Identification and Molecular
Characterization of the Chromosomal
Exopolysaccharide Biosynthesis Gene
Cluster from Lactococcus lactis subsp.
cremoris SMQ-461
N. Dabour and G. LaPointe
Appl. Environ. Microbiol. 2005, 71(11):7414. DOI:
10.1128/AEM.71.11.7414-7425.2005.
Downloaded from http://aem.asm.org/ on February 12, 2013 by guest
Updated information and services can be found at:
http://aem.asm.org/content/71/11/7414
These include:
REFERENCES This article cites 60 articles, 21 of which can be accessed free
at: http://aem.asm.org/content/71/11/7414#ref-list-1
CONTENT ALERTS Receive: RSS Feeds, eTOCs, free email alerts (when new
articles cite this article), more
Information about commercial reprint orders: http://journals.asm.org/site/misc/reprints.xhtml
To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 2005, p. 7414 7425 Vol. 71, No. 11
0099-2240/05/$08.00 0 doi:10.1128/AEM.71.11.7414 7425.2005
Copyright 2005, American Society for Microbiology. All Rights Reserved.
Identi cation and Molecular Characterization of the Chromosomal
Exopolysaccharide Biosynthesis Gene Cluster from
Lactococcus lactis subsp. cremoris SMQ-461
N. Dabour1,2 and G. LaPointe1*
STELA Dairy Research Centre and Institute for Nutraceuticals and Functional Foods, Universite Laval,
Quebec, QC, Canada G1K 7P4,1 and Department of Dairy Science and Technology,
Faculty of Agriculture, University of Alexandria, Alexandria, Egypt 2
Received 9 March 2005/Accepted 20 July 2005
Downloaded from http://aem.asm.org/ on February 12, 2013 by guest
The exopolysaccharide (EPS) capsule-forming strain SMQ-461 of Lactococcus lactis subsp. cremoris, isolated
from raw milk, produces EPS with an apparent molecular mass of >1.6 106 Da. The EPS biosynthetic genes
are located on the chromosome in a 13.2-kb region consisting of 15 open reading frames. This region is anked
by three IS1077-related tnp genes (L. lactis) at the 5 end and orfY, along with an IS981-related tnp gene, at the
3 end. The eps genes are organized in speci c regions involved in regulation, chain length determination,
biosynthesis of the repeat unit, polymerization, and export. Three (epsGIK) of the six predicted glycosyltrans-
ferase gene products showed low amino acid similarity with known glycosyltransferases. The structure of the
repeat unit could thus be different from those known to date for Lactococcus. Reverse transcription-PCR
analysis revealed that the eps locus is transcribed as a single mRNA. The function of the eps gene cluster was
con rmed by disrupting the priming glycosyltransferase gene (epsD) in Lactococcus cremoris SMQ-461, gen-
erating non-EPS-producing reversible mutants. This is the rst report of a chromosomal location for EPS
genetic elements in Lactococcus cremoris, with novel glycosyltransferases not encountered before in lactic acid
bacteria.
Lactic acid bacteria (LAB) are widely used in fermented dairy NIZO B40; it comprises 14 plasmid-encoded genes (56). Since
products, mainly for lactic acid formation but also for the pro- then, partial sequences of eps gene clusters have been identi-
duction of minor avor and preservation components. Some LAB ed in L. lactis subsp. cremoris NIZO B891 and NIZO B35
are also able to produce exopolysaccharides (EPS), which are strains (58). Recently, a large cluster consisting of 23 putative
either excreted in the growth medium as slime (ropy form) or EPS biosynthetic determinants has been identi ed on plasmid
remain attached to the bacterial cell wall forming capsular EPS pCI658 in L. lactis subsp. cremoris HO2 (20).
(6, 32). In the dairy industry, EPS-producing LAB, including the Increased sequence information about the eps gene clusters
genera Streptococcus, Lactobacillus, and Lactococcus, are used in of lactococci is important to identify strains that are able to
situ to improve the textural characteristics of fermented dairy produce novel EPS. The study of a new EPS is a fastidious
products, especially low-fat yoghurt and cheese. LAB are food- process, requiring extraction, puri cation, and chemical anal-
grade bacteria that can produce a wide variety of structurally yses for determining the sugar composition and structure of
different EPS with potential uses for new applications, for exam- the EPS produced (7, 14, 58). Furthermore, Lactococcus
ple, in replacement of polysaccharides such as gellan, pullulan, strains generally produce a rather low quantity of EPS (15, 16,
xanthan, and bacterial alginates that are presently produced by 25, 61). Therefore, the identi cation of new wild-type Lacto-
non-food-grade bacteria (12, 25, 48). coccus strains producing unique EPS polymers is a consider-
EPS-producing LAB, including strains of Lactococcus spp., able challenge that would bene t from rapid screening meth-
have been shown to express at least two distinct phenotypic ods of the genetic elements involved. Restriction fragment
forms of EPS, either ropy and/or capsular forms (32). More- length polymorphism has been used to classify lactococcal
over, they produce EPS with considerable diversity in structure EPS-producing strains in relation to the monosaccharide com-
and composition (14, 54, 55, 58, 64). This diversity in EPS position of the repeat unit of the EPS into three major groups,
composition indicates that LAB contain a vast pool of glyco-
along with a minor unique group (58). Recently, Deveau et al.
syltransferases with a wide range of sugar and linkage speci-
(14) applied restriction fragment length polymorphism using
cities. EPS-producing lactococci in particular have received
two different enzymes (AcyI and HindII) for grouping seven
growing attention in recent years, especially for the analysis of
EPS-producing lactococcal strains. A novel group of EPS-pro-
genes encoding EPS biosynthesis. The rst lactococcal eps lo-
ducing lactococci was identi ed, including L. lactis subsp. cre-
cus identi ed was that of Lactococcus lactis subsp. cremoris
moris SMQ-461. EPS biosynthesis by this strain has been
shown to have an impact on reduced-fat Cheddar cheese pro-
duction and physical characteristics, notably by increasing
* Corresponding author. Mailing address: STELA Dairy Research
moisture retention and cheese yield (13).
Centre, Room 1316, Pavillon Paul-Comtois, Universite Laval, Quebec,
The present study reveals the quantity and molecular
QC, Canada G1K 7P4. Phone: 418-***-****, ext. 3100. Fax: (418)
mass of the EPS produced by L. lactis subsp. cremoris strain
656-3353. E-mail: abqndw@r.postjobfree.com.
7414
VOL. 71, 2005 CHROMOSOMAL EPS GENE CLUSTER FROM L. CREMORIS 7415
TABLE 1. Bacterial strains, plasmids, and oligonucleotide primers used in the present study
Strain, plasmid, Relevant characteristic(s) Reference, source,
or sequence (5 to 3 )a
or primer or target
Strains
L. lactis subsp. cremoris SMQ-461 Raw milk isolate, EPS ; rosy white colonies on RRM17 medium 15
L. lactis subsp. cremoris ND461M SMQ-461 derivative with 4.5-kb pND9 integrated into the chromosome; This study
colonies with red phenotype on RRM17 medium with Em5
L. lactis subsp. cremoris ND461R ND461M derivative with pND9 excised from the chromosome; This study
colonies reverted to wild-type rosy white on RRM17 medium; Ems
L. lactis subsp. cremoris ND461D1 ND461M derivative with pND9 excised from the chromosome; This study
red colonies on RRM17 medium; Ems
L. lactis subsp. cremoris ND461D2 ND461M derivative with pND9 excised from the chromosome; This study
red colonies on RRM17 medium; Ems
L. lactis subsp. cremoris IL1403 Plasmid free 10
E. coli JM109 Cloning host (F proAB lacIqZ M15) Promega, Inc.
E. coli TOP10 Cloning host (F mcr 80lacZ M15 lacX74 recA1 deoR araD139 Invitrogen Life
Downloaded from http://aem.asm.org/ on February 12, 2013 by guest
(ara-leu)7697 galU galK resL(Strr) endA1 nupG) Technologies
Plasmids
PCR cloning vector, Kmr Amr
pCR4-TOPO Invitrogen
PCR cloning vector, Amr
pGEM-T Easy Invitrogen
Thermosensitive shuttle vector, 4.6 kb; Emr
pGh9 38
pND2 2.0-kb fragment (epsBCD) cloned into the EcoRI site of pUC18 This study
4.3-kb fragment (tnp and epsRXA) cloned into pCR4-Topo; Amr
pND4 This study
3.5-kb fragment (epsABCD) cloned into pCR4-Topo; Amr
pND5 This study
3.5-kb fragment (epsML) cloned into pCR4-Topo; Amr
pND6 This study
9.7-kb fragment (epsD to L) cloned into pGEM-T; Amr
pND7 This study
Amr; pCR4-Topo containing epsD
pND8 This study
Emr; 4.5 kb; pGh9 with epsD cloned into the PstI-XhoI site
pND9 This study
Primers
HD2 CGTACGATTCGTACGACCAT 14; PCR for epsB
HD15 TGACCAGTGACACTTGAAGC 14; PCR for epsB
SMQ461ER AATCCCTCCTAGATTAATCGC PCR for epsE
AB40F TTAAATGCTTCGGGGAATAAGGTTTGGCTAGATTA GenomeWalker-PCR
AB40FN TGGAGAAGAAATGCAGGAAACACAGGAACAGACGA GenomeWalker-PCR
AB40R TCGTCTGTTCCTGTGTTTCCTGCATTTCTTCTCCA GenomeWalker-PCR
AB40RN TAATCTAGCCAAACCTTATTCCCCGAAGCATTTAA GenomeWalker-PCR
LB40R TATTTCATCACAATATAATCCGGTACGGCTCGATCATCTT GenomeWalker-PCR
LB40RN GACTAGCAACAATCGTTTTACCATTGACAGATAGT GenomeWalker-PCR
wccf tttctgcagTAACAGCTTCGAGTGTCACTGGTCA PCR epsD
wdnr ctcgagGGCATGGTAGGCAGCTTTAATTTCTGGA PCR epsD
wdcf gctgcctaccatgccGCGCCTCTCTTACTTACTCATGTGT PCR epsD
wecr ctcgagGGCTACCGCAGCACCACTCG PCR epsD
SMQ461DFF CAGCTTTGAAGATTTGTATTGA RT-PCR for epsF
SMQ461FR GTGAGTTCCAACAGTTACAAAA RT-PCR for epsF
SMQ461GR CTTCACCAATATGTTTAACTTC RT-PCR for epsG
SMQ461HF GGTATGATGTCAACTTTTTCTT RT-PCR for epsH
SMQ461HR TGATCCTCCCATGACATTTTTT RT-PCR for epsH
SMQ461JF AGTGAATTAATAGGCAAAGATA RT-PCR for epsJ
SMQ461JR GACGCAAAATATCTAATCATCA RT-PCR for epsJ
SMQ461MF TATTGCAAGTATTTTTAGGAGC RT-PCR for epsM
SMQ461MR TAGTTCTAGAAATATATGGTGC RT-PCR for epsM
SMQ461LR CGAGCTGTTTGTTTTTGTATAA RT-PCR for epsL
SMQ461YR CAACTGGTAAAAATAATTCT RT-PCR for orfY
a
Restriction sites of primers are indicated in lowercase.
SMQ-461, as well as the organization and transcription of the L. lactis strains were grown at 30 C in M17 broth (Quelab, Montreal, Quebec,
Canada) supplemented with 0.5% glucose (GM17) or lactose (LM17), unless spec-
novel chromosomal eps gene cluster. The function of the eps
i ed otherwise. Media were solidi ed with 1.5% agar (Quelab). To distinguish be-
gene cluster in EPS biosynthesis is experimentally demon-
tween EPS-producing and nonproducing (mutant) cells, GM17 agar medium containing
strated by disrupting the priming glycosyltransferase gene 0.08% ruthenium red (RRM17) was used. A stock solution of ruthenium red (Sigma
(epsD), generating reversible non-EPS-producing mutants. Chemical Co, St. Louis, MO) at 10% (wt/vol) in water was sterilized through a 0.45- m
lter (Sartorius AG, Gottingen, Germany), and an appropriate volume was added to the
molten GM17 agar just prior to pouring it into petri plates. The presence or absence of
MATERIALS AND METHODS capsule on cells isolated from RRM17 was determined using India ink and crystal violet
counterstaining according to the capsule staining method described by Collins and Lyne
Bacterial strains and media. The bacterial strains and plasmids used in this study
are listed in Table 1. All strains were maintained in 20% glycerol stock at 80 C. (11). Strains of Escherichia coli were routinely cultured in Luria-Bertani (LB; Quelab)
7416 DABOUR AND LAPOINTE APPL. ENVIRON. MICROBIOL.
broth (40) and incubated at 37 C with aeration. When required, antibiotics were added Inc., Palo Alto, CA) and Expand High Fidelity polymerase (Roche Diagnostics
at the following concentrations: chloramphenicol hydrochloride or erythromycin (Sigma) GmbH-Boehringer Mannheim, Mannheim, Germany), under standard condi-
for L. lactis strains, 5 g ml 1, and ampicillin for E. coli strains, 100 g ml 1. tions as recommended by the manufacturer. Reaction speci city was controlled
Exopolysaccharide extraction and puri cation. An overnight culture of L. using the rst PCR ampli cations as template DNA with nested primers. Nested
lactis subsp. cremoris SMQ-461 in GM17 broth was standardized at an optical ampli cations which gave fragment sizes of 3 to 4 kb were puri ed with Micro-
density at 650 nm of 0.5, used to inoculate 500 ml of sterilized skim milk (11% con-PCR columns (Millipore, Microcon, and Milli-Q), ligated into pCR4-TOPO
[wt/wt] total solids) at a level of 2% (vol/vol), and then incubated at 30 C or 25 C plasmid (Invitrogen Life Technologies, Carlsbad, Calif.) or pGEM-T Easy
for 24 or 48 h. The EPS was extracted following the method of Cerning et al. (8), (Promega, Madison, Wisc.), and then transformed into E. coli strain TOP10
modi ed as follows. The fermented skim milk cultures were heated at 90 C for or JM109.
15 min to inactivate enzymes potentially able to degrade the EPS polymer. The Nucleotide sequence analysis. Automated DNA sequence analysis was carried
pH of milk samples was adjusted to 7.5 with 1 M NaOH and digested by pronase out on both strands by the DNA sequencing service of Laval University (Life and
E (EC 3.4.24.31; Sigma) in sterilized distilled water at a nal concentration of Health Sciences Pavillion, Quebec, Quebec, Canada) with an ABI Prism 3100
1/50 (wt/wt) for 20 h at 40 C with shaking in the presence of 1/2,000 (wt/vol) apparatus. Sequence data were assembled and analyzed with the Genetics Com-
merthiolate (Sigma) to inhibit bacterial growth. The bacterial cells were removed puter Group (Wisconsin) package 10 (Accelrys, San Diego, CA). Database
by centrifugation at 12,000 g for 15 min at 4 C, and trichloroacetic acid was similarity searches (46) were performed with the FASTA network service at the
added to the supernatant at a nal concentration of 12% (wt/vol). Precipitated National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov),
proteins were removed by centrifugation at 12,000 g for 20 min at 4 C. The EPS National Institutes of Health, Bethesda, MD (2). The BLASTX program (http:
Downloaded from http://aem.asm.org/ on February 12, 2013 by guest
in the supernatant was precipitated with 1 volume of acetone. After being //www.ncbi.nlm.nih.gov/BLAST) was used to translate the sequence of both
allowed to stand overnight at 4 C, the EPS was collected by centrifugation at DNA strands in all six open reading frames (ORFs) and to conduct similarity
16,000 g for 20 min at 4 C. The precipitated EPS was dissolved in deionized searches of the nucleotide and protein databases. Hydrophobicity plots were
water, dialyzed for 5 days against deionized water at 4 C (two water changes per generated with the TMpred program at ch.EMBnet.org (http://www.ch.embnet
day), and freeze-dried. To purify the EPS, the lyophilized samples were dissolved .org/software/TMPRED_form.html), which allows a prediction of membrane-
in water, extracted two times with an equal volume of phenol-chloroform- spanning regions and their orientation.
isoamylalcohol (25:24:1 [vol/vol/vol]), and precipitated overnight with an equal Transcription analysis of eps genes. RNA was extracted from 3-ml cultures of
volume of acetone. The precipitated EPS was dissolved in water, dialyzed, and L. lactis subsp. cremoris SMQ-461 after 6, 12, and 24 h of incubation, corre-
lyophilized as described above. The EPS concentration in glucose equivalents sponding to early, late exponential, and stationary growth phases, respectively,
was determined by the phenol-sulfuric acid method described by Dubois et al. with the RNeasy Mini Kit (QIAGEN), according to the manufacturer s instruc-
(19). D,L-Glucose (Sigma) was used as a standard. For LM17 cultures, the same tions. The extracted RNA was treated with RNase-free DNaseI (QIAGEN) at
methods of EPS extraction and puri cation were applied, without the pronase E 25 C for 40 min, followed by a second puri cation step. Reverse transcription-
digestion step.
PCR (RT-PCR) was performed on 100 ng, 1 g, and 2 g of RNA with primers
Size exclusion high-performance liquid chromatography (HPLC) analysis.
(Table 1) derived from the L. lactis subsp. cremoris SMQ-461 eps cluster to cover
The molecular mass of EPS was determined by gel permeation chromatography
the intergenic spaces of the eps locus. As controls, each fragment was ampli ed
with a Waters HPLC system (600 controller with Waters TM 600 pump; Waters
with the same primers by using chromosomal DNA from strain SMQ-461 as a
Limited, Mississauga, Ontario, Canada) at 25 C. Two columns were connected in
template, and negative controls were performed under the same reaction con-
a series: TSK 4000SW (600 mm by 7.5 m) with a silica base (Beckman Coulter,
ditions without the reverse transcription stage.
Mississauga, Ontario, Canada) and TSK Gel G40000PWXL (300 mm by 7.8 mm)
Insertional inactivation of epsD. For epsD inactivation, primers wccf and wdnr
with a polymer base (Tosohaas Keystone, Montgomeryville, PA). The mobile
(Table 1) were used to amplify by PCR the rst 330 bp of epsD and 302 bp of the
phase used was 0.1 M ammonium acetate (Sigma) at pH 7.2 with a ow rate of
adjacent upstream anking sequence with the chromosomal DNA of strain
0.5 ml min 1. A volume of 50 l of EPS sample (concentration, 1 mg ml 1) was
SMQ-461 as a template. The wdcf and wecr primers were used to amplify the last
injected. The detection was performed with the SEDEX model 75 light-scatter-
183 bp of epsD and 277 bp of downstream anking DNA. The two puri ed PCR
ing detector (Scienti c Products & Equipment, Concord, Ontario, Canada) at
products were diluted 1/10, mixed, and ampli ed with primers wccf and wecr to
45 C. The results were analyzed by HPLC system Millennium 32 software,
create a nal PCR product carrying a 165-bp internal deletion of epsD by overlap
version 3.25, with comparison to a dextran standard series of 5 103, 1 104,
PCR. The resultant 1,092-bp PCR fragment was cloned into pCR4-TOPO
2 104, 5 104, 1 105, 2 105, 4 105, 8 105, and 1.6 106 Da (Showa
(Invitrogen). After PstI-XhoI digestion, the fragment was ligated into the tem-
Denko America, Inc., New York), which were migrated simultaneously.
perature-sensitive shuttle vector pGh9, also digested with PstI-XhoI, replacing
DNA isolation and manipulation. Lactococcal genomic DNA was extracted as
the ISS1 sequence. This construct was designated pND9, which was veri ed by
described by Hill et al. (26), with some modi cations. Cells from 2 ml of late-
PstI-XhoI digestion and PCR using wccf and wecr as primers.
log-phase culture were harvested; suspended in 50 mM Tris-HCl (pH 8.0) buffer
The pND9 construct was transformed into L. lactis subsp. lactis IL-1403 to
containing 1 mM EDTA (pH 8.0), 6.7% sucrose, and 50 g of lysozyme (Sigma);
produce supercoiled plasmid to facilitate its transfer to L. lactis subsp. cre-
and incubated 20 min on ice. Lysis was achieved by addition of 50 l of 10%
moris SMQ-461 by electroporation. Transformants were selected by growth at
sodium dodecyl sulfate, after which the lysate was treated with 20 l of a 20-mg
30 C on GM17 agar medium with 5 g ml 1 erythromycin (Sigma). A single
ml 1 stock solution of proteinase K (Sigma) at 65 C for 20 min. Two extractions
erythromycin-resistant colony was then inoculated into liquid culture (GM17)
with phenol-chloroform-isoamyl alcohol (25:24:1 [vol/vol/vol]) were carried out,
containing erythromycin (5 g ml 1). After overnight growth, the saturated
followed by chloroform extraction. The DNA was precipitated with 2 volumes of
culture was diluted 100 fold in GM17 broth medium without erythromycin
95% ethanol and a one-tenth volume of 3 M potassium acetate (pH 4.8) at
and incubated 150 min at 28 C to allow exponential growth to resume. The
20 C. The pellet was washed twice with ethanol (70% [vol/vol]) and solubilized
same culture was shifted to 38 C for 150 min to inhibit plasmid replication
in 100 l of 10 mM Tris-HCl (pH 8.0).
and allow integration. Serial dilutions were then performed with 0.1% (wt/
Plasmid extraction from lactococcal strains was performed according to
vol) peptone water and plated on GM17 agar with added erythromycin (to
O Sullivan and Klaenhammer (44). Small-scale isolation of E. coli plasmids was
detect the integrants) and on GM17 agar without antibiotic (to determine the
performed with QIAGEN (Mississauga, Ontario, Canada) columns as instructed
viable cell count) with incubation at 38 C. Dilutions were also plated on
by the manufacturer. Restriction endonucleases (Roche Diagnostics, Laval,
RRM17 agar with erythromycin to select red colonies, indicating integration
Quebec, Quebec, Canada) were used as recommended by the manufacturer.
into the eps locus affecting the phenotype. In all cases, plates were incubated
DNA hybridization was performed by transferring plasmid and chromosomal
at 38 C for 24 h. To allow excision of the plasmid from the chromosome via
DNA onto positively charged nylon membranes (Roche) by capillary blotting
a second recombination event, leaving either the deletion copy or reconsti-
(50). Probes were constructed by labeling with the Dig High-Prime labeling kit
tuting the wild-type genotype, integrants were serially passaged at least ve
(Roche), the PstI/XhoI fragment of pGh9, and the internal PCR fragment of
times in GM17 broth at 30 C in the absence of erythromycin and plated on
epsB, which was generated using primers HD2 and HD15 (Table 1). Prehybrid-
RRM17 after each passage. Erythromycin-sensitive colonies were screened
ization, hybridization, washes, and detection by chemiluminescence were per-
for the expected deletion by PCR ampli cation using primers (HD2 and
formed as suggested by the manufacturer (Roche).
SMQ461ER) that anneal in the eps locus (epsB and epsE) outside the se-
Chromosomal walking strategy for isolating eps genes. PCR was performed
quence region included in the pND9 construct.
with primary and nested primers designed from the highly conserved regions
Nucleotide sequence accession number. The sequence obtained in this study is
(epsA and epsL) of lactococcal strains NIZO B40 and NIZO B891 (Table 1).
Ampli cations were obtained with the Universal Genome Walker kit (Clontech, available under GenBank accession no. AY741550.
VOL. 71, 2005 CHROMOSOMAL EPS GENE CLUSTER FROM L. CREMORIS 7417
and 124 mg liter 1 was measured from skim milk cultures at
30 C after 24 h and 48 h of incubation, respectively, under
batch fermentation conditions. After incubation at 25 C for
both 24 and 48 h, production of 115 mg liter 1 EPS was
measured. The production of EPS in LM17 (containing 2%
lactose [wt/vol]) was found to be 152 and 100 mg liter 1,
respectively, after 24 and 48 h of incubation at 30 C. Stationary
phase in LM17 was attained between 20 h and 24 h (optical
density at 650 nm 2.22). The apparent molecular mass of
puri ed EPS from skim milk and LM17 cultures was 1.6
106 Da, as determined by high-performance size exclusion
chromatography.
Identi cation and cloning of the eps gene locus. To test
whether EPS production by L. lactis subsp. cremoris SMQ-461
Downloaded from http://aem.asm.org/ on February 12, 2013 by guest
is linked to plasmid or chromosomal DNA, Southern hybrid-
ization was carried out on plasmid and total DNA with a probe
from the internal gene fragment of epsB, which was generated
by PCR. No hybridization signals were detected with digested
and nondigested plasmid, whereas a strong signal was always
associated with total DNA (Fig. 2). This result indicates that
the eps locus of strain SMQ-461 is located on the chromosome.
FIG. 1. Phase contrast microscopy of India ink-stained cells of Lac-
Principal and nested primers in the most conserved regions
tococcus lactis subsp. cremoris SMQ-461, showing the cell-associated
capsular exopolysaccharide layer. Arrows indicate cells surrounded by of epsRXABC and epsL from known lactococcal eps operons
a clear zone representing the capsular layer. were designed (Tables 1 and 2), and used in a genome walking
PCR strategy with L. lactis subsp. cremoris SMQ-461 genomic
DNA, resulting in cloned PCR fragments of 2 to 9.7 kb (Table 1).
RESULTS
The inserts of these clones were completely sequenced on both
strands for further analysis. The G C moles percent content
Determination of EPS production and molecular mass.
of the eps locus was 28%, which is identical to the reported eps
L. lactis subsp. cremoris SMQ-461 produced a capsular EPS
gene cluster from L. lactis subsp. cremoris NIZO B40 (56) but
(Fig. 1) and did not form any visible ropy strings longer than
below that of typical G C mole percent content of 38 to 40%
5 mm when colonies on GM17 or LM17 were touched and
pulled with sterilized toothpicks. The EPS production of 160 reported for Lactococcus spp. (27).
FIG. 2. Detection of the eps genes from Lactococcus lactis subsp. cremoris SMQ-461 by Southern hybridization. Digested and undigested
plasmid extraction and total DNA (A); hybridization with the epsB probe (B). Lanes 1 and 2, plasmid extraction digested with SacI and HindIII,
respectively; lane 3, undigested plasmid; lanes 4 and 5, total DNA digested with SacI and HindIII, respectively; and lane 6, undigested total DNA.
7418 DABOUR AND LAPOINTE APPL. ENVIRON. MICROBIOL.
TABLE 2. Gene organization of identi ed ORFs, predicted physical properties of the hypothetical proteins encoded by the
eps gene cluster from Lactococcus lactis subsp. cremoris SMQ-461, and percentage of identity of the predicted proteins
with those involved in EPS biosynthesis in other bacteria
Highest % of identity with proteins from L. lactis subsp.
No. of cremoris or other bacteria
GC Putative Predicted Predicted
Gene amino
RBSa
mol% pI function
NIZO B891b NIZO B35b
HO2 NIZO B40
acids Others
(AF142639) (AF036485) (AF100298) (AF100297)
epsR 30 GAGGA 105 5.17 99 (EpsR) 98 (EpsR) Transcription regulator
epsX 30 AGGGAG 255 5.41 91 (EpsX) 98 (EpsX) Unknown protein
epsA 34 GGAG 259 9.52 93 (EpsA) 92 (EpsA) Chain length determination
epsB 36 AGGAG 231 7.83 89 (EpsB) 89 (EpsB) 87 (EpsB) Chain length determination
50 (CapC)c
epsC 33 GGAG 254 5.65 97 (EpsC) 95 (EpsC) 96 (EpsC) 92 (EpsC) Phosphotyrosine-protein
phosphatase
epsD 36 GGAG 228 9.06 97 (EpsD) 90 (EpsD) 89 (EpsD) 92 (EpsD) Glycosyltransferase
epsE 32 GGA 149 9.66 97 (EpsE) Glycosyltransferase
epsF 33 GGAGG 161 8.45 37 (EpsH) 40 (EpsF) 69 (EpsF) Glycosyltransferase
epsG 29 GAGGA 187 6.91 22 (EpsG) Galactosyltransferase
Downloaded from http://aem.asm.org/ on February 12, 2013 by guest
22 (PBPRA2679)d
epsH 26 AAGA 384 9.49 Possible polymerase
23 (Eps6J)e
epsI 24 AAG 232 9.31 Glycosyltransferase
24 (PppA) f
epsJ 26 GAA 357 9.30 Protein involved in EPS
biosynthesis
epsK 27 AGG 316 9.46 27 (EpsG) Galactosyltransferase
epsM 27 AGG 482 9.78 38 (EpsK) Repeat unit transporter
22 (EpsL) g
epsL 34 GGAGG 300 6.61 93 (EpsL) 90 (EpsL) 93 (EpsL) Protein involved in EPS
biosynthesis
orfY 32 GGAG 300 9.57 97 (OrfY) 96 (OrfY) 97 (OrfY) Unknown protein
a
Sequence of the 3 end of the lactococcal 16S rRNA (3 -UCUUUCCUCC-5 ) (9).
b
Partial sequence only available in GenBank.
c
CapC from Bacillus cereus ATCC 14579 (GenBank accession no. AAP12140.1; locus tag BC5276).
d
PBPRA2679 from Photobacterium profundum (GenBank accession no. CAG21057; predicted membrane protein).
e
Eps6J from S. thermophilus (accession no. AAN63722; putative glycosyltransferase).
f
PppA from Thermoanaerobacter thermohydrosulfuricus (GenBank accession no. CAB92958; putative pyruvyltransferase).
g
EpsL from Streptococcus suis (GenBank accession no. ZP_00331573; COG4632: exopolysaccharide biosynthesis protein).
from Streptococcus thermophilus (GenBank accession no.
Sequence analysis of the eps genes and gene products. The com-
AAN63779). As found for NIZO strain B40, orfY was found at
plete nucleotide sequence of 17.5 kb of chromosomal DNA
the 3 end of the eps gene cluster and was oriented in the
was determined, revealing 19 open reading frames (Fig. 3).
Within this region, the eps operon included 15 ORFs covering opposite transcriptional sense (Fig. 3). This gene is followed by
13.2 kb and oriented in the same transcriptional sense. The a truncated ORF, whose product has 30% identity with trans-
posase yuiI found in the genome of L. lactis subsp. lactis strain
upstream anking region contained four features, of which the
rst was a partial ORF; the remaining three were oriented in IL-1403 (GenBank accession no. AAK06125) and 29% identity
the opposite transcriptional sense from the subsequent eps with Orf1 of IS981 (GenBank accession no. M33933).
operon. These four ORFs are potentially involved in DNA Based on predicted amino acid similarity, putative functions
recombination and mobility functions. The rst partial ORF could be assigned to 11 of 15 ORFs identi ed (Table 2). The
predicted functions of eps genes divide the eps operon into
had only 11% amino acid identity with a putative integrase-
recombinase from Ralstonia eutropha (GenBank accession no. regions covering regulation (epsR), chain length determination
(epsABC), biosynthesis of the repeating unit (epsDEFG, epsI,
AAP85809). The second and third ORFs were partial trans-
and epsK), polymerization (epsH), and export (epsM). No pu-
posases revealing 70 to 82% identity with the transposase of
IS1077E from L. lactis subsp. lactis IL-1403 (4). The fourth tative function can yet be assigned to EpsX, EpsJ, and EpsL.
ORF (tnp) was a complete transposase sharing 27% identical EpsL shares a highly signi cant identity of 90% with related
amino acids with the transposon-related ygcE (GenBank ac- sequences from lactococcal strains (Table 2).
cession no. AAK04736) from L. lactis subsp. lactis IL-1403 The genes encoding proteins involved in regulation and
and 20% identity with the IS1253-like transposase protein chain length determination are highly conserved among the
FIG. 3. Genetic organization of the eps gene cluster of Lactococcus lactis subsp. cremoris SMQ-461. Arrows represent potential ORFs, and gene
designations are indicated under the arrows. P, polymerization; CLD, chain length determination; GTF, glycosyltransferase. The ag and hairpin
indicate the putative promoter and terminator, respectively. The complete nucleotide sequence is available in GenBank under accession number
AY741550.
VOL. 71, 2005 CHROMOSOMAL EPS GENE CLUSTER FROM L. CREMORIS 7419
EPS-producing lactococci. EpsR displays a high level of iden- proteins are characterized by the presence of ATP-binding
tity of 98% with the corresponding protein from L. lactis subsp. motifs and a tyrosine-rich region at the C terminus. Sequence
cremoris NIZO B40 and HO2 (20, 56). This gene product has analysis of EpsB revealed the presence of a Walker A ATP-
55% amino acid identity with yqbH (GenBank accession no. binding site, AGKS (residues 56 to 59), and a Walker B-site,
AAK05674) from the genome of L. lactis subsp. lactis strain VVLID (residues 166 to 180), which may be required for
IL-1403, which has ve paralogs of this type of transcription functional phosphorylation of EpsB. Furthermore, the
regulator. Also, EpsR shows an amino acid identity of 21% tyrosine-rich region at the C terminus was identi ed at resi-
with Xre, the transcription repressor of PBSX prophage from dues 205 to 208. Sequence analysis supports the hypothesis
Bacillus subtilis (GenBank accession no. AAA22894) (63). that EpsB functions as the phosphotyrosine protein kinase and
These proteins contain a DNA-binding domain and belong to interacts with EpsA to form an EpsA-EpsB complex to facilitate
the LysR family of transcriptional regulators (51, 56). Hence, EPS polymerization with the same mechanism as previously illus-
EpsR could be involved in the regulation of eps gene expres- trated for the CpsC-CpsD complex from S. pneumoniae, which
sion. The hydrophobicity plot for EpsR did not show any trans- was experimentally demonstrated by mutation (42).
membrane segments. However, the hydrophobicity pro le of The predicted protein EpsC shows 50% amino acid iden-
Downloaded from http://aem.asm.org/ on February 12, 2013 by guest
EpsX shows two highly hydrophobic segments (residues 5 to 23 tity with the phosphotyrosine protein phosphatase BC5276
(GenBank accession no. AAP12140) from Bacillus cereus
and 165 to 188), which could function as a membrane anchor.
To date, there is no available biochemical or mutational ex- ATCC 14579. Its N terminus has a DxHCH sequence, which is
periments to assess the function of EpsX. OrfY located just a highly conserved motif among other comparable proteins
downstream of the eps operon is also highly conserved among with similar function. Hence, EpsC may play an important role
L. lactis eps operons and appears also to have been transferred in EpsB dephosphorylation to facilitate EPS polymerization, as
to some S. thermophilus strains (98% identical to cpsW from proposed for CpsB from S. pneumoniae (41). These results
strain MR-1C [GenBank accession no. AAM93404] and to suggest that EpsA, EpsB, and EpsC from strain SMQ-461 all
eps4Q from the type IV eps operon [GenBank accession no. have a role in the control of polysaccharide chain length and,
AAN63692]). The orfY gene product from strain SMQ-461 has consequently, polysaccharide molecular weight.
Nucleotide sequence comparisons between the eps operon
24% identity with the trancription regulator protein LytR from
Bacillus subtilis (GenBank accession no. Q02115) (36). Regu- of strain SMQ-461 and other eps loci that were previously
identi ed from Lactococcus spp. revealed that the 3,460-bp
latory proteins of the LytR group have been found in other
region including epsRXABC seems to be a highly conserved
LAB loci encoding EPS biosynthesis. They contain three pu-
tative transmembrane segments at the N terminus (31) and region and could come from a common ancestor of lactococcal
represent a different regulatory mechanism from EpsR that strains. This region has the same organization in all lactococci
also has not been investigated to date. and shows 95% identity (Fig. 4) with the corresponding se-
The gene products of epsA, epsB, and epsC are predicted to quence of eps loci from L. lactis subsp. cremoris HO2 (20) and
be responsible for EPS chain length determination. They share NIZO B40 (56). However, there are some notable sequence
variations (Fig. 4). The predicted protein product of epsC has
a high identity (89 to 97%) with EpsA, EpsB, and EpsC from
L. lactis subsp. cremoris NIZO B40 (GenBank accession no. an N terminus that is 24 aa shorter for strain B891 than for
strains SMQ-461 and B40. Also, in comparison with the eps
NP_053033, NP_053032, and NP_053031) and HO2 (GenBank
operon of SMQ-461, a deletion of 167 bp was detected in epsX
accession no. AAP32715, AAP32716, and AAP32717) (20, 56).
EpsA also shows signi cant identity with EpsA (36%) from from strain NIZO B40 (Fig. 4), leading to a shorter protein
Bacillus cereus (GenBank accession no. BC5278) and with (139 aa compared to 255 aa for EpsX from SMQ-461). The
EpsC (26%) from S. thermophilus (GenBank accession no. nucleotide sequences of epsR and epsX from lactococci did not
AAC44010) (52), which have been classi ed as chain length- show any signi cant identity to any genes involved in polysac-
regulating proteins in capsular or exopolysaccharide biosynthe- charide biosynthesis reported for lactic acid bacteria, other
than the lactococci (56). The rst 226 bp of epsR and 115 bp of
sis, possessing two transmembrane domains. Prediction of
epsX from L. lactis subsp. cremoris strains share up to 97%
membrane-spanning regions using the TMpred program shows
identity with a related part of orf4B, the intergenic space and
that EpsA has two putative transmembrane helices (amino
pseudogene (orf4C) from the type IV operon which was iden-
acids [aa] 24 [inside] to 44 [outside] and aa 175 [outside] to 194
ti ed in S. thermophilus (GenBank accession no. AF454495.1).
[inside]). Moreover, a consensus sequence motif (SPKPKL
YLAISVIAGLVLG) was identi ed at the C terminus (resi- This result suggests a horizontal transfer between lactococci
and streptococci. The epsR gene from lactococci is anked by
dues 170 to 188) of EpsA from strain SMQ-461. This motif
IS982 at its 5 end for strains HO2 and NIZO B40. This
(SPKX11GX3G) has been shown to be involved in determining
insertion sequence is lactococcal in origin and is known to
O-antigen chain length in several gram-negative bacteria such
be horizontally transferred from L. lactis to S. thermophilus
as Escherichia coli (WzzB or rol) (3).
(22, 23). Therefore, the transfer direction of epsR and epsX is
EpsB displays 35% identity with Cps19fD from Streptococ-
cus pneumoniae (GenBank accession no. AAC44961) (24), suggested to be from lactococci to streptococci.
The 687-bp region encoding epsD shares a high identity of
which was recently identi ed as an autophosphorylating pro-
90% to 91% between lactococcal strains. The 5 end of epsD
tein tyrosine kinase (42) and 47% with the tyrosine protein
kinase BC5277 (GenBank accession no. AAP12141) from Ba- from strain NIZO B40 contains a 6-bp deletion that does not
cillus cereus ATCC 14579 (2
Copyright © 2005, American Society for Microbiology. All Rights Reserved.