F., Millière, J.-B., Revol-Junelles, A.-M., CailliezGrimal, C., 2010. Carnobacterium maltaromaticum: identification, isolation tools, ecology and technological aspects in dairy products. Food Microbiol. 27, 573–579. http:// dx.doi.org/10.1016/j.fm.2010.03.019. Afzal, M.I., Gonzalez Ariceaga, C.C., Lhomme, E., Kamel Ali, N., Payot, S., Burgain, J., Gaiani, C., Borges, F., Revol-Junelles, A.M., Delaunay, S., Cailliez-Grimal, C., 2013. Characterization of Carnobacterium maltaromaticum LMA 28 for its positive technological role in soft cheese making. Food Microbiol. 36, 223–230. Azuma, T., Bagenda, D.K., Yamamoto, T., Kawai, Y., Yamazaki, K., 2007. Inhibition of Listeria monocytogenes by freeze-dried piscicocin CS526 fermentate in food. Lett. Appl. Microbiol. 44, 138–144. http://dx.doi.org/10.1111/j.1472-765X.2006.02054.x. Benson, G., 1999. Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res. 27, 573–580. http://dx.doi.org/10.1093/nar/27.2.573. Bourdichon, F., Casaregola, S., Farrokh, C., Frisvad, J.C., Gerds, M.L., Hammes, W.P., Harnett, J., Huys, G., Laulund, S., Ouwehand, A., Powell, I.B., Prajapati, J.B., Seto, Y., Schure, E.T., Van Boven, A., Vankerckhoven, V., Zgoda, A., Tuijtelaars, S., Hansen, E.B., 2012. Food fermentations: microorganisms with technological beneficial use. Int. J. Food Microbiol. 154, 87–97. Brillet, A., Pilet, M.F., Prevost, H., Bouttefroy, A., Leroi, F., 2004. Biodiversity of Listeria monocytogenes sensitivity to bacteriocin-producing Carnobacterium strains and application in sterile cold-smoked salmon. J. Appl. Microbiol. 97, 1029–1037. Cailliez-Grimal, C., Chaillou, S., Anba-Mondoloni, J., Loux, V., Afzal, M.I., Rahman, A., Kergourlay, G., Champomier-Vergès, M.-C., Zagorec, M., Dalgaard, P., Leisner, J.J., Prévost, H., Revol-Junelles, A.-M., Borges, F., 2013. Complete chromosome sequence of Carnobacterium maltaromaticum LMA 28. Genome Announc. 1. http://dx.doi.org/ 10.1128/genomeA.00115-12. Chmelar, D., Matusek, A., Korger, J., Durnova, E., Steffen, M., Chmelarova, E., 2002. Isolation of Carnobacterium piscicola from human pus–case report. Folia Microbiol. 47, 455–457. Claisse, O., Lonvaud-Funel, A., 2012. Development of a multilocus variable number of tandem repeat typing method for Oenococcus oeni. Food Microbiol. 30, 340–347. Delorme, C., Bartholini, C., Bolotine, A., Ehrlich, S.D., Renault, P., 2010. Emergence of a cell wall protease in the Streptococcus thermophilus population. Appl. Environ. Microbiol. 76, 451–460. Dimovski, K., Cao, H., Wijburg, O.L.C., Strugnell, R.A., Mantena, R.K., Whipp, M., Hogg, G., Holt, K.E., 2014. Analysis of Salmonella enterica serovar Typhimurium variable-number tandem-repeat data for public health investigation based on measured mutation rates and whole-genome sequence comparisons. J. Bacteriol. 196, 3036–3044. http:// dx.doi.org/10.1128/JB.01820-14. Duffes, F., Corre, C., Leroi, F., Dousset, X., Boyaval, P., 1999. Inhibition of Listeria monocytogenes by in situ produced and semipurified bacteriocins of Carnobacterium spp. on vacuum-packed, refrigerated cold-smoked salmon. J. Food Prot. 62, 1394–1403. Edima, H.C., Cailliez-Grimal, C., Revol-Junelles, A.M., Rondags, E., Milliere, J.B., 2008. Short communication: impact of pH and temperature on the acidifying activity of Carnobacterium maltaromaticum. J. Dairy Sci. 91, 3806–3813.

FDA, 2005. Agency Response Letter GRAS Notice No. GRN 000159. Feil, E.J., Li, B.C., Aanensen, D.M., Hanage, W.P., Spratt, B.G., 2004. eBURST: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J. Bacteriol. 186, 1518–1530. Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98. Hunter, P.R., Gaston, M.A., 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26, 2465–2466. Mathieu, F., Michel, M., Lebrihi, A., Lefebvre, G., 1994. Effect of the bacteriocin carnocin CP5 and of the producing strain Carnobacterium piscicola CP5 on the viability of Listeria monocytogenes ATCC 15313 in salt solution, broth and skimmed milk, at various incubation temperatures. Int. J. Food Microbiol. 22, 155–172. Nadon, C.A., Trees, E., Ng, L.K., Møller Nielsen, E., Reimer, A., Maxwell, N., Kubota, K.A., Gerner-Smidt, P., MLVA Harmonization Working Group, 2013. Development and application of MLVA methods as a tool for inter-laboratory surveillance. Eurosurveillance 18, 20565. Passerini, D., Beltramo, C., Coddeville, M., Quentin, Y., Ritzenthaler, P., Daveran-Mingot, M.L., Le Bourgeois, P., 2010. Genes but not genomes reveal bacterial domestication of Lactococcus lactis. PLoS One 5, e15306. Rahman, A., Cailliez-Grimal, C., Bontemps, C., Payot, S., Chaillou, S., Revol-Junelles, A.-M., Borges, F., 2014a. High genetic diversity among strains of the unindustrialized lactic acid bacterium Carnobacterium maltaromaticum in dairy products as revealed by multilocus sequence typing. Appl. Environ. Microbiol. 80 (13), 3920–3929. http:// dx.doi.org/10.1128/AEM.00681-14 (Jul, Epub 2014 Apr 18). Rahman, A., Gleinser, M., Lanhers, M.C., Riedel, C.U., Foligne, B., Hanse, M., Yen, F.T., Klouj, A., Afzal, M.I., Back, A., Mangavel, C., Cailliez-Grimal, C., Revol-Junelles, A.M., Borges, F., 2014b. Adaptation of the lactic acid bacterium Carnobacterium maltaromaticum LMA 28 to the mammalian gastrointestinal tract: from survival in mice to interaction with human cells. Int. Dairy J. 34, 93–99. Schobitz, R., Suazo, V., Costa, M., Ciampi, L., 2003. Effects of a bacteriocin-like inhibitory substance from Carnobacterium piscicola against human and salmon isolates of Listeria monocytogenes. Int. J. Food Microbiol. 84, 237–244. Vallenet, D., Belda, E., Calteau, A., Cruveiller, S., Engelen, S., Lajus, A., Le Fevre, F., Longin, C., Mornico, D., Roche, D., Rouy, Z., Salvignol, G., Scarpelli, C., Thil Smith, A.A., Weiman, M., Medigue, C., 2013. MicroScope–an integrated microbial resource for the curation and comparative analysis of genomic and metabolic data. Nucleic Acids Res. 41, D636–D647. Van Belkum, A., 2007. Tracing isolates of bacterial species by multilocus variable number of tandem repeat analysis (MLVA). FEMS Immunol. Med. Microbiol. 49, 22–27. http:// dx.doi.org/10.1111/j.1574-695X.2006.00173.x. Wan, J., Harmark, K., Davidson, B.E., Hillier, A.J., Gordon, J.B., Wilcock, A., Hickey, M.W., Coventry, M.J., 1997. Inhibition of Listeria monocytogenes by piscicolin 126 in milk and Camembert cheese manufactured with a thermophilic starter. J. Appl. Microbiol. 82, 273–280.