University of Groningen Lactococcus lactis bacteriophages: phage-host interaction and phage transduction Marcelli, Barbara

(1)

Lactococcus lactis bacteriophages: phage-host interaction and phage transduction

Marcelli, Barbara

DOI:

10.33612/diss.118088081

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2020

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Marcelli, B. (2020). Lactococcus lactis bacteriophages: phage-host interaction and phage transduction. Rijksuniversiteit Groningen. https://doi.org/10.33612/diss.118088081

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

540959-L-bw-Marcelli 540959-L-bw-Marcelli 540959-L-bw-Marcelli 540959-L-bw-Marcelli Processed on: 7-2-2020 Processed on: 7-2-2020 Processed on: 7-2-2020

Processed on: 7-2-2020 PDF page: 29PDF page: 29PDF page: 29PDF page: 29 29

Analyzing the genetic determinants

of the host-range of 22 newly

identified lactococcal bacteriophages

Barbara Marcelli

a

_{, Thomas Janzen}

b

_{, Oscar P. Kuipers}

a

_{, Jan Kok}

a

a _{Department of Molecular Genetics, Groningen Biomolecular Sciences and Biotechnology} Institute, University of Groningen, The Netherlands

b _{Bacterial Physiology and Improvement, R&D discovery, Chr. Hansen A/S,} Hørsholm,Denmark

(3)

ABSTRACT

Lactococcus lactis is a lactic acid bacterium extensively used in the dairy industry for the

production of different fermented foods. The infection of lactococcal starter cultures by bacteriophages represents one of the major causes of fermentation failure and consequent industrial economic losses. The analysis and characterization of the genetic determinants of the host range of lactococcal phages is of vital importance in order to understand the dynamics of bacteriophages spread in dairy process lines and to overcome this issue via strain improvement and starter culture rotation plans. Although quite some information has already been gathered on this subject, several aspects of phage-host interactions in dairy environments still remain to be unraveled. In this study we aimed at broadening the information and knowledge available on the host recognition genetic determinants of several lactococcal bacteriophages belonging to the c2 and 936 species, two of the most commonly isolated species from dairy production plants.

(4)

INTRODUCTION

Lactococcus lactis is a Gram-positive bacterium belonging to the order of the lactic acid

bacteria. It is widely used in the dairy industry as starter culture for the production of fermented foods, like buttermilk, quark, and especially a large variety of cheeses (1). In these fermentation processes, L. lactis initiates the acidification of the milk, contributes to the development of flavor and texture of the end products, and prevents the proliferation of spoilage bacteria (2, 3).

Infection of starter cultures by bacteriophages (phages) can lead to fermentation inhibition or failure with consequent great economic losses in dairying (4–6). Because of the threat that they represent for the industry, bacteriophages infecting L. lactis have been the object of intense study in the past four decades (7). All the lactococcal phages isolated so far possess a double-strand DNA genome and so they belong to the Caudovirales order; a further classification divides them into Siphoviridae (with long non-contractile tails, and isometric or prolate capsid) or Podoviridae (with short non-contractile tails and icosahedral capsid) (8). Ten different species, namely: 936, c2, P335, KSY1, Q45, 1358, P087, 949, P034 and 1706, have been identified so far on the basis of DNA-DNA hybridization, genome sequence comparison, and electron microscopy observations (9, 10). Members of the 936, c2 and P335 species are most commonly isolated in dairy environments (11). The first two comprise strictly lytic phages with conserved genome sequences and genome organizations. On the contrary, phages belonging to the P335 species can be either lytic or temperate and their genome sequences are more heterogeneous (12).

Knowledge as to how lactococcal bacteriophages recognize their hosts is of vital importance for the establishment of adequate starter culture rotation plans (13), and hence several studies have focused on gathering detailed insights into this matter. The majority of these studies regarded phages belonging to the 936, P335 and c2 species (14–18). Bacteriophages of the first two species recognize sugar receptors on the cell wall polysaccharide pellicle (CWP) surrounding their lactococcal hosts (19, 20). The structure and mechanistic details of receptor binding have been unraveled for two representatives of the P335 species: phages Tuc2009 and TP901-1 (21). It was shown that, although these phages share high structural similarities, the presence of an extra protein and of two different carbohydrate recognition sites in the Tuc2009 baseplate structure plays a major role in differentiating their ability to recognize their host (22, 23). 936 phages have been recently classified into five subgroups on the basis of the amino acid sequences of their receptor binding protein (RBP). A correlation was observed between the sequence of a phage RBP and the CWP type of its sensitive hosts (24, 25). Phages of the c2 species, on the other hand, adsorb to their hosts in a two-step process in which they first reversibly attach to a sugar molecule of the CWP, and subsequently irreversibly

(5)

bind to a proteinaceous molecule (26). To date, two different lactococcal proteins have been identified as c2 phages receptors, namely Pip and YjaE (27). A previous study classified the c2 phages into the two eponymous subgroups c2 and bIL67 on the basis of the nucleotide sequences of three structural genes involved in host recognition and adsorption (14). It was demonstrated that members of the c2 subgroup recognize Pip as the host receptor while members of the latter recognize YjaE (27).

In the current study we extended the current knowledge regarding L. lactis phage-host interaction by analyzing a collection of 27 new lactococcal bacteriophages. First, the phages were classified and assigned to the species they belonged to, and their host range was assessed. Subsequently, members of the c2 and 936 species (a total of 22 isolates) were selected for further detailed phage-host interaction analysis using combined phenotypic and in silico analysis approaches.

MATERIALS AND METHODS

Bacterial strains, phages and culture conditions

The bacteriophages and lactococcal strains used in this study are listed in Table 1 and Table 2, respectively. All L. lactis strains were grown without shaking at 30 °C in M17 medium (BD - Becton, Dickinson and Company, Franklin Lakes, NJ) supplemented with 0.5 % glucose (GM17). Bacteriophages were initially propagated on their sensitive industrial strains by infecting, with a single plaque, a 10 ml culture in its early exponential growth phase (Optical density at 600nm; OD600 0.3 – 0.5 ) in M17 media supplemented with 0.5 % lactose (LM17) containing 10 mM CaCl2 and 10 mM MgCl2.Samples were incubated at 30 °C until visible culture lysis occurred, filter-sterilized using a 0.45 µm filter (Ministart NML , Sartorius, Germany) to eliminate cells and cell debris, and stored at 4° C until further use. Bacteriophage host range assay

The host range of the phages was assessed via a spot assay as previously described (28) with the following modifications: bottom and top agar layers contained 1% and 0,4 % agar, respectively. CaCl2 was added to the media at the final concentration of 10 mM. Glycine was added at the final concentration of 0.5 % (wt/vol) to facilitate plaque visualization as previously reported (28). 100-fold diluted samples of the phage lysates obtained as described above ( diluted in TBT buffer (100 mM NaCl, 50 mM Tris-HCl, 10 mM MgCl2, [pH7]) were spotted on a solidified top agar layer containing 200 µl of an overnight sample of the bacterial strain. The plates were incubated overnight at 30° C and then checked for the presence of lysis halos in the spot area. In the case of the appearance of spots of clearance, the strain under study was considered as possibly sensitive to the phage. In that case, the phage and bacterial strain were subjected to the plaque assay

(6)

described above in order to check for the appearance of phage-derived plaques and to confirm the ability of the phage to propagate on the bacterial strain All tests were conducted in triplicate.

Total bacteriophage and bacterial DNA isolation

Phage DNA was isolated starting from 5 ml of phage lysate obtained as described above. The lysate was mixed with 10 % w/v polyethylene glycol MW 800 and 0.5 M NaCl, and incubated for 16 h at 4°C to allow the phage particles to precipitate. The sample was subsequently centrifuged for one h at 11000 x g at 4 °C in an Eppendorf table top centrifuge 5810R (Eppendorf, Hamburg, Germany), and the pellet was resuspended in 400 µl DNAseI buffer (10 mM Tris-HCl pH 7.6, 2.5 mM MgCl2, 0.5 mM CaCl2). Residual host DNA and RNA were degraded by incubation at 37 °C for at least 30 min with 1µg/ml each of DNAseI and RNAseI (Merk KGaA, Darmstadt, Germany). Ethylendiaminetetraacetic acid (EDTA) was then added at a final concentration of 5 mM and the tube was incubated at 65 °C for 15 min to inactivate the enzymes. Phage capsids were degraded by incubating the mixture at 56 °C for 20 min with proteinase K (Merk KGaA) at a final concentration of 2 µg/ml and subsequently by adding 2.5 % v/v of sodium dodecyl sulfate (SDS) and incubation at 65 °C for 10 min. Phage DNA was purified by two consecutive phenol/chloroform extractions: the sample (approx. 500 µl) was mixed with an equal amount of a mixture of phenol:chloroform:isoamyl alcohol (25:24:1) in a 2-ml 5Prime phase lock gel-light tube (Quanta BioScience, Beverly, MA, USA), and centrifuged at 13000 x g for 8 min in an Eppendorf table top centrifuge 5418. Subsequently, the upper DNA-containing aqueous phase was mixed with 0.1 volume of 3 M sodium acetate (pH 4.8) and 2.5 volumes of 80 % ice-cold ethanol. The content of the tube was mixed by inversion and incubated at -20 °C for at least one h to allow the DNA to precipitate. DNA was concentrated by centrifugation at 13,000 x g for 15 min in an Eppendorf table top centrifuge 5418. DNA was washed twice with 1 ml of 80 % ice-cold ethanol. The supernatant was discarded and the pellet was air dried for at least one h before resuspending in 50 µl TE buffer (10 mM Tris-HCl, pH 8, 1 mM EDTA). DNA samples were stored at 4 °C until use.

Total bacterial DNA was isolated starting from 4 ml of overnight culture. The culture was centrifuged at 14,000 x g for 1 minute on a Eppendorf table top centrifuge 5810R (Eppendorf) and resuspended in 300 µl of Birnboim solution A (20 % sucrose, 10 mM Tris HCl pH 8 , 10 mM EDTA, 50 mM NaCl) containing 5 mg/ml of lysozyme and 10 g/ ml of RNase. The sample was incubated at 37 °C for 30 min to initiate bacterial cell wall lysis and subsequently at 55 °C for 10 min to assure contaminating RNA degradation. Bacterial cells were further lysed by adding 500 µl of Nuclei lysis solution (Component of Wizard®, Wizard® SV and Wizard® SV 96 Genomic DNA Purification Systems available as a standalone item, Promega, Medison WI, USA) and incubating at 80 °C for 5 min. Subsequently 200 µl of Protein lysis solution (Component of Wizard®, Wizard®

(7)

SV and Wizard® SV 96 Genomic DNA Purification Systems available as a standalone item, Promega, Medison WI, USA) was added and the sample was incubated in ice for 10 min. To start purification, the lysed bacterial cells were then centrifuged at 14000 x g on a Eppendorf table top centrifuge 5810R and the supernatant containing the DNA was transferred to a clean 2ml Eppendorf containing 250 µl of demineralized water and 250 µl of phenol. The sample was mixed by inversion and incubated at room temperature for 10 min before adding 250 µl of chloroform. At this point the water/phenol/ chloroform mixture was transferred to a 2 ml 5Prime phase lock gel-heavy tube (Quanta BioScience, USA) and centrifuged at 15,000 x g for 10 min on a Eppendorf table top centrifuge 5810R. The upper water-phase obtained from centrifugation was then transferred to a new 2-ml 5Prime phase lock gel-heavy tube already containing 500 µl of chloroform. The sample was again centrifuged at 15,000 x g for 10 min a Eppendorf table top centrifuge 5810R. The water-phase was, at this point, decanted in a clean 2ml Eppendorf tube and DNA was precipitated by adding 100 µl of NaAc, pH 5.2 and 700 µl of isopropanol. The sample was mixed by inversion until precipitated DNA was visible and then centrifuged at 15,000 x g for 5 min. The supernatant was discarded and the DNA pellet was washed twice with 1 ml of 70 % of ice-old ethanol. The DNA pellet was then air-dried for 15 min at room temperature and finally dissolved in 100 µl of TE buffer. DNA samples were stored at 4 °C until use.

Bacteriophages and strains typing

C2, 936 and P335 bacteriophages species was assessed using two previously described multiplex PCR methods (29, 30). The remaining bacteriophages species were assigned based on total genome alignment with phages genomes available in the GenBank database. Cell wall polysaccharide type of the bacterial strains was assessed using the previously described multiplex PCR method (24).

Bioinformatics analyses

Bacteriophage and bacterial genomic DNA were sequenced on the Illumina MiSeq platform with 2 x 150-bp paired-end sequencing (Illumina, San Diego, CA, USA). The bacteriophages genomes were deep sequenced reaching contigs reading coverage ranging from 100 K to 2 M reads. Nucleotide sequences were trimmed using Trimmomatic (31), and subsequently analyzed and assembled using the A5-miseq pipeline (32).The assembled contigs were annotated using the RASTtk server (33). Sequence alignments were generated using ClustalO with the standard settings (34). Pip and YjaE alignments were visualized with Jalview 2.10.5 (35) and subsequently manually curated. Transmembrane domain prediction was conducted using the TMHMM server 2.0 (http://www.cbs.dtu.dk/services/ TMHMM/) and protein domain prediction was conducted using the Pfam protein domain database (36).The bacteriophage genomes accession number can be found in Table 1.

(8)

The whole genome shotgun project accession number for all the bacteriophage genomes (except L. lactis UC509.9) can be found in Table 2.

Table 1. List of bacteriophages used in this study.

L. lactis

bacteriophagea Published _nameb _speciesPhage c Origin _{isolation Reference}Year of GenBank Accession _number

MP111 CHPC52 936 Unknown 1997 This work MN689519

MP112 CHPC116 C2 USA 1989 This work MN689507

MP113 CHPC122 C2 UK 1990 This work MN689512

MP114 CHPC129 936 UK 1990 This work MN689514

MP115 CHPC134 C2 UK 1990 This work MN689515

MP116 CHPC148 Bk5t UK 1990 This work MN689516

MP120 CHPC781 936 Denmark 1997 This work MN689520

MP121 CHPC958 936 Australia 1997 This work MN689522

MP125 CHPC974 Bk5t USA 2002 This work MN689530

MP127 CHPC1020 C2 Australia 2004 This work MN689505

MP136 CHPC971 1706 USA 2002 (37) MK779875

MP138 CHPC1175 Bk5t USA 2009 This work MN689509

MP139 CHPC1242 C2 Germany 2013 This work MN689513

MP141 CHPC836 Bk5t France 1998 This work MN689521

MP143 CHPC959 936 USA 2002 This work MN689523

a ,b_{: med via in silico genome analysis.}

(9)

Table 2. List of lactococcal strains used in this study.

L.lactis

strain speciesSub Description CWPa Reference

Genbank accession

numberb

MG1363 cremoris

Laboratory model strain. Plasmid free derivative of the dairy

isolate NCDO712

NODE_1 (79332 –

103380) (38) WJVF00000000

AM1 cremoris Dairy starter strain NODE_6 (45335 – 18855) (39) WJVE00000000 SMQ86 cremoris Dairy starter strain NODE_5 (20715 – ₄₈₂₇₆₎ (40) WJVD00000000 SMQ384 cremoris Dairy starter strain NODE_5 (83727 – ₅₇₂₄₇₎ (41) WJVC00000000 SMQ385 cremoris Dairy starter strain NODE_5 (20715 – 48276) (41) WJVB00000000 SMQ450 cremoris Dairy starter strain NODE_8 (32810 – 60371 ₎ (10) WJVA00000000 SMQ562 cremoris Dairy starter strain NODE_10 (45744 – ₁₉₂₆₄₎ (10) WJUZ00000000

ML8 cremoris Dairy starter strain NODE_132 (3085 – 4806)NODE_146 (1-3846)

NODE_44 (1 – 16982) (42) WJUY00000000 3107 cremoris Dairy starter strain NODE_1 (61546 – 38671) (43) WJUX00000000 158 cremoris Dairy starter strain NODE_ 1 (43702 – 71219) (44) WJUW0000000 UC509.9 cremoris Dairy starter strain (190635 – 218148) (24)

SK11 cremoris Dairy starter strain NODE_1 (71763 – 96656) (45) WJUV00000000

184 lactis Dairy starter strain - (24) WJUU00000000

UL8 lactis Dairy starter strain NODE_5 (20715 – 48276) (24) WJUT00000000 IL1403 lactis

Laboratory model strain. Plasmid-free derivative of the dairy

isolate CNRZ157

NODE_1 (176294 –

153338) (24) WJUS00000000

C10 lactis Dairy starter strain NODE_14 (45783 – ₁₉₃₀₃₎ (24) WJUR00000000 229 lactis Dairy starter strain NODE_5 (20730 – 1) (24) WJUQ00000000 Bu260 Lactis Dairy starter strain NODE_7 (55072 - 79115) (46) WJUP00000000 a :_{Cell wall polysaccharide operon coding region information. The nucleotide residues coordinates and the node} where they can be found are reported. This information refers to the whole genome shotgun sequencing projects deposited at the GenBank referring to the specific strains used in this study. Strain UC509.9 was not sequenced in the contest of this study, hence the indicated coordinated refer to the published sequence reported in (24). No coordinates are reported for strain 184 as the CWP operon of this strain is still unknown.

b _{: Whole genome shotgun sequencing project accession number.}

RESULTS

Characterization of L. lactis bacteriophages and host strains

The 27 bacteriophages analyzed in this study were isolated over the past two decades in dairy factories located worldwide (Table 1). They were chosen for further characterization as, after an initial host-range assessment against a panel of several industrial lactococcal

(10)

strains, they proved to have quite diverse host ranges even within species (data not shown). The library of lactococcal host strains employed here is composed of 18 publicly available dairy isolates and laboratory strains with quite diverse genotypic and phenotypic dairy-relevant characteristics (Tables 2 and 3). This combination of strains and phages allowed for a broad overview of lactococcal response to bacteriophage infection. Both bacteriophages and bacteria genomes were (re)sequenced and annotated to proceed with further analyses.

First, the bacteriophage species of all the isolates was determined. Using two previously described multiplex PCR methods (29, 30), and total genome comparative analysis, we identified 22 phages that belong to the most commonly isolated bacteriophage species in dairy environments: c2 ( thirteen phages), and 936 (nine phages) (Table 3). Next, the genomes of the remaining unclassified phages were aligned to publicly available lactococcal phage genome sequences. The phages MP116, MP125, MP138 and MP141 could thus be identified as belonging to BK5-t, a P335 species, and phage MP136 as member of the 1706-species. This classification is based on both high (>72%) nucleotide (nt) identities and conserved genome organization with the genomes of the reference phages Bk5-t and 1706, respectively. Given the scarcity of phages belonging to the species P335 and 1706, we decided to focus on and more deeply analyze the results of the c2 and 936 isolates in our pool of phages.

Subsequently, the L. lactis strains were analyzed for characteristics known to play a role in phage host-range determination. One of the most important determinants in this respect is the cell wall polysaccharide pellicle (CWP) surrounding lactococcal cells. Three types have been described to date that vary in sugar composition and are encoded by corresponding specific gene clusters (24). Since it is known that many bacteriophages, including isolates belonging to the P335 and 936 species, use sugar molecules in the CWP of their hosts as receptors (24), the CWP type of all strains from our library was determined using multiplex PCR (24) (Table 3). Moreover, the results were confirmed via in silico analysis of the nucleotide (nt) and amino acid (aa) sequences of their CWP encoding genes and encoded proteins (Table 2). Ten of the strains code for a type-A CWP, four specify a type-B pellicle and three a type-C. For only one strain, L. lactis 184, the multiplex PCR failed at producing any result, an outcome already reported in a previous study (24).

The next important features that we analyzed are the two proteins Pip and Yjae that are known to be utilized as receptors by bacteriophages of the c2 species (27). Both proteins have unknown functions, are predicted to be integral membrane proteins with 5 to 7 transmembrane domains (47), and are expressed in all lactococcal strains sequenced so far. Pip is an orthologue of the membrane protein YueB of the Gram-positive soil

(11)

Table 3. Host r

ange o

f the bacteriophages analyzed in this study

. a : P re di ct ed A bo rt iv e i nf ec tio n ( Ab i) g en es p re se nt i n t he g en om e o f t he a na ly ze d s tr ai ns . T he n um be r a nd t he t yp e, w ith in b ra ck et s, o f a nn ot at ed A bi s ys te m s a re r ep or te d. N o en tr y: n o a nn ot at ed A bi s ys te m g en es p re se nt i n t he c or re sp on di ng g en om e. b: P re di ct ed R es tr ic tion /Mo di fic at ion (R /M ) g en es p re se nt in th e g enom e o f t he a na ly ze d s tr ai ns . T he n um be r a nd ty pe , w ith in b ra ck et s, o f t he a nno ta te d R /M sy st em g en es a re re po rt ed . c : C el l Wa ll P ol ys ac ch ar id e ( CW P) ty pe en co de d b y t he an al yz ed st ra in s. T he ty pi ng o f t he C W P w as d on e b y m ul tip le x P CR as p re vi ou sly d es cr ib ed (2 4) an d c on fir m ed b y in sil ic o an al ys is o f t he C W P c od in g g en e c lu st er . X: T he p ha ge c an p ro pa ga te o n t he s tr ai n; n o e nt ry : t he p ha ge c an no t p ro pa ga te o n t he s tr ai n. L. la ct is s tr ai n/ ph ag e SM Q3 84 SM Q8 6 18 4 SMQ 385 UL 8 229 AM1 SM Q4 50 SM Q5 62 C10 IL 14 03 BU 26 0 M L8 UC 50 9.9 15 8 SK 11 M G1 36 3 310 7 St ra in sub sp ec ie s lac tis lac tis lac tis lac tis lac tis lac tis cr emor is lac tis lac tis lac tis lac tis lac tis lac tis cr emor is cr emor is cr emor is cr emor is cr emor is PI P p ro fil e ∆1 ∆1 ∆1 ∆1 ∆2 ∆3 ∆4 ∆5 ∆5 ∆5 ∆5 ∆5 ∆5 ∆6 ∆6 ∆7 Re f ∆7 Yj aE p ro fil e Re f Re f Re f Re f Re f Re f Re f Re f Re f Re f Re f Re f Re f ∆A ∆B ∆C ∆D ∆E Ab i S ys te m a 2 ( F) 2 ( F) 2 ( F) 2 ( F) 2 ( F) 1 ( E) 1 (F) 2 ( F) R/ M s ys te m b 2 ( I) 1 ( I) 3 ( I) 1( III ) 2 ( I) 2 ( I) 3 ( I) 1 ( I) 5 ( I) 1 ( I) 2 ( I) 1 ( I) 1 ( I) 2 ( I) 2 ( I) 3 ( I) 1 ( I) 3 (III) 1 ( I) 3 ( I) CW P t yp ec A A Un -kn own A A B A A A A B B B A A C C C c2 p ha ge s ( c2 t yp e) M P11 5 X X X X X X M P1 33 X X X M P1 31 X X X X X X X M P1 23 X X X X X M P11 2 X X X M P11 3 X X X X M P1 35 X X X X M P13 9 X X M P1 24 X M P13 0 c2 p ha ge s ( bI L6 7 ty pe) M P1 32 X X M P13 4 X M P1 27 93 6 P ha ge s ( gr ou p I ) M P1 14 M P1 17 X M P1 20 M P1 21 M P1 43 X 93 6 P ha ge s ( un kn ow n gr oup ) M P111 _MP11 8 M P1 44 X M P1 45 P3 35 p ha ge s ( Bk 5t gr oup ) M P1 16 M P1 25 M P13 8 M P1 41 17 06 p ha ge s M P13 6 X X X X X

(12)

bacterium Bacillus subtilis. YueB is part of a type-VII protein secretion system and is the host receptor of bacteriophage SPP1 (48, 49). An in silico analysis of YjaE revealed that it possesses an ABC-2-type transporter domain. Since strains carrying a mutated form either one of these two proteins are often resistant to c2 phages without showing any fitness loss, they are normally employed as starter cultures to specifically avoid or reduce spreading of c2 or bIL67 phages in dairy production lines (50). We thoroughly analyzed the aa sequence of these two proteins encoded by every strain of our collection, and the detailed results of this analysis are reported in the next section of the manuscript. Furthermore, we searched in every strain for the presence of putative Abortive infection (Abi) and Restriction/Modification (R/M) systems since these might play a crucial role in determining sensitivity of lactococcal strains towards phages of the c2 and 936 species (see Table 3).

Lastly, all bacteriophages were tested for their ability to infect each of the lactococcal strains in our panel. All of this phenotypic and genetic data was used to assess whether our phage/strain cohort complies with the available knowledge regarding phage-host interaction in the dairy environment.

Classification of c2 phages and their host receptors

First, the cohort of c2 bacteriophages and the receptors of their sensitive lactococcal hosts were examined using currently available knowledge and our in silico data.

Previous studies have shown that three structural genes situated in the distal part of the genome of c2 bacteriophages are involved in host recognition (14). The three genes, named l14, l15 and l16 in the reference c2 phage genome (accession number: L48605.1), are positioned in consecutive order. The specific functions of the encoded proteins are still unknown. In silico analysis revealed that all three proteins contain domains typically found in proteins of other lactococcal phage tail structures involved in phage-host interaction (27). While the products of l14 and l16 are annotated as hypothetical structural proteins and both contain carbohydrate binding domains, the l15 product has been proposed to be involved in receptor binding and is also annotated as minor structural protein (27). Based on the aa sequence of these three gene products, bacteriophages of the c2 species have been divided into two subgroups, namely c2 and bIL67 (14). The aa sequences of the putative proteins involved in host recognition (encoded by ORFs l14, l15 and l16 of bacteriophage c2 and ORFs 702, 703, and 704 of the bIL67-type bacteriophage D4410, accession number: NC_031071.1) were aligned with the corresponding ORF products sequences of the phages in our collection to classify them into the two aforementioned subgroups. The subgroup assignment was based on the percentage of aa identity currently used for this classification (27). The middle parts of the corresponding proteins encoded

(13)

by ORFs l15 and 703 are very variable in length in all the known c2 bacteriophages and since the involvement of this region in host recognition is still unclear (27), only their C-termini (65 aa residues) and N-termini (100 aa residues) were used in the comparison. Ten c2-type phages and 3 bIL67-type phages were identified in this manner (Figure 1 and Table 3). For simplicity, we will refer to phages belonging to either of the subgroups as c2- or bIL67-type throughout the rest of the manuscript.

Comparative analysis of the host range of members of the two subgroups showed that c2 phages use the Pip protein as receptor, while bIL67 phages recognize protein YjaE (27). Hence, we examined the predicted aa sequences of the Pip and YjaE proteins encoded by all bacterial strains in our collection to assess if their different sequences influenced the phages host range. To comply with the criteria adopted in a previous study (27), we used the predicted sequences of the L. lactis MG1363 Pip protein and of L. lactis IL1403 YjaE protein as references. The strains were divided into different groups, based on the pattern of mutations in their Pip and YjaE proteins compared to the references, taking only gaps in the aa sequence alignment or non-conservative aa substitutions into consideration. The specific portions of Pip and YjaE involved in phage receptor binding are not known, and only few studies have been conducted to elucidate this aspect of phage-host interaction involving proteinaceous receptors. B. subtilis phage SPP1 and E. coli phage K3 both recognize membrane-anchored host receptors, namely YueB and OmpA, respectively. Only the surface exposed portions of these proteins are recognized and used for attachment by the bacteriophage receptor binding proteins (RBP) during infection (51, 52). Based on this knowledge, we assume that the predicted outer-membrane regions of Pip and YjaE are also involved in phage-host interaction. Therefore, only those mutations that are present in the predicted extracellular regions of both proteins were used in the generation of the mutation patterns. Patterns ∆1to ∆7were identified for Pip, and ∆Ato ∆Dwere recognized for YjaE (Figures 2 and S1; and Table 3). The majority of the strains used in this study possess a YjaE protein identical to that of the reference strain L. lactis IL1403, while the remaining strains show several aa changes both in the first and second extracellular loop of the protein. The Pip proteins of our set of strains, on the other hand, is very variable with not one being identical to that of the reference L. lactis MG1363. The aa changes are mainly present in the first large extracellular loop of the Pip proteins.

(14)

Figure 1. Identity matrices of the nucleotide sequence alignments of the three structural genes involved in host recognition of c2 bacteriophages.

The ORF numbers of the two phages used as reference, c2 and D4410, are given at the top of each matrix. The numbers in the matrix represent the % of nucleotide identity between the analyzed genes of all the tested bacteriophages.

(15)

The receptor recognized by c2 bacteriophages is not always predictable by in silico analyses

The classification results presented above were used to assess whether the correlation between c2-phage subgroups and their corresponding protein receptors also applies to our collection of phages and strains. Given the high amount of genotypic and phenotypic data available, it was possible to examine the involvement of the mutation patterns in Pip and/or YjaE in phage host range determination by in silico analysis.

Phage resistance caused by mutations in Pip have been shown to be a consequence of protein truncation (53). Premature stop codons leading to protein truncation have also been described for YjaE, as well as insertions leading to frameshifts or deletions (especially between aa residues 670-697) and mutations in the proximal half of the protein (54). Although several lactococcal strains in our collection are resistant to all the bacteriophages tested here, neither of their two receptor proteins appears to be truncated. Several gaps in the respective aa sequences were, however, identified. L. lactis

Figure 2. Mutation patterns of the PIP and YjaE proteins of the strains analyzed in this study.

Extracellular loops are depicted as solid boxes, transmembrane and intracellular portions of the proteins are depicted as black lines. Gaps in the aa sequence, compared to the reference protein sequence, are showed as dashed diagonal lines. Amino acid substitutions, compared to the reference protein sequence, are shown as red vertical lines. The name of the mutation patterns are given on the left of each diagram.

Panel A: PIP proteins alignment. Panel B: YjaE proteins alignment.

(16)

SK11 YjaE shows two gaps (aa residues 167-169 and 174-177, respectively, of the reference protein) while its Pip protein presents a 50-aa gap between position 140 and 190 of the reference Pip. Strain SK11 is resistant to all the c2- and bIL67-type phages tested here, which is in accordance with a previous study which showed that strains with mutations in both Pip and YjaE are resistant to both subspecies of c2 phages (27). The two aa gaps in the YjaE protein of strain SK11 are also present in YjaE of strains 158 and 3107, while YjaE of strain UC509.9 only contains the first one. Strains 158 and UC509.9 are the only two possessing a Pip ∆6 mutation pattern and, like strain SK11, cannot be infected by any c2 or bIL67 phage from our collection. It is tempting to speculate that, also in this case, (some of) the mutations in the two receptors of strains 158 and UC590.9 make them resistant to phages of both subspecies. In addition to the two gaps mentioned above, the YjaE protein of strain 3107 contains a third gap (aa residues 668-669) that reduces the length of the second predicted extracellular loop (Figures 2 and S1). This might explain why strain 3107 is resistant towards bIL67 phages. Even if only 3 of the phages in our collection belong to the bIL67 subgroup, making it impossible to further scrutinize the identified YjaE mutation patterns, it is worth mentioning that none of these phages can infect the reference strain L. lactis IL1403. However, strains SMQ384 and C10, encoding a YjaE protein identical to that of strain IL1403, can both be infected by phage MP132. This is not what would be expected as L. lactis IL1403 is a plasmid-free strain (hence doesn’t encode any of the plasmid-encoded phage resistance mechanisms that other lactococcal strains normally possess) and possesses less Abi and R/M annotated systems than strains SMQ384 and C10 (Table 3). All the c2 bacteriophages first bind to carbohydrates composing the CWP of their host and only then bind to their proteinaceous receptors (26). Since both SMQ384 and C10 possess a type-A CWP, while strain IL1403 carries a type-B (Table 3), the observed phage sensitivity differences might be due to a different affinity of phage MP132 for the sugars on the CWP of these strains. Finally, YjaE of L.

lactis MG1363 has a higher number of mutations than all the other Yjae variants, with the

changes being mainly present in the first extracellular loop of the protein. It is noteworthy to mention that the bIL67-type phage MP134 only infects L. lactis MG1363. This might be due to the adaptation of this phage’s RBP to recognize this specific YjaE protein, or to the ability of this phage to recognize Pip as receptor instead of YjaE as would be expected for a bIL67-type phage.

Mutations in Pip profiles ∆1 to ∆5 are much more spread throughout the length of the first extracellular loop of the protein than they are in profiles ∆6 and ∆7. As some of the c2 phages can infect strains with different Pip mutation profiles this result suggests that many of the identified single aa changes most probably don’t affect host recognition by the phage. Pip profiles ∆1 to ∆3 are very similar to each other except for a 6-aa gap between positions 453 and 459 of the reference protein in profile ∆2 and one aa changes (Q520L) in profile ∆3. None of the strains carrying these Pip derivatives can be infected by any of our

(17)

c2-type phages (except for SMQ384) suggesting that at least some of these mutations can prevent the phages from attaching to the mutated protein. The striking phage sensitivity of L. lactis SMQ384 is most probably caused by other characteristics (e.g. lack of specific resistance mechanisms) that were not identified by our analysis.

Pip profiles ∆4 and ∆5 are identical with the exception of one aa change in profile ∆4 (A200V). Their main shared characteristic is a 30-aa gap at the N terminus of the protein. A transmembrane domain prediction analysis indicated that, as a consequence, the N-terminal of the protein is on the outer side of the cell membrane, while the N-terminal domain of the Pip proteins with other mutation profiles and of the reference protein is buried inside the cell membrane and in the cytoplasm (Figures 2 and S1). Some of the strains possessing Pip with a ∆4 or ∆5 profile are more sensitive to c2-type phages, even more than L. lactis MG1363. It is tempting to speculate that the enlarged exposed area of their Pip proteins contributes to this sensitivity eg, by offering a wider or better binding area for some of the phage RBPs.

The role of the middle region of the RBP of c2 phages in host recognition

As mentioned above, the N- and C- termini of the putative c2 isolates’ RBPs are normally used for classification of these phages. The middle region of these proteins (the product of ORF l15 in phage c2) varies in length among all the known isolates of this phage species. Its role in host receptor recognition, if any, is still unclear. By exchanging the middle regions of the RBPs of phage CHL92 and bIL67, recombinant bacteriophages with modified host ranges were created (14). However, a correlation between the sequence of the middle variable part of this protein and phage host range could not be corroborated using a larger number of c2- and bIL67-type phages (27). Thus, this region might not be directly involved in host receptor recognition or it might help in recognizing different receptors. The middle region in the RBPs of the phages MP113, MP124, MP130, MP131, MP132, and MP139 are all relatively small (≤ 11 aa residues), while their sensitive hosts display a high variability in terms of the Pip/YjaE variants and CWP type they express (Table 3). This is especially true for phages MP113, MP130 and MP135 as the variable region in their RBPs are identical while they greatly differ with respect to their host range (Table 3).

The middle region of the RBPs of the remaining bacteriophages varies greatly in size (from 61 aa residues in MP115 to 366 aa residues in MP123) and aa sequence. The highest sequence similarities are between the variable region of the RBP of phage MP123 and those of phages MP134 and MP112 (53.5 % and 49%, respectively), and between those of phages MP112 and MP127 (49%). These five phages have very different host ranges and belong to different c2 subspecies (Table 3).

(18)

Given all of the above, we conclude that the middle region of the RBPs of the c2-phages is not involved in specific receptor binding.

A new class of RBPs might exists among 936 bacteriophages

Bacteriophages belonging to the 936 species are among the most commonly isolated lactococcal phages in dairy industry and, consequently, their interactions with their hosts have been extensively investigated (55). The 936 phages bind to sugar molecules that form the CWP surrounding their hosts cells (20). These phages have recently been divided into 5 groups on the basis of the aa sequence of the C-terminus of their RBPs. This grouping, in turn, correlates with the CWP type of the phages’ hosts (25).

Three out of the nine phages of the 936-species in our collection, are capable of infecting only one of the examined strains, while the remaining six are unable to propagate on any of the strains (Table 3). This narrower host range compared with that of the c2 phages is in agreement with a previous report (12). We classified our 936 isolates by blasting the C-terminus of their predicted RBPs against those of publicly available 936 phages. The C-terminus of the RBPs of phages MP114, MP117 and MP143 display an aa identity > 90% with that of phage Phi5.12 (25), while the C-termini of the RBPs of phages MP120 and MP121 share >70 % identity with those of phages PhiG (25) and 712 (56), respectively. Since these three known 936 bacteriophages belong to the RBP group I (25), these results allow classify phages MP114, MP117 and MP143 as belonging to this group as well. Phages MP117 and MP143 can both infect L. lactis 3107. This strain possesses a type-C CWP and belongs to the subspecies cremoris, characteristics previously identified as preferred by phages of RBP group I (24, 25). The low number of strains with CWP type C among the lactococcal hosts examined here might be the reason for the narrow host range of the remaining three phages of this group.

The C- termini of the RBPs of the remaining four 936 phages of our collection (MP111, MP118, MP144 and MP145) display a low aa identity (e.g. less than 40%) with the same region of the RBPs of all the reference phages for the 5 known RBP groups (25). However, higher aa identity (> 90%) was observed in the C-termini of the RBPs of these 4 phages with the same region in the RBPs of other 936 bacteriophages present on the NCBI but not included in the original classification study (25). Specifically, this high aa identity was found between phages MP111 and MP143 and the known phages Q23, Q64, and Q62 (57); between phage MP118 and the known phages ϕ HD17, ϕ HD18, and ϕ HD19 (58); and finally between phage MP154 and the known phage ϕ16802 (accession number: ASZ71120.1). These findings suggest that the RBPs of these publicly available phages, including those of phages MP111, MP118, MP143 and MP145 might entail a new and yet uncharacterized 936 bacteriophages group. Among this group, only phage MP143 can propagate on L. lactis SK11 (type-C CWP), while the other isolates are unable to infect

(19)

any of the strains tested. A more extensive host range analysis is needed to identify the preferred CWP type recognized by this group of phages and to explain the narrow host range of phages MP111, MP118 and MP145.

DISCUSSION

In this study we utilized a combined phenotypic and in silico analysis approach to evaluate the host range of lactococcal bacteriophages isolated from the dairy environment, using a panel of L. lactis dairy strains with different genotypes. Thirteen and nine bacteriophages of the c2 and 936 species, respectively, were evaluated: the sequence of their RBPs and of the corresponding proteinaceous or saccharidic receptors encoded by the lactococcal strains tested were analyzed in silico to examine the genetic determinants of the bacteriophage host range.

c2 bacteriophages first reversibly attach to sugar molecules on the surface of lactococcal cells, hence these sugars play an important role in host recognition. However, for the phage to irreversibly bind to the cell surface, binding must occur with either the Pip or YjaE protein. Here we identified mutations patterns in these two c2 phage receptors in all the strains of our collection. Some of the mutation profiles seem to broaden the sensitivity of the strain to a certain subgroup of phages, as in the case of some strains with a ∆5 Pip profile. Although the primary biological role of Pip and YjaE in the host cell is not known, we hypothesize that certain mutation patterns, as for example Pip ∆5, might improve the specific functionality of the protein and are therefore maintained even if they lead to higher phage sensitivity. Analysis of the mutation patterns also revealed that the second and third extracellular loops of the Pip protein and the third extracellular loop of YjaE never show aa changes. Thus, these loops are either not properly exposed for phage RBP recognition and binding, or mutations in these loops are not compatible with the main biological function of the proteins. These hypotheses might suggest that the interaction of the c2 phages with their receptors only takes place on the first extracellular loop of Pip and on the first and/or second loop on YjaE.

Our results also show that the correlation between c2 phage subgroups and the receptor they recognize (27) does not apply to all the phages and strains studied here. Previous c2 phage-host interaction studies that aimed at analyzing the receptors recognized by these phages have been conducted on bacteriophage insensitive mutants (BIMs) derived from a restricted pool of industrial starter strains isolated upon an infection in the production line (27). It is not surprising that, in this case, the number of mutations that the BIMs display in their Pip and/or YjaE proteins is low. This, in turn, makes connecting a specific mutation to the observed resistance phenotype more direct. Given the broader

(20)

heterogeneity in our lactococcal strains collection, we could identify at least one exception to the current knowledge regarding c2 phage-host interaction, derived by the results obtained for bIL67-type phage MP134. This shows that host recognition for this group of phages is perhaps more intricate than previously reported.

Lastly our in silico analysis of the middle region of the putative RBP of all the examined c2 bacteriophages suggested that this portion of the protein is most probably not directly involved in host recognition specificity. However, genetic engineering of this RBP region in different c2 phages would be needed to definitively answer this question.

The aa sequences of the RBPs of our 936 phages were used to sort them on the basis of a correlation between the C-terminus RBP sequence of the phages and the cell wall polysaccharide (CWP) type of their hosts (25). Five isolates were identified as belonging to the RBP group I, composed of phages binding to sugars in the type-C CWP. The remaining four phages did not fall into any of the other RBP groups although their RBP showed high similarity to that of other 936 phages currently not classified in any known group. Possibly, the RBPs of these isolates form a new group for which no information is available regarding the CWP type they recognize and that might even be of an unknown type, as in the case of the lactococcal strain 184.

The 936 bacteriophages tested here had very narrow host ranges, which is in agreement with previous results (12). This narrow host range may be further explained by the low number of type-C CWP expressing strains composing our library, that would be the preferred type of pellicle that phages MP114, MP117, MP120, MP121 and MP134 would utilize as host receptors. Further explanation for this results is given by the postulated hypothesis that the remaining phages examined in this study recognize receptors embedded in CWP types that none of the lactococcal strains in our library possess. With this study we contributed to expand the current knowledge regarding lactococcal phage-host interaction in dairy environments and our findings show that this represents a dynamic and multifaceted topic that needs continuous investigation and input of new data.

(21)

SUPPLEMENTAL MATERIAL

(22)

Figure S1. Alignments of proteins Pip and YjaE sequences of all the strains analyzed in this study. Panel A: Pip protein alignment. Panel B: YjaE protein alignment

The Pip protein encoded by L.lactis MG1363 and the YjaE protein encoded by L.lactis IL1403 are used as references and their entire aa sequence is shown at the top of the corresponding diagrams. The name of the L.lactis strains are showed on the left of the picture, the mutation pattern categories to which the strains belong are reported only once, on the top left margin of the alignment. The residue numbers above the alignments refer to the reference protein sequence. Amino acid that, in each analyzed protein, differ from the reference sequence are presented with the one letter code. Dots indicate identical residues, dashes represent gaps. The predicted outer membrane portions of the proteins are indicated with grey boxes

B

(23)

ACKNOWLEDGMENTS

We thank Douwe van Sinderen for kindly providing lactococcal strains UC509.9, 184, ML8, 229, UL8, C10, and 3107; and Sylvain Moineau for kindly providing lactococcal strains SMQ86, SMQ384, SMQ385, SMQ450, and SMQ562. This work was carried out within the BE-Basic R&D Program, which was granted an FES subsidy from the Dutch Ministry of Economic Affairs.

(24)

REFERENCES

1. Leroy F, De Vuyst L. 2004. Lactic acid bacteria as functional starter cultures for the food fermentation industry. Trends Food Sci Technol 15:67–78.

2. Steele J, Broadbent J, Kok J. 2013. Perspectives on the contribution of lactic acid bacteria to cheese flavor development. Curr Opin Biotechnol 24:135–141.

3. Song AA-L, In LLA, Lim SHE, Rahim RA. 2017. A review on Lactococcus lactis: from food to factory. Microb Cell Factories 16:55.

4. Garneau JE, Moineau S. 2011. Bacteriophages of lactic acid bacteria and their impact on milk fermentations. Microb Cell Factories 10 Suppl 1:S20.

5. Samson JE, Moineau S. 2013. Bacteriophages in food fermentations: new frontiers in a continuous arms race. Annu Rev Food Sci Technol 4:347–368.

6. Pujato SA, Quiberoni A, Mercanti DJ. 2019. Bacteriophages on dairy foods. J Appl Microbiol 126:14–30. 7. Brussow H. 2001. Phages of dairy bacteria. Annu Rev Microbiol 55:283–303.

8. Mahony J, Van Sinderen D. 2014. Current taxonomy of phages infecting lactic acid bacteria. Virology 5:7. 9. Braun V, Hertwig S, Neve H, Geis A, Teuber M. 1989. Taxonomic Differentiation of Bacteriophages of Lactococcus

lactis by Electron Microscopy, DNA-DNA Hybridization, and Protein Profiles. Microbiology 135:2551–2560. 10. Deveau H, Labrie SJ, Chopin M-C, Moineau S. 2006. Biodiversity and Classification of Lactococcal Phages. Appl

Environ Microbiol 72:4338–4346.

11. Muhammed MK, Kot W, Neve H, Mahony J, Castro-Mejía JL, Krych L, Hansen LH, Nielsen DS, Sørensen SJ, Heller KJ, van Sinderen D, Vogensen FK. 2017. Metagenomic Analysis of Dairy Bacteriophages: Extraction Method and Pilot Study on Whey Samples Derived from Using Undefined and Defined Mesophilic Starter Cultures. Appl Environ Microbiol 83:e00888–17.

12. Oliveira J, Mahony J, Hanemaaijer L, Kouwen TRHM, van Sinderen D. 2018. Biodiversity of bacteriophages infecting Lactococcus lactis starter cultures. J Dairy Sci 101:96–105.

13. Mahony J, Bottacini F, van Sinderen D, Fitzgerald GF. 2014. Progress in lactic acid bacterial phage research. Microb Cell Factories 13:S1.

14. Stuer-Lauridsen B, Janzen T, Schnabl J, Johansen E. 2003. Identification of the host determinant of two prolate-headed phages infecting Lactococcus lactis. Virology 309:10–17.

15. Vegge CS, Vogensen FK, Mc Grath S, Neve H, van Sinderen D, Brøndsted L. 2006. Identification of the lower baseplate protein as the antireceptor of the temperate lactococcal bacteriophages TP901-1 and Tuc2009. J Bacteriol 188:55–63.

16. Veesler D, Spinelli S, Mahony J, Lichière J, Blangy S, Bricogne G, Legrand P, Ortiz-Lombardia M, Campanacci V, van Sinderen D, Cambillau C. 2012. Structure of the phage TP901-1 1.8 MDa baseplate suggests an alternative host adhesion mechanism. Proc Natl Acad Sci U S A 109:8954–8958.

17. Bebeacua C, Tremblay D, Farenc C, Chapot-Chartier M-P, Sadovskaya I, van Heel M, Veesler D, Moineau S, Cambillau C. 2013. Structure, adsorption to host, and infection mechanism of virulent lactococcal phage p2. J Virol 87:12302–12312.

18. Hayes S, Duhoo Y, Neve H, Murphy J, Noben J-P, Franz CMAP, Cambillau C, Mahony J, Nauta A, Van Sinderen D. 2018. Identification of Dual Receptor Binding Protein Systems in Lactococcal 936 Group Phages. Viruses 10:668.

19. Chapot-Chartier M-P, Vinogradov E, Sadovskaya I, Andre G, Mistou M-Y, Trieu-Cuot P, Furlan S, Bidnenko E, Courtin P, Péchoux C, Hols P, Dufrêne YF, Kulakauskas S. 2010. Cell surface of Lactococcus lactis is covered by a protective polysaccharide pellicle. J Biol Chem 285:10464–10471.

20. Ainsworth S, Sadovskaya I, Vinogradov E, Courtin P, Guerardel Y, Mahony J, Grard T, Cambillau C, Chapot-Chartier M-P, Sinderen D van. 2014. Differences in Lactococcal Cell Wall Polysaccharide Structure Are Major Determining Factors in Bacteriophage Sensitivity. mBio 5:e00880-14.

21. Collins B, Bebeacua C, Mahony J, Blangy S, Douillard FP, Veesler D, Cambillau C, van Sinderen D. 2013. Structure and functional analysis of the host recognition device of lactococcal phage tuc2009. J Virol 87:8429–8440. 22. Stockdale SR, Mahony J, Courtin P, Chapot-Chartier M-P, van Pijkeren J-P, Britton RA, Neve H, Heller KJ, Aideh

B, Vogensen FK, van Sinderen D. 2013. The lactococcal phages Tuc2009 and TP901-1 incorporate two alternate forms of their tail fiber into their virions for infection specialization. J Biol Chem 288:5581–5590.

23. Legrand P, Collins B, Blangy S, Murphy J, Spinelli S, Gutierrez C, Richet N, Kellenberger C, Desmyter A, Mahony J, van Sinderen D, Cambillau C. 2016. The Atomic Structure of the Phage Tuc2009 Baseplate Tripod Suggests that Host Recognition Involves Two Different Carbohydrate Binding Modules. mBio 7(1):e01781-15.

24. Mahony J, Kot W, Murphy J, Ainsworth S, Neve H, Hansen LH, Heller KJ, Sørensen SJ, Hammer K, Cambillau C, Vogensen FK, van Sinderen D. 2013. Investigation of the relationship between lactococcal host cell wall polysaccharide genotype and 936 phage receptor binding protein phylogeny. Appl Environ Microbiol 79:4385– 4392.

25. Murphy J, Bottacini F, Mahony J, Kelleher P, Neve H, Zomer A, Nauta A, van Sinderen D. 2016. Comparative genomics and functional analysis of the 936 group of lactococcal Siphoviridae phages. Sci Rep 6:21345.

(25)

26. Valyasevi R, Sandine WE, Geller BL. 1991. A membrane protein is required for bacteriophage c2 infection of Lactococcus lactis subsp. lactis C2. J Bacteriol 173:6095–6100.

27. Millen AM, Romero DA. 2016. Genetic determinants of lactococcal C2viruses for host infection and their role in phage evolution. J Gen Virol 97:1998–2007.

28. Lillehaug D. 1997. An improved plaque assay for poor plaque-producing temperate lactococcal bacteriophages. J Appl Microbiol 83:85–90.

29. Labrie S, Moineau S. 2000. Multiplex PCR for detection and identification of lactococcal bacteriophages. Appl Environ Microbiol 66:987–994.

30. del Rio B, Binetti AG, Martín MC, Fernández M, Magadán AH, Alvarez MA. 2007. Multiplex PCR for the detection and identification of dairy bacteriophages in milk. Food Microbiol 24:75–81.

31. Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinforma Oxf Engl 30:2114–2120.

32. Coil D, Jospin G, Darling AE. 2015. A5-miseq: an updated pipeline to assemble microbial genomes from Illumina MiSeq data. Bioinforma Oxf Engl 31:587–589.

33. Brettin T, Davis JJ, Disz T, Edwards RA, Gerdes S, Olsen GJ, Olson R, Overbeek R, Parrello B, Pusch GD, Shukla M, Thomason JA, Stevens R, Vonstein V, Wattam AR, Xia F. 2015. RASTtk: a modular and extensible implementation of the RAST algorithm for building custom annotation pipelines and annotating batches of genomes. Sci Rep 5:8365.

34. Madeira F, Park YM, Lee J, Buso N, Gur T, Madhusoodanan N, Basutkar P, Tivey ARN, Potter SC, Finn RD, Lopez R. 2019. The EMBL-EBI search and sequence analysis tools APIs in 2019. Nucleic Acids Res 47:W636–W641. 35. Waterhouse AM, Procter JB, Martin DMA, Clamp M, Barton GJ. 2009. Jalview Version 2—a multiple sequence

alignment editor and analysis workbench. Bioinformatics 25:1189–1191.

36. Finn RD, Mistry J, Schuster-Böckler B, Griffiths-Jones S, Hollich V, Lassmann T, Moxon S, Marshall M, Khanna A, Durbin R, Eddy SR, Sonnhammer ELL, Bateman A. 2006. Pfam: clans, web tools and services. Nucleic Acids Res 34:D247–D251.

37. Marcelli B, de Jong A, Karsens H, Janzen T, Kok J, Kuipers OP. 2019. A specific sugar moiety in the Lactococcus lactis cell wall pellicle is required for infection by CHPC971, a member of the rare 1706 phage species. Appl Environ Microbiol 85:e01224–19.

38. Gasson MJ. 1983. Plasmid complements of Streptococcus lactis NCDO 712 and other lactic streptococci after protoplast-induced curing. J Bacteriol 154:1–9.

39. Kelly WJ, Ward LJH, Leahy SC. 2010. Chromosomal Diversity in Lactococcus lactis and the Origin of Dairy Starter Cultures. Genome Biol Evol 2:729–744.

40. Labrie S, Moineau S. 2002. Complete genomic sequence of bacteriophage ul36: demonstration of phage heterogeneity within the P335 quasi-species of lactococcal phages. Virology 296:308–320.

41. Samson JE, Moineau S. 2010. Characterization of Lactococcus lactis phage 949 and comparison with other lactococcal phages. Appl Environ Microbiol 76:6843–6852.

42. Crow VL, Martley FG, Coolbear T, Roundhill SJ. 1995. The influence of phage-assisted lysis of Lactococcus lactis subsp. lactis ML8 on cheddar cheese ripening. Int Dairy J 5:451–472.

43. Erazo Garzon A, Mahony J, Bottacini F, Kelleher P, van Sinderen D. 2019. Complete Genome Sequence of Lactococcus lactis subsp. cremoris 3107, Host for the Model Lactococcal P335 Bacteriophage TP901-1. Microbiol Resour Announc 8:e01635-18.

44. Jarvis AW. 1984. Differentiation of lactic streptococcal phages into phage species by DNA-DNA homology. Appl Env Microbiol 47:343–349.

45. Siezen RJ, Renckens B, Swam I van, Peters S, Kranenburg R van, Kleerebezem M, Vos WM de. 2005. Complete Sequences of Four Plasmids of Lactococcus lactis subsp. cremoris SK11 Reveal Extensive Adaptation to the Dairy Environment. Appl Environ Microbiol 71:8371–8382.

46. Neve H, Geis A, Teuber M. 1984. Conjugal transfer and characterization of bacteriocin plasmids in group N (lactic acid) streptococci. J Bacteriol 157:833–838.

47. Mooney DT, Jann M, Geller BL. 2006. Subcellular location of phage infection protein (Pip) in Lactococcus lactis. Can J Microbiol 52:664–672.

48. São-José C, Baptista C, Santos MA. 2004. Bacillus subtilis Operon Encoding a Membrane Receptor for Bacteriophage SPP1. J Bacteriol 186:8337–8346.

49. Bottai D, Gröschel MI, Brosch R. 2017. Type VII Secretion Systems in Gram-Positive Bacteria. Curr Top Microbiol Immunol 404:235–265.

50. Derkx PMF, Janzen T, Sørensen KI, Christensen JE, Stuer-Lauridsen B, Johansen E. 2014. The art of strain improvement of industrial lactic acid bacteria without the use of recombinant DNA technology. Microb Cell Factories 13,S5.

51. Koebnik R. 1999. Structural and functional roles of the surface-exposed loops of the beta-barrel membrane protein OmpA from Escherichia coli. J Bacteriol 181:3688–3694.

52. São-José C, Lhuillier S, Lurz R, Melki R, Lepault J, Santos MA, Tavares P. 2006. The ectodomain of the viral receptor YueB forms a fiber that triggers ejection of bacteriophage SPP1 DNA. J Biol Chem 281:11464–11470.

(26)

2

53. Kraus J, Geller BL. 1998. Membrane Receptor for Prolate Phages is Not Required for Infection of Lactococcus

lactis by Small or Large Isometric Phages. J Dairy Sci 81:2329–2335.

54. Stuer-Lauridsen B, Janzen T. 2007. Bacteriophage resistant Lactic Acid Bacteria, EP1838839 (B1).

55. Mahony J, Murphy J, van Sinderen D. 2012. Lactococcal 936-type phages and dairy fermentation problems: from detection to evolution and prevention. Front Microbiol 3:335.

56. Mahony J, Deveau H, Mc Grath S, Ventura M, Canchaya C, Moineau S, Fitzgerald GF, van Sinderen D. 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.

57. Deveau H, Van Calsteren M-R, Moineau S. 2002. Effect of Exopolysaccharides on Phage-Host Interactions in Lactococcus lactis. Appl Environ Microbiol 68:4364–4369.

58. Tremblay DM, Tegoni M, Spinelli S, Campanacci V, Blangy S, Huyghe C, Desmyter A, Labrie S, Moineau S, Cambillau C. 2006. Receptor-binding protein of Lactococcus lactis phages: identification and characterization of the saccharide receptor-binding site. J Bacteriol 188:2400–2410.

(27)