The present invention relates to novel Lactococcus lactis subsp. lactis lactic acid bacterium strains, having improved texturizing properties. The present invention also relates to methods of using the strains for making food products.

BACKGROUND OF THE INVENTION

Lactic acid bacteria (LAB) are used extensively by the food industry for fermentation of food.

Conversion of fresh milk to fermented milk by LAB is a way of extending the life time of the milk and provides taste as well as texture.

Thus, important features of the strains used for milk fermentation include fast acidification, stable (no/low) post-acidification, long shelf-life and good texture. Good texture is typically high mouth thickness (measured as high shear stress using a rheometer) and high gel firmness.

Some LAB strains contribute significantly to an improved texture associated with their ability to produce exo- (or extracellular) polysaccharides (EPS), which can be capsular (remain attached to the cell in the form of capsules) or secreted into the media. EPS consists of either a single type of sugar or repeating units made of different sugars. EPS- producing LAB are of interest, since EPS act as natural viscosifiers and texture enhancers of fermented foods. Furthermore, EPS from food-grade LAB with defined Theological properties have potential for development and exploitation as food additives.

Fermented milk can be produced by mesophilic LAB, e.g. Lactococcus sp. leading to e.g. sour milk, or thermophilic LAB, e.g. Streptococcus thermophilus and Lactobacillus delbruckii subsp. bulgaricus for yoghurt.

Dairy products, such as fresh cheese, butter milk, sour milk and sour cream, prepared with mesophilic starter cultures, such as combinations of Lactococcus lactis subsp lactis strains and Lactococcus lactis subsp. cremoris strains, are in popular demand with consumers.

Lactococcus sp. strains generally produce low quantities of EPS.

It is expected to find more texturizing Lactococcus lactis subsp. cremoris strains than texturizing Lactococcus lactis subsp. lactis strains, since strains of the subsp. cremoris are more specialized to the milk environment, as they are often isolated from milk products, while strains of the subsp. lactis can be isolated from, e.g. plants.

Despite EPS production has been reported for some L. lactis subsp. lactis strains (Pan and Mei 2010, Suzuki et al 2013), the structure of their eps clusters have not been elucidated. Pan and Mei (2010) characterized EPS produced by L. lactis subsp. lactis, which was isolated from Chinese pickled cabbage, but it is not known if this strain is able to acidify milk and contribute to its texture. No eps genes were reported for this strain (Pan and Mei, 2010). Suzuki et al (2013) reported the sequences of a highly conserved epsD gene and a strain-specific epsE gene in five lactococcal strains, two from the subsp. lactis biovar diacetylactis and two from the subsp. cremoris. However, neither information on a complete eps gene cluster nor if the EPS produced by these strains is able to enhance milk texture is available. It is worth noticing that not all of the EPS-producing LAB strains are able to acidify milk, but also the ability of LAB strains to produce EPS does not ensure their enhanced texturing properties in milk, which are related to milk gel viscosity and ropiness. The type of EPS and their interaction with milk proteins is the determining factor for texture development. EPS can affect formation of casein gel structure by acting as filler. Thus the effect of EPS on protein matrix and structure formation depends on their concentration, interactions with the protein, and molecular and rheological characteristics. As an example, a production of yoghurt using EPS-producing S. thermophilus, Lactobacillus casei and L. lactis subsp. lactis has been reported (Ai et al, patent CN101331900). It is not clear how the viscosity has been measured, and what is the viscosity of milk fermented with non-EPS producing strains under the same conditions would be, which would indicate the basis level of fermented milk viscosity under the fermentation conditions used, but the viscosity of milk fermented with KS4 ( .. lactis subsp. lactis) is the lowest in comparison with the milk fermented with the remaining strains tested, such as Tx (S. thermophilus), KL1 and Jl (Lactobacillus casei) (Table 4 of CN101331900). There seems to be a negative correlation between the amount of EPS produced by these strains (Table 3 of CN101331900) and the resulting milk viscosity (Table 4 of CN101331900). For instance, the S. thermophilus strain Tx produced the lowest amount of EPS, but resulted in milk with the highest viscosity, while the L. lactis subsp. lactis strain KS4 produced the highest amount of EPS, but resulted in milk with the lowest viscosity (Table 3 and Table 4 of CN101331900). These results confirm that the EPS structure and interaction with milk components is at least as important for the milk texture development as the amount of EPS produced. No information on the eps gene clusters has been reported for the strains mentioned (CN101331900).

None of the 13 L. lactis subsp. lactis strains, for which complete genomes or contigs or scaffolds were available in the NCBI database on April 27th 2015, were reported as texturing (see Example 3 for details). Since mesophilic cultures are used for fermented milk products, and texture is an important parameter, there is a growing interest from the industry for texturizing mesophilic strains, e.g. Lactococcus lactis subsp. lactis.

SUMMARY OF THE INVENTION

It is an object of the present invention to provide novel texturizing lactic acid bacterium strains suitable for use in preparation of food products. In particular, it would be beneficial to provide texturizing Lactococcus lactis subsp lactis strains suitable for use in preparation of mesophilic food products.

This object has been solved with a texturizing Lactococcus lactis subsp. lactis strain comprising a novel eps gene cluster as described herein.

As discussed in working examples herein (see e.g. Figure 1) - herein disclosed novel lactococcus lactis subsp. lactis CHCC11848 (deposited as DSM 29291) has excellent texturing properties.

The present inventors analyzed the eps gene cluster of CHCC11848 and identified novel gene sequences which are believed to be involved in the production of exopolysaccharide (EPS) and thereby involved in the creation of the excellent texturizing properties of lactococcus lactis subsp. lactis CHCC11848 strain.

The sequences as represented by SEQ ID NO: 8-13 and 16 disclosed herein are novel over the prior art.

Without being limited to theory there is no substantial reason to believe that it would not be plausible that another lactococcus lactis subsp. lactis strain (i.e. different from the specific CHCC11848 strain) that comprises eps gene cluster genes/sequences similar to the novel herein discussed characterizing eps gene cluster genes/sequences of the CHCC11848 strain, would not also have improved texturing properties. Accordingly, a first aspect of the invention relates to a texturizing Lactococcus lactis subsp. lactis lactic acid bacterium strain comprising an active eps gene cluster capable of producing exopolysaccharide (EPS); wherein the eps gene cluster is characterized by that it comprises at least one nucleotide sequence selected from the group consisting of:

(a) : a nucleotide sequence encoding a polypeptide having polymerase activity and wherein the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :9 (herein termed wzy);

(b) : a nucleotide sequence encoding a polypeptide having polysaccharide transporter activity and wherein the polypeptide has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 12 (herein termed wzx); and

(c) : at least one a nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) activity selected from the group consisting of:

(cl) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO:8 (herein termed GT1); (c2) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 10 (herein termed GT2); and

(c3) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 11 (herein termed GT3).

As discussed herein - SEQ ID NO: 16 sets out the 7097-bp nucleotide sequence of the variable part of the eps cluster of CHCC11848 corresponding to nucleotide 4171 to 11267 of SEQ ID NO: l. The SEQ ID numbers of the first aspect are all present in the variable part (i.e. SEQ ID NO: 16) of the eps cluster of CHCC11848.

In relation to lactic acid bacterium strain the term "exopolysaccharide (EPS)" is well known and the skilled person can routinely determine if a lactic acid bacterium of interest produces EPS.

As known and understood by the skilled person a lactic acid bacterium of interest, which produces EPS, will comprise an active eps gene cluster. As known to the skilled person an active eps gene cluster comprises genes involved in regulation and modulation of EPS biosynthesis and genes involved in the biosynthesis of an oligosaccharide repeat unit and export, including a glycosyltransferase (GT), a polymerase and a transporter. In short and as understood by the skilled person, since the lactic acid bacterium strain of the first aspect is capable of producing and exporting exopolysaccharide (EPS) then it will comprise an active eps gene cluster.

In relation to item (a) of the first aspect, it is routine work for the skilled person to determine if a polypeptide of interest has the required polymerase activity.

At the filing date of the present application and in relation to SEQ ID NO:9 - the present inventors believed that the closest prior art published functional similar sequence had less than 45% identify to SEQ ID NO:9.

In relation to item (b) of the first aspect - it is routine work for the skilled person to determine if a polypeptide of interest has the required polysaccharide transporter activity. At the filing date of the present application and in relation to SEQ ID NO: 12 - the present inventors believed that the closest prior art published functional similar sequence had less than 75% identify to SEQ ID NO: 12. In relation to item (b) of the first aspect - the polypeptide having polysaccharide transporter activity may alternatively be referred to as a polypeptide having polysaccharide export activity.

In relation to item (c) of the first aspect - it is routine work for the skilled person to determine if a polypeptide of interest has the required glycosyltransferase (GT) activity. At the filing date of the present application and in relation to SEQ ID NO:8 - the present inventors believed that the closest prior art published functional similar sequence had less than 55% identify to SEQ ID NO :8.

At the filing date of the present application and in relation to SEQ ID NO: 10 - the present inventors believed that the closest prior art published functional similar sequence had less than 50% identify to SEQ ID NO: 10.

At the filing date of the present application and in relation to SEQ ID NO: 11 - the present inventors believed that the closest prior art published functional similar sequence had less than 55% identify to SEQ ID NO: 11.

A second aspect of the invention relates to a method of producing a food product comprising at least one stage in which at least one lactic acid bacterium strain according to the first aspect or any herein discussed embodiments thereof is used.

DEFINITIONS

All definitions of herein relevant terms are in accordance of what would be understood by the skilled person in relation to the herein relevant technical context.

By "texturizing strain" in the present specification and claims is meant a strain which preferably generates fermented milks having under the conditions described below and as exemplified in Example 1 herein, a shear stress preferably greater than 40 Pa measured at shear rate 300 s ^"1 .

A strain of Lactococcus lactis subsp. lactis can be defined as strongly texturizing in that it generates fermented milks having, under the same conditions, a shear stress greater than 50 Pa measured at shear rate 300 s ^"1 .

The texturizing lactic acid bacterium strain of the invention may be an isolated strain, e.g., isolated from a naturally occurring source, or may be a non-naturally occurring strain, e.g. made recombinantly. Recombinant strains will differ from naturally occurring strains by at least the presence of the nucleic acid construct(s) used to transform or transfect the mother strain. The term "Sequence Identity" relates to the relatedness between two nucleotide sequences or between two amino acid sequences.

For purposes of the present invention, the degree of sequence identity between two nucleotide sequences or two amino acid sequences is determined using the sequence alignment method of ClustalW version 2 (ClustalW2) for nucleotide sequence (DNA) or amino acid sequence (protein), respectively, pairwise alignment as described by Larkin et al. (2007, Bioinformatics 23 :2947-2948) and Goujon et a/. (2010, Nucleic acids research 38 Suppl :W695-699) with default parameters (Alignment Type: Slow; DNA Weight Matrix: IUB; Protein Weight Matrix : Gonnet; Gap Open : 10; Gap Extension : 0.1), available through CLC software.

In the present context, the terms "strains derived from" and "derived strain" should be understood as a strain derived from a strain of the invention by means of e.g. genetic engineering, radiation and/or chemical treatment, and/or selection, adaptation, screening, etc. It is preferred that the derived strain is a functionally equivalent mutant, e.g. a strain that has substantially the same, or improved, properties with respect to texturizing capacity as the mother strain. Such a derived strain is a part of the present invention. Especially, the term "derived strain" refers to a strain obtained by subjecting a strain of the invention to any conventionally used mutagenization treatment including treatment with a chemical mutagen such as ethane methane sulphonate (EMS) or N-methyl-N'-nitro-N-nitroguanidine (NTG), UV light or to a spontaneously occurring mutant. A mutant may have been subjected to several mutagenization treatments (a single treatment should be understood one mutagenization step followed by a screening/selection step), but it is presently preferred that no more than 20, no more than 10, or no more than 5, treatments are carried out. In a presently preferred derived strain, less than 1%, or less than 0.1%, less than 0.01%, less than 0.001% or even less than 0.0001% of the nucleotides in the bacterial genome have been changed (such as by replacement, insertion, deletion or a combination thereof) compared to the mother strain.

The term "thermophilic" herein refers to microorganisms that thrive best at temperatures above 35°C. The industrially most useful thermophilic bacteria include Streptococcus spp. and Lactobacillus spp. The term "thermophilic fermentation" herein refers to fermentation at a temperature above about 35°C, such as between about 35°C to about 45°C. The term "thermophilic fermented milk product" refers to fermented milk products prepared by thermophilic fermentation of a thermophilic starter culture and include such fermented milk products as set-yoghurt, stirred-yoghurt and drinking yoghurt, e.g. Yakult.

The term "mesophilic" herein refers to microorganisms that thrive best at moderate temperatures (15°C-35°C). The industrially most useful mesophilic bacteria include Lactococcus spp. and Leuconostoc spp. The term "mesophilic fermentation" herein refers to fermentation at a temperature between about 22°C and about 35°C. The term "mesophilic food products" refers to food products prepared by mesophilic fermentation of a mesophilic starter culture. The term "mesophilic fermented milk product" refers to fermented milk products prepared by mesophilic fermentation of a mesophilic starter culture and include such fermented milk products as buttermilk, sour milk, cultured milk, smetana, sour cream, Kefir and fresh cheese, such as quark, tvarog and cream cheese.

The term "mesophilic starter culture" herein refers to any starter cultures culture containing at least one mesophiulic bacterium strain. Mesophilic starter cultures, such as combinations of Lactococcus lactis subsp lactis strains and Lactococcus lactis subsp. cremoris strains, are used to produce fermented milk products, such as fresh cheese, butter milk, sour milk and sour cream. The terms "fermented milk" and "dairy" are used interchangeably herein.

The terms "a" and "an" and "the" and similar referents as used in the context of describing the invention are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. Thus, "a" and "an" and "the" may mean at least one, or one or more.

In connection with the present invention, shear stress may be measured by the following method : The day after incubation, the fermented milk product was brought to 13°C and manually stirred gently by means of a stick fitted with a perforated disc until homogeneity of the sample. The rheological properties of the sample were assessed on a rheometer (Anton Paar Physica Rheometer with ASC, Automatic Sample Changer, Anton Paar® GmbH, Austria) by using a bob-cup. The rheometer was set to a constant temperature of 13 °C during the time of measurement. Settings were as follows:

Holding time (to rebuild to somewhat original structure)

5 minutes without any physical stress (oscillation or rotation) applied to the sample.

Oscillation step (to measure the elastic and viscous modulus, G' and G", respectively, therefore calculating the complex modulus G*)

Constant strain = 0.3 %, frequency (f) = [0.5...8] Hz

6 measuring points over 60 s (one every 10 s)

Rotation step (to measure shear stress at 300 1/s)

Two steps were designed :

Shear rate = [0.3-300] 1/s and 2) Shear rate = [275-0.3] 1/s. Each step contained 21 measuring points over 210 s (on every 10 s).

The shear stress at 300 1/s was chosen for further analysis, as this correlates to mouth thickness when swallowing a fermented milk product.

5 Alternatively, the shear stress may be measured by the following method : Shear stress data were obtained by inoculating the same microbial cultures in semi-fat milk (1.5 % fat) enriched with 2 % skim milk powder; milk was heated at 90 °C for 20 min and cooled down to the inoculation temperature, prior to inoculation with 1 % overnight microbial culture. The inoculation took place for 12-15 h at 30 °C in 200-ml scale until pH~4.55 followed by 10 cooling to 4 °C and storage for 5 days at 4 °C. After the storage, the fermented milk was stirred gently by means of a stick fitted with a bored disc until homogeneity of the sample. Shear stress of the samples was assessed on a rheometer (Anton Paar Physica Rheometer with ASC, Automatic Sample Changer, Anton Paar® GmbH, Austria) using the following settings:

15

Wait time (to rebuild to somewhat original structure)

5 minutes without oscillation or rotation

Rotation (to measure shear stress at 300 s ^"1 etc.)

- Y' = [0.2707-300] s ^"1 and γ' = [275-0.2707] s ^"1

20

21 measuring points over 210 s (on every 10 s) going up to 300 s ^"1

and 21 measuring points over 210 s (one every 10 s) going down to 0.2707 s ^"1

For the data analysis, the shear stress at shear rate 300 s ^"1 was chosen.

25

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 illustrates TADM (pressure curve derived results) and rheometer data for selected L. lactis subsp. lactis strains from Chr. Hansen's strain collection. No eps cluster was found

30 in the genome of Strain G, thus this strain is not shown on Figs 2-3. TADM data were obtained from 1-ml samples made in micro-titer plates, which were inoculated for 20 hours at 30°C; thus the sample end pH varies (pH≤4.55), depending on how fast different strains are in acidifying milk. Hamilton liquid handling unit was used to measure pressure during aspiration and dispense (TADM). Shear stress data were obtained by inoculating strains in

35 200-ml scale until they reach pH~4.55 followed by Theological measurements using a rheometer. "Milk" refers to B-milk, which was not inoculated with any strain.

Figure 2 depicts an overview of eps clusters of L. lactis subsp. lactis strains from Chr. Hansen's culture collection and several publicly available L. lactis genomes, described on 40 Figure 3 B. ORFs are annotated according to their proven or predicted functions, based on BLAST analysis on NCBI web-page against refseq protein database using default parameters. Because of differences in the eps gene nomenclature, some automatic annotations belonging to the publicly available sequences were renamed here to enable the eps gene cluster comparisons. The ORFs of eps clusters from Chr. Hansen's culture collection strains structurally similar to those from publicly available sequences were annotated in the same way, for instance the eps cluster of Strain F was annotated as in H02. GT, glycosyltransferase; IS, transposase; hypot, hypothetical protein.

Figure 3A shows the phylogenetic tree of eps clusters of L. lactis subsp. lactis strains from Chr. Hansen's culture collection and several publicly available L. lactis genomes constructed using "Create Tree" tool within "Alignments and Trees" set of tools of the CLC Main Workbench 7 software (construction method : UPGMA; nucleotide substitution model : Jukes Cantor; bootstrapping was performed using 200 replicates). Figure 3B depicts the localization of the eps clusters in publicly available sequences. Figure 3C shows the percent identity matrix of the eps clusters depicted on Figure 2 and 3 A on the nucleotide level.

Figure 4 shows the genetic organization of the eps clusters of L. lactis strains NIZO B40, CHCC11848, S. thermophilus LMD-9, and S. pneumoniae D39. Gene functional grouping marked with different colors. Relative localizations of eps genes with similar functions are indicated with connection lines. Arrows represent genes oriented in the opposite transcriptional sense. Genes with unknown functions are in white; genes not likely being involved in the EPS biosynthesis are in grey; transposase-like ORF are in black; truncated genes are shown as short boxes. GT, glycosyltransferase; init, initial; polym, polymerization; transp, transport; NDP-sugar, nucleotide diphospho-sugar.

Figure 5 depicts the gene organization of identified ORFs, predicted properties of the hypothetical proteins encoded by the eps gene cluster from CHCC11848, and comparison of the predicted proteins with those in other bacteria. BLAST analysis performed on May 13 ^th 2015 using blastp tool of NCBI (default parameters) against Non-redundant (nr) protein, refseq_protein, and S. pneumoniae (taxid : 1313) databases; the top scores from each analyses are shown. Protein domains were identified using InterPro and Pfam tools. However, the results of blastp analysis against nr are shown only when they were different from those obtained from the refseq_protein database. E value (expectation value) is the number of alignments that are expected to occur by chance in a database search with similarity scores to the query equal to or better than that of the result sequence; it is indicated as an exponent of 10. GT, glycosyltransferase; IS, transposase; hypot, hypothetical protein; EPS, exopolysaccharide; aa, amino acid.

BRIEF DESCRIPTION OF THE SEQUENCE LISTING SEQ ID NO: l sets out the 13097-bp nucleotide sequence of the EPS cluster of CHCC11848. SEQ ID NO: 2 sets out the open reading frame (ORF) of the epsR gene corresponding to nucleotides 1 to 318 of SEQ ID NO: l.

SEQ ID NO: 3 sets out the ORF of the epsX gene corresponding to nucleotides 352 to 1119 of SEQ ID NO: l.

SEQ ID NO:4 sets out the ORF of the epsA gene corresponding to nucleotides 1159 to 1938 of SEQ ID NO: l.

SEQ ID NO: 5 sets out the ORF of the epsB gene corresponding to nucleotides 1948 to 2643 of SEQ ID NO: l.

SEQ ID NO: 6 sets out the ORF of the epsC gene corresponding to nucleotides 2698 to 3462 of SEQ ID NO: l.

SEQ ID NO: 7 sets out the ORF of the epsD gene corresponding to nucleotides 3484 to 4170 of SEQ ID NO: l.

SEQ ID NO:8 sets out the ORF coding a putative GTl protein corresponding to nucleotides 4177 to 5298 of SEQ ID NO: l and to nucleotides 7 to 1128 of SEQ ID NO: 16.

SEQ ID NO:9 sets out the ORF of a putative wzy gene corresponding to nucleotides 5562 to 6665 of SEQ ID NO: l and to nucleotides 1392 to 2495 of SEQ ID NO: 16.

SEQ ID NO: 10 sets out the ORF coding a putative GT2 protein corresponding to nucleotides 6686 to 7561 of SEQ ID NO: l and to nucleotides 2516 to 3391 of SEQ ID NO: 16.

SEQ ID NO: 11 sets out the ORF coding a putative GT3 protein corresponding to nucleotides 7558 to 8454 of SEQ ID NO: l and to nucleotides 3388 to 4284 of SEQ ID NO: 16.

SEQ ID NO: 12 sets out the ORF of a putative wzx gene corresponding to nucleotides 8605 to 10026 of SEQ ID NO: l and to nucleotides 4435 to 5856 of SEQ ID NO: 16.

SEQ ID NO: 13 sets out the ORF of a putative ugd gene corresponding to nucleotides 10080 to 11246 of SEQ ID NO: l and to nucleotides 5910 to 7076 of SEQ ID NO: 16.

SEQ ID NO: 14 sets out the ORF of the epsL gene corresponding to nucleotides 11268 to 12170 of SEQ ID NO: l.

SEQ ID NO: 15 sets out the nucleotide sequence, corresponding to nucleotides 12195 to

1397 of SEQ ID NO: l, of the complementary strand to the ORF of the orfY gene.

SEQ ID NO: 16 sets out the 7097-bp nucleotide sequence of the variable part of the eps cluster of CHCC11848 corresponding to nucleotide 4171 to 11267 of SEQ ID No.l.

SEQ ID NO: 17 sets out the amino acid sequence encoded by SEQ ID NO:2.

SEQ ID NO: 18 sets out the amino acid sequence encoded by SEQ ID NO:3.

SEQ ID NO: 19 sets out the amino acid sequence encoded by SEQ ID NO:4.

SEQ ID NO:20 sets out the amino acid sequence encoded by SEQ ID NO: 5.

SEQ ID NO:21 sets out the amino acid sequence encoded by SEQ ID NO:6.

SEQ ID NO:22 sets out the amino acid sequence encoded by SEQ ID NO:7.

SEQ ID NO:23 sets out the amino acid sequence encoded by SEQ ID NO:8.

SEQ ID NO:24 sets out the amino acid sequence encoded by SEQ ID NO:9.

SEQ ID NO:25 sets out the amino acid sequence encoded by SEQ ID NO: 10.

SEQ ID NO:26 sets out the amino acid sequence encoded by SEQ ID NO: 11. SEQ ID NO:27 sets out the amino acid sequence encoded by SEQ ID NO: 12.

SEQ ID NO:28 sets out the amino acid sequence encoded by SEQ ID NO: 13.

SEQ ID NO:29 sets out the amino acid sequence encoded by SEQ ID NO: 14.

SEQ ID NO:30 sets out the amino acid sequence encoded by the complementary strand of SEQ ID NO: 15.

DETAILED DESCRIPTION OF THE INVENTION

A texturizing Lactococcus lactis subsp. lactis lactic acid bacterium strain

As discussed above, a first aspect of the invention relates to a texturizing Lactococcus lactis subsp. lactis lactic acid bacterium strain comprising an active eps gene cluster capable of producing exopolysaccharide (EPS); wherein the eps gene cluster is characterized by that it comprises at least one nucleotide sequence selected from the group consisting of:

(a) : a nucleotide sequence encoding a polypeptide having polymerase activity and wherein the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :9 (herein termed wzy);

(b) : a nucleotide sequence encoding a polypeptide having polysaccharide transporter activity and wherein the polypeptide has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 12 (herein termed wzx); and

(c) : at least one a nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) activity selected from the group consisting of:

(cl) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO:8 (herein termed GT1);

(c2) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 10 (herein termed GT2); and

(c3) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: l l (herein termed GT3).

Preferably, the texturizing lactic acid bacterium of the first aspect is a LAB, wherein

(a) : the nucleotide sequence encoding a polypeptide having polymerase activity has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO:9 (herein termed wzy);

(b) : the nucleotide sequence encoding a polypeptide having polysaccharide transporter activity has at least 85% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 12 (herein termed wzx);

(c) : the nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) :

(cl) : has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO:8 (herein termed GT1); (c2) : has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 10 (herein termed GT2) ; and

(c3) : has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 11 (herein termed GT3) .

Preferably, the texturizing lactic acid bacterium of the first aspect is a LAB, wherein

(a) : the nucleotide sequence encoding a polypeptide having polymerase activity has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :9 (herein termed wzy) ;

(b) : the nucleotide sequence encoding a polypeptide having polysaccharide transporter activity has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 12 (herein termed wzx) ;

(c) : the nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) :

(cl) : has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :8 (herein termed GT1) ;

(c2) : has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 10 (herein termed GT2) ; and

(c3) : has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 11 (herein termed GT3) .

Preferably, the texturizing lactic acid bacterium of the first aspect is a LAB, wherein

(a) : the nucleotide sequence encoding a polypeptide having polymerase activity has at least 80% identity (preferably at least 85% identity, more preferably at least 90% identity, more preferably at least 95% identity, more preferably at least 98% identity, most preferably at least 100% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :9 (herein termed wzy) .

Preferably, the texturizing lactic acid bacterium of the first aspect is a LAB, wherein

(b) : the nucleotide sequence encoding a polypeptide having polysaccharide transporter activity has at least 85% identity (preferably at least 90% identity, more preferably at least

95% identity, more preferably at least 98% identity, most preferably at least 100% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 12 (herein termed wzx) . Preferably, the texturizing lactic acid bacterium of the first aspect is a LAB, wherein

(c) : the nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) :

(cl) : has at least 80% identity (preferably at least 85% identity, more preferably at least 90% identity, more preferably at least 95% identity, more preferably at least 98% identity, most preferably at least 100% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :8 (herein termed GT1) . Preferably, the texturizing lactic acid bacterium of the first aspect is a LAB, wherein

(c) : the nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) :

(c2) : has at least 80% identity (preferably at least 85% identity, more preferably at least 90% identity, more preferably at least 95% identity, more preferably at least 98% identity, most preferably at least 100% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 10 (herein termed GT2) .

Preferably, the texturizing lactic acid bacterium of the first aspect is a LAB, wherein

(c) : the nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) :

(c3) : has at least 80% identity (preferably at least 85% identity, more preferably at least 90% identity, more preferably at least 95% identity, more preferably at least 98% identity, most preferably at least 100% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 11 (herein termed GT3).

Preferably, the texturizing lactic acid bacterium comprises more than one of the nucleotide sequences specified in the first aspect.

Accordingly, preferably the texturizing lactic acid bacterium of the first aspect and herein described embodiments thereof (i .e. the texturizing lactic acid bacterium as described herein) is a LAB, wherein the eps gene cluster comprises at least:

(a) : the nucleotide sequence encoding a polypeptide having polymerase activity of item (a) of the first aspect and herein described embodiments thereof; and

(c) : at least one (preferably at least two and more preferably at least three) of the nucleotide sequences encoding a polypeptide having glycosyltransferase (GT) activity of item (c) of the first aspect and herein described embodiments thereof.

Preferably, the texturizing lactic acid bacterium of the invention comprises at least one nucleotide sequence selected from the group consisting of:

(a) : a nucleotide sequence encoding a polypeptide having polymerase activity and wherein the polypeptide has at least 70%, preferably at least 80%, more preferably 95%, identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 9 (herein termed wzy) ; and

(c) : at least one a nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) activity selected from the group consisting of:

(cl) : the polypeptide has at least 70%, preferably at least 80%, more preferably 95%, identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :8 (herein termed GT1) ; (c2) : the polypeptide has at least 70%, preferably at least 80%, more preferably 95%, identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 10 (herein termed GT2) ; and

(c3) : the polypeptide has at least 70%, preferably at least 80%, more preferably 95%, identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : l l (herein termed GT3) .

Preferably, the texturizing lactic acid bacterium of the invention comprises at least one nucleotide sequence selected from the group consisting of:

(a) : a nucleotide sequence encoding a polypeptide having polymerase activity and wherein the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :9 (herein termed wzy) ; and

(c) : at least one a nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) activity selected from the group consisting of:

(cl) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :8 (herein termed GT1) ;

(c2) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 10 (herein termed GT2) ; and

(c3) : the polypeptide has at least 70% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 11 (herein termed GT3) .

Preferably, the texturizing lactic acid bacterium of the invention comprises at least one nucleotide sequence selected from the group consisting of:

(a) : a nucleotide sequence encoding a polypeptide having polymerase activity and wherein the polypeptide has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :9 (herein termed wzy) ; and

(c) : at least one a nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) activity selected from the group consisting of:

(cl) : the polypeptide has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :8 (herein termed GT1) ;

(c2) : the polypeptide has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 10 (herein termed GT2); and

(c3) : the polypeptide has at least 80% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO : 11 (herein termed GT3) .

Preferably, the texturizing lactic acid bacterium of the invention comprises at least one nucleotide sequence selected from the group consisting of:

(a) : a nucleotide sequence encoding a polypeptide having polymerase activity and wherein the polypeptide has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :9 (herein termed wzy) ; and (c) : at least one a nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) activity selected from the group consisting of:

(cl) : the polypeptide has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO:8 (herein termed GT1);

(c2) : the polypeptide has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 10 (herein termed GT2); and

(c3) : the polypeptide has at least 95% identity with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 11 (herein termed GT3). Preferably, the texturizing lactic acid bacterium of the first aspect and herein described embodiments thereof is a LAB, wherein the eps gene cluster comprises the following nucleotide sequences:

(a) : a nucleotide sequence encoding a polypeptide having polymerase activity and wherein the polypeptide has at least 70% identity (preferably at least 85% identity, more preferably at least 90% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO :9 (herein termed wzy); and

(b) : a nucleotide sequence encoding a polypeptide having polysaccharide transporter activity and wherein the polypeptide has at least 80% identity (preferably at least 90% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 12 (herein termed wzx); and

(c) : a nucleotide sequence encoding a polypeptide having glycosyltransferase (GT) activity selected from the group consisting of:

(cl) : the polypeptide has at least 70% identity (preferably at least 85% identity, more preferably at least 90% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO:8 (herein termed GT1);

(c2) : the polypeptide has at least 70% identity (preferably at least 85% identity, more preferably at least 90% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 10 (herein termed GT2); and

(c3) : the polypeptide has at least 70% identity (preferably at least 85% identity, more preferably at least 90% identity) with the amino acid sequence encoded by the nucleotide sequence of SEQ ID NO: 11 (herein termed GT3).

In relation to the embodiment immediately above, it is preferred that the eps gene cluster comprises:

(c) : the three nucleotide sequences encoding a polypeptide having glycosyltransferase (GT) activity of item (c) of the embodiment immediately above.

As discussed herein - SEQ ID NO: 16 sets out the 7097-bp nucleotide sequence of the variable part of the eps cluster of CHCC11848 corresponding to nucleotide 4171 to 11267 of SEQ ID NO: l. In a preferred embodiment, the texturizing lactic acid bacterium strain as described herein is a LAB, wherein the eps gene cluster comprises:

(d) : a nucleotide sequence having at least 85% identity (preferably at least 90% identity 5 and more preferably at least 95% identity) with the nucleotide sequence of SEQ ID NO: 16 (herein termed the variable part of the eps cluster of CHCC11848).

In a preferred embodiment, the texturizing lactic acid bacterium strain as described herein is a LAB, wherein the eps gene cluster comprises:

10 (d) : a nucleotide sequence having at least 85% identity (preferably at least 90% identity and more preferably at least 95% identity) with the nucleotide sequence of SEQ ID NO: l (herein termed the eps cluster of CHCC11848).

As discussed in working examples herein (see e.g. Figure 1) - herein disclosed novel 15 lactococcus lactis subsp. lactis CHCC11848 (deposited as DSM 29291) has excellent texturing properties.

Preferably, the texturizing lactic acid bacterium strain as described herein is a LAB, wherein the texturizing lactic acid bacterium strain is a strain which generates fermented milks

20 having a shear stress greater than 40 Pa, more preferably 41 Pa, more preferably 42 Pa, more preferably 43 Pa, more preferably 44 Pa, more preferably 45 Pa, more preferably 46 Pa, more preferably 47 Pa, more preferably 48 Pa, more preferably 49 Pa, and most preferably 50 Pa, measured at shear rate 300 s ^"1 , measured under following conditions: 200 ml semi-fat milk (1.5% fat) enriched with 2 g skim milk powder is heated to 90°C for

25 20 min, followed by cooling to inoculation temperature, and inoculated with 2 ml of an overnight culture of the lactic acid bacterium strain, and left at inoculation temperature until pH 4.55 followed by storage at 4°C for 5 days, followed by gently stirring and measuring the shear stress at shear rate 300 s ^"1 , wherein the inoculation temperature is 30°C. The shear stress is measured using the method indicated in Example 1.

30

The present invention also relates to a Lactococcus lactis subsp. lactis strain selected from the group consisting of the strain deposited with DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Inhoffenstr. 7B, D-38124 Braunschweig, on August 21, 2014 under the accession no. DSM 29291 and strains derived from DSM 29291, 35 wherein the derived strain is characterized as having at least the same texturizing capability as DSM 29291.

In the present context, the term "derived strain" should be understood as a strain derived from a mother strain by means of e.g. genetic engineering, radiation and/or chemical 40 treatment, and/or selection, adaptation, screening, etc. In specific embodiments the derived strain is a functionally equivalent mutant, e.g. a mutant that has substantially the same, or improved, properties (e.g. regarding texture) as the mother strain. Such a derived strain is part of the present invention. The term "derived strain" includes a strain obtained by subjecting a strain of the invention to any mutagenization treatment including treatment with a chemical mutagen such as ethane methane sulphonate (EMS) or N-methyl-N'-nitro- N-nitroguanidine (NTG), UV light or to a spontaneously occurring mutant. A derived strain may have been subjected to several mutagenization treatments (a single treatment should be understood one mutagenization step followed by a screening/selection step), but typically no more than 20, no more than 10, or no more than 5, treatments are carried out. In specific embodiments of derived strains, less than 1%, or less than 0.1%, less than 0.01%, less than 0.001% or even less than 0.0001% of the nucleotides in the bacterial genome have been changed (such as by replacement, insertion, deletion or a combination thereof) compared to the mother strain. Composition comprising a LAB of the invention :

The invention also relates to a composition comprising at least one lactic acid bacterium strain according to the invention.

In a specific embodiment the composition comprises at least 1x10 ^s CFU (cell forming units)/ml of the at least one lactic acid bacterium strain according to the invention. It may be preferred that the composition comprises at least lxlO ⁸ CFU (cell forming units)/ml of the at least one lactic acid bacterium strain according to the invention.

Method of producing a food product:

The present invention further relates to a method of producing a food product comprising at least one stage in which at least one lactic acid bacterium strain according to the invention is used.

It is evident that a composition comprising at least one lactic acid bacterium strain according to the invention may be used.

In specific embodiments the food product is a dairy product and the method comprises fermenting a milk substrate with the at least one lactic acid bacterium strain according the invention.

Preferably, the method comprises fermenting a milk substrate with a composition comprising at least 1x10 ^s CFU (preferably at least lxlO ⁸ CFU) (cell forming units)/ml of the at least one lactic acid bacterium strain according to the invention.

In some embodiments, the food product has an improved texture as compared to a comparable composition without the strain. The invention also relates to a food product comprising at least one lactic acid bacterium strain according to the invention. In a specific embodiment the food product is a dairy product, a meat product, a vegetable product, a fruit product or a cereal product. In specific embodiments, the food product is a dairy product.

The term "dairy product" as used herein refers to a food product produced from milk. In the context of the present application, the term "milk" is broadly used in its common meaning to refer to liquids produced by the mammary glands of animals or by plants. In accordance with the present invention the milk may have been processed and the term "milk" includes whole milk, skim milk, fat-free milk, low fat milk, full fat milk, lactose-reduced milk, or concentrated milk. Fat-free milk is non-fat or skim milk product. Low-fat milk is typically defined as milk that contains from about 1% to about 2% fat. Full fat milk often contains 2% fat or more. The term "milk" is intended to encompass milks from different mammals and plant sources. Mammal sources of milk include, but are not limited to cow, sheep, goat, buffalo, camel, llama, mare and deer. Plant sources of milk include, but are not limited to, milk extracted from soy bean, pea, peanut, barley, rice, oat, quinoa, almond, cashew, coconut, hazelnut, hemp, sesame seed and sunflower seed. In a specific embodiment, the milk is cow's milk.

As typical dairy products according to the invention there can be mentioned a fermented milk product and cheese.

In a specific embodiment the dairy product is a mesophilic dairy product.

The production of the dairy product is carried out by methods known to the person skilled in the field, and in particular involves the fermentation of milk by at least one strain according to the invention.

Discussion of eps gene cluster of CHCC11848:

Only about 1% of ~1000 Lactococcus lactis strains from the Chr. Hansen culture collection were found to enhance the texture of fermented milk, as shown by TADM (total aspiration and dispense monitoring)-based screening using a Hamilton liquid handling unit in 1-ml scale followed by confirmatory rheology assays in 200-ml scale. Most of the texturizing strains were of the subsp. cremoris and had similar eps clusters to those from the published Lactococcus lactis subsp. cremoris strains. However, one of the texturizing strains, CHCC11848, was of the subsp. lactis and was found to have a unique eps cluster. The eps cluster of CHCC11848 was found to be unique, as it was found to contain novel glycosyltransferases and other genes likely involved in the modification of the oligosaccharide unit, and the EPS as a product of this clusters is likely to contribute to the enhanced texturing properties of this strain in milk. The eps cluster of CHCC11848 is likely being chromosome-associated, as the eps cluster-containing contig of CHCC11848 was above 414 Kbp, which is approximately 10 times larger than a typical lactococcal plasmid. Lactococcal plasmids found up-to-date are between 1.6 and 80.6 Kbp in size. Moreover, the content of the eps-cluster-containing contig of CHCC11848 was found to be highly similar to the chromosome-associated genetic content of several publicly available L. lactis subsp. lactis strains, e.g. KF147 and KLDS 4.0325. It is believed that the eps cluster of CHCC11848 is unique in being chromosome-associated as opposed to plasmid-associated. Also, a strain containing a chromosome-associated eps cluster has the advantage over a strain containing a plasmid-associated eps cluster that it has a much more stable EPS producing property, because a plasmid is easily lost from a bacterium cell as opposed to chromosomal genes. In particular, a strain containing a chromosome-associated eps cluster provides a more stable mother strain for use as a basis for developing mutant strains.

The genetic loci for polysaccharide biosynthesis by the Wzy-dependent mechanism are similar in all bacteria and are well studied in S. pneumonia. Genetic analysis of the CPS locus from 90 pneumococcal serotypes demonstrated a striking feature of the polysaccharide operon : the presence of many highly divergent forms of each of the key enzyme classes. Thus, there were found 40 homology groups for polysaccharide polymerases, 13 groups of lipases, and a great diversity of glycosyltransferases. The presence of multiple non-homologous or highly divergent forms of these enzymes, together with often different G+C content of the region in which these are encoded, supports the view that these genes have been imported on multiple occasions from different and unknown sources. Many eps gene clusters have undergone rearrangement mediated by insertion sequence (IS) elements and received genes from other organisms by a horizontal gene transfer. Typical of eps operon organization is the presence of IS elements flanking or within the operon. The plethora of glycosyltransferases observed in the loci for polysaccharide production provides an opportunity to continually generate new strains producing unique EPS by gene shuffling. As EPS show an enormous diversity in monosaccharide building blocks, anomeric configuration, conformation, and stereochemistry, the resulting diversity of EPS structures is uncanny: for instance, two glucose residues can be joined together in 30 different ways. According to Carbohydrate- Active enZymes (CAZy) database (cazy.org), glycosyltransferases are currently classified into 97 families (June 2015), which can help in predicting their mode of action. Nevertheless, this does not mean that all enzymes of a family recognize the same donor and acceptor, as polyspecificity is common among glycosyltransferase families, and thus one should be prudent with the over-interpretation of predictions based purely on this classification. In most cases, the factors determining the specificity of glycosyltransferases remain elusive, and it is difficult to predict their mode of action based merely on sequence analysis. Thus, a crucial area for further research is the search for factors or motifs determining the substrate specificity of glycosyltransferases, both in the glycosyltransferases and in the substrates. A complicating aspect, however, is that some glycosyltransferases show promiscuity toward different substrates. The three predicted glycosyltransferase gene products of CHCC11848, which might act sequentially to build the oligosaccharide repeat unit, showed low amino acid similarity with known glycosyltransferases. The structure of the repeat unit could thus also be different from those known to date for Lactococcus. The sugar specificity of the glycosyltransferases needed for EPS biosynthesis can probably be predicted according to the sugars present in the EPS. However, the structure of EPS produced by CHCC11848 has not been resolved yet.

L. lactis is found in many environments, although the original niche for L. lactis is now widely accepted to be plant based. Lactococcal strains that are used in the dairy industry appear to have undergone extensive adaptation to the nutrient-rich dairy environment through a process of reductive evolution, which, when compared to lactococcal strains isolated from plant material, appears to have resulted in a smaller genome size, a higher number of pseudogenes and acquisition of a much more extensive plasmid complement. As most nondairy isolates belong to the lineage containing strains of L. lactis subsp. lactis, and represent molecular diversity not found within the dairy strains, whereas L. lactis subsp. cremoris is typically found in dairy fermentations, it is expected result to find more texturizing strains from the subsp. cremoris than from the subsp. lactis. Thus CHCC11848 is a unique strain from the subsp. lactis able to positively contribute to the milk texturing properties.

A typical lactococcal eps operon consists of a conserved part including genes epsR, epsX, epsA, epsB, epsC, and epsD, a variable part, which includes a polymerase, a transporter and one or more glucosyltransferases or other polymer modifying genes, and another conserved part including genes epsL and orfY (Dabour and LaPointe 2005, Forde and Fitzgerald 2003, Nierop Groot and Kleerebezem 2007, van Kranenburg 1997, van Kranenburg 1999). The predicted functions of eps genes divide the eps operon into regions covering regulation (epsR), modulation of EPS synthesis including chain length determination (epsABC), biosynthesis of the oligosaccharide repeat unit including linkage of the first sugar to the lipid carrier (epsD) and subsequent addition of sugar to lipid-linked sugar or other moieties modifying the repeat unit performed by the genes of the variable part of the cluster, as well as polymerization (wzy), and export (wzx). No putative function could be yet assigned to epsX and epsL. NIZO B40 epsL can be disrupted by single crossover using an internal gene fragment or overproduced without any effect on EPS production (van Kranenburg 1999). However it might be that the second copy of epsL, which we have found in a putative cluster for glycosylated techoic acids in several lactococcal strains (see Example 4 for details), takes over, if the one from the eps cluster is not functional.

EpsR is responsible for EPS biosynthesis regulation, and thus certain mutations would affect the EPS production. EpsABC and ATP are believed to form a stable complex acting as a tyrosine kinase - phosphatase system, which controls EPS synthesis, likely through the phosphorylation of EpsD, a glycosylphospho-transferase that catalyzes the first step in the assembly of the EPS repeat unit, and defines the type of sugar added to the lipid carrier for the formation of EPS. All three genes responsible for tyrosine phosphorylation are essential for the complete encapsulation of the pneumococcus, with CpsC being a major virulence factor, crucial via its role in the regulation of the CPS biosynthesis (Whittall et al 2015). In L. lactis, the product of the cpsC-like gene, EpsA, is classified as chain length determinant protein, while EpsB is a putative tyrosine protein kinase, and EpsC, a putative tyrosine protein phosphatase. In L. lactis, EpsA and EpsB were found to be essential for the EPS production, while EpsC was not strictly required, as the effect of its deletion was the reduced amount of EPS produced (Nierop Groot and Kleerebezem 2007). Gene epsD encoding the initial glycose phosphate transferase, which does not catalyze glycosidic linkage, but is involved in linking the first sugar of the repeat unit to the lipid carrier, was shown to be essential for polysaccharide biosynthesis in L. lactis, as its disruption abolished EPS production (Dabour and LaPointe 2005, van Kranenburg et al 1997).

Subsequently, the following genes of the eps cluster typically encoding glycosyltransferases, polymerases and transporters are situated in a variable part of the cluster, and do often have a low degree of similarity to already characterized genes, which makes the prediction of their putative functions difficult. Comparison of polysaccharide synthesis operons from 90 pneumococcal serotypes, where polysaccharide biosynthesis is well studied, revealed that central genes responsible for the synthesis and polymerization of the repeat unit are highly variable and often non-homologous between serotypes (Bentley et al 2006). Wzy- dependent CPS biosynthesis in S. pneumoniae resembles peptidoglycan synthesis, whereby repeat units are built on the inner face of the cytoplasmic membrane, transported to the outer face of the membrane by a Wzx transporter, also called flippase, and polymerized by a Wzy polymerase. The polysaccharide polymerase wzy links individual repeat units to form lipid-linked CPS. In S. pneumonia, 40 homology groups for polysaccharide polymerases were found. The initial sugar of the repeat oligosaccharide unit is also the donor sugar in the polymerization of the repeat units, and the specificity of the Wzy polymerase determines the linkage type. The predictions for initial sugars, and subsequent repeat-unit polymerization linkage, correlate well with the polymerase homology groups. In S. pneumonia, there are 32 polymerase homology groups associated with WchA, five with Weil, four with WcjG and one with WcjH. These associations are mostly exclusive, with only five polymerase homology groups associated with two initial transferases, which indicates a high specificity of the initial transferases (Bentley et al 2006) .

Engineering strategies for increased production of EPS aiming at increasing pool of sugar nucleotides (i .e., EPS precursors) to enhance the carbon flux toward the final polymer, overexpression of genes involved in the EPS assembly (e.g., Wzx, Wzy, glycosyltransferases), targeted engineering of regulatory protein, disruption of pathways competing for precursors, or single gene knock-outs were successful for some EPS producers, but failed in other cases (for a review see Schmid et al 2015) . Overexpression of the entire eps gene cluster in L. lactis NIZO B40 strain resulted in significantly reduced growth rate, suggesting that increased EPS production generates a significant metabolic burden due to the required high-level production of sugar nucleotides, which are utilized in both EPS production and growth (Boels et al 2003) . The optimization of manufacturing process parameters might be more promising than engineering EPS biosynthesis for many established industrial EPS producers (Schmid et al 2015) .

The nucleotide sequence of SEQ ID NO : l includes the eps gene cluster of CHCC11848.

Any combination of the above-described elements, aspects and embodiments in all possible variations thereof is encompassed by the invention unless otherwise indicated herein or otherwise clearly contradicted by context.

Embodiments of the present invention are described below, by way of examples only. EXAMPLES

Example 1. High-throughput screening for texturing strains and measurement of milk gel texture

Milk (liquid) is typically converted into milk gel (soft solid) when fermented with lactic acid bacteria typically belonging to Streptococcus thermophilus, Lactobacillus spp. and Lactococcus lactis spp. Rheometer or texture analyzer are typically used to assess rheological properties of fermented milk gels, such as shear stress. Shear stress measurements are related to perceived mouth thickness, when the texture of milk gels is assessed by a sensory panel . High mouth thickness is considered an important quality factor of fermented milk gels such as yoghurt, and consumer acceptance is often very closely linked to the texture properties such as mouth thickness, which is a function of shear stress.

A liquid handling station, Hamilton Robotics MicroLab Star, equipped with pressure sensor inside the air displacement barrel of the individual pipettes was used in the following experiment. The liquid handler has a pressure sensor located in the headspace of each pipetting channel. Pressure data from each sensor was collected by TADM (Total Aspiration Dispense Monitoring) software of the Hamilton Robotics MicroLab Star liquid handler (Hamilton Robotics).

Fig. 1 shows aspiration pressure values at 1 s (Pa) measured by pipetting using Hamilton liquid handling unit plotted against shear stress (Pa) at shear rate 300 s ^"1 data measured using rheometer for selected milk gel samples obtained by fermenting milk using eight different L. lactis subsp. lactis strains. "Milk" refers to B-milk, which was not inoculated with any strain, and was used as a control. In the present context, 9.5 % B-milk is boiled milk made with reconstituted low fat skim milk powder to a level of dry matter of 9.5 % and pasteurized at 99 °C for 30 min, followed by cooling to 40 °C.

Pressure versus time data (TADM) were obtained from 1-ml samples made in a 96-well micro-titer plate, where B-milk was inoculated for 20 h at 30 °C in the presence of different strains (1 % of inoculum) unless otherwise stated, and then stored at 4 °C for 1 day. Hamilton liquid handling unit was used to measure pressure during aspiration, and pressure values of each sample corresponding to the time point of 1 s were plotted on the y-axis of Fig. 4. A volume of 500 μΙ was aspirated (350 μ-,/ε).

Shear stress data were obtained by inoculating the same microbial cultures in semi-fat milk (1.5 % fat) enriched with 2 % skim milk powder; milk was heated at 90 °C for 20 min and cooled down to the inoculation temperature, prior to inoculation with 1 % overnight microbial culture. The inoculation took place for 12-15 h at 30 °C in 200-ml scale until pH~4.55 followed by cooling to 4 °C and storage for 5 days at 4 °C. After the storage, the fermented milk was stirred gently by means of a stick fitted with a bored disc until homogeneity of the sample. Shear stress of the samples was assessed on a rheometer (Anton Paar Physica Rheometer with ASC, Automatic Sample Changer, Anton Paar® GmbH, Austria) using the following settings:

Wait time (to rebuild to somewhat original structure)

5 minutes without oscillation or rotation

Rotation (to measure shear stress at 300 s ^"1 etc.)

Y' = [0.2707-300] s ^"1 and γ' = [275-0.2707] s ^"1

21 measuring points over 210 s (on every 10 s) going up to 300 s ^"1

and 21 measuring points over 210 s (one every 10 s) going down to 0.2707 s ^"1

For the data analysis, the shear stress at shear rate 300 s ^"1 was chosen. Strain CHCC11848 had the highest shear stress out of the eight different L. lactis subsp. lactis strains tested, both according to the TADM and the rheometer measurements (Fig. 1). A significant correlation (R ² of 0.97) was observed between the TADM and the rheometer measurements.

Example 2. Sequencing genomes of lactococcal strains

Total DNA was purified from a culture grown overnight at 30°C in M17 medium containing 1% lactose and 1% glucose using DNeasy Blood & Tissue Kit (Qiagen). The DNA quality was checked using gel electrophoresis, the DNA concentration was estimated using Nanodrop 2000 spectrophotometer, and 15 μg DNA (approximately 150 ng/μΙ) were used for sequencing at BGI (HongKong, China) using Illumina HiSeq equipment with pair-end channel module with 2x100 bp read length and an insert size of 500 bp. The assembly of raw reads into contigs was performed using CLC Genomics workbench 7.0 software with default parameters resulted in 48 contigs (average length 57325 bp, average coverage 317), which were used for eps gene mining.

Example 3. Characterization of the eps gene cluster of lactococcal strains Since an enhanced texture is associated with the production of EPS, the texturing and non- texturing lactococcal strains from the CHCC culture collection presented on Fig. 1 were genome-sequenced. Mining for eps gene clusters was performed, and the eps clusters found were compared to those of the publicly available lactococcal genome sequences. We focused on the subsp. lactis, as none of the strains from this subsp., for which genome sequences were available, are reported as texturing, indicating that CHCC11848 is a unique strain from the subsp. lactis with enhanced texturing properties.

In order to find eps gene clusters in the lactococcal strains under investigation, "BLAST" tool of the software "CLC Main Workbench 7" was applied using standard parameters. 13 Lactococcus lactis subsp. lactis strains, for which complete genomes or contigs were available in the NCBI database on April 27th 2015, were used for eps gene cluster mining : KLDS 4.0325 (GenBank CP006766), 111403 (GenBank AE005176), CNCM 1-1631 (GenBank AGHX00000000), CV56 (GenBank CP002365), Dephy 1 (GenBank CBUJ000000000), KF147 (GenBank CP001834), YF11 (GenBank APAV00000000), IO-l (GenBank AP012281), A12 (GenBank CBLU000000000), 1AA59 (GenBank AZQT00000000), JCM 5805 = NBRC 100933 (GenBank BBSI00000000), K214 plasmid pK214 (GenBank NC_009751), NCD02118 (GenBank CP009054), and NCDO 2118 plasmid pNCD02118 (GenBank CP009055). No eps clusters were found in CV56, II 1403, IO-l, JCM 5805, A12, Dephy 1, K214 plasmid pK214, NCDO 2118 genome and plasmid pNCD02118. The eps clusters found in the remaining five strains, together with the eps clusters from five different L. lactis subsp. cremoris strains such as NIZO B40 (GenBank AF036485), SMQ-461 (GenBank AY741550.2), Ropy352 (GenBank EF192213), H02 (GenBank AF142639), and A76 (GenBank CP003132), were used as eps cluster references for BLAST analysis of the lactococcal strains from the CHCC culture collection (Figure 3B). Mining for eps gene clusters in the lactococcal strains from CHCC culture collection was performed first using the assembled genomes (containing 5 contigs), and in the case no eps genes were found in the assembled genomes, "Map Reads to reference" tool of CLC Main Workbench 7 software was applied using standard parameters. The latter was performed to ensure that no raw reads representing eps genes that were not assembled into contigs were present in the raw genome data.

10 Both BLAST analysis results and results of alignment of the eps gene clusters from different lactococcal strains to each other using "Create Alignment" tool within "Alignments and Trees" set of tools of the CLC Main Workbench 7 software were used to make annotations of ORF of the strains from the CHCC culture collection. "Create Tree" tool within "Alignments and Trees" set of tools of the CLC Main Workbench 7 software (construction method :

15 UPGMA; nucleotide substitution model : Jukes Cantor; bootstrapping was performed using 200 replicates) was used to create a phylogenetical tree based on L. lactis subsp. lactis and cremoris alignment files (Figure 3A). The identity of sequences was calculated using percent identity matrix by ClustalW2 (available through CLC software) (Figure 3C). InterPro (http://www.ebi.ac.uk/interpro/) and Pfam (http://pfam.xfam.org/) tools were used for

20 functional characterization of proteins by classifying them into families, predicting hydrophobicity profile, domains and important sites.

A large eps cluster diversity was observed in the lactococcal strains under investigation (Fig. 2). Only two strains from the subsp. lactis, KF147 and YF11, were found to have very

25 similar (99 % identical from eps to epsL) eps clusters, based on their nucleotide sequence analysis by ClustalW2 (Figure 3A, 3C), but containing a significant number of single nucleotide polymorphisms (SNP), e.g. 13 SNP in epsX (10 resulting in amino acid changes), 13 SNP in epsA (four resulting in amino acid changes), four in epsB (one resulting in amino acid change), six in epsC (three resulting in amino acid changes), 29 SNP in epsD (14

30 amino acid changes). Despite the eps cluster of CHCC11848 seemed somewhat similar to those of publicly available lactococcal strains and those from the CHCC culture collection (Fig. 2), the similarity was highest in the conserved regions of the cluster, from epsR to epsD and from epsL to orfY. When comparing the eps cluster of CHCC11848 with that of KLDS 4.0325, the identity of the conserved region of the eps cluster from eps to epsD was

35 97 %, from epsL to orfY, 98 %, while the identity of the variable region, between epsD and epsL was 71 % on the nucleotide level.

40 Example 4. Characterization of the eps gene cluster of CHCC11848

The eps operon of CHCC11848 included 14 open reading frames (ORF) covering 13 kb and oriented in the same transcriptional sense except for the last gene of the cluster, orfY (Fig. 4, Fig. 5) . As for many lactococcal strains including NIZO B40, H02 and SMQ-461, orfY was found at the 3' end of the eps gene cluster and was oriented in the opposite transcriptional sense (Fig. 4). This gene was followed by a putative cadmium resistance protein cadA. Based on the amino acid similarity, putative functions could be assigned to 12 out of 14 ORFs identified (Fig. 5). The predicted functions of eps genes divide the eps operon into regions covering regulation (epsR), modulation of EPS synthesis including chain length determination (epsABC), biosynthesis of the oligosaccharide repeat unit including linkage of the first sugar to the lipid carrier (epsD) and subsequent addition of sugar or other moieties modifying the repeat unit (GT1, GT2, GT3, ugd), as well as polymerization (wzy), and export (wzx). No putative function could be yet assigned to eps and epsL.

The organization of the eps gene cluster from L. lactis is similar to that of the gene clusters encoding EPS biosynthesis in S. thermophilus and CPS biosynthesis in S. pneumoniae (Fig. 4) . We have named the conserved genes of CHCC11848 according to the nomenclature commonly used for the lactococcal eps genes (Dabour and LaPointe 2005, Forde and Fitzgerald 2003, Nierop Groot and Kleerebezem 2007, van Kranenburg 1997, van Kranenburg 1999). However, eps genes with the same names often have different functions in different organisms, as the genes are often designated alphabetically by order of occurrence in a given locus and not based on their functions. Thus, epsB is L. lactis encodes tyrosine protein kinase, while the corresponding gene in S. thermophilus is named epsD, and in S. pneumoniae, epsD (or wxe) (Fig. 4) . In NIZO strain B40, the EPS polymeraze is named epsl, and the export gene, epsK, while in SMQ-461, the genes with corresponding functions are names epsH and epsM. The original nomenclature for S. pneumoniae capsules utilizes eps followed by the serotype number and gene designation, where genes are designated alphabetically by order of occurrence in a given locus, while an alternative nomenclature is based on functions (e.g. wzg), but lacks the ability to readily distinguish serotype. The nomenclature of the S. thermophilus eps genes is closely related to that of the pneumococcal one; however the genes are generally named eps and not eps. In order to annotate the genes of CHCC11848, we made protein functional characterization using InterPro and Pfam tools, additionally to the BLAST analysis (Fig. 5).

CHCC11848 was found to contain highly conserved epsR, which was assigned for EPS biosynthesis regulation, as it contains a DNA binding domain, and is identical to a lactococcal putative transcriptional regulator from XRE family (Fig. 5) . The corresponding transcriptional regulators of EPS synthesis in e.g. S. thermophilus LMD-9 and S. pneumoniae D39, epsA and cpsA, respectively, belong to the LytR family of transcriptional regulators, similarly to orfY from CHCC11848. While EpsA and CpsA contained three putative transmembrane segments at the N terminus and a large hydrophobic segment, OrfY had a similar hydrophobicity profile, but only one transmembrane segment. LytR group of transcriptional regulators represent a different regulatory mechanism from EpsR that has not been investigated to date.

Both EpsX and EpsL shared highly significant similarity (98 and 99% identity, respectively) with related sequences from lactococcal strains (Fig . 5) . However, there are no available experiments to assess their functions to date. NIZO B40 epsL can be disrupted by single crossover using an internal gene fragment or overproduced without any effect on EPS production (van Kranenburg 1999) . The hydrophobicity plot of EpsL of CHCC11848 showed a large hydrophobic, a cytoplasmic, and a transmembrane segment. It was predicted that EpsL is a periplasmic protein containing DUF2233 domain of unknown function (Fig . 5) . Curiosly we found epsL-Wke genes as a part of lactococcal eps clusters, when present, but also in techoic acid clusters. For instance, L. lactis strain KF147 was found to contain two epsL-Wke genes in its genome, one in the putative techoic acid cluster (protein id ABX75721, locus tag LLKF_0940 of GenBank sequence CP001834) and one in the eps gene cluster (protein id ABX75689, locus tag LLKF_0142) . Strains CV56 and 111403, which do not seem to have an eps cluster, were found to contain one epsL-Wke gene each in their putative techoic acid clusters. The hydrophobicity profile of EpsX showed that it possesses a large cytoplasmic, a non-cytoplasmic, and a transmembrane segment; the latter could function as a membrane anchor. The cytoplasmic region of EpsX included a domain with similarity to GDSL-like Lipase / acylhydrolase family of presumed lipases and related enzymes, and to SGNH-hydrolase superfamily enzymes belonging to a diverse family of lipases and esterases. The role of epsX and epsL in the EPS biosynthesis remains to be established .

Strains CV56 and 111403 both have putative (lipo)techoic acid clusters. CV56 contains a gene encoding O-antigen export system ATP-binding protein in its putative lipotechoic acid cluster; it represents a lipopolysaccharide exporter (locus tag CVCAS_0185 of the GenBank sequence CP002365, protein id ADZ62853, protein name RFBl aka RfbB, accession nr Q48476 in the transporter classification database tcdb.org), which is required for translocation of lipopolysaccharide O-antigen side-chains across the cytoplasmic membrane. The ABC transporter found within the putative lipotechoic acid cluster of 111403 (protein id AAK04301, locus tag L4342 of the NCBI sequence NC_002662) was 79 % identical to that of CV56, based on ClustalW2 analysis, and was classified as ABC transporter required for O- antigen biosynthesis and multicellular development, RfbAB (accession nr Q50863 in tcdb.org) . Techoic acid ABC transporter ATP binding protein found within putative techoic acid clusters of CV56 (protein id ADZ63531, locus tag CVCAS_0878 of the GenBank sequence CP002365) and 111403 (protein id AAK05013, locus tag L137446 of the NCBI sequence NC_002662) were 99.8 % identical, based on ClustalW2 analysis, and were classified as techoic acid exporters TagGH (accession nr P42954 in tcdb.org).

EpsABC and ATP are believed to form a stable complex acting as a tyrosine kinase - phosphatase system, which controls EPS synthesis, likely through the phosphorylation of EpsD, a glycosylphospho-transferase that catalyzes the first step in the assembly of the EPS repeat unit, and defines the type of sugar added to the lipid carrier for the formation of EPS. All three genes responsible for tyrosine phosphorylation are essential for the complete encapsulation of the pneumococcus, with CpsC being a major virulence factor, crucial via its role in the regulation of the CPS biosynthesis. In CHCC11848, the product of the cpsC-like gene, EpsA, has been classified as chain length determinant protein, while EpsB was found to be a putative tyrosine protein kinase, and EpsC, a putative tyrosine protein phosphatase. The three proteins have a high degree of similarity to the corresponding proteins in other organisms (Fig. 5), but have a different order in the eps cluster of CHCC11848 than in S. thermophilus LMD-9 and S. pneumoniae D39 (Fig. 4). Gene epsD is likely to encode the initial glycose phosphate transferase, which does not catalyze glycosidic linkage, but is involved in linking the first sugar of the repeat unit to the lipid carrier. The initial glycose phosphate transferase was shown to be essential for polysaccharide biosynthesis, as its disruption abolished EPS production. While the majority of pneumococcal serotypes contain wchA, and where wchA is absent, the products of the fifth cps gene fall into homology groups Weil, WcjG or WcjH, the epsD of CHCC11848 containing a large cytoplasmic segment and one transmembrane segment was found to belong to the WcjG group (Fig. 5).

Subsequently, the following genes of the eps cluster of CHCC11848 encoding glycosyltransferases may transfer various nucleotide sugars including UDP-glucose, UDP- galactose, dTDP-rhamnose, UDP-GlcNAc and UDP-galactofuranose to form the repeating units in a glycosidic linkage-dependent manner. All three predicted glycosyltransferase gene products of CHCC11848 showed relatively low amino acid similarity with known glycosyltransferases (Fig. 5) . The three glycosyltransferases GT1, GT2, GT3 together with ugd encoding UDP-glucose 6-dehydrogenase are potentially involved in sequential building of the repeat unit, although their specific functions and therefore order of action have not been demonstrated. The protein encoded by ugd was previously found in L. lactis pCI658- encoded eps operon. It converts UDP-glucose to UDP-glucuronic acid, which is possibly one of the components of the EPS structure of CHCC11848. However, chemical structure and sugar composition of the EPS repeat unit of CHCC11848 remain to be determined. Comparison of polysaccharide synthesis operons from 90 pneumococcal serotypes revealed that central genes responsible for the synthesis and polymerization of the repeat unit are highly variable and often non-homologous between serotypes (Bentley et al 2006). Wzy- dependent CPS biosynthesis in S. pneumoniae resembles peptidoglycan synthesis, whereby repeat units are built on the inner face of the cytoplasmic membrane, transported to the outer face of the membrane by a Wzx flippase, and polymerized by a Wzy polymerase. The polysaccharide polymerase wzy links individual repeat units to form lipid-linked CPS. In S. thermophilus LMD-9, epsK encoding polysaccharide polymerase has nine transmembrane segments, and the corresponding polymerase in S. pneumoniae D39, epsH or wzy, has 11 transmembrane segments, both representing an O-antigen ligase like membrane protein. Based on sequence identity analysis by ClustalW2, the putative polymerase of CHCC11848, Wzy, had only 12 and 14 % identity to the polymerases from LMD-9 and D39, respectively. The hydrophobicity plot for Wzy from CHCC11848 showed 10 transmembrane segments and seems related to a protein family that includes EpsG from Bacillus subtilis and Wzy from Acinetobacter baumannii (Fig. 5). Despite being classified as belonging to the same protein type by BLAST analysis and InterPro and Pfam tools, polymerases from LMD-9 and D39 were found only being 14 % identical to each other on the amino acid level. In S. pneumonia, 40 homology groups for polysaccharide polymerases were found. The initial sugar of the repeat oligosaccharide unit is also the donor sugar in the polymerization of the repeat units, and the specificity of the Wzy polymerase determines the linkage type. The predictions for initial sugars, and subsequent repeat-unit polymerization linkage, correlate well with the polymerase homology groups. In S. pneumonia, there are 32 polymerase homology groups associated with WchA, five with Weil, four with WcjG and one with WcjH. These associations are mostly exclusive, with only five polymerase homology groups associated with two initial transferases, which indicates a high specificity of the initial transferases (Bentley et al 2006). The gene product of wzx of CHCC11848 shares moderate similarity with other predicted lactococcal flippases and has 12 predicted transmembrane regions. DEPOSIT AND EXPERT SOLUTION

The applicant requests that a sample of micro-organisms deposited for the present application as described below may only be made available to an expert, until the date on which the patent is granted. Lactococcus lactis subsp. lactis CHCC11848 was deposited with DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Inhoffenstr. 7B, D-38124 Braunschweig, on August 21, 2014 under the accession no. DSM 29291.

The deposit was made according to the Budapest treaty on the international recognition of the deposit of microorganisms for the purposes of patent procedure. REFERENCES

1. Ai Q, Liu H, Zhang H, Zhong D. New Lactobacillus strains that produce bile salt hydrolase (BSH) and exopolysaccharide (EPS), useful for preparing yogurt that can lower blood cholesterol and has improved stability due to the exopolysaccharide. Patent CN 101 331 900 A

2. Boels IC, Van Kranenburg R, Kanning MW, Chong BF, De Vos WM, Kleerebezem M. 2003. Increased exopolysaccharide production in Lactococcus lactis due to increased levels of expression of the NIZO B40 eps gene cluster. Appl Environ Microbiol. 69: 5029-5031.

3. Bentley SD, Aanensen DM, Mavroidi A, Saunders D, Rabbinowitsch E, Collins M, Donohoe K, Harris D, Murphy L, Quail MA, Samuel G, Skovsted IC, Kaltoft MS, Barrell B, Reeves PR, Parkhill J, Spratt BG. 2006. Genetic analysis of the capsular biosynthetic locus from all 90 pneumococcal serotypes. PLoS Genet. 2:e31.

4. Dabour N, LaPointe G. 2005. Identification and Molecular Characterization of the Chromosomal Exopolysaccharide Biosynthesis Gene Cluster from Lactococcus lactis subsp. cremoris SMQ-461. Appl Environ Microbiol. 71 : 7414-7425.

5. Forde A, Fitzgerald GF. 2003. Molecular organization of exopolysaccharide (EPS) encoding genes on the lactococcal bacteriophage adsorption blocking plasmid, pCI658. Plasmid. 49: 130-142. 6. Goujon M, McWilliam H, Li W, Valentin F, Squizzato S, Paern J, Lopez R. 2010. A new bioinformatics analysis tools framework at EMBL-EBI. Nucleic Acids Res. 38: W695-W699.

7. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ, Higgins DG. 2007. Clustal W and Clustal X version 2.0. Bioinformatics. 23 : 2947-2948.

8. Nierop Groot MN, Kleerebezem M. 2007. Mutational analysis of the Lactococcus lactis NIZO B40 exopolysaccharide (EPS) gene cluster: EPS biosynthesis correlates with unphosphorylated EpsB. J Appl Microbiol. 103 : 2645-2656

9. Pan D, Mei X. 2010. Antioxidant activity of an exopolysaccharide purified from Lactococcus lactis subsp. lactis 12. Carbohydrate Polymers 80: 908-914

10. Schmid J, Sieber V, Rehm B. 2015. Bacterial exopolysaccharides: biosynthesis pathways and engineering strategies. Front Microbiol. 6: 496. 11. Suzuki C, Kobayashi M, Kimoto-Nira H. Novel exopolysaccharides produced by Lactococcus lactis subsp. lactis, and the diversity of epsE genes in the exopolysaccharide biosynthesis gene clusters. 2013. Biosci Biotechnol Biochem. 77: 2013-2018.

5

12. van Kranenburg R, Marugg JD, van Swam II, Willem NJ, de Vos WM. 1997. Molecular characterization of the plasmid-encoded eps gene cluster essential for exopolysaccharide biosynthesis in Lactococcus lactis. Mol Microbiol. 2: 387-397

10 13. van Kranenburg R, Vos HR, van Swam II, Kleerebezem M, de Vos WM. 1999. Functional analysis of glycosyltransferase genes from Lactococcus lactis and other gram-positive cocci : complementation, expression, and diversity. J Bacteriol. 181 : 6347-6353

14. Whittall JJ, Morona R, Standish AJ. 2015. Topology of Streptococcus pneumoniae CpsC, 15 a Polysaccharide co-polymerase and BY-kinase adaptor protein. J Bacteriol 197: 120.