Skip to main content
NCBI home page
Applied and Environmental Microbiology logoLink to Applied and Environmental Microbiology
. 2012 Jul;78(13):4646–4653. doi: 10.1128/AEM.00931-12

Multilocus Sequence Typing Scheme for the Characterization of 936-Like Phages Infecting Lactococcus lactis

Maxim Moisan 1, Sylvain Moineau 1,✉
PMCID: PMC3370485  PMID: 22522686

Abstract

Lactococcus lactis phage infections are costly for the dairy industry because they can slow down the fermentation process and adversely impact product safety and quality. Although many strategies have been developed to better control phage populations, new virulent phages continue to emerge. Thus, it is beneficial to develop an efficient method for the routine identification of new phages within a dairy plant to rapidly adapt antiphage tactics. Here, we present a multilocus sequence typing (MLST) scheme for the characterization of the 936-like phages, the most prevalent phage group infecting L. lactis strains worldwide. The proposed MLST system targets the internal portion of five highly conserved genomic sequences belonging to the packaging, morphogenesis, and lysis modules. Our MLST scheme was used to analyze 100 phages with different restriction fragment length polymorphism (RFLP) patterns isolated from 11 different countries between 1971 and 2010. PCR products were obtained for all the phages analyzed, and sequence analysis highlighted the high discriminatory power of the MLST system, detecting 93 different sequence types. A conserved locus within the lys gene (coding for endolysin) was the most discriminative, with 65 distinct alleles. The locus within the mcp gene (major capsid protein) was the most conserved (54 distinct alleles). Phylogenetic analyses of the concatenated sequences exhibited a strong concordance of the clusters with the phage host range, indicating the clonal evolution of these phages. A public database has been set up for the proposed MLST system, and it can be accessed at http://pubmlst.org/bacteriophages/.

INTRODUCTION

Lactococcus lactis is a lactic acid bacterium widely used for the production of fermented milk products. Since lactococcal phages are ubiquitous in dairy environments, most cheese plants (if not all) have experienced some problems with phage contamination (8). L. lactis phages are currently classified into 10 genetically distinct groups (6), and they all belong to the order Caudovirales due to their genome composition (double-stranded DNA) and the presence of a tail connected to their capsids (1). Members of three lactococcal phage groups, 936, c2, and P335, are mostly found in dairy environments. A multiplex PCR method is available to promptly classify lactococcal phages into one of these three groups (19). The 936-like phages are clearly the most prevalent group associated with phage attacks in many countries (2, 17, 22, 27, 28, 33, 38). Phages of the 936 group possess a long noncontractile tail and a small isometric capsid (morphotype B1), which are characteristics of the family Siphoviridae. Until now, only strictly virulent representatives of this phage group have been isolated. Several complete genomes of the 936-like phages are also publicly available (4, 5, 11, 24, 34, 36). Genome sizes vary from 27,453 bp (phage jj50) to 32,182 bp (phage CB13). A core genome was recently determined for this lactococcal phage group, and it contained 33 open reading frames (ORFs) (34), confirming the high level of relatedness among isolates. The reason for their predominance is unknown but is likely related to the industrial use of similar high-performing L. lactis strains.

Many strategies have been developed to keep lactococcal phage populations under control, but new virulent phages are still frequently isolated. The detection of new phage “variants” is of importance to optimize L. lactis strain selection as well as starter culture rotation. For this purpose, the phage DNA restriction profile method is one of the tools used (29). This typing method is laborious and does not allow unambiguous data comparisons. Also, phage DNA extraction requires a phage lysate with a high titer (>108 PFU/ml), which can be difficult to obtain in some cases. A randomly amplified polymorphic DNA (RAPD)-PCR technique was developed, which produces unique and reproducible band patterns from phages (9). This approach has the advantage of the use of lysates rather than DNA but does not discriminate closely related phages (9). Moreover, this method does not create data that can be easily shared between laboratories.

Multilocus sequence typing (MLST) was first proposed in 1998 as a portable and universal method for the characterization of bacteria (26). Because it is based on the sequencing of four to seven housekeeping gene loci, this typing method is highly discriminatory and provides unambiguous data that can be compared. Publicly accessible MLST databases have been developed for many prokaryotes and eukaryotes and for IncN plasmids (http://pubmlst.org/). MLST schemes are already available for L. lactis strains (http://www-mlst.biotoul.fr/) (31, 32). MLST outperforms restriction- or other PCR-based typing methods, because it provides information about key features of the evolutionary history, population structure, and long-term epidemiology of microbial species (25, 43). However, to our knowledge, no MLST scheme has been elaborated for the characterization of a group of phages.

Here, we present the first MLST system to characterize phages of the 936 group infecting L. lactis. This novel typing approach appeared to be highly discriminative at differentiating the 100 phages analyzed in this study, and we suggest that it could be applied worldwide for the routine monitoring of these lactococcal phages in the dairy industry.

MATERIALS AND METHODS

Bacterial strains and phages.

L. lactis host strains and phages are listed in Table 1. Twelve reference phages from the 936 group were retrieved from the Félix d'Hérelle Reference Center for Bacterial Viruses (http://www.phage.ulaval.ca/). We also used a group of 80 phages of the 936 group isolated from whey samples from a Canadian cheese plant between 2001 and 2010. Eight additional 936-like phages, isolated from other countries between 1991 and 1997, were obtained from Danisco (Dangé St-Romain, France). Thus, a total of 100 lactococcal 936-like phages with different restriction fragment length polymorphism (RFLP) patterns were selected for MLST analyses. L. lactis host strains were grown at 30°C in M17 broth (Oxoid) supplemented with 0.5% glucose or lactose (41). Phages were propagated on their respective hosts as described elsewhere (14).

Table 1.

Bacteria and phages used in this study

Phage Yr of isolation Country of isolation Host straina ST Allelic profile
 Reference or source
terL mcp mtp tmp lys
Reference phages
    sk1 <1976 Australia MG1363 1 1 1 1 1 1 6
    bIL170 1973 France IL1403 2 2 2 2 2 2 7
    712 <1988 New Zealand MG1363 3 3 3 3 3 3 29
    jj50 1985 Denmark MG1363 4 4 3 4 4 1 29
    P008 1971 Germany IL1403 5 5 4 5 5 4 29
    bIBB29 <2007 Poland IL1403 6 6 5 6 6 5 15
    CB13 2003 Canada SMQ-404 7 7 6 7 7 6 40
    CB14 2003 Canada SMQ-404 8 8 7 8 8 6 40
    CB19 2003 Canada SMQ-404 9 8 8 9 9 6 40
    CB20 2003 Canada SMQ-404 10 7 8 9 9 6 40
    SL4 1996 Canada SMQ-404 11 9 9 10 10 7 40
    p2 <1988 United States MG1363 12 4 3 11 11 1 42
European phages
    D912 1991 France SMQ-1161 65 58 52 48 63 59 Danisco
    D1699 1995 France SMQ-1162 69 53 53 49 64 60 Danisco
    D1930 1997 France SMQ-1163 19 14 15 17 40 41 Danisco
    D1972 1997 Switzerland SMQ-1164 17 54 54 50 18 61 Danisco
    D2023 1997 France SMQ-1165 53 57 55 51 62 58 Danisco
    D2024 1997 Czech Republic SMQ-1166 21 13 56 54 53 62 Danisco
    D2047 1997 United Kingdom SMQ-1167 54 55 57 52 6 63 Danisco
    D2078 1997 United Kingdom SMQ-1168 56 56 30 53 51 64 Danisco
Canadian phages
    LS1 2001 Canada SMQ-794 29 20 23 25 24 21 This study
    LS3 2001 Canada SMQ-795 30 21 24 26 25 22 This study
    LS4 2001 Canada SMQ-796 31 22 24 26 25 23 This study
    LS5 2002 Canada SMQ-1158 32 23 1 27 26 24 This study
    LS6 2002 Canada SMQ-1159 (SMQ-743) 33 24 25 28 27 25 This study
    LS9 2002 Canada SMQ-1151 34 25 26 29 28 26 This study
    LS10 2002 Canada SMQ-1151 35 26 27 30 29 27 This study
    LS11 2002 Canada SMQ-847 36 27 28 31 30 28 This study
    LS12 2002 Canada SMQ-1155 37 28 29 32 31 29 This study
    LS14 2002 Canada SMQ-1156 39 29 25 33 33 31 This study
    LS15 2002 Canada SMQ-1157 40 30 31 15 19 11 This study
    CB1 2003 Canada SMQ-1160 (SMQ-743) 45 24 25 17 27 25 This study
    CB3 2003 Canada SMQ-859 46 27 28 31 30 33 This study
    CB5 2003 Canada SMQ-795 (SMQ-796) 84 22 24 26 46 23 This study
    CB6 2003 Canada SMQ-796 (SMQ-795) 84 22 24 26 46 23 This study
    CB7 2003 Canada SMQ-797 66 38 23 43 49 65 This study
    CB8 2003 Canada SMQ-1152 67 39 39 29 48 44 This study
    CB9 2003 Canada SMQ-1152 68 39 39 29 28 44 This study
    CB10 2003 Canada SMQ-1152 68 39 39 29 28 44 This study
    CB11 2003 Canada SMQ-1152 68 39 39 29 28 44 This study
    CB15 2003 Canada SMQ-445 81 40 40 26 47 45 This study
    CB16 2003 Canada SMQ-445 81 40 40 26 47 45 This study
    CB18 2003 Canada SMQ-441 73 41 41 19 52 46 This study
    CB21 2003 Canada SMQ-747 71 10 20 9 9 34 This study
    CB22 2003 Canada SMQ-439 47 10 20 36 12 34 This study
    CB24 2003 Canada SMQ-745 48 32 33 37 36 35 This study
    CB25 2003 Canada SMQ-745 60 33 34 38 37 42 This study
    CB28 2003 Canada SMQ-746 49 34 35 39 38 14 This study
    LS17 2003 Canada SMQ-858 41 31 32 34 34 32 This study
    LS18 2003 Canada SMQ-859 42 31 32 35 34 32 This study
    LS19 2003 Canada SMQ-860 43 31 32 34 35 32 This study
    LS20 2003 Canada SMQ-860 44 31 32 35 35 32 This study
    CB29 2004 Canada SMQ-439 50 10 10 12 12 36 This study
    CB30 2004 Canada SMQ-1001 51 14 14 40 27 30 This study
    GR5 2004 Canada SMQ-1019 72 42 42 37 42 47 This study
    GR8 2005 Canada SMQ-1017 14 11 11 13 13 9 This study
    GR12 2005 Canada SMQ-1002 (SMQ-1001) 82 14 14 16 54 30 This study
    GR13 2005 Canada SMQ-1024 (SMQ-1018) 83 49 37 42 55 48 This study
    GR14 2005 Canada SMQ-1024 (SMQ-1018) 86 49 37 42 43 48 This study
    JS1 2005 Canada SMQ-1018 61 37 37 42 43 40 This study
    JS2 2005 Canada SMQ-1001 62 14 38 16 44 41 This study
    JS3 2005 Canada SMQ-1001 59 14 38 17 45 41 This study
    JS4 2005 Canada SMQ-1001 64 14 38 17 44 41 This study
    JS5 2005 Canada SMQ-1023 (SMQ-1001) 55 14 15 17 27 38 This study
    GL2 2006 Canada SMQ-746 79 34 35 39 38 55 This study
    GL3 2006 Canada SMQ-1020 15 12 12 14 14 10 This study
    GL4 2006 Canada SMQ-1001 91 50 15 17 32 56 This study
    GL7 2006 Canada SMQ-747 13 10 10 12 12 8 This study
    JG1 2006 Canada SMQ-430 57 35 36 41 41 39 This study
    JG2 2006 Canada SMQ-745 63 32 33 37 42 35 This study
    JG4 2006 Canada SMQ-1023 (SMQ-1001) 58 36 14 16 32 30 This study
    JG5 2006 Canada SMQ-747 13 10 10 12 12 8 This study
    JG6 2006 Canada SMQ-1001 80 14 49 16 16 30 This study
    JG7 2006 Canada SMQ-1001 92 14 50 16 54 30 This study
    JG9 2006 Canada SMQ-435 93 51 51 15 39 57 This study
    DM1 2007 Canada SMQ-1005 52 11 3 15 39 10 This study
    DM2 2007 Canada SMQ-430 74 35 43 41 41 39 This study
    DM3 2007 Canada SMQ-1025 75 12 19 22 21 17 This study
    DM4 2007 Canada SMQ-1154 76 43 44 44 57 50 This study
    DM5 2007 Canada SMQ-1154 77 16 17 45 58 51 This study
    DM7 2007 Canada SMQ-1001 78 14 14 16 32 30 This study
    DM8 2007 Canada SMQ-1025 24 17 19 22 21 17 This study
    DM10 2007 Canada SMQ-1154 70 44 45 46 59 52 This study
    DM11 2007 Canada SMQ-1025 38 19 22 22 21 43 This study
    DM12 2007 Canada SMQ-1154 28 52 21 46 50 37 This study
    JG10 2007 Canada SMQ-1025 90 48 48 22 56 49 This study
    AF1 2008 Canada SMQ-1153 87 45 46 21 20 16 This study
    AF2 2008 Canada SMQ-1025 26 19 22 22 23 19 This study
    AF4 2008 Canada SMQ-337 25 18 13 23 22 18 This study
    AF5 2008 Canada SMQ-1005 16 12 13 15 15 11 This study
    GL1 2008 Canada SMQ-747 13 10 10 12 12 8 This study
    GL9 2008 Canada SMQ-1025 27 19 22 24 23 20 This study
    DM13 2008 Canada SMQ-1154 88 46 17 45 60 53 This study
    DM14 2008 Canada SMQ-1154 89 47 47 47 61 54 This study
    AF6 2009 Canada SMQ-1001 18 14 14 16 16 12 This study
    AF7 2009 Canada SMQ-746 20 15 16 18 17 13 This study
    AF8 2009 Canada SMQ-1154 22 16 17 20 19 15 This study
    AF9 2009 Canada SMQ-1153 23 12 18 21 20 16 This study
    RM1 2009 Canada SMQ-747 13 10 10 12 12 8 This study
    RM8 2010 Canada SMQ-746 85 15 16 17 18 14 This study
Open in a new tab
a

Alternative host strains are in parentheses.

MLST method.

In silico comparative analyses of 12 full-genome sequences of 936-like phages were used to identify conserved loci (phages sk1 [GenBank accession number AF011378], bIL170 [accession number AF009630], 712 [accession number DQ227763], jj50 [accession number DQ227764], P008 [accession number DQ054536], bIBB29 [accession number EU221285], SL4 [accession number FJ848881], CB13 [accession number FJ848882], CB14 [accession number FJ848883], CB19 [accession number FJ848884], CB20 [accession number FJ848885], and p2 [accession number GQ979703]). These loci were selected according to their high degree of nucleotide identity, the putative functions of the proteins which they encode, their genomic module, and their length. Ten loci were first chosen (terS, terL, por, mcp, mtp, tmp, bpl, rbp, lys, and orf33p2) to develop a preliminary MLST method. Degenerate primers (Table 2) with the same annealing temperature were designed based on highly conserved genomic regions within each locus. The selected conserved primers were also flanking a relatively polymorphic region of 200 to 500 bp in length.

Table 2.

Ten MLST loci used in this study and primers used to amplify and sequence them

Locusa Geneb Gene size (bp) Gene product Primer Sequence (5′→3′)c Tm (°C)d Positionse PCR product size (bp)
terS orf1 525 Terminase small subunit terS-F TAAGATGTAYGAAGAACCGC 50.9 312–716 405
terS-R TTAAGGTCATKAGCGCTTG 51.4
terL orf2 1,623 Terminase large subunit terL-F GATGATTTTAGGCGGACAAT 50.5 1149–1732 584
terL-R AGAACTTATTCTGTAACGCTG 50.1
por orf4 1,137 Portal protein por-F ATTCAAACTAAGCTGGAACA 49.5 3185–3646 462
por-R TTCTTTCAAAGTTGCAAACTT 49.1
mcp orf6 1,182 Major capsid protein mcp-F GCAAAACTTGCTGAAAATGG 51.0 4675–5259 585
mcp-R CTCATCTAACAAGGCTTTACG 51.0
mtp orf11 906 Major tail protein mtp-F TARCTGGTTTAGTATCRGTTGG 51.9 6942–7296 355
mtp-R ACTGAWTCTGTTTCTGATTCTT 49.7
tmp orf14 3,000 Tail measure protein tmp-F AAYTTACAAACGCAGTTGG 50.0 8767–9358 592
tmp-R CAAAYGCTTGGTTAGCTTTTA 50.5
bpl orf16 1,128 Baseplate protein bpl-F GACACGGAAACAATARYAGAAC 51.7 13178–13602 425
bpl-R GGYTTGTCTCCACTTTCTAC 51.7
rbp orf18 795 Receptor-binding protein rbp-F CCAGTCGGTKCTAATAATGA 50.5 13936–14788 853
rbp-R GGTTCATYGCTTCTCTATCTT 51.0
lys orf20 762 Endolysin lys-F AACAAACGGAGGATAAAAAAG 49.0 15032–15511 480
lys-R AGTTCTGYRATAAARTATGAGC 49.4
orf33 orf33 216 Unknown orf33-F TCGCCAAGAATTTCATCAG 50.4 19647–20145 499
orf33-R GAATCAGGCGARTACGTAAA 51.1
Open in a new tab
a

Loci shown in boldface type compose the final MLST scheme.

b

Gene name according to the phage p2 genome.

c

A, deoxyadenosine; C, deoxycytidine; G, deoxyguanosine; T, thymidine; K, T+G; R, A+G; W, A+T; Y, C+T.

d

Tm, melting temperature.

e

Numbers denote the positions of the first and last bases of the locus on the phage p2 genome.

PCR and sequencing.

PCRs were performed by the addition of 1 μl of phage lysate (106 PFU/ml or more) as the template to 49 μl of a reaction mixture containing 125 μM each deoxynucleoside triphosphate (dNTP), 1 μM each primer (Invitrogen), 2.5 U of Taq DNA polymerase (Invitrogen), 1× Taq buffer, 1.5 mM MgCl2, and PCR-grade H2O (Sigma-Aldrich, Mississauga, Ontario, Canada). A negative control (without a template) and a positive control (phage p2 lysate) were included. PCR conditions were optimized on a Robocycler Gradient 40 apparatus (Stratagene, La Jolla, CA) and were set as follows: an initial denaturation step at 94°C for 5 min; 35 cycles at 94°C for 30 s, 50°C for 30 s, and 72°C for 30 s; and a final extension step at 72°C for 5 min. The PCR products were separated on a 2% agarose gel in 40 mM Tris-acetate–1 mM EDTA (TAE) buffer, stained with EZ-Vision Three (Amresco, Solon, OH), and visualized under UV light. PCR products were sent to the Plateforme de Séquençage et de Génotypage des Génomes at the CHUL/CHUQ Research Center for sequencing using the ABI 3730XL DNA analyzer. Both DNA strands of the amplicons were sequenced by using the same primer pairs as those used for PCR amplification. Forward and backward sequences were aligned, analyzed, and trimmed by using the Staden 1.6.0 package (37).

Allele and sequence type assignment.

The nucleotide sequences obtained for each locus were aligned and compared by using the ClustalW multiple-alignment tool (42) implemented in BioEdit software, version 7.0.9.0. The consensus length and position of each locus were then determined and assigned to the appropriate reading frame. An arbitrary number was assigned to each distinct allele at a locus. Each unique allelic profile, defined by the allele numbers of each locus analyzed, was assigned a sequence type (ST). The same ST was used for different isolates if they shared the same allelic profile. The data were deposited into the MLST database (16).

Descriptive and phylogenetic analyses of MLST data.

Genetic diversity information such as the mean percent G+C content, the number of polymorphic sites, the nucleotide diversity (i.e., the average number of nucleotide differences per site between any two DNA sequences chosen randomly from the sample population) (π), and the mean number of nucleotide differences per sequence (k) were calculated for each locus by using DnaSP, version 5.10.1 (21). The same software was also used to perform Tajima's D neutrality test (39). The dN/dS ratio described previously by Nei and Gojobori (30), where dN is the number of nonsynonymous substitutions per nonsynonymous site and dS is the number of synonymous substitutions per synonymous site, was calculated by using START2 (15) as a test for selection. Phylogenetic trees were compiled by using MEGA, version 5.05 (40). Dendrograms were constructed by the neighbor-joining (NJ) method with the Kimura two-parameter model. The percent bootstrap confidence levels for internal branches were calculated from 1,000 random resamplings. MEGA trees were then converted into Newick files and edited with iTOL, version 2.1 (20).

Recombination tests.

Evidence of linkage disequilibrium (intergenic recombination) between alleles at all loci was estimated by using the standardized index of association (IAS) (10). This statistic, which is expected to be zero when the alleles are in linkage equilibrium (free recombination), was calculated by using START2 (15). Indications of intragenic recombination were investigated by using two different approaches. Sequence alignments of each locus and the concatenated terL, mcp, mtp, tmp, and lys loci were converted into FASTA files and used to generate split decomposition trees with 1,000 bootstrap runs by using SPLITSTREE, version 4.11.3 (13). Evidence of recombinational exchanges was also examined by performing Sawyer's test (35) with START2 (15).

RESULTS

MLST scheme and characterization of 936-like phages.

Ten highly conserved phage loci (terS, terL, por, mcp, mtp, tmp, bpl, rbp, lys, and orf33) were first selected based on various factors. However, we kept only five of them (terL, mcp, mtp, tmp, and lys) for the final MLST scheme. These loci were successfully amplified and sequenced for all 100 phages listed in Table 1. PCR products could not be obtained with some phage isolates for the rejected bpl, rbp, and orf33 loci, while insertions were also observed for the orf33 locus (not shown). The terS and por loci were discarded because of their very low discriminatory power (not shown). No amplification products were obtained with phages bIL67 (c2 group) and P335 (P335 group), validating the primers' specificity for the 936 phage group (not shown). Allelic profiles and sequence types (STs) were determined from the nucleotide sequence analysis, and an ST was assigned to each phage (Table 1). The MLST scheme defined alleles between 210 bp (mtp) and 438 bp (mcp) in length and generated between 54 (mtp) and 65 (lys) distinct alleles (out of the 100 phages analyzed) per locus (Table 3). Based on the allelic profiles with the five selected loci, 93 STs were obtained from the 100 distinct phages analyzed, suggesting that the proposed MLST scheme is highly discriminative (Table 1). A public database was set up and can be accessed at http://pubmlst.org/ (16). Additional data can be uploaded for further comparisons.

Table 3.

Descriptive analysis of the MLST results obtained with five loci

Locus Fragment length (bp) No. of alleles Mean G+C content (%) No. of polymorphic sites (%) Mean π ± SDa kb dN/dSc Tajima's D valued
terL 423 58 35.6 100 (23.6) 0.054 ± 0.001 22.9 0.009 0.205
mcp 438 57 37.8 128 (29.2) 0.058 ± 0.002 25.3 0.106 −0.314
mtp 210 54 39.2 58 (27.6) 0.064 ± 0.003 13.5 0.040 0.198
tmp 435 64 40.2 180 (41.4) 0.087 ± 0.004 37.8 0.116 −0.024
lys 306 65 42.4 126 (41.2) 0.105 ± 0.003 32.3 0.080 0.747
Concatenate 1,812 39.0 592 (32.7)
Open in a new tab
a

Average number of nucleotide differences per site between any two DNA sequences chosen randomly from the sample population (nucleotide diversity). Standard deviations were included to evaluate the dispersion of diversity.

b

Mean number of nucleotide differences per sequence.

c

dN/dSrepresents the ratio of nonsynonymous to synonymous substitutions, which is indicative of selective pressure on loci.

d

D values were not significantly different from zero (P > 0.10).

Comparison of MLST and RFLP typing methods.

All 100 phages used in this study were previously characterized as being different based on RFLP patterns. According to sequencing data obtained by MLST, only four sequence types were assigned to more than one phage: ST-13 (GL1, GL7, JG5, and RM1), ST-68 (CB9, CB10, and CB11), ST-81 (CB15 and CB16), and ST-84 (CB5 and CB6). These phages have the same host range (ST-13, SMQ-747; ST-68, SMQ-1152; ST-81, SMQ-445; ST-84, SMQ-795/SMQ-796) and are probably derived from each other, since only a point mutation can give rise to a new RFLP pattern. On the other hand, when we analyzed 15 phage samples with the same RFLP pattern (GL7) by MLST, three distinct STs were obtained (not shown).

Descriptive analysis of genetic diversity at five loci.

The mean percent G+C contents of the five loci varied from 35.6% (terL) to 42.4% (lys) and corresponded to an average of 39.0% for the concatenated fragment (Table 3). Interestingly, all the loci had slightly higher percent G+C contents than the average percent G+C contents found for the complete genomes of 936-like phages available (33.9% to 35.2%) (34). Based on the 100 phages analyzed here, the numbers of polymorphic sites ranged from 58 for mtp to 180 for tmp, which correspond to 27.6% and 41.4% of the sites present in the alleles of these loci, respectively (Table 3). The nucleotide diversity among the diverse alleles (the average number of nucleotide differences per site between any two DNA sequences chosen randomly from the sample population, denoted by π) was relatively high for the five loci, as π values ranged from 0.054 (terL) to 0.105 (lys) (Table 3). On the other hand, the mean numbers of pairwise nucleotide differences per distinct allelic sequence (k) ranged from 13.5 (mtp) to 37.8 (tmp) (Table 3), demonstrating that the highest average number of nucleotide differences was within the internal region of the gene coding for the tape measure protein (tmp).

dN/dS ratios were calculated to estimate the level of selection applied to each locus (Table 3). All values were low, indicating that there was strong selective pressure against amino acid changes, as typically observed for core genes, particularly for the terL locus (Table 3). There were also approximately 10 times (mcp and tmp) to 100 times (terL) more synonymous substitutions than nonsynonymous substitutions, indicating that the selected loci were appropriate for population studies. Finally, Tajima's D test (39) was used to examine the neutrality in population genetics. The D values obtained support a neutral evolution of the distinct alleles (i.e., there was no balancing or purifying selection of certain alleles at these loci) (Table 3).

Phylogeny based on MLST data.

A neighbor-joining tree was constructed from the 1,812-bp concatenated sequence of the five loci to show the genetic relatedness among the 936-like phages investigated in this study (Fig. 1). The concatenated tree revealed two major phylogroups, strongly supported by bootstrap values, one of which contained all phages isolated from a unique Canadian cheese plant, while the second one grouped phages isolated from other countries. Only phages D1930 (ST-19) and D1972 (ST-17), isolated from France and Switzerland, respectively, belonged to another phylogroup. The tree also exhibited strong agreement with the phage host range, suggesting a clonal evolution of phages isolated within the same cheese plant and/or infecting the same strain(s). This tree was in agreement with previous genomic analyses which showed that lactococcal phages jj50, p2, and sk1 (24) were highly related, as were phages bIL170, bIBB29, and P008 (11) and phages CB13, CB19, CB20, and SL4 (34). Phages within these three subgroups were indeed clustered together. Interestingly, phage CB14, which is known to share high levels of nucleic acid identity with phages CB19 (90.3%) and CB20 (89.9%), did not cluster with them. Phage CB14 was isolated in the summer of 2003, while phages CB19 and CB20 were isolated in the fall of 2003.

Fig 1.

Fig 1

Open in a new tab

Neighbor-joining phylogenetic tree of 100 phages of the 936 phage group constructed from the concatenated sequences of the five loci included in this study. Colored labels represent the phages that have the same host strain or host range. Black branches indicate phages that were isolated from a unique Canadian cheese plant, and blue branches correspond to phages that were isolated elsewhere. Bootstrap values above 70% are indicated. The bar scale shows the number of nucleotide substitutions per site.

Intergenic and intragenic recombination analyses.

The standardized index of association (IAS) (10) was used to evaluate the relative contributions of mutation and recombination to the evolution of the loci analyzed in this study. The IAS value is expected to be close to zero for linkage equilibrium (i.e., free recombination), and a significant deviation from this value indicates a degree of linkage disequilibrium. An IAS value of 0.3623 was obtained after 1,000 computer randomizations, revealing linkage disequilibrium between the five loci used for the MLST analysis. This value also suggests that the genetic diversity seen within these five loci of lactococcal phages is due mainly to mutation (i.e., clonal population) rather than recombination. The split-decomposition method was used to examine the impact of intragenic recombination in the five loci and in the concatenated sequence. A tree-like structure is obtained when the descent is clonal, but a network-like structure or a parallelogram will be created if recombination has been involved in the evolution of the analyzed sequence. Split graphs of each locus and the concatenated sequence showed a tree-like structure, indicating that alleles of the five selected loci were not affected by intragenic recombination (see Fig. S1 in the supplemental material). The absence of intragenic recombination was also confirmed by Sawyer's test (35), implemented in START2 (15; not shown).

DISCUSSION

L. lactis cultures are regularly exposed to 936-like phages during milk fermentation processes (2, 22, 27, 38). These phages are evolving, and new phage variants keep emerging in dairy factories, possibly due to the use of antiphage strategies. The industry needs a rapid and reliable typing method to monitor the emergence of distinct phages to update control strategies. We propose here a new tool for the characterization of 936-like lactococcal phages. The MLST scheme is based on the internal regions of five genes coding for the large subunit of the terminase, the major capsid protein, the major tail protein, the tail tape-measure protein, and the endolysin. All five genes are part of the core genome of 936-like phages (34).

According to Hunter and Gaston's index (12), the discriminatory power of our current MLST method is 0.998. This value represents a probability of 99.8% that two randomly selected phages are discriminated. An index greater than 0.90 suggests that the typing results are to be interpreted with confidence (12). Therefore, our MLST method appears to be highly discriminative and reliable (12). This conclusion is also supported by the high number of distinct alleles, the nucleotide diversity (π), and the mean number of pairwise nucleotide differences per sequence (k) observed for all loci (Table 3). The MLST method succeeded in amplifying all loci of phages isolated from 11 different countries between 1971 and 2010 that infected 48 L. lactis strains.

Host range and RFLP analyses are currently the main methods used to differentiate between lactococcal phages from the same genetic group (19, 29). Host range analysis offers limited discriminatory power for closely related phages, as they often infect the same strains. The main limitation of the RFLP method is that the DNA patterns may vary with experimental conditions, which limits data reproducibility and comparison. The MLST approach provides portable nucleic acid sequences that can be shared and compared easily between different laboratories. The MLST sequence data (1,812-bp concatenated sequence), which represent approximately 6% of the complete phage genome sequence, can also differentiate phages with the same RFLP pattern.

MLST has also been used for population analyses to estimate recombination and mutation rates as well as to investigate evolutionary relationships (44). We investigated the genetic relatedness among the 936-like phages by constructing a phylogenetic tree from the concatenated sequence of the five loci. The tree revealed that the 93 STs form two major distinct phylogroups of 80 and 13 STs. The first one regrouped all phages isolated from the same Canadian cheese plant, with two additional phages isolated from France (D1930 [ST-19]) and Switzerland (D1972 [ST-17]). It is not surprising that distinct phages from the same cheese plant are genetically related, since a dairy plant is a specialized ecological niche where phages coevolve with their abundant and repeatedly used L. lactis hosts. The second phylogroup was more heterogeneous and contained only lactococcal phages isolated from other countries, which are likely using different manufacturing conditions as well as distinct starter cultures.

Although horizontal gene transfer plays a pivotal role in phage evolution, many phage genomes are clearly shaped by vertical evolution (3). This seems particularly true for the 936-like phages, because the phylogenetic tree exhibited a strong concordance of the clusters with the host range, indicating clonal evolution. Moreover, the five selected loci are expressed late during the phage infection process (4). Late genes encode mostly proteins for virion assembly and cell lysis (7, 23). Those genes tend to diverge by point mutation, in comparison to early and middle genes (5, 11, 24, 34). This was also recently observed for Staphylococcus aureus siphophages (18). Accordingly, the split graphs showed a tree-like structure, confirming that the population is not affected by extensive intragenic recombination at these loci.

MLST should be useful for phage monitoring in the dairy industry, as it can be performed directly with filtered phage lysates. Phage concentration, DNA extraction, and DNase I treatment are not required. Because the primers were designed to have the same annealing temperature, the five MLST PCRs can be accomplished in a single PCR run. As more phage samples are tested, it is possible that some loci may not be amplified by PCR due to sequence variability. A reduction of the hybridization temperature could be the easiest way to circumvent the problem, but modification of the primer(s) at the 3′ end could constitute another alternative.

In conclusion, this MLST scheme is applicable for characterizing 936-like phages infecting L. lactis strains worldwide. The chosen loci have been shown to be robust molecular markers to differentiate between closely related phages from the same cheese plants as well as diversified phages. This represents the first MLST scheme for the characterization of a group of phages.

Supplementary Material

Supplemental material
supp_78_13_4646__index.html (1.1KB, html)

ACKNOWLEDGMENTS

We thank Claudia Bergeron, Marie-Ève Dupuis, Audrey Fleury, Josiane Garneau, Simon Labrie, Geneviève Lacasse, Bruno Martel, Rym Menasria, Geneviève Rousseau, and Julie Samson for lactococcal phage isolation over the years. We are grateful to Maryse Lamoureux (Agropur) for providing whey samples, Christophe Fremaux (Danisco) for some phages, Pascal Le Bourgeois for discussion, and Keith Jolley for setting up the MLST database.

M.M. is the recipient of scholarships from the Natural Sciences and Engineering Research Council (NSERC) of Canada and from the Fonds de Recherche sur la Nature et les Technologies (FQRNT). S.M. acknowledges funding from the NSERC through its strategic program. S.M. holds a tier 1 Canada Research Chair in bacteriophages.

Footnotes

Published ahead of print 20 April 2012

Supplemental material for this article may be found at http://aem.asm.org/.

REFERENCES

  • 1. Ackermann HW. 1998. Tailed bacteriophages: the order Caudovirales. Adv. Virus Res. 51:135–201 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Bissonnette F, Labrie S, Deveau H, Lamoureux M, Moineau S. 2000. Characterization of mesophilic mixed starter cultures used for the manufacture of aged cheddar cheese. J. Dairy Sci. 83:620–627 [DOI] [PubMed] [Google Scholar]
  • 3. Ceyssens PJ, et al. 2011. Phenotypic and genotypic variations within a single bacteriophage species. Virol. J. 8:134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Chandry PS, Moore SC, Boyce JD, Davidson BE, Hillier AJ. 1997. Analysis of the DNA sequence, gene expression, origin of replication and modular structure of the Lactococcus lactis lytic bacteriophage sk1. Mol. Microbiol. 26:49–64 [DOI] [PubMed] [Google Scholar]
  • 5. Crutz-Le Coq AM, Cesselin B, Commissaire J, Anba J. 2002. Sequence analysis of the lactococcal bacteriophage bIL170: insights into structural proteins and HNH endonucleases in dairy phages. Microbiology 148:985–1001 [DOI] [PubMed] [Google Scholar]
  • 6. Deveau H, Labrie SJ, Chopin MC, Moineau S. 2006. Biodiversity and classification of lactococcal phages. Appl. Environ. Microbiol. 72:4338–4346 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Duplessis M, Russell WM, Romero DA, Moineau S. 2005. Global gene expression analysis of two Streptococcus thermophilus bacteriophages using DNA microarray. Virology 340:192–208 [DOI] [PubMed] [Google Scholar]
  • 8. Emond E, Moineau S. 2007. Bacteriophages and food fermentations, p 93–124 In McGrath S, van Sinderen D. (ed), Bacteriophage: genetics and molecular biology. Horizon Scientific Press/Caister Academic Press, Norfolk, United Kingdom [Google Scholar]
  • 9. Gutierrez D, et al. 2011. Typing of bacteriophages by randomly amplified polymorphic DNA (RAPD)-PCR to assess genetic diversity. FEMS Microbiol. Lett. 322:90–97 [DOI] [PubMed] [Google Scholar]
  • 10. Haubold B, Travisano M, Rainey PB, Hudson RR. 1998. Detecting linkage disequilibrium in bacterial populations. Genetics 150:1341–1348 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Hejnowicz MS, Golebiewski M, Bardowski J. 2009. Analysis of the complete genome sequence of the lactococcal bacteriophage bIBB29. Int. J. Food Microbiol. 131:52–61 [DOI] [PubMed] [Google Scholar]
  • 12. 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 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Huson DH, Bryant D. 2006. Application of phylogenetic networks in evolutionary studies. Mol. Biol. Evol. 23:254–267 [DOI] [PubMed] [Google Scholar]
  • 14. Jarvis AW. 1978. Serological studies of a host range mutant of a lactic streptococcal bacteriophage. Appl. Environ. Microbiol. 36:785–789 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Jolley KA, Feil EJ, Chan MS, Maiden MC. 2001. Sequence type analysis and recombinational tests (START). Bioinformatics 17:1230–1231 [DOI] [PubMed] [Google Scholar]
  • 16. Jolley KA, Maiden MC. 2010. BIGSdb: scalable analysis of bacterial genome variation at the population level. BMC Bioinformatics 11:595 doi:10.1186/1471-2105-11-595 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Josephsen J, et al. 1994. An ecological study of lytic bacteriophages of Lactococcus lactis subsp. cremoris isolated in a cheese plant over a five year period. Int. Dairy J. 4:123–140 [Google Scholar]
  • 18. Kahankova J, et al. 2010. Multilocus PCR typing strategy for differentiation of Staphylococcus aureus siphoviruses reflecting their modular genome structure. Environ. Microbiol. 12:2527–2538 [DOI] [PubMed] [Google Scholar]
  • 19. Labrie S, Moineau S. 2000. Multiplex PCR for detection and identification of lactococcal bacteriophages. Appl. Environ. Microbiol. 66:987–994 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Letunic I, Bork P. 2011. Interactive Tree of Life v2: online annotation and display of phylogenetic trees made easy. Nucleic Acids Res. 39:W475–W478 doi:10.1093/nar/gkr201 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Librado P, Rozas J. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25:1451–1452 [DOI] [PubMed] [Google Scholar]
  • 22. Madera C, Monjardin C, Suarez JE. 2004. Milk contamination and resistance to processing conditions determine the fate of Lactococcus lactis bacteriophages in dairies. Appl. Environ. Microbiol. 70:7365–7371 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Madsen PL, Hammer K. 1998. Temporal transcription of the lactococcal temperate phage TP901-1 and DNA sequence of the early promoter region. Microbiology 144:2203–2215 [DOI] [PubMed] [Google Scholar]
  • 24. Mahony J, et al. 2006. Sequence and comparative genomic analysis of lactococcal bacteriophages jj50, 712 and P008: evolutionary insights into the 936 phage species. FEMS Microbiol. Lett. 261:253–261 [DOI] [PubMed] [Google Scholar]
  • 25. Maiden MC. 2006. Multilocus sequence typing of bacteria. Annu. Rev. Microbiol. 60:561–588 [DOI] [PubMed] [Google Scholar]
  • 26. Maiden MC, et al. 1998. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc. Natl. Acad. Sci. U. S. A. 95:3140–3145 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Moineau S, et al. 1996. Isolation and characterization of lactococcal bacteriophages from cultured buttermilk plants in the United States. J. Dairy Sci. 79:2104–2111 [Google Scholar]
  • 28. Moineau S, Fortier J, Ackermann HW, Pandian S. 1992. Characterization of lactococcal bacteriophages from Quebec cheese plants. Can. J. Microbiol. 38:875–882 [Google Scholar]
  • 29. Moineau S, Pandian S, Klaenhammer TR. 1994. Evolution of a lytic bacteriophage via DNA acquisition from the Lactococcus lactis chromosome. Appl. Environ. Microbiol. 60:1832–1841 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Nei M, Gojobori T. 1986. Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol. Biol. Evol. 3:418–426 [DOI] [PubMed] [Google Scholar]
  • 31. Passerini D, et al. 2010. Genes but not genomes reveal bacterial domestication of Lactococcus lactis. PLoS One 5:e15306 doi:10.1371/journal.pone.0015306 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Rademaker JLW, et al. 2007. Diversity analysis of dairy and nondairy Lactococcus lactis isolates, using a novel multilocus sequence analysis scheme and (GTG)5-PCR fingerprinting. Appl. Environ. Microbiol. 73:7128–7137 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Raiski A, Belyasova N. 2009. Biodiversity of Lactococcus lactis bacteriophages in the Republic of Belarus. Int. J. Food Microbiol. 130:1–5 [DOI] [PubMed] [Google Scholar]
  • 34. Rousseau GM, Moineau S. 2009. Evolution of Lactococcus lactis phages within a cheese factory. Appl. Environ. Microbiol. 75:5336–5344 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Sawyer S. 1989. Statistical tests for detecting gene conversion. Mol. Biol. Evol. 6:526–538 [DOI] [PubMed] [Google Scholar]
  • 36. Sciara G, et al. 2010. Structure of lactococcal phage p2 baseplate and its mechanism of activation. Proc. Natl. Acad. Sci. U. S. A. 107:6852–6857 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Staden R. 1996. The Staden sequence analysis package. Mol. Biotechnol. 5:233–241 [DOI] [PubMed] [Google Scholar]
  • 38. Szczepanska AK, Hejnowicz MS, Kolakowski P, Bardowski J. 2007. Biodiversity of Lactococcus lactis bacteriophages in Polish dairy environment. Acta Biochim. Pol. 54:151–158 [PubMed] [Google Scholar]
  • 39. Tajima F. 1989. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123:585–595 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Tamura K, et al. 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28:2731–2739 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Terzaghi BE, Sandine WE. 1975. Improved medium for lactic streptococci and their bacteriophages. Appl. Microbiol. 29:807–813 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Thompson JD, Higgins DG, Gibson TJ. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673–4680 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Turner KM, Feil EJ. 2007. The secret life of the multilocus sequence type. Int. J. Antimicrob. Agents 29:129–135 [DOI] [PubMed] [Google Scholar]
  • 44. Urwin R, Maiden MC. 2003. Multi-locus sequence typing: a tool for global epidemiology. Trends Microbiol. 11:479–487 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental material
supp_78_13_4646__index.html (1.1KB, html)
supp_78.13.4646_AEM-AEM00931-12-s01.pdf (1.1MB, pdf)

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES