• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. Oct 2010; 76(20): 6843–6852.
Published online Aug 27, 2010. doi:  10.1128/AEM.00796-10
PMCID: PMC2953028

Characterization of Lactococcus lactis Phage 949 and Comparison with Other Lactococcal Phages[down-pointing small open triangle]

Abstract

The virulent Lactococcus lactis phage 949 was isolated in 1975 from cheese whey in New Zealand. This phage is a member of the Siphoviridae family and of a rare lactococcal phage group that bears its name (949 group). It has an icosahedral capsid (79-nm diameter) and a very long noncontractile tail (length, 500 nm; width, 12 nm). It infected 7 of 59 tested L. lactis strains, a somewhat expanded host range for a rare lactococcal phage. The abortive phage infection defense mechanisms AbiQ and AbiT strongly inhibited the multiplication of phage 949, but AbiK and AbiV did not. Its double-stranded DNA (dsDNA) genome of 114,768 bp is, to date, the largest among lactococcal phages. Its GC content was calculated at 32.7%, which is the lowest reported for a lactococcal phage. Its 154 open reading frames (ORFs) share limited identity with database sequences. In addition, terminal redundancy was observed as well as the presence of six tRNAs, one group I intron, and putative recombinases. SDS-PAGE coupled with mass spectrometry identified 13 structural proteins. The genomes of the members of the 10 currently known L. lactis phage groups were used to construct a proteomic tree. Each L. lactis phage group separated into distinct genetic clusters, validating the current classification scheme. Of note, members of the polythetic P335 groups were clearly separated into subgroups.

Lactococcus lactis is widely used to produce an array of fermented dairy products. Virulent phages ubiquitous in dairy factories are still the main cause of milk fermentation failures or inconsistencies leading to economic losses and low-quality products (34). Many strategies were developed to control phage populations in this ecological niche populated with high host densities. However, new lactococcal phages are still frequently isolated. They also persist for a long time in dairy environments (35, 43). Their fast evolution rate and their high adaptation capacity are accountable for the vast diversity observed among these bacterial viruses (27).

Lactococcal phages isolated to date are members of the Caudovirales order since their genomes are composed of double-stranded DNA (dsDNA) and their capsids are connected to a tail (1). They are also included in two of the three families of the Caudovirales order, namely the Siphoviridae (long noncontractile tail) and Podoviridae (short noncontractile tail) families. Overall, they are separated into 10 genetically distinct groups based on their morphology and DNA-DNA hybridization (10). All these lactococcal groups contain virulent phages except the P335 group, which contains virulent and temperate phages. Only three phage groups, 936, c2, and P335, are regularly associated with milk fermentation problems worldwide (5, 23, 33, 36, 38). Excluding remnant prophages identified in sequenced L. lactis genomes, over 30 complete lactococcal phage genomes have been sequenced and are available in public databases. Twelve of them are from the 936 group, two from the c2 group, 12 from the P335 group, while the rest come from lactococcal phages that belong to rarely isolated groups, namely KSY1, P087, Q54, 1358, 1706, and ascc[var phi]28 (the P034 group). In total, the genomic sequence is available for at least one member of nine out of the 10 lactococcal phage groups.

Comparative genome analysis has confirmed phage diversity and has hinted at evolutionary processes. Members of the 936 and c2 groups are genetically similar, while the P335 group qualifies as a polythetic species (10, 43). Indeed, the latter group is composed of interconnected isolates with shared properties or modules, but no single feature is shared by all known members of this group (10, 43). Phage Q54 has apparently acquired genetic modules from the 936- and c2-like phages (19), while KSY1 possesses a transcriptional system similar to that of T7-like phages (8). Lactococcal phages P087 (48) and 1706 (20) possess similarities with prophages found in other Firmicutes. Phage ascc[var phi]28 (P034) (24) has homology with Streptococcus and Bacillus phages. Finally, the recently characterized lactococcal phage 1358 (14) shares protein similarities with Listeria phages. Thus, members of these rare phage groups were likely the result of illegitimate recombination between different lactococcal phages and phages infecting other bacteria (14), suggesting that the phage gene pool is globally connected. In fact, this plasticity of phage genomes is likely needed to adapt to changing environments (40). Common lactococcal phages (such as 936-like phages) are highly optimized and adapted to rapidly propagate during large-scale milk fermentation. It is assumed that evolutionary pressure has led to refined replication and assembly machineries to allow such rapid amplification (17). On the other hand, rare lactococcal phages are less fit in this domesticated ecosystem (14). The remarkable diversity of lactococcal phages was likely driven by the use of a wide variety of host strains that were selected for specific industrial phenotypes.

Here, we report the complete genome sequence and analysis of the virulent phage 949, a representative of the 10th lactococcal phage group currently known.

MATERIALS AND METHODS

Bacterial strains and phage.

Phage 949 was obtained from the Felix d'Hérelle Reference Center for Bacterial Viruses (www.phage.ulaval.ca). L. lactis strains were growth at 30°C in M17 broth (Oxoid or Difco) supplemented with 0.5% glucose. For phage 949 propagation, host cells (L. lactis SMQ-385) were grown at 30°C to an optical density at 600 nm of 0.1, and then approximately 105 phages and 10 mM CaCl2 (final concentration) were added. Incubation was continued at room temperature until complete bacterial lysis, and the resulting lysate was filtered using a 0.45-μm syringe filter. To facilitate phage enumeration, 0.5% glycine was added to the top agar and the petri dishes were incubated overnight at room temperature (30). To obtain highly concentrated phage preparations, 1 liter of phage lysate was mixed with polyethylene glycol (10%) and separated on a discontinuous CsCl gradient followed by a continuous CsCl gradient, as described previously (44). The first centrifugation was performed at 35,000 rpm for 3 h in a Beckman SW41 Ti rotor and the second at 60,000 rpm for 18 h in a Beckman NVT65 rotor.

Microbiological assays.

A one-step growth curve assay was carried out in triplicate as previously described (37) with a multiplicity of infection of 0.05 and at a temperature of 22°C. The burst size was calculated by dividing the average phage titer after the exponential phase by the average titer before the infected cells began to release virions (37). The host range of phage 949 was determined by spotting 10 μl of a phage preparation (diluted at 10−2 to obtain a concentration of approximately 107 PFU/ml) on soft agar containing an L. lactis strain. In total, 59 laboratory and industrial strains were tested. High-copy-number plasmids containing either AbiQ (15), AbiT (7), AbiK (16), or AbiV (21) were transformed into L. lactis SMQ-385 by electroporation (12). The sensitivity of phage 949 to these phage resistance mechanisms was evaluated by calculating the efficiency of plaquing (EOP) using the following formula: the titer of the phage on the Abi-positive strain divided by the titer of the phage on the strain without the mechanism.

DNA sequencing and sequence analysis.

Phage DNA was isolated using the Lambda Maxi kit (Qiagen) with modifications reported elsewhere (11). Genome sequencing was performed at the Genome Quebec Center (McGill University, Montreal, Quebec, Canada) using a GS-FLX apparatus (Roche) and the 454 pyrosequencing technique. Overall, 22,924 reads were generated and assembled into a single contig with a coverage of 47×. The genomic sequence was completed by primer sequencing at the genomic platform of the Centre Hospitalier de l'Université Laval using an ABI Prism 3100 apparatus. Contig assembly and sequence editing were performed using the Staden package program (http://staden.sourceforge.net/) and BioEdit software (http://www.mbio.ncsu.edu/BioEdit/bioedit.html). Open reading frames (ORFs) were identified using the GenMark program (32) and ORFinder (http://www.ncbi.nlm.nih.gov/projects/gorf/). An ORF was considered valid if it had AUG, UUG, or GUG as the starting codon, possessed at least 29 amino acids (aa), and was preceded by an L. lactis Shine-Dalgarno sequence (AGAAAGGAGGT) (6). Function was attributed to an ORF by comparison of the translated products with the BLASTp (NCBI, http://blast.ncbi.nlm.nih.gov/Blast.cgi) and ACLAME (http://aclame.ulb.ac.be/) databases. Those results were reinforced with the search for protein functional domains using the NCBI Conserved Domain Database (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) and EMBL InterProScan sequence search (http://www.ebi.ac.uk/Tools/InterProScan/). Theoretical molecular masses (MM) and isoelectric points (pI) of the proteins were obtained using the ProtParam tool available on the ExPASy Web page (http://ca.expasy.org/tools/protparam.html). tRNAs were identified using the tRNAscan-SE server (31) and confirmed using the ARAGORN program (29) as well as by nucleotide comparison using BLASTn (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Codon usage was determined using the codon usage tool accessible through the DNA 2.0 Web server (DNA 2.0, Menlo Park, CA) and the Countcodon program available on the Kazusa DNA Research Institute Web page (http://www.kazusa.or.jp/codon/). The frequency per thousand codons was determined for each ORF individually as well as for the complete genome of 949. The bacterial codon usage for the host L. lactis IL1403 was obtained from the Kazusa DNA Research Institute database.

Electron microscopy.

A 1.5-ml sample of phage lysate (titer of at least 109 PFU/ml) was centrifuged at 23,500 × g for 1 h at 4°C. The supernatant was removed, leaving approximately 100 μl in the tube. The phage pellet was washed twice with 1.5 ml of ammonium acetate (0.1 M, pH 7.5). The residual volume (100 μl) was used to prepare the observation grid as follows. Ten microliters of the staining solution (2% phosphotungstic acid, pH 7.0) was deposited on a Formvar carbon-coated grid (200 mesh; Pelco International). After 30 s, 10 μl of the washed lysate was mixed with the stain by pipetting up and down. After 90 s, the residual liquid was removed from the grid by touching the edge with blotting paper. Phages were observed at 80 kV using a JEOL 1230 transmission electron microscope.

Structural protein identification.

Purified phages (~1011 PFU/ml) were mixed with 4× loading buffer (0.250 M Tris-HCl [pH 6.8], 40% [vol/vol] glycerol, 8% [wt/vol] SDS, 20% [vol/vol] β-mercaptoethanol, and 0.1% [wt/vol] bromophenol blue) and boiled for 5 min. The sample was sonicated for 5 s (output control, 4; duty cycle, 40%) with an ultrasonic Sonifier W-350 cell disrupter (Branson Sonic Power Co.). Proteins were then separated by migration on a 12% SDS-polyacrylamide gel (1.5 mm thick). Protein bands were detected using Coomassie blue staining and cut from the gel for protein identification. Trypsin digestion of the proteins followed by liquid chromatography/tandem mass spectrometry (LC/MS-MS) was performed at the proteomic platform of the Centre de Génomique du Québec (Quebec City, Quebec, Canada).

Proteomic tree design.

Lactococcal phage genomes were downloaded from the NCBI databases, and all the proteins were extracted using the published annotations. Each protein concatemer was arranged to start with the terminase whenever possible, and the order of proteins was identical to the gene order as it is presented in the phage genome. Multiple alignments were performed using MEGA 4.1 software and a BLOSSUM matrix (25). A phylogenetic tree was constructed using the neighbor-joining method and the number of differences model. The tree was constructed to be unrooted, and the parameters used to test the phylogeny were bootstraps from 100 replicates with a random seed.

Nucleotide sequence accession number.

A sequence was submitted to the GenBank database under the accession number HM029250.

RESULTS AND DISCUSSION

Characteristics of phage 949 and its lytic cycle.

Phage 949 was isolated in 1975 from cheese whey in New Zealand (22). This dsDNA phage is a member of the Siphoviridae family and has a B1 morphotype (Fig. (Fig.1).1). It possesses an icosahedral capsid of 79 ± 7 nm (mean ± standard deviation) and a long noncontractile tail of 500 ± 27 nm in length and 12 ± 1 nm in width (10). By far, phage 949 has the longest tail observed for a lactococcal phage. To our knowledge, very few phages have such a long tail. Those that do include Lactobacillus plantarum phage B2 (Siphoviridae) (500-nm noncontractile tail), Bacillus cereus phage Bace-11 (Myoviridae) (485-nm extended-tail length), and Bacillus megaterium phage G (Myoviridae) (tail length of 455 nm) (2, 3, 13, 39).

FIG. 1.
Phage electron micrographs. (A) Lactococcal phage 949; (B) phage 949 lysate intentionally spiked with lactococcal phage p2 (936 group).

The infection cycle of phage 949 was evaluated during growth on its host strain, L. lactis SMQ-385, which is also known as L. lactis ML8. The latency period was calculated at 71 ± 4 min, and the burst size was 78 ± 6 PFU per infected cell. These data indicate that the burst size of phage 949 is similar to that of other lactococcal phages (14) but its latency period is somewhat longer compared to those of the most common L. lactis phages, which generally have a latency period ranging from 20 to 60 min. This extended latency period may contribute, at least in part, to its paucity in industrial dairy environments.

The lytic spectrum of 949 was then determined by testing phage growth on 59 L. lactis strains, including 43 strains currently used by a Canadian cheese manufacturer. The 16 other L. lactis strains were obtained from international collaborators and are either sensitive or used as the host for reference lactococcal phages covering the 10 known DNA homology groups (10). Phage 949 infects seven of those 59 strains, including only one industrial strain. The other phage-sensitive strains included its host (SMQ-385) as well as L. lactis SMQ-86 and NCK203, which are also sensitive to members of the P335 group. Phage 949 also infects L. lactis SMQ-562 and SMQ-384, which are also sensitive to rare lactococcal phages Q54 and P087, respectively. Of note, the laboratory model strain L. lactis IL1403 (6), which is also sensitive to 936- and c2-like phages, was also sensitive to phage 949, opening up possibilities for further in-depth studies of its biology.

Taken together, the virulent phage 949 infects the host strains of five different lactococcal phage groups and an industrial strain. A similar host range diversity was observed for the lactococcal rare phage 1706 (20). However, this is in sharp contrast to common lactococcal phages of the 936 group, which are usually highly specific to a few host strains (46, 47). Considering this somewhat expanded host range, it is conceivable that recombination events may have occurred during coinfection which could have altered the genetic makeup of phage 949.

Natural phage resistance mechanisms can be used by the dairy industry to control phage populations, particularly to keep at bay members of the three main phage groups (936, c2, and P335) (28). We were interested in determining whether some of these intracellular antiphage barriers could block the lytic cycle of phage 949. High-copy-number plasmids containing different abortive infection mechanisms (Abi) were transformed into L. lactis SMQ-385, and an EOP was calculated to determine the efficacy of these systems against phage 949. AbiQ (15) and AbiT (7) were very efficient against 949, with EOPs of 10−8 and 10−5, respectively. Conversely, AbiK (16) and AbiV (21) were not effective against 949 (EOP = 1). A summary of the efficacies of four Abi systems on lactococcal phage groups is presented in Table Table11 .

TABLE 1.
Efficacy of four Abi systems on lactococcal phage groups

Genome analysis.

The complete genome of phage 949 was determined to be 114,768 bp, which is, to date, the largest lactococcal phage genome. The molecular GC content was calculated at 32.7%, which is lower than the GC content (35.4%) of one of its host strains, L. lactis IL1403 (6). In fact, it is the lowest reported GC content for a lactococcal phage genome, as prior to this study a GC content of 33.7% was observed for both phages 1706 (20) and asccø28 (24). Terminal redundancy was also observed, suggesting that this phage uses a pac-type mode of encapsidation, as observed for some but not all L. lactis phages.

The ORFs were identified using two bioinformatics software programs (GenMark and ORFinder) and were limited to those encoding proteins of more than 29 amino acids. Whenever possible, a Shine-Dalgarno sequence was identified and compared with the L. lactis complementary ribosome binding site (RBS) consensus sequence. In total, 154 ORFs were identified that covered 88% of the genome length, which is somewhat lower than usual for lactococcal phages (Table (Table2;2; Fig. Fig.2).2). The predominant starting codon was AUG (89%), followed by UUG (8%) and GUG (3%).

FIG. 2.
Genetic map of the phage 949 genome. Each arrow represents an open reading frame. The putative functions of the ORFs are written above the arrows, while the putative tRNAs and introns are indicated below. The gray shading represents the structural proteins ...
TABLE 2.
ORF identification, position in the phage genome, putative function, and comparison with sequences available in public databases

ORF function assignment and genomic organization.

The ORF functions were determined by comparing (BLASTp) putative protein sequences with GenBank and ACLAME databases. Those results were complemented with the identification of functional protein domains using the NCBI Conserved Domain Database and InterProScan (EMBL) bioinformatics tools. Only the ORFs that have a significant hit with proteins in databases are presented in Table Table2.2. Overall, a function could be attributed to only 37 of the 154 ORFs (24%). Limited similarity with known proteins was found for a few other phage ORFs. In fact, 84 ORFs (54.5%) had no significant identity with proteins in the databases, indicating that new viral genes/proteins were revealed by the characterization of this phage. Of note, four gene clusters (orf1 to orf28, orf29 to orf39, orf40 to orf122, and orf123 to orf154) oriented in different transcriptional directions revealed an unusual genomic organization for a lactococcal phage (Fig. (Fig.2).2). Large noncoding regions were identified between orf28 and orf29, orf40 and orf41, and orf88 and orf89. Some of those noncoding sequences were preceded by a putative endonuclease or corresponded to the orientation change of gene transcription. These transition points may also represent molecular fossils of illegitimate (nonhomologous) recombination events in the ancestry of 949 (49). While point mutations are usually considered fine-tuning tools with limited influence on the performance of the organism (40), the modular nature of phage genomes facilitates swapping by recombination, which contributes to the significant diversification of the genome (18).

DNA replication and nucleotide biosynthesis genes.

Many genes associated with DNA replication and nucleotide biosynthesis were found in the phage 949 genome. The products of genes orf116 and orf149 shared identity with putative DNA polymerases, suggesting that they represent two different subunits of this protein. Strangely, they were found in two separate gene clusters, but still it suggests that this phage may encode its own DNA polymerase instead of relying on its host.

ORF60 was similar to a replicative DNA helicase, ORF61 was related to a DNA primase, ORF62 had 97% identity with a putative single-stranded DNA (ssDNA) exonuclease of lactococcal phage CB13 (936 group), and orf94 and orf95 coded for the two subunits of a DNA gyrase. The products of six genes (orf101, orf102, orf103, orf107, orf108, and orf109) were associated with ribonucleotide diphosphate or triphosphate reductase, while many other genes could play a role in nucleotide transport (orf100), modification (orf96, orf99, and orf110), and degradation (orf52 and orf111). Interestingly, two classes of ribonucleotide reductase proteins were observed, interspersed by a group I intron gene (orf104): members of the class II ribonucleotide reductase proteins included ORF107, ORF109, and the activator ORF108, while class III included ORF101, ORF102, and the activator ORF103. The presence of two different classes of ribonucleotide reductases has already been observed in phages. The nrdB and nrdD genes found in coliphage T4 coded for class I and class III ribonucleotide reductases, respectively (45, 50). Surprisingly, both of these genes contained a group I intron. Moreover, phage 949 possessed two potential homing endonucleases (ORF40 and ORF88) normally associated with introns. Multiple endonucleases have also been identified in the lactococcal phage bIL170, but their functions in the phage multiplication cycle remain unclear (9). It was proposed that they could play a role in the replication, recombination, repair, and packaging of the phage DNA.

DNA packaging and lysis genes.

ORF147 had several characteristics in common with the phage terminase large subunit because it shared 25% identity with a similar Bacillus phage protein and possessed a typical domain associated with terminase proteins. Usually, a phage terminase is composed of two subunits (large and small), and their genes are found beside each other in the phage genome (41). Unfortunately, a gene coding for the small terminase subunit was not found in the 949 genome. However, due to the location of its gene as well as its molecular size, it is tempting to speculate that ORF148 might be the small terminase subunit.

The endolysin function was attributed to ORF55 as it shared 80% identity with the endolysin of lactococcal phage KSY1 and had a characteristic amidase domain. No protein similar to known holins was found beside this gene. Conversely, ORF124 possessed all the proprieties usually associated with a holin: an N-terminal transmembrane domain, a hydrophilic C-terminal region, and a holin domain identified by bioinformatics analysis. If ORF55 and ORF124 are the two proteins responsible for phage lysis, the distant position of their respective genes in the genome is atypical compared to other phages, but it has been seen before. Indeed, this unusual gene location has already been noticed for L. lactis phages 1706 (20) and P087 (48).

Structural proteins of 949.

As it is now a standard procedure for the characterization of new phage isolates, we identified the phage 949 structural proteome. A high-titer purified phage sample was first sonicated and then separated on a 12% SDS-polyacrylamide gel followed by Coomassie blue staining. Eleven colored bands were observed and sent for identification by mass spectrometry (Fig. (Fig.3).3). In addition, a sample of the complete phage was analyzed by LC/MS-MS to detect the proteins present in small amounts and that may not have been visualized on the gel. In the three highest bands on the SDS-PAGE gel, only two proteins were identified, ORF130 and ORF128. Since these two proteins have very high theoretical molecular masses of 357.7 kDa and 214.3 kDa, respectively, separation of both proteins on the gel could not be achieved under the conditions used. ORF130 is most likely the tail tape measure protein, and its large size is related to the very long and distinctive tail of phage 949 (Fig. (Fig.1).1). On the other hand, ORF128 most probably plays a role in host specificity and has homology with the putative receptor binding protein of the virulent lactococcal phage mutant ul36.t1 (27). As previously observed, phage 949 can infect the host strain of phage ul36 and its mutants (L. lactis SMQ-86 and NCK203).

FIG. 3.
Analysis of the phage 949 structural proteins. (A) Migration of the phage 949 proteins on a 12% SDS-PAGE gel followed by Coomassie blue staining. The numbers on the left indicate the molecular mass of the ladder (protein ladder, 10 to 250 kDa; ...

In bands 4, 7, and 8, the same protein, ORF143, was detected by mass spectrometry (Fig. (Fig.3).3). The first 88 amino acids were never identified in any of the mass spectra, indicating that this protein is probably processed. The molecular mass of the truncated protein was evaluated to be 29 kDa and could explain the presence of ORF143 in bands 7 and 8. Considering the possibility that this protein could form multimers, this could explain its presence in band 4, except that the molecular masses determined by SDS-PAGE and calculated based on the amino acid sequence diverge significantly. No putative function could be attributed to this ORF by bioinformatics analysis, although this protein is present in large amounts in the phage. The major capsid protein has the potential to form multimers and is often processed, as observed for other phages (19). Therefore, it is tempting to attribute this function to ORF143.

As for the other bands (band 5 [ORF127], band 6 [ORF146], band 9 [ORF138], band 9 [ORF136], band 10 [ORF144], band 11 [0RF141]), proteins with the expected molecular masses were identified and thus considered to be structural proteins. In the analysis of the complete phage structural proteome by LC/MS-MS, four additional proteins (ORF129, ORF132, ORF139, ORF145) were detected and annotated in the same way (Table (Table2).2). In total, 13 structural proteins were identified for phage 949.

Presence of tRNA in the 949 genome.

The presence of tRNAs in the genome of phage 949 was identified using tRNAscan-SE software and supported with the informatics tool ARAGORN and BLASTn (NCBI). Six tRNA sequences were found that encoded the amino acids Met (CAT), Arg (TCT), Asp (GTC), Pro (TGG), Sup (CTA), and Trp (CCA). To date, this is the highest number of tRNAs identified in a lactococcal phage. Previously, it was found that L. lactis phage P087 encoded 5 tRNAs (48) and KSY1 possessed 3 tRNAs (8). All the 949 tRNAs were located in a short genomic region (positions 42403 to 47064; Fig. Fig.2)2) but interrupted by ORFs, indicating that they were not likely acquired via recombination of a specific genomic block as was proposed for lactococcal phage P087 (48). One of the interspaced genes is orf88, which encodes a putative HNH endonuclease. The tRNAs encoded in the L. lactis KSY1 genome were also interspaced by other ORFs (8). Moreover, the 949 tRNAs mapped in a genomic region encoding nucleotide biosynthesis and genome replication, indicating that they are probably involved in translation or in transcriptional regulation.

To determine if the identified tRNAs may contribute to the lytic cycle of phage 949, the codon usage of the amino acids encoded by the tRNAs and their corresponding codons was investigated for the complete genome of 949 and for each individual ORF. Those data were compared to the codon usage of L. lactis strain IL1403, for which the complete genome is available and which is sensitive to phage 949.

Overall, the anticodon of the tRNAs found in the phage 949 genome were not necessarily matching the most common codons used by the phage (Table (Table3).3). For example, the GAC codon, which is coding for aspartic acid (Asp) and is encoded by a tRNA in the 949 genome, is the less used of the two possible codons for this amino acid (Table (Table3).3). On the other hand, the CCA (proline) and AGA (arginine) codons were slightly more used by phage 949 than their analogous codons that encode similar amino acids. Thus, we observed no significant difference between the codon usages of the amino acids encoded by the 949 tRNAs compared with the corresponding amino acid codons used by the phage.

TABLE 3.
Codon usage of phage 949 and host IL1403 for amino acids encoded by the 949 tRNAs

Conversely, some phage proteins contain more amino acids encoded by the phage tRNAs. For example, ORF100 (nicotinamide mononucleotide transporter) and ORF85 (hypothetical protein) have a frequency per thousand of 45.5 and 61.5, respectively, for the Trp codon, which is higher than the value of 10.2 found for the complete phage codon usage. Similarly, ORF62 (ssDNA exonuclease) has a frequency per thousand for Asp of 40.4, while the two HNH endonucleases ORF40 and ORF88 have a high number of Arg codons, 36.6 and 49.0 per thousand, respectively. Those values are higher than what we calculated for the complete phage codon usage for the respective amino acids (Table (Table3).3). Of note, the ORFs that contain a higher content for the amino acids encoded by the tRNAs are distributed randomly within the phage genome. Therefore, it is possible that phage 949 encoded tRNAs to increase the expression of some proteins.

Comparison of these data with the codon usage of the host L. lactis IL1403 revealed that AGA (Arg) and GAC (Asp) were used slightly more often by the phage, while the usages of other codons determined by the phage 949 tRNAs were somewhat similar between the phage and its host (Table (Table3).3). Consequently, it seems that phage 949 has a proportion of codons encoded by its tRNAs that is similar to that of the bacteria. Apparently, there is no significant reason for the phage to carry those tRNAs except if the cell is not able to provide enough tRNA during the phage propagation.

The presence of tRNAs in the phage genome is not unusual, especially in a large genome. It was previously suggested that phages encode tRNAs corresponding to codons that are less used by the host bacteria to increase specific phage protein expression (4). In the case of phage 949, this was not observed, since the codon usage is similar between both the host and the phage. On the other hand, particular phage proteins contain more of those rare codons and the phage probably uses its tRNAs to control its protein production (4). Phage 949 possesses 6 tRNAs that are overrepresented in some proteins, suggesting that it helps the phage to control or increase its protein expression. Since it is difficult to identify their origin, we can only speculate that the acquisition of the unique tRNAs was required for phage adaptation to its host and that this is relatively recent in the evolutionary history of the phage.

Comparison with other lactococcal phage genomes.

Because 949 was the last phage genome needed to have the complete set of 10 currently known lactococcal phage groups, a multiple alignment was performed for all available lactococcal phage genomes. A proteomic phylogenetic tree was constructed using MEGA 4.1 software and the neighbor-joining method (Fig. (Fig.4).4). The proteomic tree (42) gave sharper distinctions between phage groups and P335 subgroups (Fig. (Fig.44).

FIG. 4.
Comparison of the lactococcal phage complete proteome and phylogenic tree constructed using MEGA 4.1 software and the neighbor-joining method. The model used was the number of differences, the tree was constructed to be unrooted, and the parameters used ...

As expected, all 936-like phages grouped together since they shared high sequence similarity. These data indicate that 936-like phages probably evolved from a common ancestor with the ability to rapidly multiply in their respective host strains. The two c2-like phages (c2 and bIL67) also grouped together as expected but shared similarity with phage Q54 (Fig. (Fig.4).4). The result from this proteomic tree agreed with previous analyses of phage Q54, which among others shares morphological features with c2-like phages (19).

The polythetic nature of the P335 group was also illustrated with the proteomic tree (Fig. (Fig.4).4). Indeed, even if they are currently regrouped in a unique taxon, these P335-like phages diverged significantly from each other. It was previously proposed to divide the P335 members into subgroups according to their morphological properties and comparison of their structural proteins (26). As suggested and confirmed here, phages P335, Tuc2001, TP901-1, and ul36 could form the first subgroup, BK5-T, 4268, and bIL286 could form the second, while r1T and phiLC3 could represent a third subgroup. Moreover, the proteomic tree demonstrated that the P335-like prophages bIL285, bIL309, and ϕSMQ-86 diverged significantly from each other as well as from all the other P335 phages. As more genomes become available, they may represent subgroups of their own.

Other lactococcal phage groups were separated into unique clusters, including phage 949, which confirms the current classification of phages infecting Lactococcus lactis strains (10, 43).

Conclusions.

The microbiological and molecular characterizations of a member of the 10th and last presently recognized lactococcal phage group are reported here. Phage 949 has several exclusive features that distinguish it from other lactococcal phages. It has (i) the longest noncontractile tail, (ii) the largest dsDNA genome, (iii) the lowest GC content, (iv) an unusual genomic organization, and (v) the highest number of tRNAs. These unique characteristics as well as our phylogenetic analyses confirm the status of phage 949 as a unique group of L. lactis phage. The evolution of this phage possibly included recombinations with other phages which led, at least, to the acquisition of a host recognition module and tRNAs for adequate expression of its proteins. Finally, the availability of the genome for at least one representative of each of the 10 lactococcal phage groups should facilitate the characterization of newly isolated phages through morphological and molecular methods.

Acknowledgments

We thank Barbara-Ann Conway for editorial assistance, Jessica Labonté for helpful discussion, as well as Sylvie Bourassa, Audrey Fleury, and Siham Ouennane for their technical support.

J.E.S. is recipient of a scholarship from the Fonds de Recherche sur la Nature et les Technologies (FQRNT). This work was funded by a strategic grant from the Natural Sciences and Engineering Research Council (NSERC) of Canada.

Footnotes

[down-pointing small open triangle]Published ahead of print on 27 August 2010.

REFERENCES

1. Ackermann, H. W. 1998. Tailed bacteriophages: the order Caudovirales. Adv. Virus Res. 51:135-201. [PubMed]
2. Ackermann, H. W., S. Yoshino, and S. Ogata. 1995. A Bacillus phage that is a living fossil. Can. J. Microbiol. 41:294-297.
3. Ahmed, R., P. Sankar-Mistry, S. Jackson, H. W. Ackermann, and S. S. Kasatiya. 1995. Bacillus cereus phage typing as an epidemiological tool in outbreaks of food poisoning. J. Clin. Microbiol. 33:636-640. [PMC free article] [PubMed]
4. Bailly-Bechet, M., M. Vergassola, and E. Rocha. 2007. Causes for the intriguing presence of tRNAs in phages. Genome Res. 17:1486-1495. [PMC free article] [PubMed]
5. Bissonnette, F., S. Labrie, H. Deveau, M. Lamoureux, and S. Moineau. 2000. Characterization of mesophilic mixed starter cultures used for the manufacture of aged cheddar cheese. J. Dairy Sci. 83:620-627. [PubMed]
6. Bolotin, A., P. Wincker, S. Mauger, O. Jaillon, K. Malarme, J. Weissenbach, S. D. Ehrlich, and A. Sorokin. 2001. The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res. 11:731-753. [PMC free article] [PubMed]
7. Bouchard, J. D., E. Dion, F. Bissonnette, and S. Moineau. 2002. Characterization of the two-component abortive phage infection mechanism AbiT from Lactococcus lactis. J. Bacteriol. 184:6325-6332. [PMC free article] [PubMed]
8. Chopin, A., H. Deveau, S. D. Ehrlich, S. Moineau, and M. C. Chopin. 2007. KSY1, a lactococcal phage with a T7-like transcription. Virology 365:1-9. [PubMed]
9. Crutz-Le Coq, A. M., B. Cesselin, J. Commissaire, and J. Anba. 2002. Sequence analysis of the lactococcal bacteriophage bIL170: insights into structural proteins and HNH endonucleases in dairy phages. Microbiology 148:985-1001. [PubMed]
10. Deveau, H., S. J. Labrie, M. C. Chopin, and S. Moineau. 2006. Biodiversity and classification of lactococcal phages. Appl. Environ. Microbiol. 72:4338-4346. [PMC free article] [PubMed]
11. Deveau, H., M. R. Van Calsteren, and S. Moineau. 2002. Effect of exopolysaccharides on phage-host interactions in Lactococcus lactis. Appl. Environ. Microbiol. 68:4364-4369. [PMC free article] [PubMed]
12. de Vos, W. M. 1987. Gene cloning and expression in lactic streptococci. FEMS Microbiol. Rev. 46:281-295.
13. Donelli, G., F. Guglielm, and L. Paoletti. 1972. Structure and physico-chemical properties of bacteriophage G. 1. Arrangement of protein subunits and contraction process of tail sheath. J. Mol. Biol. 71:113-125. [PubMed]
14. Dupuis, M. E., and S. Moineau. 2010. Genome organization and characterization of the virulent lactococcal phage 1358 and its similarities with Listeria phages. Appl. Environ. Microbiol. 76:1623-1632. [PMC free article] [PubMed]
15. Emond, E., E. Dion, S. A. Walker, E. R. Vedamuthu, J. K. Kondo, and S. Moineau. 1998. AbiQ, an abortive infection mechanism from Lactococcus lactis. Appl. Environ. Microbiol. 64:4748-4756. [PMC free article] [PubMed]
16. Emond, E., B. J. Holler, I. Boucher, P. A. Vandenbergh, E. R. Vedamuthu, J. K. Kondo, and S. Moineau. 1997. Phenotypic and genetic characterization of the bacteriophage abortive infection mechanism AbiK from Lactococcus lactis. Appl. Environ. Microbiol. 63:1274-1283. [PMC free article] [PubMed]
17. Ferguson, P. L., and D. H. Coombs. 2000. Pulse-chase analysis of the in vivo assembly of the bacteriophage T4 tail. J. Mol. Biol. 297:99-117. [PubMed]
18. Filee, J., E. Bapteste, E. Susko, and H. M. Krisch. 2006. A selective barrier to horizontal gene transfer in the T4-type bacteriophages that has preserved a core genome with the viral replication and structural genes. Mol. Biol. Evol. 23:1688-1696. [PubMed]
19. Fortier, L. C., A. Bransi, and S. Moineau. 2006. Genome sequence and global gene expression of Q54, a new phage species linking the 936 and c2 phage species of Lactococcus lactis. J. Bacteriol. 188:6101-6114. [PMC free article] [PubMed]
20. Garneau, J. E., D. M. Tremblay, and S. Moineau. 2008. Characterization of 1706, a virulent phage from Lactococcus lactis with similarities to prophages from other Firmicutes. Virology 373:298-309. [PubMed]
21. Haaber, J., S. Moineau, L. C. Fortier, and K. Hammer. 2008. AbiV, a novel antiphage abortive infection mechanism on the chromosome of Lactococcus lactis subsp. cremoris MG1363. Appl. Environ. Microbiol. 74:6528-6537. [PMC free article] [PubMed]
22. Jarvis, A. W. 1977. The serological differentiation of lactic streptococcal bacteriophages. N. Z. J. Dairy Sci. Technol. 12:176-181.
23. Josephsen, J., N. Andersen, E. Behrndt, E. Brandsborg, G. Christiansen, M. B. Hansen, S. Hansen, E. W. Nielsen, and F. K. Vogensen. 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.
24. Kotsonis, S. E., I. B. Powell, C. J. Pillidge, G. K. Limsowtin, A. J. Hillier, and B. E. Davidson. 2008. Characterization and genomic analysis of phage asccphi28, a phage of the family Podoviridae infecting Lactococcus lactis. Appl. Environ. Microbiol. 74:3453-3460. [PMC free article] [PubMed]
25. Kumar, S., M. Nei, J. Dudley, and K. Tamura. 2008. MEGA: a biologist-centric software for evolutionary analysis of DNA and protein sequences. Brief. Bioinform. 9:299-306. [PMC free article] [PubMed]
26. Labrie, S. J., J. Josephsen, H. Neve, F. K. Vogensen, and S. Moineau. 2008. Morphology, genome sequence, and structural proteome of type phage P335 from Lactococcus lactis. Appl. Environ. Microbiol. 74:4636-4644. [PMC free article] [PubMed]
27. Labrie, S. J., and S. Moineau. 2007. Abortive infection mechanisms and prophage sequences significantly influence the genetic makeup of emerging lytic lactococcal phages. J. Bacteriol. 189:1482-1487. [PMC free article] [PubMed]
28. Labrie, S. J., J. E. Samson, and S. Moineau. 2010. Bacteriophage resistance mechanisms. Nat. Rev. Microbiol. 8:317-327. [PubMed]
29. Laslett, D., and B. Canback. 2004. ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide sequences. Nucleic Acids Res. 32:11-16. [PMC free article] [PubMed]
30. Lillehaug, D. 1997. An improved plaque assay for poor plaque-producing temperate lactococcal bacteriophages. J. Appl. Microbiol. 83:85-90. [PubMed]
31. Lowe, T. M., and S. R. Eddy. 1997. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 25:955-964. [PMC free article] [PubMed]
32. Lukashin, A. V., and M. Borodovsky. 1998. GeneMark.hmm: new solutions for gene finding. Nucleic Acids Res. 26:1107-1115. [PMC free article] [PubMed]
33. Madera, C., C. Monjardin, and J. E. Suarez. 2004. Milk contamination and resistance to processing conditions determine the fate of Lactococcus lactis bacteriophages in dairies. Appl. Environ. Microbiol. 70:7365-7371. [PMC free article] [PubMed]
34. McGrath, S., G. F. Fitzgerald, and D. van Sinderen. 2007. Bacteriophages in dairy products: pros and cons. Biotechnol. J. 2:450-455. [PubMed]
35. Moineau, S., and C. Lévesque. 2005. Control of bacteriophages in industrial fermentation, p. 286-296. In E. Kutter and A. Sulakvelidze (ed.), Bacteriophages: biology and applications. CRC Press, Boca Raton, FL.
36. Moineau, S., M. Borkaev, B. J. Holler, S. A. Walker, J. K. Kondo, E. R. Vedamuthu, and P. A. Vandenbergh. 1996. Isolation and characterization of lactococcal bacteriophages from cultured buttermilk plants in the United States. J. Dairy Sci. 79:2104-2111.
37. Moineau, S., E. Durmaz, S. Pandian, and T. R. Klaenhammer. 1993. Differentiation of two abortive mechanisms by using monoclonal antibodies directed toward lactococcal bacteriophage capsid proteins. Appl. Environ. Microbiol. 59:208-212. [PMC free article] [PubMed]
38. Moineau, S., J. Fortier, H. W. Ackermann, and S. Pandian. 1992. Characterization of lactococcal bacteriophages from Quebec cheese plants. Can. J. Microbiol. 38:875-882.
39. Nes, I. F., J. Brendehaug, and K. O. von Husby. 1988. Characterization of the bacteriophage B2 of Lactobacillus plantarum ATCC 8014. Biochimie 70:423-427. [PubMed]
40. Petri, R., and C. Schmidt-Dannert. 2004. Dealing with complexity: evolutionary engineering and genome shuffling. Curr. Opin. Biotechnol. 15:298-304. [PubMed]
41. Rao, V. B., and M. Feiss. 2008. The bacteriophage DNA packaging motor. Annu. Rev. Genet. 42:647-681. [PubMed]
42. Rohwer, F., and R. Edwards. 2002. The phage proteomic tree: a genome-based taxonomy for phage. J. Bacteriol. 184:4529-4535. [PMC free article] [PubMed]
43. Rousseau, G. M., and S. Moineau. 2009. Evolution of Lactococcus lactis phages within a cheese factory. Appl. Environ. Microbiol. 75:5336-5344. [PMC free article] [PubMed]
44. Sambrook, J., and D. W. Russell. 2001. Molecular cloning, a laboratory manual, 3rd edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
45. Sjoberg, B. M., S. Hahne, C. Z. Mathews, C. K. Mathews, K. N. Rand, and M. J. Gait. 1986. The bacteriophage T4 gene for the small subunit of ribonucleotide reductase contains an intron. EMBO J. 5:2031-2036. [PMC free article] [PubMed]
46. Spinelli, S., A. Desmyter, C. T. Verrips, H. J. de Haard, S. Moineau, and C. Cambillau. 2006. Lactococcal bacteriophage p2 receptor-binding protein structure suggests a common ancestor gene with bacterial and mammalian viruses. Nat. Struct. Mol. Biol. 13:85-89. [PubMed]
47. Tremblay, D. M., M. Tegoni, S. Spinelli, V. Campanacci, S. Blangy, C. Huyghe, A. Desmyter, S. Labrie, S. Moineau, and C. Cambillau. 2006. Receptor-binding protein of Lactococcus lactis phages: identification and characterization of the saccharide receptor-binding site. J. Bacteriol. 188:2400-2410. [PMC free article] [PubMed]
48. Villion, M., M. C. Chopin, H. Deveau, S. D. Ehrlich, S. Moineau, and A. Chopin. 2009. P087, a lactococcal phage with a morphogenesis module similar to an Enterococcus faecalis prophage. Virology 388:49-56. [PubMed]
49. Weisberg, R. A., M. E. Gottesmann, R. W. Hendrix, and J. W. Little. 1999. Family values in the age of genomics: comparative analyses of temperate bacteriophage HK022. Annu. Rev. Genet. 33:565-602. [PubMed]
50. Young, P., M. Ohman, M. Q. Xu, D. A. Shub, and B. M. Sjoberg. 1994. Intron-containing T4 bacteriophage gene sunY encodes an anaerobic ribonucleotide reductase. J. Biol. Chem. 269:20229-20232. [PubMed]

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

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...