AU2001284558A1 - Lactobacillus rhamnosus polynucleotides, polypeptides and methods for using them - Google Patents

Description

LACTOBACILLUS RHAMNOSUS POLYNUCLEOTIDES,

POLYPEPTIDES AND METHODS FOR USING THEM

Technical Field of the Invention

This invention relates to polynucleotides isolated from lactic acid bacteria, namely Lactobacillus rhamnosus, including full-length sequences encoding polypeptides, as well as to probes and primers specific to the polynucleotides;

DNA constructs comprising the polynucleotides; biological materials, including microorganisms and multicellular organisms, incorporating the polynucleotides; polypeptides encoded by the polynucleotides; and methods for using the polynucleotides and polypeptides.

Background of the Invention

The present invention relates to polynucleotides isolated from a specific strain of lactic acid bacteria, namely Lactobacillus rhamnosus HN001 (L. rhamnosus HN001). Lactic acid bacteria, and their enzymes, are the major determinants of flavor and fermentation characteristics in fermented dairy products, such as cheese and yogurt. Flavors are produced tlirough the action of bacteria and their enzymes on proteins, carbohydrates and lipids.

Lactobacillus rhamnosus strain HN001 are heterofermentative bacteria that are Gram positive, non-motile, non-spore forming, catalase negative, facultative anaerobic rods exhibiting an optimal growth temperature of 37+1 °C and an optimum pH of 6.0 - 6.5. Experimental studies demonstrated that dietary supplementation with Lactobacillus rhamnosus strain HN001 induced a sustained enhancement in several aspects of both natural and acquired immunity (See PCT International Publication No. WO 99/10476). In addition, L. rhamnosus HN001 , and certain other Gram-positive bacteria can specifically and directly modulate human and animal health (See, for example, Tannock et al., Applied Environ. Microbiol. 66:2578-2588, 2000; Gill et al, Brit. J. Nutrition 83:167-176; Quan Shu et al, Food and Chem. Toxicol. 38:153-161, 2000; Quan Shu et al, Intl. J. Food Microbiol. 56:87-96, 2000; Quan Shu et al, Intl. Dairy J. 9:831-836, 1999; Prasad et al, Intl. Dairy J. 8:993-1002, 1998; Sanders and Huis in't Veld,

Antonie van Leeuwenhoek 76:293-315, 1999; Salminen et al, 1998. In: Lactic Acid Bacteria, Salminen S and von Wright A (eds)., Marcel Dekker Inc, New York, Basel, Hong Kong, pp. 203 -253; Delcour et al, Antonie van Leeuwenhoek 76: 159-184, 1999; Blum et al, Antonie van Leeuwenhoek 76:199-205, 1999; Yasui et al, Antonie van Leeuwenhoek 76:383-389, 1999; Hirayama and Rafter,

Antonie van Leeuwenhoek 76:391-394, 1999; Ouwehand, 1998. In: Lactic Acid Bacteria, Salminen S and von Wright A (eds)., Marcel Dekker Inc, New York, Basel, Hong Kong, pp. 139-159; Isolauri et al, S 1998. In: Lactic Acid Bacteria, Salminen S and von Wright A (eds)., Marcel Dekker Inc, New York, Basel, Hong Kong, pp. 255-268; Lichtenstein and Goldin, 1998. In: Lactic Acid Bacteria,

Salminen S and von Wright A (eds)., Marcel Dekker Inc, New York, Basel, Hong Kong, pp. 269-277; El-Nezami and Ahokas, 1998. In: Lactic Acid Bacteria, Salminen S and von Wright A (eds)., Marcel Dekker Inc, New York, Basel, Hong Kong, pp. 359-367; Nousianen et al, 1998. In: Lactic Acid Bacteria, Salminen S and von Wright A (eds)., Marcel Dekker Inc, New York, Basel, Hong Kong, pp.

437-473; Meisel and Bockelmann, Antonie van Leeuwenhoek 76:207-215, 1999; Christensen et al, Antonie van Leeuwenhoek 76:217-246, 1999; Dunne et al, Antonie van Leeuwenhoek 76:279-292, 1999).

Beneficial health effects attributed to these bacteria mclude the following: • Increased resistance to enteric pathogens and anti-infection activity, including treatment of rotavirus infection and infantile diarrhea - due to increases in antibody production caused by an adjuvant effect, increased resistance to pathogen colonization; alteration of intestinal conditions, such as pH; and the presence of specific antibacterial substances, such as bacteriocins and organic acids.

• Aid in lactose digestion - due to lactose degradation by bacterial lactase enzymes (such as beta-galactosidase) that act in the small intestine.

• Anti-cancer (in particular anti-colon cancer) and anti-mutagenesis activities - due to anti-mutagenic activity; alteration of procancerous enzymatic activity of colonic microbes; reduction of the carcinogenic enzymes azoreductase, beta-glucuronidase and nitroreductase in the gut and/or faeces; stimulation of immune function; positive influence on bile salt concentration; and antioxidant effects.

• Liver cancer reduction - due to aflatoxin detoxification and inhibition of mould growth. • Reduction of small bowel bacterial overgrowth - due to antibacterial activity; and decrease in toxic metabolite production from overgrowth flora.

• Immune system modulation and treatment of autoimmune disorders and allergies - due to enhancement of non-specific and antigen-specific defence against infection and tumors; enhanced mucosal immunity; adjuvant effect in antigen-specific immune responses; and regulation of Thl/Th2 cells and production of cytokines.

• Treatment of allergic responses to foods- due to prevention of antigen translocation into blood stream and modulation of allergenic factors in food. • Reduction of blood lipids and prevention of heart disease - due to assimilation of cholesterol by bacteria; hydrolysis of bile salts; and antioxidative effects.

• Antihypertensive effect - bacterial protease or peptidase action on milk peptides produces antihypertensive peptides. Cell wall components act as ACE inhibitors

• Prevention and treatment of urogenital infections - due to adhesion to urinary and vaginal tract cells resulting in competitive exclusion; and production of antibacterial substances (acids, hydrogen peroxide and biosurfactants).

• Treatment of inflammatory bowel disorder and irritable bowel syndrome - due to immuno-modulation; increased resistance to pathogen colonization; alteration of intestinal conditions such as pH; production of specific antibacterial substances such as bacteriocins, organic acids and hydrogen peroxide and biosurfactants; and competitive exclusion.

• Modulation of infective endocarditis - due to fibronectin receptor- mediated platelet aggregation associated with Lactobacillus sepsis. • Prevention and treatment of Helicobacter pylori infection - due to competitive colonization and antibacterial effect.

• Prevention and treatment of hepatic encephalopathy - due to inhibition and/or exclusion of urease-producing gut flora.

• Improved protein and carbohydrate utilisation and conversion - due to production of beneficial products by bacterial action on proteins and carbohydrates.

Other beneficial health effects associated with L. rhamnosus include improved nutrition; regulation of colonocyte proliferation and differentiation; improved lignan and isoflavone metabolism; reduced mucosal permeabilit detoxification of carcinogens and other harmful compounds; relief of constipation and diarrhea; and vitamin synthesis, in particular folate.

Peptidases are enzymes that break the peptide bonds linking the amino group of one amino acid with the carboxy group (acid group) of an adjacent amino acid in a peptide chain. The bonds are broken in a hydrolytic reaction. There is a large family of peptidase enzymes that are defined by their specificity for the particular peptides bonds that they cleave (Barrett A J, Rawlings N D and

Woessner J F (Eds.) 1998. Handbook of proteolytic enzymes, Academic Press,

London, UK). The two main families are exopeptidases and endopeptidases.

Exopeptidases cleave amino acids from the N- or C- terminus of a peptide chain, releasing free amino acids or short (di- and tripeptides). Different types of exopeptidases include:

• Aminopeptidases - release a free amino acid from the N-terminus of a peptide chain;

• dipeptidyl-peptidase (also known as dipeptidyl-aminopeptidases) - release a dipeptide from the N-terminus of a peptide chain;

• tripeptidyl-peptidases (also known as tripeptidyl-aminopeptidases) - release a tripeptide from the N-terminus of a peptide chain);

• carboxypeptidases - release a free amino acid from the C-terminus of a peptide chain; • peptidyl-dipeptidase - release a dipeptide from the C-terminus of a peptide chain;

• dipeptidases - release two free amino acids from a dipeptide; and

• tripeptidases - release a free amino acid and a dipeptide from a tripeptide. Endopeptidases hydrolyze peptide bonds internally within a peptide and are classified on the basis of their mode of catalysis:

• serine-endopeptidases - depend on serine (or threonine) as the nucleophile in the catalytic reaction;

• cysteine-endopeptidases - depend on the sulphydryl group of cysteine as the nucleophile in the catalytic reaction;

• aspartic-endopeptidases - contain aspartate residues that act as ligands for an activated water molecule which acts as the nucleophile in the catalytic reaction; and

• metallo-endopeptidases - contain one or more divalent metal ions that activate the water molecule that acts as the nucleophile in the catalytic reaction. Peptidases are important enzymes in the process of cheese ripening and the development of cheese flavor. The hydrolysis of milk caseins in cheese results in textural changes and the development of cheese flavors. The raft of proteolytic enzymes that cause this hydrolysis come from the lactic acid bacteria that are bound up in the cheese - either starter cultures that grow up during the manufacture of the cheese, or adventitious and adjunct non-starter lactic acid bacteria that grow in the cheese as it ripens (Law and Haandrikman, Int. Dairy J. 7:1-11, 1997). Many other enzymes can also influence dairy product flavor, and functional and textural characteristics, as well as influencing the fermentation characteristics of the bacteria, such as speed of growth, acid production and survival (Urbach, Int. Dairy J. 5:877-890, 1995; Johnson and Somkuti, Biotech. Appl. Biochem._. 13:196-204, 1991; El Soda and Pandian, J. Dairy Sci. 74:2317- 2335, 1991; Fox et al, In Cheese: chemistry, physics and microbiology. Volume 1, General aspects, 2^nd edition, P Fox (ed) Chapman and Hall, London;

Christensen et al, Antonie van Leeuwenhoek 76:217-246, 1999; Stingle et al, J.

Bacteriol. 20:6354-6360, 1999; Stingle et al, Mol. Microbiol. 32:1287-1295,

1999; Lemoine et al, Appl. Environ. Microbiol. 63:1512-3518, 1997). Enzymes influencing specific characteristics and/or functions include the following: ■ Lysis of cells. These enzymes are mostly cell wall hydrolases, including amidases; muramidases; lysozymes, including N-acetyl muramidase; muramidase; N-acetylglucosaminidase; and N-acetylmuramoyl-L-alanine amidase. DEAD-box helicase proteins also influence autolysis.

■ Carbohydrate utilization. Lactose, citrate and diacetyl metabolism, and alcohol metabolism are particularly important. The enzymes involved include beta-galactosidase, lactate dehydrogenase, citrate lyase, citrate permease, 2,3 butanediol dehydrogenase (acetoin reductase), acetolactate decaboxylase, acetolactate synthase, pyruvate decarboxylase, pyruvate formate lyase, diacetyl synthase, diacetyl reductase, alcohol decarboxylase, lactate dehydrogenase, pyruvate dehydrogenase, and aldehyde dehydrogenase.

■ Lipid degradation, modification or synthesis. Enzymes involved include lipases, esterases, phospholipases, serine hydrolases, desaturases, and linoleate isomerase.

■ Polysaccharide synthesis. Polysaccharides are important not only for potential immune enhancement and adhesion activity but are important for the texture of fermented dairy products. The enzymes involved are a series of glucosyl transferases, including beta-(l-3) glucosyl transferase, alpha-N acetylgalactosaminyl transferase, phosphogalactosyl transferase, alpha- glycosyl transferase, UDP-N-acetylglucosamine C4 epimerase and UDP-N- acetylglucosamine transferase.

■ Amino acid degradation. Enzymes include gluta ate dehydrogenase, aminotransferases, amino acid decarboxylases, and enzymes involved in sulphur amino acid degradation including cystothione beta-lyase. Sequencing of the genomes, or portions of the genomes, of numerous organisms, including humans, animals, microorganisms and various plant varieties, has been and is being carried out on a large scale. Polynucleotides identified using sequencing techniques may be partial or full-length genes, and may contain open reading frames, or portions of open reading frames, that encode polypeptides. Polypeptides may be identified based on polynucleotide sequences and further characterized. The sequencing data relating to polynucleotides thus represents valuable and useful information.

Polynucleotides and polypeptides may be analyzed for varying degrees of novelty by comparing identified sequences to sequences published in various public domain databases, such as EMBL. Newly identified polynucleotides and corresponding polypeptides may also be compared to polynucleotides and polypeptides contained in public domain information to ascertain homology to known polynucleotides and polypeptides. In this way, the degree of similarity, identity or homology of polynucleotides and polypeptides having an unknown function may be determined relative to polynucleotides and polypeptides having known functions.

Information relating to the sequences of isolated polynucleotides may be used in a variety of ways. Specified polynucleotides having a particular sequence may be isolated, or synthesized, for use in in vivo or in vitro experimentation as probes or primers. Alternatively, collections of sequences of isolated polynucleotides may be stored using magnetic or optical storage medium and analyzed or manipulated using computer hardware and software, as well as other types of tools.

Summary of the Invention

The present invention provides isolated polynucleotides comprising a sequence selected from the group consisting of: (a) sequences identified in the attached Sequence Listing as SEQ ID NOS: 1-33; (b) complements reverse sequences and reverse complements of SEQ ID NOS: 1-33 and fragments of SEQ ID NOS: 1-33; (c) open reading frames contained in SEQ ID NOS: 1-33 and their variants; (d) functional domains contained in SEQ ID NOS: 1-33; and (e) sequences comprising at least a specified number of contiguous residues of a sequence of SEQ ID NOS: 1-33 (x-mers). Oligonucleotide probes and primers corresponding to the sequences set out in SEQ ID NOS: 1-33, and their variants are also provided. All of these polynucleotides and oligonucleotide probes and primers are collectively referred to herein, as "polynucleotides of the present invention."

The polynucleotide sequences identified as SEQ ID NOS: 1-33 were derived from a microbial source, namely from fragmented genomic DNA of Lactobacillus rhamnosus, strain HN001, described in PCT International Publication No. WO 99/10476. Lactobacillus rhamnosus strain HN001 are heterofermentative bacteria that are Gram positive, non-motile, non-spore forming, catalase negative, facultative anaerobic rods exhibiting an optimal growth temperature of 37+1 °C and an optimum pH of 6.0 - 6.5. A biologically pure culture of Lactobacillus rhamnosus strain HN001 was deposited at the Australian Government Analytical Laboratories (AGAL), The New South Wales

Regional Laboratory, 1 Sual in Street, Pymble, NSW 2073, Australia, as Deposit No. NM97/09514, dated 18 August 1997.

The polynucleotide sequences disclosed herein are primarily "full-length" sequences, in that they represent a full-length gene encoding a full-length polypeptide and confirm an open reading frame. Similarly, RNA sequences, reverse sequences, complementary sequences, antisense sequences and the like, corresponding to the polynucleotides of the present invention, may be routinely ascertained and obtained using the polynucleotides identified as SEQ ID NOS: 1- 33. The present invention further provides isolated polypeptides, including polypeptides encoded, or partially encoded, by the polynucleotides disclosed herein. In certain specific embodiments, the polypeptides of the present invention comprise a sequence selected from the group consisting of sequences identified as SEQ ID NO: 42-75, and variants thereof. Polypeptides encoded by the polynucleotides of the present invention may be expressed and used in various assays to determine their biological activity. Such polypeptides may be used to raise antibodies, to isolate corresponding interacting proteins or other compounds, and to quantitatively determine levels of interacting proteins or other compounds. The polypeptides of the present invention may also be used as nutritional additives and as additives in dairy processing and fermentation processing. Several polypeptides of the present invention also have human and animal health related benefits.

Genetic constructs comprising the inventive polynucleotides are also provided, together with transgenic host cells comprising^* such constructs and transgenic organisms, such as microbes, comprising such cells. The present invention also contemplates methods for modulating the polynucleotide and/or polypeptide content and composition of an organism, such methods involving stably incorporating into the genome of the organism a genetic construct comprising a polynucleotide of the present invention. Such modulation may involve up regulating or down regulating expression from one or more polynucleotides of the present invention. Up regulation may be accomplished, for example, by providing multiple gene copies, modulating expression by modifying regulatory elements or the like. Similarly, down regulation may be accomplished using known antisense and gene silencing techniques. In one embodiment, the target organism is a microbe, preferably a microbe used in fermentation, more preferably a microbe of the genus Lactobacillus, and most preferably

Lactobacillus rhamnosus, or other closely microbial related species used in the dairy industry. In a related aspect, methods for producing a microbe having an altered genotype and/or phenotype is provided, such methods comprising transforming a microbial cell with a genetic construct of the present invention to provide a transgenic cell, and cultivating the transgenic cell under conditions conducive to growth and multiplication. Organisms having an altered genotype or phenotype as a result of modulation of the level or content of a polynucleotide or polypeptide of the present invention compared to a wild-type organism, as well as components and progeny of such organisms, are contemplated by and encompassed within the present invention. The isolated polynucleotides of the present invention may be usefully employed for the detection of lactic acid bacteria, preferably L. rhamnosus, in a sample material, using techniques well known in the art, such as polymerase chain reaction (PCR) and DNA hybridization, as detailed below.

The inventive polynucleotides and polypeptides may also be employed in methods for the selection and production of more effective probiotic bacteria; as

"bioactive" (health-promoting) ingredients and health supplements, for immune function enhancement; for reduction of blood lipids such as cholesterol; for production of bioactive material from genetically modified bacteria; as adjuvants; for wound healing; in vaccine development, particularly mucosal vaccines; as animal probiotics for improved animal health and productivity; in selection and production of genetically modified rumen microorganisms for improved animal nutrition and productivity, better flavor and improved milk composition; in methods for the selection and production of better natural food bacteria for improved flavor, faster flavor development, better fermentation characteristics, vitamin synthesis and improved textural characteristics; for the production of improved food bacteria through genetic modification; and for the identification of novel enzymes for the production of, for example, flavors or aroma concentrates.

The isolated polynucleotides of the present invention also have utility in genome mapping, in physical mapping, and in positional cloning of genes of more or less related microbes. Additionally, the polynucleotide sequences identified as

SEQ ID NOS: 1-33, and their variants, may be used to design oligonucleotide probes and primers. Oligonucleotide probes and primers have sequences that are substantially complementary to the polynucleotide of interest over a certain portion of the polynucleotide. Oligonucleotide probes designed using the polynucleotides of the present invention may be used to detect the presence and examine the expression patterns of genes in any organism having sufficiently similar DNA and RNA sequences in their cells, using techniques that are well known in the art, such as slot blot DNA hybridization techniques. Oligonucleotide primers designed using the polynucleotides of the present invention may be used for PCR amplifications. Oligonucleotide probes and primers designed using the polynucleotides of the present invention may also be used in connection with various microarray technologies, including the microarray technology of Affymetrix (Santa Clara, CA).

The polynucleotides of the present invention may also incorporate regulatory elements such as promoters, gene regulators, origins of DNA replication, secretion signals, cell wall or membrane anchors for genetic tools

(such as expression or integration vectors).

The polynucleotide sequences, encoded polypeptides and genetic constructs of this invention are useful for improving the properties of microbes that are used in the manufacture of milk-derived products, such as cheeses, yogurt, fermented milk products, sour milks, and buttermilk. Microbial metabolism during fermentation, which results in the breakdown of proteins, lipids and lactose in milk, influences the speed of ripening, the texture and consistency of fermented milk products, and the development of flavors and aromas during ripening. Undesirable flavors in milk products are produced, for example, by the food of milk-producing animals, microbial action, and enzymatic activity during fermentation, and require removal. The present invention provides polynucleotides and polypeptides and methods for their use in modifying the flavor, aroma, texture and health-related benefits of milk-derived products. Methods are described for modulating the polynucleotide content or composition of microbes used in the dairy industry by transforming the microbes with one or more polynucleotides sequences of Lactobacillus rhamnosus strain HN001. The inventive polynucleotides alsoinclude sequences encoding polypeptides that increase the survivability of microbes during industrial fermentation processes, wherein exposure to osmotic, temperature and other stresses can lead to reduced microbial viability, impaired metabolic activity and suboptimal fermentation conditions. While the present invention is described with particular reference to milk-derived products, it will be recognized that microbes such as Lactobacillus, which are used in the dairy industry, are also used in the production of other foods and beverages (e.g., fermented vegetables, beer, wines, juices, sourdough breads). It is expected that the polynucleotides described herein and their methods of use can be used for the processing of these foods and beverages as well.

This invention also provides transgenic microbial populations comprising expressible polynucleotide sequences of Lactobacillus rhamnosus strain HN001 which health-related benefits. For example, the polypeptides encoded by the inventive sequences include enzymes that detoxify carcinogens, degrade allergenic proteins and lactose, and produce bioactive peptides and biogenic amines. Microbes transformed with these polynucleotide sequences can be taken internally as a probiotic composition or alternatively, the microbes or their encoded polypeptides can be added to products to provide health-related benefits. Nonpathogenic bacteria, preferably lactic-acid producing species of Bacillus,

Lactobacillus, Sporolactobacillus or Bifidiobacterium, that are able to colonize the gastrointestinal tract, preferably the gastrointestinal tract of a mammal, are useful for preventing or reducing pathogen colonization of the gastrointestinal mucosa, and for replacing normal flora that are depleted, for example, by drug therapy. The polynucleotide sequences of this invention can be used to transform microbes for use in a therapeutic composition that- is effective for treating or preventing a gastrointestinal condition or disorder caused by the presence of pathogenic microbes in the gastrointestinal tract or by the absence of normal intestinal microbes in the intestinal tract. Such probiotic compositions can be administered alone or in combination with another pharmaceutical agent, depending on the condition that is to be treated.

All references cited herein, including patent references and non-patent publications, are hereby incorporated by reference in their entireties.

Brief Description of the Drawings

Fig. 1 shows the nucleotide sequence containing L. rhamnosus strain HN001 esterase gene AA7 showing ATG initiation and translation stop codons (boxed).

Fig. 2 shows the amino acid sequence of HNOOl esterase AA7. Fig. 3 demonstrates the esterase activity of the AA7 fusion protein.

Production of ethyl butyrate from j?Ωrø-nitrophenyl butyrate substrate was measured by change in OD at 410 nm. While buffer only (♦) and the HNOOl non- esterase fusion protein (•) showed minimal esterase activity, the STl esterase from Streptococcus thermophilus (A) and the AA7 esterase fusion protein (■) showed strong activity.

Fig 4 shows the dose-response of the AA7 fusion protein. While buffer- only (•) showed no esterase activity, increasing amounts of His-patch/Thio/AA7 fusion protein; 5 μl (♦), 10 μl (A) and 20 μl (■) purified protein showed increasing rates of substrate hydrolysis. The increase in substrate hydrolysis was proportional to amount of AA7 fusion protein added.

Fig. 5 shows the effect of the serine esterase inhibitor PMSF on esterase AA7 activity. Esterase activity of the His-patch/Thio/AA7 fusion protein was assessed in the absence (■) and presence (A) of 10 mM PMSF. A buffer-only reaction (•) was used as a negative control. The presence of PMSF reduced HNOOl esterase AA7 enzyme activity.

Fig. 6 shows the nucleotide sequence containing L. rhamnosus strain HNOOl autoaggregation gene AG5 showing ATG initiation and translation stop codons (boxed).

Fig. 7 shows the amino acid sequence of HNOOl autoaggregation protein AG5.

Figs. 8A and 8B are images of phase contrast photomicrographs. Fig. 8A illustrates an image of a phase-contrast photomicrograph (exposure 1/8 sec, final magnification x 240) showing obvious clumping of washed L. rhamnosus strain HNOOl cells in the presence of AG5 autoaggregation protein tagged with GST. Fig. 8B illustrates an image of a phase-contrast photomicrograph (exposure 1/8 sec, final magnification x 240) showing no clumping of washed L. rhamnosus strain HNOOl cells in the presence of an irrelevant (non-adhesion) HNOOl protein tagged with GST, as a negative control. Fig. 9 shows the nucleotide sequence containing L. rhamnosus strain

HNOOl malic enzyme gene AA5 showing ATG initiation and translation stop codons (boxed).

Fig. 10 shows the amino acid sequence of HNOOl malic enzyme AA5. Fig 11 demonstrates malate enzyme activity measured as rate of pyruvate reduction by crude lysate preparations of EJ1321 cell transformants. ■ PBS buffer-only; A 3.5 μg wild-type EJ1321 cell lysate; ♦ 3.5 μg cell lysate of EJ1321 transformed with pGEX-6P-3 construct encoding an irrelevant HNOOl protein (AD5); • 3.5 μg cell lysate of EJ1321 transformed with pGEX-6P-3 construct encoding HNOOl malic enzyme AA5. Fig 12 shows data illustrating the effect of increasing amounts of EJ1321 crude lysate on malic enzyme activity. ■ 5 μl wild-type EJ1321 cell lysate; A 5 μl cell lysate of EJ1321 transformed with pGex-6P-3 encoding AA5; ♦ 50 μl cell lysate of EJ1321 transformed with pGex-6P-3 encoding AA5; • 200 μl cell lysate ofEJ1321 transformed with pGex-6P-3 encoding AA5. Fig. 13 shows the nucleotide sequence containing L. rhamnosus strain

HNOOl malate dehydrogenase gene_4G3 showing TTG initiation and translation stop codons (boxed).

Fig. 14 shows the amino acid sequence of HNOOl malate dehydrogenase AG3. Fig. 15 shows the nucleotide sequence containing L. rhamnosus strain

HNOOl dihydrodipicolinate synthase gene AI2 showing ATG initiation and translation stop codons (boxed).

Fig. 16 shows the amino acid sequence of HNOOl dihydrodipicolinate synthase AI2. Fig. 17 shows the nucleotide sequence containing L. rhamnosus strain aspartate aminotransferase gene AH9 showing GTG initiation and translation stop codons (boxed).

Fig. 18 shows the amino acid sequence of HNOOl aspartate aminotransferase AH9. Fig. 19 shows the nucleotide sequence containing L. rhamnosus strain

HNOOl serine dehydratase subunits α (AFT) and β (AF8). ATG translation initiation codons and termination codons are shown, boxed for AF8, shaded for AFT.

Fig. 20 shows the percentage serine utilisation by HNOOl strain in liquid culture with 5 mM initial serine concentration. ■ HNOOl transformed with vector only; ♦ pTRKH2 construct containing HNOOl serine dehydratase.

Fig. 21 shows the percentage serine utilisation by HNOOl strain in liquid culture with 12 mM initial serine concentration. ■ HNOOl transformed with vector only, ♦ pTRKH2 construct containing HNOOl serine dehydratase. Fig. 22A shows the amino acid sequence of J. rhamnosus strain HNOOl serine dehydratase subunit β (AF7), and Fig. 22B shows the ammo acid sequence of L. rhamnosus strain HNOOl serine dehydratase subunit (AF8).

Fig. 23 shows the nucleotide sequence containing L. rhamnosus strain HNOOl histidinol-phosphate aminotransferase gene AG2 showing ATG initiation and translation stop codons (boxed).

Fig. 24 shows the amino acid sequence of HNOOl histidinol-phosphate aminotransferase AG2.

Fig. 25 shows the nucleotide sequence containing L. rhamnosus strain HNOOl malY-aminotransferase gene AJ6 showing ATG initiation and translation stop codons (boxed).

Fig. 26 shows the amino acid sequence of HNOOl malY-aminotransferase AJ6.

Fig. 27 shows the ucleotide sequence containing L. rhamnosus strain HNOOl malY-aminotransferase gene AJ7 showing ATG initiation and translation stop codons (boxed).

Fig. 28 shows the amino acid sequence of HNOOl malY-aminotransferase AJ7.

Fig. 29 shows the nucleotide sequence containing L. rhamnosus strain HNOOl cystathione β-lyase gene AC8 showing ATG initiation and translation stop codons (boxed). Fig. 30 shows the amino acid sequence of HNOOl cystathione β-lyase

AC8.

Fig. 31 shows experimental results demonstrating cystathione β-lyase activity measured as rate of mercaptide formation. ♦ 10 μl purified HNOOl cystathione β-lyase AC8 fusion protein; ■ 10 μl purified CAT fusion protein; A 10 μl H₂O only; • 10 μl elution buffer only.

Fig. 32 shows the experimentally determined dose-response of the AC 8 fusion protein. Cystathione β-lyase activity of increasing amounts of His- patch/Thio/AC8 fusion protein; 10 μl (♦), 25 μl (■) and 50 μl (A) purified protein showed increasing rates of mercaptide formation. The increase in mercaptide formation was proportional to amount of AC 8 fusion protein added.

Fig. 33 shows the nucleotide sequence containing L. rhamnosus strain HNOOl phosphoenolpyruvate hydratase AK4 showing ATG initiation and translation stop codons (boxed).

Fig. 34 shows the amino acid sequence of J. rhamnosus strain HNOOl phosphoenolpyruvate hydratase AK4.

Fig. 35 shows the nucleotide sequence containing L. rhamnosus strain HNOOl tagatose bisphosphate aldolase AK1 showing ATG initiation and translation stop codons (boxed).

Fig. 36 shows the amino acid sequence of L. rhamnosus strain HNOOl tagatose bisphosphate aldolase AKl .

Fig. 37 shows the nucleotide sequence containing L. rhamnosus strain HNOOl phosphoglycerate ldnase AK6 showing TTG initiation and translation stop codons (boxed).

Fig. 38 shows the amino acid sequence of L. rhamnosus strain HNOOl phosphoglycerate kinase AK6.

Fig. 39 shows the nucleotide sequence containing L. rhamnosus strain HNOOl triosephosphate isomerase AK5 showing ATG initiation and translation stop codons (boxed). Fig. 40 shows the amino acid sequence of J. rhamnosus strain HNOOl triosephosphate isomerase AK5.

Fig. 41 shows the nucleotide sequence containing L. rhamnosus strain HNOOl phosphoryl carrier protein ΗPRAA9 showing ATG initiation and translation stop codons (boxed). Fig. 42 shows the amino acid sequence of L. rhamnosus strain HNOOl phosphoryl carrier protein HPR AA9.

Fig. 43 shows the nucleotide sequence containing L. rhamnosus strain HNOOl glyceraldehyde-3-phosphate dehydrogenase AKT showing ATG initiation and translation stop codons (boxed). Fig. 44 shows the amino acid sequence of L. rhamnosus strain HNOOl glyceraldehyde-3-phosphate dehydrogenase AK7.

Fig. 45 shows the nucleotide sequence containing L. rhamnosus strain HNOOl sorR transcription regulator AL3 showing ATG initiation and translation stop codons (boxed). Fig. 46 shows the amino acid sequence of L. rhamnosus strain HNOOl sorR transcription regulator AL3.

Fig. 47 shows the nucleotide sequence containing L. rhamnosus strain fpg gene AL4 showing ATG initiation and translation stop codons (boxed). Fig. 48 shows the amino acid sequence of HNOOl fpg AL4. Fig. 49 shows the nucleotide sequence containing the L. rhamnosus strain

HNOOl acetoin dehydrogenase gene API showing ATG initiation and translation stop codons (boxed).

Fig. 50 shows the amino acid sequence of HNOOl acetoin dehydrogenase API. Fig. 51 illustrates the experimental results of an acetoin reductase assay as measured by oxidation of NADH co-factor by OD at 340 nm in the presence of acetoin substrate. •, elution buffer only; ■, purified irrelevant GST-fusion protein; A, purified GST protein; ♦, purified API -GST fusion protem. Fig. 52 shows the nucleotide sequence containing the L. rhamnosus strain

HNOOl aflatoxin Bi aldehyde reductase gene AIT showing ATG initiation and translation stop codons (boxed).

Fig. 53 shows the amino acid sequence of HNOOl aflatoxin Bi aldehyde reductase AI7. Fig. 54 shows the experimental results of aflatoxin _\ aldehyde reductase assay according to oxidation of the NADPH co-factor in the presence of acetoin substrate. X , water only; +, Sepharose column elution buffer only; •, irrelevant

GST-fusion protein; ■, 10 μl purified AP4-GST fusion protein; A 20 μl purified

AP4-GST fusion protein. Fig. 55 shows the experimental determination of 6-Phospho-β- galactosidase enzyme activity as measured by substrate utilisation using crude lysates of strains transformed with pGex-6P-3 encoding A05 (♦), pGex-6P-3 encoding an irrelevant protein (■), or using lysis buffer only (X).

Fig. 56 shows 6-Phospho-β-galactosidase enzyme activity as measured experimentally by substrate utilisation using increasing amounts of crude lysate from strains transformed with pGex-6P-3 encoding A05-GST fusion protein. ♦, 50 μl lysate; ■. 100 μl lysate; A, 200 μl lysate; Φ, 200 μl lysis buffer only.

Fig. 57 shows the nucleotide sequence containing the L. rhamnosus strain HNOOl aromatic aminotransferase gene_4H7 showing ATG initiation and translation stop codons (boxed).

Fig. 58 shows the amino acid sequence of ΗN001 aromatic aminotransferase AΗ7.

Fig. 59 shows the nucleotide sequence containing the L. rhamnosus strain HNOOl acetate kinase gene AP5 showing ATG initiation and translation stop codons (boxed).

Fig. 60 shows the amino acid sequence of HNOOl acetate kinase AP5.

Fig. 61 shows the nucleotide sequence containing the L. rhamnosus strain HNOOl basic surface protein gene_4 showing ATG initiation and translation stop codons (boxed). Fig. 62 shows the amino acid sequence of HNOOl basic surface protein

AC9.

Fig. 63 shows the nucleotide sequence containing the L. rhamnosus strain HNOOl aromatic outer membrane protein A AL8 showing ATG initiation and translation stop codons (boxed). Fig. 64 shows the amino acid sequence of HNOOl outer membrane protein

AL8.

Fig. 65 shows the nucleotide sequence containing the L. rhamnosus strain HNOOl aromatic extracellular matrix binding protein AM4 showing ATG initiation and translation stop codons (boxed). Fig. 66 shows the amino acid sequence of HNOOl extracellular matrix binding protein AM4.

Fig. 67 shows the nucleotide sequence containing the L. rhamnosus strain HNOOl aromatic high-molecular- weight adhesion protein AL 7 showing ATG initiation and translation stop codons (boxed). Fig. 68 shows the amino acid sequence of HNOOl high-molecular- weight adhesion protem AL7.

Fig. 69 shows the nucleotide sequence containing the L. rhamnosus strain HNOOl aromatic PEB1 AJ4 showing ATG initiation and translation stop codons (boxed). Fig. 70 shows the amino acid sequence of HNOOl PEB1 AJ4.

Fig. 71 shows the relative density of autoradiographic signals from AJ4 protein (grey bars) to dot blots of intestinal proteins, compared to a positive control (mapA, white bars) and negative control (irrelevant HNOOl protein, black bars). Results for each dot (duplicates) are shown. Fig. 72 shows the nucleotide sequence containing the L. rhamnosus strain

HNOOl dihydrodipicolinate reductase AI3 showing ATG initiation and translation stop codons (boxed).

Fig. 73 shows the amino acid sequence of HNOOl dihydrodipicolinate reductase AI3. Fig. 74 shows the nucleotide sequence containing the L. rhamnosus strain

HNOOl Fructose-bisphosphate aldolase AM8 showing ATG initiation and translation stop codons (boxed).

Fig. 75 shows the amino acid sequence of HNOOl Fructose-bisphosphate aldolase AM8. Fig. 76 shows the nucleotide sequence containing the L. rhamnosus strain

HNOOl chaperone protein dnaK AM9 showing ATG initiation and translation stop codons (boxed).

Fig. 77 shows the amino acid sequence of HNOOl chaperone protein dnaK AM9. Fig. 78 shows the nucleotide sequence containing the L. rhamnosus strain

HNOOl 6-phospho-β-galactosidase AO5 showing translation stop codon (boxed).

Fig. 79 shows the amino acid sequence of HNOOl 6-phospho-β- galactosidase AO5.

Fig. 80 shows the nucleotide sequence containing the L. rhamnosus strain HNOOl peptidase pepO showing ATG initiation and translation stop codons

(boxed).

Fig. 81 shows the amino acid sequence of HNOOl peptidase. pepO.

Detailed Description The polynucleotides disclosed herein were isolated by high throughput sequencing of DNA libraries from the lactic acid bacteria Lactobacillus rhamnosus as described in Example 1. Cell wall, cell surface and secreted components of lactic acid bacteria are known to mediate immune modulation, cell adhesion and antibacterial activities, resulting in many beneficial effects including: resistance to enteric pathogens: modulation of cancer, including colon cancer: anti-mutagenesis effects; reduction of small bowel bacterial overgrowth; modulation of auto-immune disorders; reduction in allergic disorders; modulation of urogenital infections, inflammatory bowel disorder, irritable bowel syndrome, Helicobacter pylori infection and hepatic encephalopatiiy; reduction of infection with pathogens; regulation of colonocyte proliferation and differentiation; reduction of mucosal permeability; and relief of constipation and diarrhea. These cell components include, but are not limited to, peptidoglycans, teichoic acids, lipoteichoic acids, polysaccharides, adhesion proteins, secreted proteins, surface layer or S-layer proteins, collagen binding proteins and other cell surface proteins, and antibacterial substances such as bacteriocins and organic acids produced by these bacteria. Polynucleotides involved in the synthesis of these proteins and in the synthesis, modification, regulation, transport, synthesis and/or accumulation of precursor molecules for these proteins can be used to modulate the immune effects, antibacterial, cell adhesion and competitive exclusion effects of the bacteria or of components that might be produced by these bacteria. In order to function effectively as probiotic bacteria, L. rhamnosus HNOOl must survive environmental stress conditions in the gastrointestinal tract, as well as commercial and industrial processes. Modification of particular polynucleotides or regulatory processes have been shown to be effective against a number of stresses including oxidative stress, pH, osmotic stress, dehydration, carbon starvation, phosphate starvation, nitrogen starvation, amino acid starvation, heat or cold shock and mutagenic stress. Polynucleotides involved in stress resistance often confer multistress resistance, i.e., when exposed to one stress, surviving cells are resistant to several non-related stresses. Bacterial genes and/or processes shown to be involved in multistress resistance include: Intracellular phosphate pools - inorganic phosphate starvation leads to the induction of pho regulon genes, and is linked to the bacterial stringent response. Gene knockouts involving phosphate receptor genes appear to lead to multistress resistance. Intracellular guanosine pools - purine biosynthesis and scavenger pathways involve the production of phosphate-guanosine compounds that act as signal molecules in the bacterial stringent response. Gene knockouts involving purine scavenger pathway genes appear to confer multistress resistance. Osmoregulatory molecules - small choline-based molecules, such as glycine- betaine, and sugars, such as trehalose, are protective against osmotic shock and are rapidly imported and/or synthesized in response to increasing osmolarity. Acid resistance - lactobacilli naturally acidify their environment through the excretion of lactic acid, mainly through the cit operon genes responsible for citrate uptake and utilization.

Stress response genes - a number of genes appear to be induced or repressed by heat shock, cold shock, and increasing salt through the action of specific promoters.

The isolated polynucleotides of the present invention, and genetic constructs comprising such polynucleotides may be employed to produce bacteria having desired phenotypes, including increased resistance to stress and improved fermentation properties.

Many enzymes are known to influence dairy product flavor, functional and textural characteristics as well as general fermentation characteristics such as speed of growth, acid production and survival. These enzymes include those involved in the metabolism of lipids, polysaccharides, amino acids and carbohydrates as well as those involved in the lysis of the bacterial cells.

The isolated polynucleotides and polypeptides of the present invention have been demonstrated to have the identities, functions and utilities described throughout this application and in the Examples. The polynucleotide and polypeptide SEQ ID NOS of the present invention, and corresponding identification and functional information is provided below in Table 1.

The inventive polynucleotide identified herein as SEQ ID NO: 1 shows some degree of homology to the previously identified peptidase pepO gene from Lactococcus (Tan et al, Appl. Environ. Microbiol 57:3539-3599, 1991). PepO is a 70 kDa metallo-endopeptidase that hydrolyzes a range of polypeptides, including casein fragments. Peptidase pepO is believed to be a key enzyme in the cheese ripening process and contributes to flavor development as a cheese matures. The enzyme remains active under cheese conditions of reduced pH, high salt and low water activity where many other peptidases of lactic acid bacteria are inactivated. As detailed below, the polypeptide encoded by the inventive polynucleotide of SEQ ID NO: 1 (amino acid sequence provided in SEQ ID NO: 42) is effective in the hydrolysis of the milk protein, casein. The hydrolysis of milk caseins in cheese results in textural changes and the development of cheese flavors. The polypeptide of SEQ ID NO: 42 and compositions comprising this polypeptide and/or variants thereof, may thus be effectively employed in the enhancement of cheese flavors and textures. Isolated polynucleotides of the present invention include the polynucleotides identified herein as SEQ ID NOS: 1-33; isolated polynucleotides comprising a polynucleotide sequence selected from the group consisting of SEQ ID NOS: 1-33; isolated polynucleotides comprising at least a specified number of contiguous residues (x-mers) of any of the polynucleotides identified as SEQ ID NOS: 1-33; isolated polynucleotides comprising a polynucleotide sequence that is complementary to any of the above polynucleotides; isolated polynucleotides comprising a polynucleotide sequence that is a reverse sequence or a reverse complement of any of the above polynucleotides; ^* antisense sequences corresponding to any of the above polynucleotides; and variants of any of the above polynucleotides, as that term is described in this specification.

The word "polynucleotide(s)," as used herein, means a single or double stranded polymer of deoxyribonucleotide or ribonucleotide bases and includes DNA and corresponding RNA molecules, including mRNA molecules, both sense and antisense strands of DNA and RNA molecules, and comprehends cDNA, genomic DNA and recombinant DNA, as well as wholly or partially synthesized polynucleotides. A polynucleotide of the present invention may be an entire gene, or any portion thereof. A gene is a DNA sequence which codes for a functional protein or RNA molecule. Operable antisense polynucleotides may comprise a fragment of the corresponding polynucleotide, and the definition of "polynucleotide" therefore includes all operable antisense fragments. Antisense polynucleotides and techniques involving antisense polynucleotides are well known in the art and are described, for example, in Robinson-Benion, et al., "Antisense techniques," Methods in Enzymol 254(23): 363-375, 1995; and Kawasaki, et al, Artiβc. Organs 20 (8): 836-848, 1996. The definitions of the terms "complement," "reverse complement," and

"reverse sequence," as used herein, are best illustrated by the following examples. For the sequence 5' AGG ACC 3', the complement, reverse complement, and reverse sequences are as follows: complement 3' TCCTGG 5' reverse complement 3 ' GGTCCT 5 ' reverse sequence 5' CCAGGA 3 '

Identification of genomic DNA and heterologous species DNA can be accomplished by standard DNA/DNA hybridization techniques, under appropriately stringent conditions, using all or part of a DNA sequence as a probe to screen an appropriate library. Alternatively, PCR techniques using oligonucleotide primers that are designed based on known DNA and protein sequences can be used to amplify and identify other identical or similar DNA sequences. Synthetic DNA corresponding to the identified sequences or variants thereof may be produced by conventional synthesis methods. All of the polynucleotides described herein are isolated and purified, as those terms are commonly used in the art.

The polynucleotides identified as SEQ ID NOS: 1-33 may contain open reading frames ("ORFs"), or partial open reading frames, encoding polypeptides.

Additionally, polynucleotides identified as SEQ ID NOS: 1-33 may contain non- coding sequences such as promoters and terminators that may be useful as control elements. The open reading frames contained in polynucleotides of the present invention may be isolated and/or synthesized. Expressible genetic constructs comprising the open reading frames and suitable promoters, initiators, terminators, etc., which are well Icnown in the art, may then be constructed. Such genetic constructs may be introduced into a host cell to express the polypeptide encoded by the open reading frame. Genetic constructs may be designed and constructed, as is known in the art, to enhance or silence expression of an identified polypeptide. Genetic constructs of the present invention may thus be assembled using techniques known in the art to enhance or reduce expression of polypeptides of the present invention encoded by polynucleotides of the present invention. Suitable host cells may .include various prokaryotic and eukaryotic cells. In vitro expression of polypeptides is also possible, as well Icnown in the art. As used herein, the term "oligonucleotide" refers to a relatively short segment of a polynucleotide sequence, generally comprising between 6 and 60 nucleotides, and comprehends both probes for use in hybridization assays and primers for use in the amplification of DNA by polymerase chain reaction.

As used herein, the term "x-mer," with reference to a specific value of "x," refers to a polynucleotide or polypeptide comprising at least a specified number

("x") of contiguous residues of any of the polynucleotides and polypeptides identified as SEQ ID NOS: 1-75. The value of x may be from about 20 to about 600, depending upon the specific sequence.

In another aspect, the present invention provides isolated polypeptides encoded, or partially encoded, by the above polynucleotides, including the polypeptides identified as SEQ ID NOS: 42-75. As used herein, the term "polypeptide" encompasses amino acid chains of any length, including full-length proteins, wherein the amino acid residues are linked by covalent peptide bonds. The term "polypeptide encoded by a polynucleotide" as used herein, includes polypeptides encoded by a polynucleotide which comprises an isolated polynucleotide sequence or variant provided herein. Polypeptides of the present invention may be naturally purified products, or may be produced partially or wholly using recombinant techniques. Such polypeptides may be glycosylated with bacterial, fungal, mammalian or other eukaryotic carbohydrates or may be non-glycosylated.

Polypeptides of the present invention may be produced recombinantly by inserting a polynucleotide that encodes the polypeptide into an expression vector and expressing the polypeptide in an appropriate host. Any of a variety of expression vectors known to those of ordinary skill in the art may be employed. Expression may be achieved in any appropriate host cell that has been transformed or transfected with an expression vector containing a polypeptide encoding a recombinant polypeptide. Suitable host cells include prokaryotes, yeast and higher eukaryotic cells. Preferably, the host cells employed are Escherichia coli, Lactococcus lactis, Lactobacillus, insect, yeast or a mammalian cell line such as COS or CHO. The polynucleotide(s) expressed in this manner may encode naturally occurring polypeptides, portions of naturally occurring polypeptides, or other variants thereof.

In a related aspect, polypeptides are provided that comprise at least a functional portion of a polypeptide having an amino acid sequence encoded by a polynucleotide of the present invention. As used herein, a "functional portion" of a polypeptide is that portion which contains the active site essential for affecting the function of the polypeptide, for example, the portion of the molecule that is capable of binding one or more reactants. The active site may be made up of separate portions present on one or more polypeptide chains and will generally exhibit high binding affinity. Functional portions of a polypeptide may be identified by first preparing fragments of the polypeptide by either chemical or enzymatic digestion of the polypeptide, or by mutation analysis of the polynucleotide that encodes the polypeptide and subsequent expression of the resulting mutant polypeptides. The polypeptide fragments or mutant polypeptides are then tested to determine which portions retain biological activity, using, for example, the representative assays provided below.

Portions and other variants of the inventive polypeptides may be generated by synthetic or recombinant means. Synthetic polypeptides having fewer than about 100 amino acids, and generally fewer than about 50 amino acids, may be generated using techniques that are well Icnown to those of ordinary skill in the art. For example, such polypeptides may be synthesized using any of the commercially available solid-phase techniques, such as the Merrifield solid-phase synthesis method, where amino acids are sequentially added to a growing amino acid chain (See Merrifield, J. Am. Chem. Soc. 85:2149-2154, 1963). Equipment for automated synthesis of polypeptides is commercially available from suppliers such as Perlcin Elmer/ Applied Biosystems, Inc. (Foster City, CA), and may be operated according to the manufacturer's instructions. Variants of a native polypeptide may be prepared using standard mutagenesis techniques, such as oligonucleotide-directed site-specific mutagensis (Kunkel, Proc. Natl. Acad. Sci. USA 82: 488-492, 1985). Sections of DNA sequences may also be removed using standard techniques to permit preparation of truncated polypeptides.

In general, the polypeptides disclosed herein are prepared in an isolated, substantially pure form. Preferably, the polypeptides are at least about 80% pure; more preferably at least about 90%) pure; and most preferably at least about 99% pure.

As used herein, the term "variant" comprehends polynucleotide or polypeptide sequences different from the specifically identified sequences, wherein one or more nucleotides or amino acid residues is deleted, substituted, or added. Variants may be naturally occurring allelic variants, or non-naturally occurring variants. Variant polynucleotide sequences preferably exhibit at least

40%, more preferably at least 60%, more preferably yet at least 75%, and most preferably at least 90%) identity to a sequence of the present invention. Variant polypeptide sequences preferably exhibit at least 50%o, more preferably at least 75%o, more preferably yet at least 90%, and most preferably at least 95%> identity to a sequence of the present invention. The percentage identity is determined by aligning the two sequences to be compared as described below, determining the number of identical residues in the aligned portion, dividing that number by the total number of residues in the inventive (queried) sequence, and multiplying the result by 100. Polynucleotide and polypeptide sequences may be aligned, and the percentage of identical residues in a specified region may be determined against another polynucleotide or polypeptide, using computer algorithms that are publicly available. Two exemplary algorithms for aligning and identifying the similarity of polynucleotide sequences are the BLASTN and FASTA algorithms. Polynucleotides may also be analyzed using the BLASTX algorithm, which compares the six-frame conceptual translation products of a nucleotide query sequence (both strands) against a protein sequence database. The percentage identity of polypeptide sequences may be examined using the BLASTP algorithm. The BLASTN, BLASTX and BLASTP programs are available on the NCBI anonymous FTP server ( ftp ://ncbi.nlm.nih. gov) under /blast/executables/ and are available from the National Center for Biotechnology Information (NCBI),

National Library of Medicine, Building 38A, Room 8N805, Bethesda, MD 20894, USA. The BLASTN algorithm Version 2.0.11 [Jan-20-2000], set to the parameters described below, is preferred for use in the determination of polynucleotide variants according to the present invention. The BLASTP algorithm, set to the parameters described below, is preferred for use in the determination of polypeptide variants according to the present invention. The use of the BLAST family of algorithms, including BLASTN, BLASTP and BLASTX, is described at NCBFs website at URL http://www.ncbi.nlm.nih.gov/BLAST/newblast.html and in the publication of Altschul, et al, Nucleic Acids Res. 25: 3389-3402, 1997.

The computer algorithm FASTA is available on the Internet at the ftp site ftp://ftp.virginia.edu/pub/fasta/, and from the University of Virginia by contacting David Hudson, Vice Provost for Research, University of Virginia, P.O. Box 9025, Charlottesville, VA 22906-9025, USA. FASTA Version 2.0u4 [February 1996], set to the default parameters described in the documentation and distributed with the algorithm, may be used in the determination of variants according to the present invention. The use of the FASTA algorithm is described in Pearson and Lipman, Proc. Natl. Acad. Sci. USA 85:2444-2448, 1988; and Pearson, Methods in Enzymol. 183: 63-98, 1990. The following running parameters are preferred for determination of alignments and similarities using BLASTN that contribute to the E values and percentage identity for polynucleotide sequences: Unix running command: blastall -p blastn -d embldb -e 10 -GO -E0 -r 1 -v 30 -b 30 -i queryseq -o results; the parameters are: -p Program Name [String]; -d Database [String]; -e Expectation value (E) [Real]; -G Cost to open a gap (zero invoices default behavior) [Integer]; -E Cost to extend a gap (zero invokes default behavior) [Integer]; -r Reward for a nucleotide match (BLASTN only) [Integer]; -v Number of one-line descriptions (V) [Integer]; -b Number of alignments to show (B) [Integer]; -i Query File [File In]; and -o BLAST report Output File [File Out] Optional. The following running parameters are preferred for determination of alignments and similarities using BLASTP that contribute to the E values and percentage identity of polypeptide sequences: blastall -p blastp -d swissprotdb -e 10 -G 0 -E 0 -v 30 -b 30 -i queryseq -o results; the parameters are: -p Program Name [String]; -d Database [String]; -e Expectation value (E) [Real]; -G Cost to open a gap (zero invoices default behavior) [Integer]; -E Cost to extend a gap (zero invoices default behavior) [Integer]; -v Number of one-line descriptions (v) [Integer]; -b Number of alignments to show (b) [Integer]; -I Query File [File In]; - o BLAST report Output File [File Out] Optional. The "hits" to one or more database sequences by a queried sequence produced by BLASTN, FASTA, BLASTP or a similar algorithm, align and identify similar portions of sequences.

The hits are arranged in order of the degree of similarity and the length of sequence overlap. Hits to a database sequence generally represent an overlap over only a fraction of the sequence length of the queried sequence.

The BLASTN, FASTA, and BLASTP algorithms also produce "Expect" values for alignments. The Expect value (E) indicates the number of hits one can

"expect" to see over a certain number of contiguous sequences by chance when searching a database of a certain size. The Expect value is used as a significance threshold for determining whether the hit to a database, such as the preferred EMBL database, indicates true similarity. For example, an E value of 0.1 assigned to a polynucleotide hit is interpreted as meaning that in a database of the size of the EMBL database, one might expect to see 0.1 matches over the aligned portion of the sequence with a similar score simply by chance. By this criterion, the aligned and matched portions of the polynucleotide sequences then have a probability of 90% of being the same. For sequences having an E value of 0.01 or less over aligned and matched portions, the probability of finding a match by chance in the EMBL database is 1% or less using the BLASTN or FASTA algorithm.

According to one embodiment, "variant" polynucleotides and polypeptides, with reference to each of the polynucleotides and polypeptides of the present invention, preferably comprise sequences producing an E value of 0.01 or less using the BLASTN, FASTA, or BLASTP algorithms set at parameters described above when compared to the polynucleotide or polypeptide of the present invention. According to a preferred embodiment, a variant polynucleotide is a sequence having the same number or fewer nucleic acids than a polynucleotide of the present invention and having an E value of 0.01 or less using the BLASTN or FASTA algorithms set at parameters described above when analyzed against a polynucleotide of the present invention. Similarly, according to a preferred embodiment, a variant polypeptide is a sequence having the same number or fewer amino acids than a polypeptide of the present invention and having an E value of 0.01 or less using the BLASTP algorithm set at the parameters described above when analyzed against a polynucleotide of the present invention.

As noted above, the percentage identity is determined by aligning sequences using one of the BLASTN, FASTA, or BLASTP algorithms, set at the running parameters described above, and identifying the number of identical nucleic or amino acids over the aligned portions; dividing the number of identical nucleic or amino acids by the total number of nucleic or amino acids of the polynucleotide or polypeptide sequence of the present invention; and then multiplying by 100 to determine the percentage identity. For example, a polynucleotide of the present invention having 220 nucleic acids has a hit to a polynucleotide sequence in the EMBL database having 520 nucleic acids over a stretch of 23 nucleotides in the alignment produced by the BLASTN algorithm using the parameters described above. The 23 nucleotide hit includes 21 identical nucleotides, one gap and one different nucleotide. The percentage identity of the polynucleotide of the present invention to the hit in the EMBL library is thus 21/220 times 100, or 9.5%>. The polynucleotide sequence in the EMBL database is thus not a variant of a polynucleotide of the present invention.

In addition to having a specified percentage identity to an inventive polynucleotide or polypeptide sequence, variant polynucleotides and polypeptides preferably have additional structure and/or functional features in common with the inventive polynucleotide or polypeptide. Polypeptides having a specified degree of identity to a polypeptide of the present invention share a high degree of similarity in their primary structure and have substantially similar functional properties. In addition to sharing a high degree of similarity in their primary structure to polynucleotides of the present invention, polynucleotides having a specified degree of identity to, or capable of hybridizing to an inventive polynucleotide preferably have at least one of the following features: (i) they contain an open reading frame or partial open reading frame encoding a polypeptide having substantially the same functional properties as the polypeptide encoded by the inventive polynucleotide; or (ii) they contain identifiable domains in common. Alternatively, variant polynucleotides of the present invention hybridize to the polynucleotide sequences recited in SEQ ID NOS: 1-33, or complements, reverse sequences, or reverse complements of those sequences under stringent conditions. As used herein, "stringent conditions" refers to prewashing in a solution of 6X SSC, 0.2% SDS; hybridizing at 65°C, 6X SSC, 0.2% SDS overnight; followed by two washes of 30 minutes each in IX SSC, 0.1% SDS at

65° C and two washes of 30 minutes each in 0.2X SSC, 0.1% SDS at 65°C.

The present invention also encompasses polynucleotides that differ from the disclosed ^■ sequences but that, as a consequence of the discrepancy of the genetic code, encode a polypeptide having similar enzymatic activity as a polypeptide encoded by a polynucleotide of the present invention. Thus, polynucleotides comprising sequences that differ from the polynucleotide sequences recited in SEQ ID NOS: 1-33, or complements, reverse sequences, or reverse complements of those sequences as a result of conservative substitutions are encompassed within the present invention. Additionally, polynucleotides comprising sequences that differ from the inventive polynucleotide sequences or complements, reverse complements, or reverse sequences as a result of deletions and/or insertions totaling less than 10% of the total sequence length are also contemplated by and encompassed within the present invention. Similarly, polypeptides comprising sequences that differ from the inventive polypeptide sequences as a result of amino acid substitutions, insertions, and/or deletions totaling less than 15% of the total sequence length are contemplated by and encompassed within the present invention, provided the variant polypeptide has similar activity to the inventive polypeptide.

The polynucleotides of the present invention may be isolated from various libraries, or may be synthesized using techniques that are well Icnown in the art. The polynucleotides may be synthesized, for example, using automated oligonucleotide synthesizers (e.g., Beckman Oligo 1000M DNA Synthesizer) to obtain polynucleotide segments of up to 50 or more nucleic acids. A plurality of such polynucleotide segments may then be ligated using standard DNA manipulation techniques that are well known in the art of molecular biology. One conventional and exemplary polynucleotide synthesis technique involves synthesis of a single stranded polynucleotide segment having, for example, 80 nucleic acids, and hybridizing that segment to a synthesized complementary 85 nucleic acid segment to produce a 5-nucleotide overhang. The next segment may then be synthesized in a similar fashion, with a 5-nucleotide overhang on the opposite strand. The "sticky" ends ensure proper ligation when the two portions are hybridized. In this way, a complete polynucleotide of the present invention may be synthesized entirely in vitro.

Polynucleotides and polypeptides. of the present invention comprehend polynucleotides and polypeptides comprising at least a specified number of contiguous residues (x-mers) of any of the polynucleotides and polypeptides identified as SEQ ID NOS: 1-75 or their variants. According to preferred embodiments, the value of x is preferably at least 20, more preferably at least 40, more preferably yet at least 60, and most preferably at least 80. Thus, polynucleotides and polypeptides of the present invention include polynucleotides comprising a 20-mer, a 40-mer, a 60-mer, an 80-mer, a 100-mer, a 120-mer, a

150-mer, a 180-mer, a 220-mer a 250-mer, or a 300-mer, 400-mer, 500-mer or 600-mer of a polynucleotide or polypeptide identified as SEQ ID NOS: 1-75 or a variant of one of the polynucleotides or polypeptides identified as SEQ ID NOS: 1-75. Oligonucleotide probes and primers complementary to and/or corresponding to SEQ ID NOS: 1-33, and variants of those sequences, are also comprehended by the present invention. Such oligonucleotide probes and primers are substantially complementary to the polynucleotide of interest. An oligonucleotide probe or primer is described as "corresponding to" a polynucleotide of the present invention, including one of the sequences set out as

SEQ ID NOS: 1-33 or a variant, if the oligonucleotide probe or primer, or its complement, is contained within one of the sequences set out as SEQ ID NOS: 1- 33 or a variant of one of the specified sequences.

Two single stranded sequences are said to be substantially complementary when the nucleotides of one strand, optimally aligned and compared, with the appropriate nucleotide insertions and/or deletions, pair with at least 80%o, preferably at least 90% to 95%, and more preferably at least 9%% to 100%, of the nucleotides of the other strand. Alternatively, substantial complementarity exists when a first DNA strand will selectively hybridize to a second DNA strand under stringent hybridization conditions. Stringent hybridization conditions for determining complementarity include salt conditions of less than about 1 M, more usually less than about 500 mM and preferably less than about 200 mM. Hybridization temperatures can be as low as 5°C, but are generally greater than about 22°C, more preferably greater than about 30°C and most preferably greater than about 37°C. Longer DNA fragments may require higher hybridization temperatures for specific hybridization. Since the stringency of hybridization may be affected by other factors such as probe composition, presence of organic solvents and extent of base mismatching, the combination of parameters is more important than the absolute measure of any one alone. DNA-DNA hybridization studies may performed using either genomic DNA or DNA derived by preparing cDNA from the RNA present in a sample to be tested.

In addition to DNA-DNA hybridization, DNA-RNA or RNA-RNA hybridization assays are also possible. In the first case, the mRNA from expressed genes would then be detected instead of genomic DNA or cDNA derived from mRNA of the sample. In the second case, RNA probes could be used. In addition, artificial analogs of DNA hybridizing specifically to target sequences could also be used.

In specific embodiments, the oligonucleotide probes and/or primers comprise at least about 6 contiguous residues, more preferably at least about 10 contiguous residues, and most preferably at least about 20 contiguous residues complementary to a polynucleotide sequence of the present invention. Probes and primers of the present invention may be from about 8 to 100 base pairs in length or, preferably from about 10 to 50 base pairs in length or, more preferably from about 15 to 40 base pairs in length. The primers and probes may be readily selected using procedures well Icnown in the art, taking into account DNA-DNA hybridization stringencies, annealing and melting temperatures, potential for formation of loops and other factors, which are well Icnown in the art. Tools and software suitable for designing probes, and especially suitable for designing PCR primers, are available on the Internet, for example, at URL http ://www.horizonpress . com/per/. In addition, a software program suitable for designing probes, and especially for designing PCR primers, is available from

Premier Biosoft International, 3786 Corina Way, Palo Alto, CA 94303-4504. Preferred techniques for designing PCR primers are also disclosed in Dieffenbach and Dyksler, PCR primer: a laboratory manual, CSHL Press: Cold Spring Harbor, NY, 1995. A plurality of oligonucleotide probes or primers corresponding to a polynucleotide of the present invention may be provided in a kit form. Such kits generally comprise multiple DNA or oligonucleotide probes, each probe being specific for a polynucleotide sequence. Kits of the present invention may comprise one or more probes or primers corresponding to a polynucleotide of the present invention, including a polynucleotide sequence identified in SEQ ID

NOS: 1-33.

In one embodiment useful for high-throughput assays, the oligonucleotide probe kits of the present invention comprise multiple probes in an array format, wherein each probe is immobilized in a predefined, spatially addressable location on the surface of a solid substrate. Array formats which may be usefully employed in the present invention are disclosed, for example, in U.S. Patents No. 5,412,087,

5,545,531, and PCT Publication No. WO 95/00530, the disclosures of which are hereby incorporated by reference.

Oligonucleotide probes for use in the present invention may be constructed synthetically prior to immobilization on an array, using techniques well known in the art (See, for example, Gait, ed., Oligonucleotide synthesis a practical approach, IRL Press: Oxford, England, 1984). Automated equipment for the synthesis of oligonucleotides is available commercially from such companies as Perkin Elmer/Applied Biosystems Division (Foster City, CA) and may be operated according to the manufacturer's instructions. Alternatively, the probes may be constructed directly on the surface of the array using techniques taught, for example, in PCT Publication No. WO 95/00530.

The solid substrate and the surface thereof preferably form a rigid support and are generally formed from the same material. Examples of materials from which the solid substrate may be constructed include polymers, plastics, resins, membranes, polysaccharides, silica or silica-based materials, carbon, metals and inorganic glasses. Synthetically prepared probes may be immobilized on the surface of the solid substrate using techniques well Icnown in the art, such as those disclosed in U.S. Patent No. 5,412,087.

In one such technique, compounds having protected functional groups, such as thiols protected with photochemically removable protecting groups, are attached to the surface of the substrate. Selected regions of the surface are then irradiated with a light source, preferably a laser, to provide reactive thiol groups. This irradiation step is generally performed using a mask having apertures at predefined locations using photolithographic techniques well Icnown in the art of semiconductors. The reactive thiol groups are then incubated with the oligonucleotide probe to be immobilized. The precise conditions for incubation, such as temperature, time and pH, depend on the specific probe and can be easily determined by one of skill in the art. The surface of the substrate is washed free of unbound probe and the irradiation step is repeated using a second mask having a different pattern of apertures. The surface is subsequently incubated with a second, different, probe. Each oligonucleotide probe is typically immobilized in a discrete area of less than about 1 mm². Preferably each discrete area is less than about 10,000 mm², more preferably less than about 100 mm². In this manner, a multitude of oligonucleotide probes may be immobilized at predefined locations on the array. The resulting array may be employed to screen for differences in organisms or samples or products containing genetic material as follows. Genomic or cDNA libraries are prepared using techniques well known in the art. The resulting target DNA is then labeled with a suitable marker, such as a radiolabel, chromophore, fluorophore or chemiluminescent agent, using protocols well known for those skilled in the art. A solution of the labeled target DNA is contacted with the surface of the array and incubated for a suitable period of time.

The surface of the array is then washed free of unbound target DNA and the probes to which the target DNA hybridized are determined by identifying those regions of the array to which the markers are attached. When the marker is a radiolabel, such as ³²P, autoradiography is employed as the detection method. In one embodiment, the marker is a fluorophore, such as fiuorescein, and the location of bound target DNA is determined by means of fluorescence spectroscopy. Automated equipment for use in fluorescence scanning of oligonucleotide probe arrays is available from Affymetrix, Inc. (Santa Clara, CA) and may be operated according to the manufacturer's instructions. Such equipment may be employed to determine the intensity of fluorescence at each predefined location on the array, thereby providing a measure of the amount of target DNA bound at each location. Such an assay would be able to indicate not only the absence and presence of the marker probe in the target, but also the quantitative amount as well.

The significance of such a high-throughput screening system is apparent for applications such as microbial selection and quality control operations in which there is a need to identify large numbers of samples or products for unwanted materials, to identify microbes or samples or products containing microbial material for quarantine purposes, etc., or to ascertain the true origin of samples or products containing microbes. Screening for the presence or absence of polynucleotides of the present invention used as identifiers for tagging microbes and microbial products can be valuable for later detecting the genetic composition of food, fermentation and industrial microbes or microbes in human or animal digestive system after consumption of probiotics, etc. In this manner, oligonucleotide probe kits of the present invention may be employed to examine the presence/absence (or relative amounts in case of mixtures) of polynucleotides in different samples or products containing different materials rapidly and in a cost-effective manner. Examples of microbial species which may be examined using the present invention, include lactic acid bacteria, such as Lactobacillus rhamnosus, and other microbial species.

Another aspect of the present invention involves collections of a plurality of polynucleotides of the present invention. A collection of a plurality of the polynucleotides of the present invention, particularly the polynucleotides identified as SEQ ID NOS: 1-33, may be recorded and/or stored on a storage medium and subsequently accessed for purposes of analysis, comparison, etc.

Suitable storage media include magnetic media such as magnetic diskettes, magnetic tapes, CD-ROM storage media, optical storage media, and the like. Suitable storage media and methods for recording and storing information, as well as accessing information such as polynucleotide sequences recorded on such media, are well Icnown in the art. The polynucleotide information stored on the storage medium is preferably computer-readable and may be used for analysis and comparison of the polynucleotide information.

Another aspect of the present invention thus involves storage medium on which are recorded a collection of the polynucleotides of the present invention, particularly a collection of the polynucleotides identified as SEQ ID NOS: 1-33.

According to one embodiment, the storage medium includes a collection of at least 20, preferably at least 50, more preferably at least 100, and most preferably at least 200 of the polynucleotides of the present invention, preferably the polynucleotides identified as SEQ ID NOS: 1-33, including variants of those polynucleotides. Another aspect of the present invention involves a combination of polynucleotides, the combination containing at least 5, preferably at least 10, more preferably at least 20, and most preferably at least 50 different polynucleotides of the present invention, including polynucleotides selected from SEQ ID NOS: 1- 33, and variants of these polynucleotides. In another aspect, the present invention provides genetic constructs comprising, in the 5 '-3' direction, a gene promoter sequence and an open reading frame coding for at least a functional portion of a polypeptide encoded by a polynucleotide of the present invention. In certain embodiments, the genetic constructs of the present invention also comprise a gene termination sequence. The open reading frame may be oriented in either a sense or antisense direction.

Genetic constructs comprising a non-coding region of a gene coding for a polypeptide encoded by the above polynucleotides or a nucleotide sequence complementary to a non-coding region, together with a gene promoter sequence, are also provided. A terminator sequence may form part of this construct. Preferably, the gene promoter and termination sequences are functional in a host organism. More preferably, the gene promoter and termination sequences are common to those of the polynucleotide being introduced. The genetic construct may further include a marker for the identification of transformed cells.

Techniques for operatively linking the components of the genetic constructs are well known in the art and include the use of synthetic linkers containing one or more restriction endonuclease sites as described, for example, by Sambrook et al, in Molecular cloning: a laboratory manual, Cold Spring Harbor Laboratories Press: Cold Spring Harbor, NY, 1989. The genetic constructs of the present invention may be linked to a vector having at least one replication system, for example, E. coli, whereby after each manipulation, the resulting construct can be cloned and sequenced and the correctness of the manipulation determined.

Transgenic microbial cells comprising the genetic constructs of the present invention are also provided by the present invention, together with microbes comprising such transgenic cells, products and progeny of such microbes, and materials including such microbes. Techniques for stably incorporating genetic constructs into the genome of target microbes, such as Lactobacillus species, Lactococcus lactis or E. coli, axe well known in the art of bacterial transformation and are exemplified by the transformation of E. coli for sequencing in Example 1, as well as the transformations described in numerous of the examples provided below.

Transgenic, non-microbial, cells comprising the genetic constructs of the present invention are also provided, together with organisms comprising such transgenic cells, and products and progeny of such organisms. Genetic constructs of the present invention may be stably incorporated into the genomes of non- microbial target organisms, such as fungi, using techniques well known in the art. In preferred embodiments, the genetic constructs of the present invention are employed to transform microbes used in the production of food products, ingredients, processing aids, additives or supplements and for the production of microbial products for pharmaceutical uses, particularly for modulating immune system function and immunological effects; and in the production of chemoprotectants providing beneficial effects, probiotics and health supplements. The inventive genetic constructs may also be employed to transform bacteria that are used to produce enzymes or substances such as polysaccharides, flavor compounds, and bioactive substances, and to enhance resistance to industrial processes such as drying and to adverse stimuli in the human digestive system.

The genes involved in antibiotic production, and phage uptake and resistance in Lactobacillus rhamnosus are considered to be especially useful. The target microbe to be used for transformation with one or more polynucleotides or genetic constructs of the present invention is preferably selected from the group consisting of bacterial genera Lactococcus, Lactobacillus, Streptococcus,

Oenococcus, Lactosphaera, Trichococcus, Pediococcus and others potentially useful in various fermentation industries selected, most preferably, from the group consisting of Lactobacillus species in the following list: Lactobacillus acetotolerans, Lactobacillus acidophilus, Lactobacillus agilis, Lactobacillus alimentarius, Lactobacillus amylolyticus, Lactobacillus amylophilus, Lactobacillus amylovorus, Lactobacillus animalis, Lactobacillus arizonae,

Lactobacillus aviarius, Lactobacillus bavaricus, Lactobacillus bifermentans, Lactobacillus brevis, Lactobacillus buchneri, Lactobacillus bulgaricus, Lactobacillus casei, Lactobacillus collinoides, Lactobacillus coryniformis, Lactobacillus crispatus, Lactobacillus curvatus, Lactobacillus delbrueckii, Lactobacillus delbrueckii subsp. bulgaricus, Lactobacillus delbrueckii subsp. lactis, Lactobacillus farciminis, Lactobacillus fermentum, Lactobacillus fructivorans, Lactobacillus gallinarum, Lactobacillus gasseri, Lactobacillus graminis, Lactobacillus hamsteri, Lactobacillus helveticus, Lactobacillus helveticus subsp. jugurti, Lactobacillus hetero, Lactobacillus hilgardii, Lactobacillus homohiochii, Lactobacillus japonicus, Lactobacillus johnsonii,

Lactobacillus kefiri, Lactobacillus lactis, Lactobacillus leichmannii, Lactobacillus lindneri, Lactobacillus mali, Lactobacillus maltaromicus, Lactobacillus manihotivorans, Lactobacillus mucosae, Lactobacillus murinus, Lactobacillus oris, Lactobacillus panis, Lactobacillus paracasei, Lactobacillus paracasei subsp. pseudoplantarum, Lactobacillus paraplantarum, Lactobacillus pentosus,

Lactobacillus plantarum, Lactobacillus pontis, Lactobacillus reuteri, Lactobacillus rhamnosus, Lactobacillus ruminis, Lactobacillus sake, Lactobacillus salivarius, Lactobacillus salivarius subsp. salicinius, Lactobacillus salivarius subsp. salivarius, Lactobacillus sanfranciscensis, Lactobacillus sharpeae, Lactobacillus thermophilus, Lactobacillus vaginalis, Lactobacillus vermiforme, Lactobacillus zeae.

In yet a further aspect, the present invention provides methods for modifying the concentration, composition and/or activity of a polypeptide in a host organism, such as a microbe, comprising stably incorporating a genetic construct of the present invention into the genome of the host organism by transforming the host organism with such a genetic construct. The genetic constructs of the present invention may be used to transfonn a variety of organisms. Organisms which may be transformed with the inventive constructs include plants, such as monocotyledonous angiosperms (e.g., grasses, corn, grains, oat, wheat and barley); dicotyledonous angiosperms (e.g., Arabidopsis, tobacco, legumes, alfalfa, oaks, eucalyptus, maple); gymnosperms, (e.g., Scots pine

(Aronen, Finnish Forest Res. Papers, Vol. 595, 1996); white spruce (Ellis et al, Biotechnology 11:84-89, 1993); and larch (Huang et al, In Vitro Cell 27:201-207, 1991); and any kind of plant amenable to genetic engineering.

Thus, in yet another aspect, transgenic plant cells comprising the genetic constructs of the present invention are provided, together with plants comprising such transgenic cells, and fruits, seeds, products and progeny of such plants. Techniques for stably incorporating genetic constructs into the genome of target organisms, such as plants, are well known in the art and include Agrobacterium tumefaciens mediated introduction, electroporation, protoplast fusion, injection into reproductive organs, injection into immature embryos, high velocity projectile introduction and the like. The choice of technique will depend upon the target plant to be transformed. For example, dicotyledonous plants and certain monocots and gymnosperms may be transformed by Agrobacterium Ti plasmid technology, as described, for example by Bevan, Nucleic Acids Res. 12:8711- 8721, 1984. Targets for the introduction of the genetic constructs include tissues, such as leaf tissue, disseminated cells, protoplasts, seeds, embryos, meristematic regions; cotyledons, hypocotyls, and the like.

Once the cells are transformed, cells having the genetic construct incorporated in their genome are selected. Transgenic cells may then be cultured in an appropriate medium, using techniques well known in the art. In the case of protoplasts, the cell wall is allowed to reform under appropriate osmotic conditions. In the case of seeds or embryos, an appropriate germination or callus initiation medium is employed. For explants, an appropriate regeneration medium is used. Regeneration of plants is well established for many species. For a review of regeneration of forest trees, see Dunstan et al, "Somatic embryogenesis in woody plants," in Thorpe, T.A., ed., In vitro embryogenesis of plants, (Current Plant Science and Biotechnology in Agriculture), 20(12):471-540, 1995. Specific protocols for the regeneration of spruce are discussed by Roberts et al. ("Somatic embryogenesis of Spruce," in Redenbaugh K., ed., Synseed: applications of synthetic seed to crop improvement, CRC Press: Ch.23:427-449, 1993). The resulting transformed plants may be reproduced sexually or asexually, using methods well known in the art, to give successive generations of transgenic plants and practically unlimited amounts of tagged plant-derived products.

Polynucleotides of the present invention may also be used to specifically suppress gene expression by methods such as RNA interference (RNAi), which may also include cosuppression and quelling. This and other techniques of gene suppression are well Icnown in the art. A review of this technique is found in Science 288:1370-1372, 2000. Traditional methods of gene suppression, employing antisense RNA or DNA, operate by binding to the reverse sequence of a gene of interest such that binding interferes with subsequent cellular processes and thereby blocks synthesis of the corresponding protein. RNAi also operates on a post-transcriptional level and is sequence specific, but suppresses gene expression far more efficiently

Studies have demonstrated that one or more ribonucleases specifically bind to and cleave double-stranded RNA into short fragments. The ribonuclease(s) remains associated with these fragments, which in turn specifically bind to complementary mRNA, i.e. specifically bind to the transcribed mRNA strand for the gene of interest. The mRNA for the gene is also degraded by the ribonuclease(s) into short fragments, thereby obviating translation and expression of the gene. Additionally, an RNA polymerase may act to facilitate the synthesis of numerous copies of the short fragments, which exponentially increases the efficiency of the system. A unique feature of this gene suppression pathway is that silencing is not limited to the cells where it is initiated. The gene-silencing effects may be disseminated to other parts of an organism and even transmitted through the germ line to several generations. Specifically, polynucleotides of the present invention are useful for generating gene constructs for silencing specific genes. Polynucleotides of the present invention may be used to generate genetic constructs that encode a single self-complementary RNA sequence specific for one or more genes of interest. Genetic constructs and/or gene-specific self-complementary RNA sequences may be delivered by any conventional method known in the art. Within genetic constructs, sense and antisense sequences flank an intron sequence arranged in proper splicing orientation making use of donor and acceptor splicing sites. Alternative methods may employ spacer sequences of various lengths rather than discrete intron sequences to create an operable and efficient construct. During post-transcriptional processing of the gene construct product, intron sequences are spliced-out, allowing sense and antisense sequences, as well as splice junction sequences, to bind forming double-stranded RNA. Select ribonucleases bind to and cleave the double-stranded RNA, thereby initiating the cascade of events leading to degradation of specific mRNA gene sequences, and silencing specific genes. Alternatively, rather than using a gene construct to express the self- complementary RNA sequences, the gene-specific double-stranded RNA segments are delivered to one or more targeted areas to be internalized into the cell cytoplasm to exert a gene silencing effect.

Using this cellular pathway of gene suppression, gene function may be studied and high-throughput screening of sequences may be employed to discover sequences affecting gene expression. Additionally, genetically modified microbes and higher order organisms may be generated.

The following examples are offered by way of illustration and not by way of limitation.

Example 1 Isolation and Characterization of DNA Sequences from Lactobacillus rhamnosus strain HNOOl

Lactobacillus rhamnosus strain HNOOl DNA libraries were constructed and screened as follows.

DNA was prepared in large scale by cultivating the bacteria in 2 x 100 ml cultures with 100 ml MRS broth (Difco Laboratories, Detroit MI) and 1 ml

Lactobacillus glycerol stock as inoculum, placed into 500 ml culture flasks and incubated at 37 °C for approx. 16 hours with shaking (220 rpm).

The cultures were centrifuged at 3500 rpm for 10 min to pellet the cells. The supernatant was removed and the cell pellet resuspended in 40 ml fresh MRS broth and transferred to clean 500 ml culture flasks. Fresh MRS broth (60 ml) was added to bring the volume back to 100 ml and flasks were incubated for a further 2 hrs at 37°C with shaking (220 rpm). The cells were pelleted by centrifugation (3500 rpm for 10 min) and supernatant removed. Cell pellets were washed twice in 20 ml buffer A (50 mM NaCl, 30 mM Tris pH 8.0, 0.5 mM EDTA).

Cells were resuspended in 2.5 ml buffer B (25%) sucrose (w/v), 50 mM Tris pH 8.0, 1 mM EDTA, 20 mg/ml lysozyme, 20 μg/ml mutanolysin) and incubated at 37 °C for 45 min. Equal volumes of EDTA (0.25 M) was added to each tube and allowed to incubate at room temperature for 5 min. 20% SDS (1 ml) solution was added, mixed and incubated at 65 °C for 90 min. 50 μl Proteinase K

(Gibco BRL, Gaithersburg, MD) from a stock solution of 20 mg/ml was added and tubes incubated at 65 °C for 15 min.

DNA was extracted with equal volumes of phenol:chloroform:isoamylalcohol (25:24:1). Tubes were centrifuged at 3500 rpm for 40 min. The aqueous phase was removed to clean sterile Oak Ridge centrifuge tubes (30 ml). Crude DNA was precipitated with an equal volume of i cold isopropanol and incubated at -20 °C overnight.

After resuspension in 500 μl TE buffer, DNase-free RNase was added to a final concentraion of 100 μg/ml and incubated at 37 °C for 30 min. The incubation was extended for a further 30 min after adding 100 μl Proteinase K from a stock solution of 20 mg/ml. DNA was precipitated with ethanol after a phenol:chloroform:isoamylalcohol (25:24:1) and a chloroform:isoamylalcohol (24:1) extraction and dissolved in 250 μl TE buffer.

DNA was digested with SauiAl at a concentration of 0.004 U/μg in a total volume of 1480 μl, with 996 μl DNA, 138.75 μl 10X REACT 4 buffer and 252.75 μl H₂O. Following incubation for 1 hour at 37 °C, DNA was divided into two tubes. 31 μl 0.5 M EDTA was added to stop the digestion and 17 μl samples were taken for agarose gel analysis. Samples were put into 15 ml Falcon tubes and diluted to 3 ml for loading onto sucrose gradient tubes.

Sucrose gradient size fractionation was conducted as follows. 100 ml of 50% sucrose (w/v) was made in TEN buffer (1M NaCl, 20 mM Tris pH 8.0, 5 mM EDTA) and sterile filtered. Dilutions of 5, 10, 15, 20, 25, 30, 35 and 40% sucrose were prepared and overlaid carefully in Beclcman Polyallomer tubes, and kept overnight at 4°C. TEN buffer (4 ml) was loaded onto the gradient, with 3 ml of DNA solution on top. The gradients were centrifuged at 26K for 18 hours at 4°C in a Centricon T-2060 centrifuge using a Kontron TST 28-38 rotor. After deceleration without braking (approx. 1 hour), the gradients were removed and fractions collected using an auto Densi-Flow (Haake-Buchler Instruments). Agarose gel was used to analyse the fractions. The best two pairs of fractions, were pooled and diluted to contain less than 10%o sucrose. TEN buffer (4 ml) was added and DNA precipitated with 2 volumes of 100%) ice cold ethanol and an overnight incubation at -20°C.

DNA pellets were resuspended in 300 μl TE buffer and re-precipitated for approx. 6 hours at -20 °C after adding 1/10 volume 3 M NaOAC pH 5.2 and 2 volumes of ethanol. DNA was pelleted at top speed in a microcentrifuge for 15 min, washed with 70% ethanol and pelleted again, dried and resuspended in 10 μl

TE buffer.

DNA was ligated into dephosphorylated _5αrnHI-digested pBluescript SK II⁺ and dephosphorylated _5αmHI-digested lambda ZAP Express using standard protocols. Packaging of the DNA was done using Gigapack III Gold packaging extract (Stratagene, La Jolla, CA) following the manufacturer's protocols.

Packaged libraries were stored at 4 °C.

Mass excision from the primary packaged phage library was done using XL 1 -Blue MRF' cells and ExAssist Helper Phage (Stratagene). The excised phagemids were diluted with NZY broth (Gibco BRL, Gaithersburg, MD) and plated out onto LB-kanamycin agar plates containing 5-bromo-4-chloro-3-indolyl- β-D-galactoside (X-gal) and isopropylthio-beta-galactoside (IPTG). After incubation, single colonies were picked for PCR size determination before the most suitable libraries were selected for sequencing.

Of the colonies picked for DNA minipreps and subsequent sequencing, the large majority contained an insert suitable for sequencing. Positive colonies were cultured in LB broth with kanamycin or ampicillin depending on the vector used, and DNA was purified by means of rapid alkaline lysis minipreps (solutions: Qiagen, Venlo, The Netherlands; clearing plates, Millipore, Bedford, MA). Agarose gels at 1%> were used to screen sequencing templates for chromosomal contamination and concentration. Dye terminator sequencing reactions were prepared using a Biomek 2000 robot (Beclcman Coulter, Inc., Fullerton, CA) and Hydra 96 (Robbins Scientific, Sunnyvale, CA) for liquid handling. DNA amplification was done in a 9700 PCR machine (Perlcin Elmer/Applied Biosystems, Foster City, CA) according to the manufacturer's protocol. The sequence of the genomic DNA fragments were determined using a

Perlcin Elmer/ Applied Biosystems Division Prism 377 sequencer. The DNA clones were sequenced from the 5' and/ or 3' end, and are identified as SEQ ID NOS: 1-33. /12506

52

This example not only shows how the sequences were obtained, but also that a bacterium (E. coli) can be stably transformed with any desired DNA fragment of the present invention for permanent marking for stable inheritance.

The determined DNA sequences were compared to and aligned with known sequences in the public databases. Specifically, the polynucleotides identified in SΕQ ID NO: 1-33 were compared to polynucleotides in the ΕMBL database as of the end of July 2001, using BLASTN algorithm Version 2.0.11 [Jan-20-2000], set to the following running parameters: Unix running command: blastall -p blastn -d embldb -e 10 -G 0 -Ε 0 -r 1 -v 30 -b 30 -i queryseq -o results. Multiple alignments of redundant sequences were used to build up reliable consensus sequences. Based on similarity to Icnown sequences, the isolated polynucleotides of the present invention identified as SΕQ ID NOS: 1-33 were identified as encoding polypeptides.

Numerous of the sequences provided in SΕQ ID NO: 1-33 were found to be "full-length" and to contain open reading frames (ORFs). These full-length sequences, the location of ORFs (by nucleotide position) contained within these sequences, and the corresponding amino acid sequences are provided in Table 2 below.

TABLE 2

The polynucleotide and polypeptide sequences of SEQ ID NOS: 1-33 and 42-75 were compared to sequences in the EMBL and SwissProt databases using the BLAST computer algorithms version 2.0.11 [Jan-20-2000]. Comparisons of polynucleotide sequences provided in SEQ ID NOS: 1-33 to sequences in the EMBL database were made as of August 2001. Comparisons of amino acid sequences provided in SEQ ID NOS: 42-75 to sequences in the SwissProt database were made as of August 2001. Analysis of six-frame translations of the polynucleotides of SEQ ID NOS: 1-33 were also compared to and aligned with the six-frame translations of polynucleotides in the SwissProt database using the BLASTX program.

BLASTN Polynucleotide Analysis The polynucleotide sequences of SEQ ID NOS: 1-3, 5-23 and 25-33 were determined to have less than 50% identity, determined as described above, to sequences in the EMBL database using the computer algorithm BLASTN, as described above. The polynucleotide sequence of SEQ ID NO: 24 was determined to have less than 90%) identity, determined as described above, to sequences in the EMBL database using BLASTN, as described above. The polynucleotide sequence of SEQ ID NO: 4 was determined to have less than 98%> identity, determined as described above, to sequences in the EMBL database using BLASTN, as described above.

BLASTP Amino Acid Analysis

The amino acid sequences of SEQ ID NOS: 43, 45-47, 51-53, 58, 60, 61, 63, 67, 68, 70, 71, 73 and 74 were determined to have less than 50% identity, determined as described above, to sequences in the SwissProt database using the BLASTP computer algorithm as described above. The amino acid sequences of SEQ ID NOS: 48-50, 55-56, 62, 64, 66, 69, 72 and 75 were determined to have less than 75% identity, determined as described above, to sequences in the .SwissProt database using the BLASTP computer algorithm as described above.

The amino acid sequences of SEQ ID NOS: 57 and 65 were determined to have less than 90%) identity, determined as described above, to sequences in the SwissProt database using the computer algorithm BLASTP, as described above. The amino acid sequence of SEQ ID NO: 54 and 59 was determined to have less than 98%) identity, determined as described above, to sequences in the SwissProt database using the computer algorithm BLASTP, as described above.

BLASTX Analysis

The six-frame translations of the polynucleotide sequences of SEQ ID NOS: 1-33 were compared to and aligned with six-frame translations of polynucleotides in the EMBL database using the BLASTX program version 2.0.11 [Jan-20-2000] set to the following running parameters: Unix running command: blastall -p blastn -d embldb -e 10 -G 0 -E 0 -v 30 -b 30 -i queryseq -o results. The translations of the polynucleotides of SEQ ID NOS: 1, 3, 5-9, 11-19, 21 and 25-32 were determined to have less than 50% identity, determined as described above, to translations of polynucleotides in the EMBL database using the computer algorithm BLASTX. The translations of the polynucleotides of SEQ ID NOS: 2, 4, 10, 20, 22, 23 and 33 were determined to have less than 75%o identity, determined as described above, to translations of polynucleotides in the EMBL database using the computer algorithm BLASTX. The translations of the polynucleotide sequence of

SEQ ID NO: 24 was determined to have less than 90% identity, determined as described above, to translations of polynucleotides in the EMBL database using the computer algorithm BLASTX.

Example 2

Isolation and Characterization of the Peptidase pepO from L. rhamnosus The full-length gene sequence of a peptidase pepO from L. rhamnosus strain HNOOl (given in SEQ ID NO: 1 and shown in Fig. 80) was isolated essentially as described in Example 1. Primers were designed to this sequence and employed to amplify pepO from L. rhamnosus HNOOl using standard PCR methodology. PepO was cloned in the vector pTRKH2 (obtained from Dr Todd E-laenhammer, North Carolina State University, North Carolina, USA) and transformed into E. coli. Competent cells of L. rhamnosus HNOOl were transformed with the pTRKH2+pepO construct to overexpress the gene in strain HNOOl. The amino acid sequence of the expressed protein is provided in SEQ ID

NO: 42 and shown in Fig. 81.

Cell extracts of the FINOOl strain constructs with enhanced levels of the peptidase enzyme showed enhanced enzyme activity on the casein peptide, αsi- casein(l-17). Specifically, αs_-casein(l-17) was incubated with non-transformed strain HNOOl (referred to as DR20 WT) and strain HNOOl transformed with the pepO construct described above (referred to as DR20 PepO:l and DR20 PepO:4) HPLC separation of the resulting peptide products was performed using a Vydac reverse phase C18 column, 4.6 mm x 250 mm. The solvent system was solvent A, 0.1% TFA in water, solvent B, 0.08%> TFA in acetonitrile and the gradient employed was 15-40%) solvent B over 20 minutes. A major peak was observed at 11 minutes, together with other non-identified minor peaks corresponding to hydrolysis products of the original substrate.

With non-transformed HNOOl (DR 20 WT), the major peak of unhydrolysed α_sι-casein(l-17) had a height of approximately 250 mAU. With each of the two transformed strains of HNOOl (DR 20 PepO:l and DR 20 PepO:4) the major peak of unhydrolysed α_sι-casein(l-17) had a height of approximately

150 mAU, demonstrating that HNOOl transformed with the pepO construct has enhanced peptidase activity compared to non-transformed HNOOl .

The pepO peptidase from strain HNOOl was not active on bradylcinin, a standard substrate for measuring pepO activity (Pritchard et al, Microbiol. 140:923-30,1994) and thus has a specificity that is significantly different to the homologous enzyme from Lactococcus.

This enzyme may be used to develop new characteristics in food products, supplements and additives, including cheese and hydrolyzed milk protein products. This enzymes may also be used to develop non-food products. The attributes that may be conferred by this enzyme include:

- flavor and aroma enhancement;

- removal of bitter peptides and undesirable flavors;

- nutritional enhancement; - enhanced texture and functionality;

- production of bioactive peptides; and

~ removal of allergenic peptides or proteins.

These attributes may be produced in food, such as dairy, systems (including milk protein hydrolysates and cheese) by directed activity of the enzyme, either in a bacterial strain (including strain HNOOl, or starter cultures) or as an enzyme preparation.

Example 3 Isolation and Characterisation of an Esterase from L. rhamnosus HNOOl

The full-length polynucleotide sequence of an esterase gene, given in SEQ ID NO: 3, was used to amplify the AAT esterase gene from L. rhamnosus HNOOl using standard PCR methodology. Fig. 1 shows the nucleotide sequence containing L. rhamnosus strain HNOOl esterase gene AAT, with the ATG initiation and translation stop codons shown boxed.

The AA7 esterase gene sequence was then cloned into the pUniBlunt/V5- HisTopo vector (Invitrogen, Auckland, NZ) and transformed into the E. coli strain PIR1 OneShot competent cells (Invitrogen). The amino acid sequence is given in

SEQ ID NO: 44. To construct an expression plasmid, the pUniBlunt/V5-HisTopo vector construct was recombined with the pBad/Thio-E vector (Invitrogen) and transformed into the E. coli strain TOP 10 competent cells ^"(Invitrogen) according to the manufacturer's instructions. The gene product was therefore cloned as a fusion protein tagged with a His-patch polypeptide and thioredoxin protein. The esterase fusion protein was expressed and purified using a Ni-NTA column (Qiagen, Auckland, NZ) according to the manufacturer's instructions and protein expression checked by SDS-PAGE. The amino acid sequence of the esterase AA7 polypeptide is shown in Fig. 2. Esterase activity was assessed using the 3αrα-nitrophenyl butyrate assay as described in Lee and Lee, Biotech. Appl. Biochem. 11 :552-563, 1989, with some modifications. Briefly, esterase activity was measured spectrophotometrically using ^-nitrophenyl butyrate (Sigma Chemical Co., St Louis, MO) as substrate. Substrate was prepared by sonicating 1 ml of 50 mM methanolic p-nitrophenyl butyrate in 18 ml 50 mM sodium phosphate buffer (pH 7.5). Ahquots of 1.9 ml were placed in cuvettes, allowed to stabilize at 30 °C, and between 5 and 20 μl of purified AA7 esterase added. Changes in optical density (OD) 410 nm were determined. Based on the results, enzyme activity was calculated, with one unit (U) of enzyme defined as the amount required to hydrolyze 1 μmol substrate per minute.

Esterase activity of the AA7 fusion protein was compared to the activity of a known esterase enzyme from Streptococcus thermophilus (STl, as described in Liu et al, Int. Dairy J. 11 :27-35, 2001), a non-esterase HNOOl enzyme also expressed as a His-patch/Thioredoxin fusion protein and buffer-only. The results are shown in Fig. 3 and the enzyme activities are given in Table 1. Fig. 3 demonstrates the production of ethyl butyrate from para- nitrophenyl butyrate substrate as measured by change in OD at 410 nm. As shown in Fig. 3, while buffer only (♦) and the HNOOl non-esterase fusion protein (•) showed minimal esterase activity, the STl esterase from Streptococcus thermophilus (A) and the AA7 esterase fusion protein (■) showed strong activity. Thus, the AA7 esterase fusion protein showed strong esterase activity, compared to the positive control, and negligible amounts of esterase was produced by the two negative controls (buffer-only and the non-esterase fusion protein).

Table 1. Esterase activity of the AA7 fusion protein

The esterase activity exhibited by the AA7 fusion protein was not due to background hydrolysis of the substrate as the buffer-only control showed little or no activity. The specific enzyme activity of the His-patch/Thio/AA7 fusion protein was 1.42 μmol/min/mg protein compared with 0.03 μmol/min/mg for the non-esterase fusion protein showing an almost 50-fold difference in esterase activity. Therefore, AA7 esterase activity was due not due to the His- patch/Thioredoxin fusion protein tag. The dose-response of the AA7 fusion protein was determined by comparing the esterase activity in a series of three two-fold dilutions of the purified enzyme. Results are shown in Fig. 4 and the rate of change in optical density at 410 nm and enzyme activities given in Table 2. As shown in Fig. 4, while buffer-only (•) showed no esterase activity, increasing amounts of His- patch/Thio/AA7 fusion protein; 5 μl (♦), 10 μl (A) and 20 μl (■) purified protein showed increasing rates of substrate hydrolysis. The increase in substrate hydrolysis was proportional to amount of AA7 fusion protein added.

Table 2. Esterase activity for increasing amounts of AA7 fusion protein

Results indicated the rate of change in OD at 410 nm was proportional to the amount of enzyme added, whilst enzyme activity remained relatively constant. Therefore, esterase activity was dependent on the amount of esterase AA7 fusion protein present.

The effect of the serine esterase inhibitor PMSF was determined using the j7-nitrophenyl butyrate assay. Esterase activity of the AA7 fiision protein was assessed in the presence and absence of 10 mM PMSF. Results are shown in Fig. 5, and the rate of change in OD at 410 nm and enzyme activities given in Table 3. Results in Fig. 5 and Table 3 indicate that the PMSF inhibitor caused a 17.9 % reduction in the esterase activity of the AA7 fusion protein. Therefore, AA7 esterase activity was inhibited by the serine esterase-specific inhibitor PMSF.

Table 3. Effect of PMSF inhibitor on AA7 fusion protein esterase activity

The enzymatic breakdown of milk fat plays an essential role in the development of flavor in cheese. Esterases and lipases catalyze the lipolysis of milk fat in dairy products such that the triglycerides are hydrolyzed to free fatty acids and glycerol or mono- and diglycerides. Although exogenous esterases and lipases of mammalian and fungal origins are often used to encourage extensive lipolysis in cheeses, esterases and lipases from cheese microorganisms may also contribute to lipolysis (reviewed in Fox and Wallace, Adv. Appl. Microbiol. 45:17-85, 1997 and McSweeney and Wallace, Lait 80:293-324, 2000). Therefore, applications of the HNOOl esterase AA7 include:

• enhanced flavor and aroma

• removal of off-flavors

• altered levels of butyric acid

• altered metabolic characteristics

Example 4 Isolation and Characterisation of Autoaggregation protein A G5 from L. rhamnosus HNOOl

The full-length polynucleotide sequence of an autoaggregation protein from L. rhamnosus strain HNOOl, given in SEQ ID NO: 10, was used to amplify the AG5 autoaggregation gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The full-length polynucleotide sequence containing L. rhamnosus strain HNOOl autoaggregation gene AG5, showing ATG initiation and translation stop codons (boxed) is shown in Fig. 6.

AG5 was then cloned into the EcoRI and -Sail sites of the pGEX-6P-3 expression vector (Pharmacia Biotech, Auckland, NZ) and transformed into the E. coli strain K12 XL-lBlue competent cells according to standard laboratory protocols. The amino acid sequence is given in SEQ ID NO: 52. The amino acid sequence of the autoaggregation protein AG5 is shown in Fig. 7.

The autoaggregation AG5 protein was expressed as a fusion protein with glutathione S-transferase (GST), isolated and purified using Glutathione Sepharose 4B resin (Pharmacia Biotech) according to the manufacturer's instructions and protein expression checked by SDS-PAGE. An assay for aggregation was adapted from Roos et al, Mol. Microbiol.

32:427-436, 1999 with modifications. A 10 ml overnight culture of L. rhamnosus strain HNOOl was grown in Man-Rogosa-Sharpe (MRS) broth (Oxoid) with glucose at a final concentration of 1%>. The bacteria were washed five times in sterile deionized water resulting in loss of endogenous aggregation. Bacteria were suspended in 1 ml PBS, and 5 μl of the purified the HNOOl autoaggregation protein AG5 fusion protein or an irrelevant (ie. non-adhesion) GST-fusion protein were added to 20 μl aliquots of the bacterial suspension, and placed on microscope slides. The slides were rocked gently for 13 min, and aggregation monitored by light microscopy.

As shown in Fig. 8A, in the presence of the AG5 autoaggregation GST- fusion protein, L. rhamnosus strain HNOOl cells readily aggregated. Fig. 8 A illustrates an image of a phase-contrast photomicrograph (exposure 1/8 sec, final magnification x 240) showing obvious clumping of washed L. rhamnosus strain HNOOl cells in the presence of AG5 autoaggregation protein tagged with GST. If an irrelevant (ie. non-adhesion) GST-fusion protein was used, no aggregation occurred. Fig. 8B illustrates an image of a phase-contrast photomicrograph (exposure 1/8 sec, final magnification x 240) showing no clumping of washed L. rhamnosus strain HNOOl cells in the presence of an irrelevant (non-adhesion) HNOOl protein tagged with GST, as a negative control. The GST-tagged HNOOl autoaggregation protein AG5 did not form observable clumps in the absence of bacterial cells (data not shown). Thus, the HNOOl autoaggregation protein AG5 mediated the autoaggregation of L. rhamnosus strain HNOOl cells.

The L. rhamnosus strain HNOOl is Icnown to have probiotic properties (see Tannock et al, Appl. Environ. Microbiol. 66:2,578-2,588, 2000; Gill et al, Br. J. Nutr. 83 : 167-176, 2000; Prasad et al, Int. Dairy J. 8:993-1002, 1998). In order to function effectively as probiotic bacteria, L. rhamnosus HNOOl must colonize (at least transiently) the gut environment, as well as exert positive health benefits, possibly through the exclusion of pathogenic bacteria from intestinal surfaces. The ability to form aggregates may be important for both and survival in the gut environment and functionality of L. rhamnosus HNOOl. The ability to autoaggregate may assist in the formation of biofilms of J. rhamnosus HNOOl and/or related species, improving the chances of colonization in the highly 2506

62

competitive gut environment, and then exclusion of competing bacteria, including pathogens. Applications for the AG5 autoaggregation protein include:

• as a prebiotic to enhance the growth of J. rhamnosus HNOOl or other Lactobacillus species in the gut;

• as an agent to promote clumping of J. rhamnosus HNOOl in media to improve survival in industrial processes; and

• as an agent to help prevent pathogenic colonization of mucosal surfaces.

Example 5 Isolation and Characterisation of Malic enzyme from L. rhamnosus HNOOl The full-length polynucleotide sequence of malic enzyme AA5, given in

SEQ ID NO: 2, was amplified from L. rhamnosus HNOOl DNA using standard PCR methodology. The polynucleotide sequence containing L. rhamnosus strain HNOOl malic enzyme gene AA5 showing ATG initiation and translation stop codons (boxed) is shown in Fig. 9. The upstream and downstream primers were tagged with EcoRI and BamRI restriction endonuclease recognition sequences to facilitate cloning.

The AA5 gene was then cloned into the EcoRI and BamBI sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain DH-5α competent cells according to standard laboratory protocols. Cells were lysed by sonication and the AA5 protein, expressed as a GST fusion protein, was checked by SDS-PAGE analysis. The polypeptide sequence is given in SEQ ID NO: 43 and shown in Fig. 10.

Malic enzyme activity was assessed determining the rate of pyruvate reduction in transformed strains of an E. coli mutant. The E. coli strain EJ1321 contains multiple mutations that affect both NAD- and NADP-dependent malic enzyme activity, as well as malic enzyme regulation (Hansen and Juni, Biochem. Biophys. Res. Comm. 65:559-566, 1975). The strain was obtained from the E. coli Genetic Stock Centre (Yale University, USA), and transformed with the pGEX-6P-3 vector construct encoding the HNOOl malic enzyme AA5. Trahsformants were selected by resistance to 100 μg/ml ampicillin on M9 plates 12506

63

supplemented with 0.5% glucose (ie. permissive growth conditions). Ampicillin resistant EJ132 colonies were picked and grown overnight at 37 °C in 10 ml LB broth with 100 μg/ml ampicillin and 2 ml then used to inoculate 100 ml LB broth with 100 μg/ml ampicillin. Cultures were incubated at 37 °C with shaking until OD at 600 nm reached approximately 0.4 whereupon expression of the AA5 protein was induced by the addition of 100 μl of 1 M IPTG. After a further 4 hours culture at 37 °C with shaking, 10 ml aliquots were talcen, spun at 4000 rpm for 5 min, supematants removed and cells resuspended in 5 ml PBS. Cultures were then sonicated to produce crude lysates. Malic enzyme activity in the crude lysates was measured according to Kobayashi et al, J. Biol. Chem. 264:3200- 3205, 1989, with modifications. Briefly, total protein contents of the lysates were quantitated using the BCA Protein Assay Reagent kit (Pierce, Rockford, IL, USA) according to the manufacturer's instructions, and 3.5 mg total protein added to 990 μl reaction solution containing 100 μM MOPS buffer (pH 6.1), 100 μM Na₂CO₃, 50 μM NADH and 5 μM MgCl₂ (Sigma). Lastly, 10 μl of 1 M sodium pyruvate was added as substrate and utilization of NADH measured as change in

OD at 340 nm.

Malic enzyme activity was compared between PBS buffer only (20 μl), crude lysate from wild type EJ1321 cells (ie. non-transformed), EJ1321 cells transformed with pGEX-6P-3 encoding an irrelevant protein (AD5), and EJ1321 cells transformed with pGEX-6P-3 encoding HNOOl malic enzyme AA5

(Fig. 11), Specific activities are given in Table 4, with a unit of enzyme was defined as μmole NADH used per min per mg protein.

Results in Fig. 11 and Table 4 indicate that although NADH was stable (ie. no change in OD in the presence of NADH and substrate), some background NADH reduction occurred when crude lysates from wild-type EJ1321 cells or

EJ1321 cells expressing an irrelevant protein. Nonetheless, clear malic enzyme activity was observed when crude lysate from EJ1321 cells expressing AA5 protein was used, with over 6-fold more enzyme activity compared to background. Therefore, AA5 encodes a malic enzyme. 2506

64

Fig. 11 shows malate enzyme activity measured as rate of pyruvate reduction by crude lysate preparations of EJ1321 cell transformants. ■ PBS buffer-only; A 3.5 μg wild-type EJ1321 cell lysate; ♦ 3.5 μg cell lysate of EJ1321 transformed with pGEX-6P-3 construct encoding an irrelevant HNOOl protein (AD5); • 3.5 μg cell lysate of EJ1321 transformed with pGEX-6P-3 construct encoding HNOOl malic enzyme AA5.

Table 4. Malic enzyme activity in crude lysates of transformed and non- transformed EJ1321 cells

The malic enzyme assay was repeated with increasing amounts of crude lysate from EJ1321 cells expressing AA5 protein to determine whether malic enzyme activity was proportional to amount of AA5 protein present (Fig. 12 and Table 5).

Results from Fig. 12 and Table 5 indicate that increased amounts of crude lysate of EJ1321 E. coli strain transformed with HNOOl malic enzyme AA5 led to increased malic enzyme activity. However, as the amount of substrate became limiting at higher amounts of lysate, the increases in activity were not strictly proportional. Nonetheless, these results support the evidence that AA5 encodes the HNOOl malic enzyme. Fig. 12 shows data illustrating the effect of increasing amounts of EJ1321 crude lysate on malic enzyme activity. ■ 5 μl wild-type EJ1321 cell lysate; A 5 μl cell lysate of EJ1321 transformed with pGex-6P-3 encoding AA5; ♦ 50 μl cell 2506

65

lysate of EJ1321 transformed with pGex-6P-3 encoding AA5; • 200 μl cell lysate of EJ1321 transformed with pGex-6P-3 encoding AA5.

Table 5. Malic enzyme activity with increasing amounts of cell lysate

The NAD-dependent malic enzyme (EC 1.1.1.38) catalyzes L-malate oxidative decarboxylation and pyruvate reductive carboxylation (Murai, T. et al,. Biochem. Biophys. Res. Comm. 43:875-881, 1971) and is central to citrate metabolism. Applications for HNOOl malic enzyme AA5 include:

• manipulation of energy production and growth in particular media; • altered survival characteristics in industrial processes;

• formation of common intermediates of various flavor compounds; and

• lactic acid production, important for antibacterial effects and acid tolerance

Example 6

Isolation and Characterisation of Malate Dehydrogenase from L. rhamnosus

HNOOl

The full-length polynucleotide sequence of malic enzyme, given in SEQ

ID NO: 9, was amplified from the AG3 malate dehydrogenase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. Fig. 13 shows the polynucleotide sequence containing L. rhamnosus strain HNOOl malate dehydrogenase gene AG3 showing the TTG initiation and translation stop codons

(boxed).

AG3 was then cloned into the pUniBlunt/V5-HisTopo vector (Invitrogen) and transformed into the E. coli strain PIR1 OneShot competent cells (Invitrogen) according to the manufacturer's instructions. To construct an expression plasmid, the pUniBlunt/V5-HisTopo vector construct was recombined with the pBad/Thio-

Ε Echo vector (Invitrogen) and transformed into the E. coli strain TOP 10 competent cells (Invitrogen) according to the manufacturer's instructions. The AG3 gene product was therefore cloned as a fusion protein tagged with a His- patch polypeptide and thioredoxin protein. The fusion protein was expressed and purified using a Ni-NTA column (Qiagen, Auckland, NZ) according to the manufacturer's instructions and protein expression checked by SDS-PAGE. The polypeptide sequence is given in SEQ ID NO: 51 and shown in Fig. 14.

Malate dehydrogenase activity was assessed by gene complementation of the mutant E. coli strain UTH4606 that lacks a functional malate dehydrogenase gene (Heard et al, J. Bacteriol. 122:329-331, 1975; Shaw et al, Mutation Res.,

18:247-250, 1973), provided by the E. coli Genetic Stock Centre (Yale University, USA). UTH4606 strain cells cannot utilize malate as a carbon source, in contrast to wild-type E. coli. pBAD-Thio-E construct containing the HNOOl malate dehydrogenase AG3 gene or empty pBAD-Thio-E vector was transformed into the UTH4606 E. coli strain and plated onto M9 media plates containing 100 μg/ml kanamycin and 0.5% glucose. Transformant colonies were picked, and plated out onto a series of selective M9 agar plates containing 100 μg/ml Kanamycin and/or 0.5% glucose or 0.5% malate. Growth of the UTH4606 transformed with pBAD-Thio-E encoding the AG3 protein was compared with wild-type UTH4606 cells and UTH4606 cells transformed with empty pBAD-

Thio-E vector. Plates were incubated aerobically at 37 °C overnight. Growth was assessed for malate dehydrogenase complementation.

Results are shown in Table 6 and indicate that wild-type UTH780 cells grew on M9 media supplemented with glucose, but not on M9 media supplemented with malate, or on media containing Kanamycin. This confirmed 2/12506

67

the phenotype of the UTH780 strain of being unable to utilize malate as a carbon source due to the loss of malate dehydrogenase function. Transformation with empty pBAD/Thio-E vector allowed growth on media containing Kanamycin, but did not complement the malate dehydrogenase mutation. Transformation with pBAD/Thio-E encoding the HNOOl malate dehydrogenase AG3 allowed growth on Kanamycin, indicating the presence of the plasmid, and on malate, indication that the AG3 protein complemented the E. coli malate dehydrogenase deficiency. Therefore, the HNOOl protein AG3 has malate dehydrogenase activity.

Table 6. Results of LB agar plate assay for malate dehydrogenase gene complementation

+: growt ; -: no growt

Malate dehydrogenase (EC 1.1.1.37) catalyzes the reversible oxidation of malate to oxaloacetate with the concomitant reduction of NAD. As lactobacilli appear not to have a functioning Krebs cycle, the enzyme may be involved in amino acid biosynthesis or L-malate utilization pathways. Applications for

HNOOl malate dehydrogenase AG3 include:

• manipulation of energy production and growth in particular media; • altered survival characteristics in industrial processes ; and

• formation of common intermediates of various flavor compounds. Example 7

Isolation and Characterisation of Dihydrodipicolinate Synthase from L. rhamnosus HNOOl

The full-length polynucleotide sequence of dihydrodipicolinate synthase from L. rhamnosus HNOOl, given in SEQ ID NO: 13 and shown in Fig. 15 with

ATG initiation and translation stop codons (boxed), was used to amplify the AI2 dihydrodipicolinate synthase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and Sail restriction endonuclease recognition sequences to facilitate cloning.

AI2 was then cloned into the EcoRI and Sail sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain K12 XL-lBlue competent cells according to standard laboratory protocols. The dihydrodipicolinate synthase AJ2 protein was expressed as a fusion protein with glutathione S-transferase (GST), bound to Glutathione Sepharose 4B resin

(Pharmacia Biotech), and PreScission protease used to cleave off dihydrodipicolinate synthase AI2 protein, according to the manufacturer's instructions. An aliquot of the purified AI2 protein was checked by SDS-PAGE analysis. The polypeptide sequence is given in SEQ ID NO: 55 and is shown in Fig. 16.

Dihydrodipicolinate synthase activity was assessed by gene complementation of the mutant E. coli strain AT997 deficient in dihydrodipicolinate synthase gene function (Bukhari and Taylor, J. Bacteriol. 105:844-854, 1971), provided by the E. coli Genetic Stock Centre (Yale University, USA). AT997 cells require diaminopimelic acid (DAP) for growth, in contrast to wild-type E. coli that is DAP-independent. pGEX-6P-3 construct containing the HNOOl dihydrodipicolinate synthase AI2 gene or empty pGEX-6P- 3 vector was transformed into the AT997 E. coli strain. Transformed AT997 cells were plated onto LB agar plates containing ampicillin (100 μg/ml) only or ampicillin and 45 μg/ml DAP, at dilutions designed to allow the visualization of distinct colonies (ie. < 200 colonies/plate). Plates were incubated aerobically at 37 °C overnight and growth assessed as the presence of distinct colonies.

Results in Table 7 indicate that while AT997 cells transformed with either empty pGEX-6P-3 or pGEX-6P-3 containing the HNOOl dihydrodipicolinate synthase AI2 grew in the presence of DAP, only cells transformed with vector containing AI2 grew without DAP. Therefore, the HNOOl dihydrodipicolinate synthase protein AI2 complemented the dihydrodipicolinate synthase gene mutation in E. coli strain AT997.

Table 7. Results of LB agar plate assay for dihydrodipicolinate synthase gene complementation.

+: growth; -: no growt

Dihydrodipicolinate synthase (EC 4.2.1.52) converts L-aspartate 4- semialdehyde and pyruvate to 1-2,3 -dihydrodipicolinate as part of the lysine biosynthesis pathway. L-aspartate 4-semialdehyde is also the first step of the glycine, serine and threonine metabolic pathways. Applications for HNOOl dihydrodipicolinate synthase protein AI2 include:

• altered amino acid content, with important flavor and metabolic impacts;

• commercial production of lysine or intermediates;

• manipulation of energy production and growth in particular media; and

• altered survival characteristics in industrial processes . Example 8

Isolation and Characterisation of Dihydrodipicolinate Reductase from L. rhamnosus HNOOl

The full-length polynucleotide sequence of dihydrodipicolinate reductase from L. rhamnosus strain HNOOl, given in SEQ ID NO: 14 and shown in Fig. 72 with ATG initiation and translation stop codons (boxed), was used amplify the AI3 dihydrodipicolinate reductase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and SaR restriction endonuclease recognition sequences to facilitate cloning.

AI3 was then cloned into the EcoRI and Sail sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain K12 XL-lBlue competent cells according to standard laboratory protocols. The polypeptide sequence of dihydrodipicolinate reductase AI3 is given in SEQ ID NO: 56 and is shown in Fig. 73. The dihydrodipicolinate reductase AI3 protein was expressed as a fusion protein with glutathione S-transferase (GST), bound to Glutathione Sepharose 4B resin (Pharmacia Biotech), and PreScission protease used to cleave off dihydrodipicolinate reductase AI3 protein, according to the manufacturer's instructions. An aliquot of the purified AI3 protein was checked by SDS-PAGE analysis.

Dihydrodipicolinate reductase activity was assessed by gene complementation of the mutant E. coli strain AT999 deficient in dihydrodipicolinate reductase gene function (Bukhari and Taylor, J Bacteriol. 105:844-854, 1971), provided by the E. coli Genetic Stock Centre (Yale University, USA). AT999 cells require diaminopimelic acid (DAP) for growth, in contrast to wild-type E. coli that is DAP-independent. pGEX-6P-3 construct containing the HNOOl dihydrodipicolinate reductase AI3 gene or empty pGEX- 6P-3 vector was transformed into the AT999 E. coli strain. Transformed AT999 cells were plated onto LB agar plates containing ampicillin (100 μg/ml) only or ampicillin and 45 μg/ml DAP, at dilutions designed to allow the visualization of distinct colonies (ie. < 200 colonies/plate). Plates were incubated aerobically at 37 °C overnight and growth assessed as the presence of distinct colonies (Table 8).

Results in Table 8 indicate that while AT999 cells transformed with either empty pGEX-6P-3 or pGEX-6P-3 containing the HNOOl dihydrodipicolinate reductase AI3 grew in the presence of DAP, only cells transformed with vector containing AJ3 grew without DAP. Therefore, the HNOOl dihydrodipicolinate reductase protein AI3 complemented the dihydrodipicolinate Reductase gene mutation in E. coli strain AT999.

Table 8. Results of LB agar plate assay for dihydrodipicolinate reductase gene complementation.

+: growth; -: no growth

Dihydrodipicolinate reductase (EC 1.3.1.26) converts L-2,3- dihydrodipicolinate to L-tetrahydropicolinate as part of the lysine biosynthesis pathway. Applications for HNOOl dihydrodipicolinate synthase protein AI3 include:

• altered amino acid content, with important flavor and metabolic impacts;

• commercial production of lysine or intermediates;

• manipulation of energy production and growth in particular media; and

• altered survival characteristics in industrial processes. Example 9

Isolation and Characterisation of Aspartate Aminotransferase from L. rhamnosus

HNOOl

The full-length gene sequence of aspartate aminotransferase from L. rhamnosus strain HNOOl, given in SEQ ID NO: 12 and shown in Fig. 17 with GTG initiation and translation stop codons (boxed), was used to amplify the AH9 aspartate aminotransferase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and Sail restriction endonuclease recognition sequences to facilitate cloning. AH9 was then cloned into the EcoRI and Sail sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain K12 XL-lBlue competent cells according to standard laboratory protocols. The aspartate aminotransferase AH9 protein was expressed as a fusion protein with glutathione S-transferase (GST), bound to Glutathione Sepharose 4B resin (Pharmacia Biotech), and PreScission protease used to cleave off the aspartate aminotransferase AH9 protein, according to the manufacturer's instructions. An aliquot of the purified AH9 protein was checked by SDS-PAGE analysis. The polypeptide sequence is given in SEQ ID NO: 54 and is shown in Fig. 18.

AH9 activity was assayed according to the previously published malate dehydrogenase-coupled method (Karmen, J Clin. Invest. 34:131-133, 1955) with modifications. Briefly, 1 ml reaction mixtures containing 100 μmol Tris hydrochloride buffer (pH 8.0), 100 μmol L-aspartate, 10 μmol of α-ketoglutarate, 0.2 μmol NADH, 0.015 μmol pyrodoxal 5'-phosphate (PLP), and 3 μg (3.6 U) malate dehydrogenase (all chemicals from Sigma Chemical Co.) were incubated at 37°C with increasing amounts (0 to 142.5 ng) of the purified HNOOl aspartate aminotransferase AH9 protein. The rationale of the assay is that aspartate aminotransferase converts α-ketoglutarate and L-aspartate to oxaloacetate and L- glutamate. The oxaloacetate is then substrate for the malate dehydrogenase, which oxidizes one molecule of NADH to NAD⁺ for every molecule of oxaloacetate converted to L-malate. As the first step is rate limiting, the amount of

NADH oxidized in the second step is directly proportional to the aspartate aminotransferase-dependent production of oxaloacetate from α-ketoglutarate in the first step. The reaction was monitored by the decrease in absorbance at 340 nm, and results used to calculate the μmol NADH oxidized per minute. One unit of enzyme was defined as the amount of enzyme that catalyzed the production of

1 μmol of oxaloacetate per minute at 37 °C.

Results in Table 9 indicate that while in the absence of purified HNOOl aspartate aminotransferase AH9 protein, there was some background oxidation of NADH, the addition of AH9 protein led to increased rates of aspartate aminotransferase-dependent NADH oxidation. Increased amounts of AH9 increased NADH oxidation in a dose-dependent manner. A similar background rate observed in reaction mixtures without the addition of AH9 protein was also observed in reaction mixtures without both AH9 protein and α-ketoglutarate substrate (data not shown), indicating that the background NADH oxidation was not aspartate aminotransferase-dependent. The addition of 142.5 ng of AH9 protein led to an over 19-fold increase in NADH oxidation. The activity of the purified HNOOl aspartate aminotransferase AH9 protein was calculated to be 31 U/mg protein. Therefore, HNOOl protein AH9 is an aspartate aminotransferase.

Table 9. Results of the malate dehydrogenase-coupled aspartate aminotransferase assay.

α-lcetoglutarate is an important chemical mediator in lactic acid bacteria, and the addition of this compound to cheese curd has positive impacts on cheese flavor (Yvon et al, Int. Dairy J. 8:889-898, 1998). The formation of α- ketoglutarate using L-glutamate as an amino donor, catalysed by aspartate aminotransferase, is an important pathway in maintaining intracellular α- ketoglutarate levels. Applications for HNOOl aspartate aminotransferase AH9 include:

• altered amino acid content, with important flavor and metabolic impacts;

• manipulation of energy production and growth in particular media; and

• altered survival characteristics in industrial processes

Example 10 Isolation and Characterisation of Serine Dehydratase subunits α and β from L. rhamnosus HNOOl

The full-length polynucleotide sequence of serine dehydratase subunits α and β, given in SEQ ID NO: 7 was used to amplify the AF8 serine dehydratase α subunit and AFT serine dehydratase β subunit from L. rhamnosus HNOOl DNA as a single operon using standard PCR methodology. The polynucleotide sequence containing L. rhamnosus strain HNOOl serine dehydratase subunits α (AF8) and β (AF7) is shown in Fig. 19, with ATG translation initiation codons and termination codons shown boxed for AF8 and shaded for AF7. The AF8 serine dehydratase α subunit and AFT serine dehydratase β subunit were amplified from L. rhamnosus HNOOl DNA as a single operon using standard PCR methodology. The AF8 and AFT genes were cloned in the vector pTRKH2 (obtained from Dr Todd Klaenhammer, North Carolina State University, North Carolina, USA) and transformed into E. coli DH5α cells. Positive transformants were selected, grown overnight and the plasmid isolated by standard laboratory techniques. Competent L. rhamnosus HNOOl cells were then transformed with the pTRKH2 construct containing the HNOOl serine dehydratase subunits α and β to overexpress the genes in strain HNOOl. The amino acid sequence of the expressed proteins are given in SEQ ID NOS: 48 and 49 is shown in Figs. 22A and 22B. Serine dehydratase enzyme activity was assessed by comparing serine utilization in liquid cultures of HNOOl strain cells transformed with either the pTRKH2 construct containing the HNOOl serine dehydratase or empty pTRKH2 vector only.

The results shown in Figs. 20 and 21 indicate that the presence of the expression plasmid encoding HNOOl serine dehydratase subunits α (AF8) and β

(AF7) significantly increased the utilization of serine by HNOOl strain cells, compared to cells transformed with empty expression vector only. Therefore the HNOOl genes AFT and AF8 encode the serine dehydratase enzyme. Fig. 20 shows the percentage serine utilization by HNOOl strain in liquid culture with 5 mM initial serine concentration. ■ HNOOl transformed with vector only; ♦ pTRKH2 construct containing HNOOl serine dehydratase. Fig. 21 shows the percentage serine utilization by HNOOl strain in liquid culture with 12 mM initial serine concentration. ■ HNOOl transformed with vector only, ♦ pTRKH2 construct containing HNOOl serine dehydratase. Serine dehydratase (EC 4.2.1.13), comprising α and β subunits, catalyzes the irreversible deamination of serine to pyruvate and ammonia (Ogawa et αl, J. Biol. Chem. 264:15818-15822, 1989; Grabowslci et αl, Trends in Biochem. Sci. 18:297-300, 1993). Applications for HNOOl serine dehydratase subunits AF7 and AF8 include:

• energy supply from amino acids present in growth media or environment;

• production of ammonia, regarded as a flavor compound;

• altered pyruvate levels - pyruvate is a highly reactive compound, and is important in a number of flavor pathways; and • altered survival characteristics in industrial processes.

Example 11 Isolation and Characterisation of Histidinol-Phosphate Aminotransferase from L. rhamnosus HNOOl

The full-length polynucleotide sequence of histidinol-phosphate aminotransferase from L. rhamnosus strain HNOOl, given in SEQ ID NO: 8 and shown in Fig. 23 with ATG initiation and translation stop codons (boxed), was used to amplify the AG2 histidinol-phosphate aminotransferase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and BamΗI restriction endonuclease recognition sequences to facilitate cloning.

AG2 was then cloned into the EcoRI and Bam l sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into E. coli strain DH-5α competent cells according to standard laboratory protocols. Cells were lysed by sonication and the presence of AG2 protein, expressed as a GST fusion protein, checked by SDS-PAGE analysis. The polypeptide sequence of AG2 is given in SEQ ID NO: 50 and shown in Fig. 24.

Histidinol-phosphate aminotransferase activity was assessed by gene complementation of the mutant E. coli strain UTH780 that lacks a functional hisC gene that encodes histidinol-phosphate aminotransferase (Goldschmidt et al, Genetics, 66:219-229, 1970), provided by the E. coli Genetic Stock Centre (Yale University, USA). UTH780 cells require L-histidine for growth, in contrast to wild-type E. coli that is L-histidine-independent. pGEX-6P-3 construct encoding HNOOl histidinol-phosphate aminotransferase AG2 was transformed into the

UTH780 E. coli strain and plated onto LB agar plates containing 100 μg/ml ampicillin. Ampicillin-resistant transformant colonies were picked and plated out onto selective media (ie. M9 media plates with and without 100 μg/ml L-histidine, with and without 100 μg/ml ampicillin). Growth of UTH780 transformed with AG2 was compared with the growth of wild-type UTH780 cells and UTH780 cells transformed with a pGex-6P-3 construct encoding a non-histidinol-phosphate aminotransferase (AE8). Plates were incubated aerobically at 37 °C overnight and growth assessed as the presence of distinct colonies.

Results in Table 10 indicate that while wild-type UTH780 cells grew in the presence of histidine, no growth was observed when ampicillin was added to the media. Therefore, ampicillin resistance in transformed UTH780 was due to the presence of pGEX-6P-3 vector. UTH780 cells transformed with either empty pGEX-6P-3 or pGEX-6P-3 encoding an irrelevant protein (AE9) grew in the presence of histidine and ampicillin, but remained auxotrophic for hisitidine, indicating that the HisC phenotype was not complemented. UTH780 cells transformed with pGEX-6P-3 encoding HNOOl histidinol-phosphate aminotransferase AG2 grew in on M9 media without histidine. Therefore, the AG2 protein complemented the hisC mutation of UTH780 strain E. coli cells.

Table 10. Results of M9 agar plate assay for histidinol-phosphate aminotransferase gene complementation.

+: growth; -: no growth

Histidinol-phosphate aminotransferase (EC 2.6.1.9) catalyzes the transamination of histidinol phosphate and 2-oxoglutarate to 3-(Imidazol-4-yl)-2- oxopropyl phosphate and glutamate, as the eighth step in histidine biosynthesis (Martin et al, J. Bio. Chem. 242:1168-1174, 1967). Some lactic acid bacteria are known to decarboxylate amino acids, such that histidine can be converted to histamine, which has undesirable physiological effects (Lonvaud-Funel, FEMS Microb. Lett. 199:9-13, 2001). Applications for HNOOl histidinol-phosphate aminotransferase AG2 include:

• altered levels of particular amino acids, leading to flavor and metabolic changes;

• affect aromatic amino acid metabolism, a source of important flavor compounds; and

• modulate production of biogenic amines.

Example 12 Isolation and Characterisation of malY-Aminotransferase from L. rhamnosus

HNOOl

The full-length polynucleotide sequence of malY aminotransferase from L. rhamnosus strain HNOOl, given in SEQ ID NO: 17 and shown in Fig. 25 with ATG initiation and translation stop codons (boxed), was used to amplify the AJ6 aminotransferase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and BamHI restriction endonuclease recognition sequences to facilitate cloning.

AJ6 was then cloned into the EcoRI and BamHI sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into E. coli strain DH-5 competent cells according to standard laboratory protocols. Cells were lysed by sonication and the presence of AJ6 protein, expressed as a GST fusion protein, checked by SDS-PAGE analysis. The polypeptide sequence of AJ6 is given in SEQ ID NO: 59 and shown in Fig. 26.

A feature of malY-aminotransferases is the ability to complement mutations of the E. coli cystathione β-lyase protein metC (Zdych et al, J. Bacteriol. 177:5035-5039, 1995). Therefore, AJ6 activity was assessed by suppression of the metC phenotype in the E. coli strain CAG18527 (Singer et al, Microbiol. Rev. 53:1-24, 1989) provided by the E. coli Genetic Stock Centre (Yale University, USA). CAG18527 cells require L-methionine for growth, in contrast to wild-type E. coli that is L-methionine-independent. A pGEX-6P-3 construct encoding the HNOOl aminotransferase AJ6 was transformed into the

CAG18527 E. coli strain and plated onto LB agar plates containing 100 μg/ml ampicillin. Ampicillin-resistant transformant colonies were picked and plated out onto selective media (M9 plates with and without 1 mM L-methionine, with and without 5 μg/ml ampicillin). Growth of the CAG18527 transformed with AJ6 was compared with the growth of wild-type CAG18527 cells and CAG18527 cells transformed with a ρGΕX-6P-3 construct encoding a non-aminotransferase irrelevant protein. Plates were incubated aerobically at 37 °C for 48 hrs and growth assessed.

Results in Table 11 indicate that while wild-type CAG 18527 cells grew in the presence of methionine, no growth was observed in the presence of ampicillin.

This confirmed the ampicillin-sensitive, methionine-auxotrophic phenotype of the CAG18527 strain. CAG18527 cells transformed with either empty pGEX-6P-3 or pGEX-6P-3 encoding an irrelevant HNOOl protein (AC9) grew in the presence of methionine and ampicillin, but not in the absence of methionine, indicating that the metC- phenotype was not suppressed. CAG18527 cells transformed with pGEX-6P-3 encoding HNOOl aminotransferase AJ6 were ampicillin resistant and grew on M9 media without methionine. Therefore, the AJ6 protein suppressed the metC mutation of CAG18527 strain E. coli cells.

Table 11. Results of M9 agar plate assay for suppression of the metC phenotype.

+: growth; -: no growt The malY/PatB pyridoxal-5'-phosphate-dependent aminotransferase family (EC 2.6.1.-) appear to have both aminotransferase and regulatory activities (Mehta and Christen, Eur. J. Biochem. 203 :373-376, 1993), including the transamination of methionine and regulation of maltose utilisation (Reidl and Boos, J Bacteriol. 173:4862-4876, 1991), as well as other activities (Chu et al., Infect. Imm. 63: 4448-4455, 1995). Applications for HNOOl malY- aminotransferase AJ6 include:

• altered levels of particular amino acids, leading to flavor and metabolic changes • altered expression of catabolite or other regulons

• modulation of hemolytic activity

• probiotic effects

Example 13 Isolation and Characterisation of malY-Aminotransferase from L. rhamnosus

HNOOl

The full-length polynucleotide sequence of a second alY- aminotransferase from L. rhamnosus strain HNOOl, given in SEQ ID NO: 18 and shown in Fig. 27 with ATG initiation and translation stop codons (boxed), was used to amplify the AJT aminotransferase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and BamHI restriction endonuclease recognition sequences to facilitate cloning.

AJT was then cloned into the EcoRI and BamHI sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into E. coli strain DH-5α competent cells according to standard laboratory protocols. Cells were lysed by sonication and the presence of AJ7 protein, expressed as a GST fusion protein, checked by SDS-PAGE analysis. The polypeptide sequence of AJ7 is given in SEQ ID NO: 60 and shown in Fig. 28. A feature of malY-aminotransferases is the ability to complement mutations of the E. coli cystathione β-lyase protein metC (Zdych et al, J. Bacteriol. 177:5035-5039, 1995). Therefore, AJ7 activity was assessed by suppression of the metC phenotype in the E. coli strain CAG18527 (Singer et al, Microbiol. Rev. 53:1-24, 1989) provided by the E. coli Genetic Stock Centre (Yale University, USA). CAG18527 cells require L-methionine for growth, in contrast to wild-type E. coli that is L-methionine-independent. pGΕX-6P-3 construct encoding the HNOOl aminotransferase AJ7 was transformed into the CAG18527 E. coli strain and plated onto LB agar plates containing 100 μg/ml ampicillin. Ampicillin-resistant transformant colonies were picked and plated out onto selective media (M9 plates with and without 1 mM L-methionine, with and without 5 μg/ml ampicillin). Growth of the CAG18527 transformed with AJ7 was compared with the growth of wild-type CAG18527 cells and CAG18527 cells transformed with a pGEX-6P-3 construct encoding a irrelevant protein. Plates were incubated aerobically at 37 °C for 48 hrs and growth assessed. Results in Table 12 indicate that while wild-type CAG18527 cells grew in the presence of methionine, no growth was observed in the presence of ampicillin. This confirmed the ampicillin-sensitive, methionine-auxotrophic phenotype of the CAG18527 strain. CAG18527 cells transformed with either empty pGEX-6P-3 or pGEX-6P-3 encoding an irrelevant HNOOl protein (AC9) grew in the presence of methionine and ampicillin, but not in the absence of methionine, indicating that the metC- phenotype was not suppressed. CAG18527 cells transformed with pGEX-6P-3 encoding HNOOl aminotransferase AJ7 were ampicillin resistant and grew on M9 media without methionine. Therefore, the AJ7 protein suppressed the metC mutation of CAG18527 strain E. coli cells.

Table 12. Results of M9 agar plate assay for suppression of the metC phenotype.

+: growth; -: no growth

The malY/PatB pyridoxal-5'-phosphate-dependent aminotransferase family (EC 2.6.1.-) appear to have both aminotransferase and regulatory activities (Mehta and Christen, Eur. J. Biochem. 203 :373-376, - 1993), including the transamination of methionine and regulation of maltose utilization (Reidl and

Boos, J. Bacteriol. 173:4862-4876, 1991), as well as other activities (Chu et al, Infect. Imm. 63: 4448-4455, 1995). Applications for HNOOl malY- aminotransferase AJ7 include:

• altered levels of particular amino acids, leading to flavor and metabolic changes;

• altered expression of catabolite or other regulons;

• modulation of hemolytic activity; and

• probiotic effects

Example 14

Isolation and Characterisation of Cystathione β-Lvase from L. rhamnosus HNOOl

The full-length polynucleotide sequence of cystathione β-lyase from L. rhamnosus strain HNOOl, given in SEQ ID NO: 5 and shown in Fig. 29 with ATG initiation and translation stop codons (boxed), was used to amplify the AC8 cystathione β-lyase gene from L. rhamnosus HNOOl DNA using standard PCR methodology.

AC8 was cloned into the pUniBlunt/V5-HisTopo vector (Invitrogen) and transformed into the E. coli strain PIR1 OneShot competent cells (Invitrogen). To construct an expression plasmid, the pUniBlunt/V5-HisTopo vector construct was recombined with the pBad/Thio-E vector (Invitrogen) and transformed into the E. coli strain TOP 10 competent cells (Invitrogen) according to the manufacturer's instructions. The AC gene product was therefore cloned as a fusion protein tagged with a His-patch polypeptide and thioredoxin protein. The AC 8 fusion protein was expressed and purified using a Ni-NTA column (Qiagen, Auckland, NZ) according to the manufacturer's instructions and protein expression checked by SDS-PAGE. The polypeptide sequence of AC8 is given in SEQ ID NO: 46 and shown in Fig. 46.

Cystathione β-lyase activity was assessed according to the method of Uren, Methods in Enzymol. 143:483-496, 1987, with modifications. Briefly, aliquots of the purified AC8-GST fusion protein were added to 1 ml cuvettes containing 780 μl of 0.1 M Tris-HCl pH 9.0, 200 μl of 10 mM L-cystathionine, and 20μl of 0.1M potassium phosphate, with pyridoxal-5' -phosphate added to a final concentration of 20 μM, on ice. Change in OD was measured at 412 nm over time, and one unit of enzyme defined as the formation of 1 μmol of mercaptide per minute at 37 °C. Cystathione β-lyase activity of the AC8 fusion protein was compared with activity of an irrelevant protein (pBAD/Thio-E/Uni- CAT expression control vector, Invitrogen), and reactions containing water or Ni-

NTA column elution buffer. Results are shown in Fig. 31, with rates of change of OD and enzyme activity given in Table 13.

Results in Fig. 31 and Table 13 indicate that similar background rates of mercaptide formation were obseived in reactions containing water only, elution buffer only or 10 μl purified CAT fusion protein. Significantly greater mercaptide formation was observed in reactions containing 10 μl purified HNOOl cystathione β-lyase AC8 fusion protein. Therefore, AC8 protein has cystathione β-lyase activity. Fig. 31 shows cystathione β-lyase activity measured as rate of mercaptide formation. ♦ 10 μl purified HNOOl cystathione β-lyase AC8 fusion protein; ■ 10 μl purified CAT fusion protein; A 10 μl H₂O only; • 10 μl elution buffer only. Table 13. Cystathione β-lyase activity of AC8 compared with irrelevant protem, H₂O and elution buffer controls.

The dose-response of the HNOOl cystathione β-lyase activity AC8 was determined by comparing mercaptide formation in a series of dilutions of the purified enzyme. Results are shown in Fig. 32, and the rate of change in optical density and enzyme activities given in Table 14.

Results in Fig. 32 and Table 14 indicate that the increased rate of mercaptide peptide was proportional to the amount of AC 8 fusion protein, supporting that AC8 encodes HNOOl cystathionine β-lyase. Fig. 32 shows the experimentally determined dose-response of the AC8 fusion protein. Cystathione β-lyase activity of increasing amounts of His-patch Thio/AC8 fusion protein; 10 μl (♦), 25 μl (■) and 50 μl (A) purified protein showed increasing rates of mercaptide formation. The increase in mercaptide formation was proportional to amount of AC8 fusion protein added.

Table 14. Cystathione β-lyase activity in increasing amounts of AC8 protein

Cystathionine β-lyase (EC 4.4.1.8) deaminates cystathionine to L- homocysteine, ammonia and pyruvate (Dwivedi et al, Biochem. 21 :3064-3069, 1982), and may also have active on L-cystine and related substrates (Uren, Methods in Enzymol. 143:483-486, 1987; Alting et al, Appl. Environ. Microbiol. 61:4037-4042,1995). Thus, cystathionine β-lyase is involved in a number of pathways including methionine metabolism and catabolism of sulphur-containing compounds. L-homocysteine has been shown to have important health impacts in humans (Nittynen et al, Ann. Med. 31:318-326, 1999; Giles et al, Am. Heart J.

139:446-453, 2000). Applications for HNOOl cystathionine β-lyase AC8 include:

• altered flavor and metabolic characteristics through changes in levels of particular amino acids;

• altered levels of important sulphur-containing flavor compounds; and

• health impacts through the modulation of L-homocysteine levels

Example 15 Isolation and Characterisation of Phosphoenolpyruvate Hydratase from L. rhamnosus HNOOl

HNOOl phosphoenolpyruvate hydratase AK4 was isolated by a series of experiments designed to identify HNOOl strain proteins that were up-regulated in response to physiological stresses encountered during industrial processes. Cells were subjected to heat or osmotic shock, proteins radiolabeled with [³⁵S]- methionine and [³⁵S]-cysteine (Amersham, USA), and cell-free extracts from shocked and non-shocked HNOOl cultures compared by 2-D analysis and N- terminal sequencing as below. Shock proteins were radiolabeled according to standard laboratory methods. Heat shock was performed by incubation at 50°C on both log phase and stationary phase HNOOl strain cultures, and salt (osmotic) shock on late log phase by HNOOl strain cultures by transfer into MRS broth containing 0.6 M sodium chloride. Immediately after heat or osmotic shock, approximately 5 μCi ml^"1 each of L-[³⁵S]-methionine and L-[³⁵S]-cysteine were added to the culture medium and incubated for 30 min, followed by the addition of excess of cold 1 mM L-cysteine hydrochloride and 1 mM L-methionine, and cultures then placed on ice. Radiolabeled cells were collected by centrifugation washed twice in washing buffer (0.1 M Tris-HCl, 1 mM EDTA, pH 7.5) and resuspended in resuspension buffer (10 mM Tris-HCl, 5 mM MgCl₂ , 2 mM PMSF, pH 7.5). About 0.5 ml cell suspension was mixed with 0.5 g of 0.17 - 0.18 mm glass beads and homogenized using Shake-it-Baby (Biospec products). After homogenization for 25 min, the suspension was centrifuged and the supernatant was collected.

2-D Gel electrophoresis was performed on the cell free extract containing 50 - 75 μg of protein. Excess chilled methanol was added and kept at -80° C for 1 hr followed by centrifugation at 13,000 rpm to collect the pellet. The pellet was vacuum-dried and resuspended in rehydration buffer (8M urea, 2%o Triton X 100, 0.5% (v/v) IPG buffer (Amersham Pharmacia Biotech, USA) and few grains of bromophenol blue). Endonuclease (Sigma) was added (150 U) to the rehydrated sample and incubated at room temperature for 20 min. The solution was then added to IPG strips and rehydrated overnight at 20°C. The rehydrated IPG strips were placed on a flat bed electrophoresis unit (Amersham Pharmacia Biotech, USA) and focused at 300 Volts for 30 min followed by 3,000 volts for 4 hrs. The focused strips were equilibrated (15 min) in equilibration buffer (50 mM Tris- HCl, pH 8.8, 6 M Urea, 30% (v/v) glycerol, 2% (w/v) SDS and few grains bromophenol blue) containing either dithioerythritol (1.0% w/v) or iodoacetamide

(2.5% w/v). After equilibration, the strips were placed on the second dimensional (vertical SDS-PAGE homogeneous) gels a using PROTEAN II xi cell (Bio-Rad). The second dimension was carried out at 20 mA per plate for 15 min and 40 mA per plate for 4 hrs. Gels were then equilibrated in protein transfer buffer (24.8 mM Tris, pH

8.3, 192 mM Glycine and 10% (v/v) methanol) and blotted on a PVDF membrane using a Trans-blot apparatus (Bio-Rad) at 24 volts overnight at 4°C. PVDF membranes were exposed to Hyperfilm-βmax (Amersham Pharmacia Biotech, USA) for up to two weeks using standard procedures. Resultant autoradiograms were scanned using the Fluor-S Multimager system (Bio-Rad) and patterns compared using PDQuest software. For N-terminal sequencing, membranes were stained with Coommassie Brilliant Blue R-250. The desired spots were excised and N-terminal sequencing carried out using a protein sequencer (Applied BioSystems, Model 476A) according to standard methods. A protein up-regulated by heat and osmotic shock was N-terminal sequenced and the amino acid sequence is given in SEQ ID NO: 83. This _, sequence was used to search an HNOOl sequence database using the TBLASTN program (NCBI) and the corresponding polynucleotide and polypeptide sequences are given in SEQ ID NOS: 20 and 62, and shown in Figs. 33 and 34, respectively. Similarity searching using BLAST software revealed closest amino acid sequence similarity to phosphoenolpyruvate hydratase sequences but with significant differences.

Phosphoenolpyruvate hydratase (EC 4.2.1.11) is a glycolytic pathway enzyme that hydrolyzes 2-phospho-D-glycerate to give phosphoenolpyruvate (Malmstroem, B.G, The Enzymes, 2nd. Ed., Boyer, P.D., Lardy, H., Myrback, K., eds., 5:471-494, 1961). Applications for HNOOl phosphoenolpyruvate hydratase

AK4 include:

• methods of enhanced survival of industrial processes; • improved colonization of human intestinal environment; and

• altered metabolic characteristics through changes in carbohydrate utilisation

Example 16 Isolation and Characterisation of Tagatose Bisphosphate Aldolase from L. rhamnosus HNOOl

HNOOl tagatose bisphosphate aldolase AKl was isolated by a series of experiments designed to identify HNOOl strain proteins that were up-regulated in response to physiological stresses encountered during industrial processes. Cells were subjected to heat or osmotic shock, proteins radiolabeled with [³⁵S]- o r methionine and [ S]-cysteιne (Amersham, USA), and cell-free extracts from shocked and non-shocked HNOOl cultures compared by 2-D analysis and N- terminal sequencing as described for Example 15 (HNOOl phosphoenolpyruvate hydratase AK4). A protein up-regulated by heat and osmotic shock was N-terminal sequenced and the polypeptide sequence is given in SEQ ID NO: 81. This was used to search an HNOOl sequence database using the TBLASTN program (NCBI) and the corresponding polynucleotide and polypeptide sequences are given in SEQ ID NOS: 19 and 61, and shown in Figs. 35 and 36, respectively. Similarity searching using BLAST software revealed closest amino acid sequence similarity to tagatose bisphosphate aldolase sequences but with significant differences.

Tagatose bisphosphate aldolase (EC 4.1.2.40) is involved in the tagatose 6-phosphate pathway of lactose catabolism, and converts D-tagatose 1,6- bisphosphate to glycerone phosphate and D-glyceraldehyde 3 -phosphate

(Anderson and Marlcwell, Methods in Enzymol. 90:232-234, 1982). Applications of HNOOl tagatose bisphosphate aldolase AK1 include:

• methods of enhanced survival of industrial processes;

• improved colonization of human intestinal environment; and • altered metabolic characteristics through changes in carbohydrate utilisation

Example 17 Isolation and Characterisation of Phosphoglycerate Kinase from L. rhamnosus HNOOl

HNOOl phosphoglycerate kinase AK6 was isolated by a series of experiments designed to identify HNOOl strain proteins that were up-regulated in response to physiological stresses encountered during industrial processes. Cells were subjected to heat or osmotic shock, proteins radiolabeled with [³⁵SJ- methionine and [ S]-cysteine (Amersham, USA), and cell-free extracts from shocked and non-shocked HNOOl cultures compared by 2-D analysis and N- terminal sequencing as described for Example 15 (HNOOl phosphoenolpyruvate hydratase AK4). A protein up-regulated by heat and osmotic shock was N-terminal sequenced and the polypeptide sequence is given in SEQ ID NO: 82. This was used to search an HNOOl sequence database using the TBLASTN program (NCBI) and the corresponding polynucleotide and polypeptide sequences are given in SEQ ID NOS: 22 and 64, and shown in Figs. 37 and 38, respectively. Similarity searching using BLAST software revealed closest amino acid sequence similarity to phosphoglycerate kinase sequences but with significant differences.

Phosphoglycerate kinase (EC 2.7.2.3) is involved in the glycolysis pathway, and catalyzes the phospho-transfer reaction of ATP and 3-phospho-D- gly cerate to ADP and 3-phospho-D-glyceroyl phosphate (bacterial enzyme reviewed in Suzuki and Imahori, Methods in Enzymol. 90:126-130, 1982).

Applications for HNOOl phosphoglycerate kinase AK6 include:

• methods of enhanced survival of industrial processes;

• improved colonization of human intestinal environment; and • altered metabolic characteristics through changes in carbohydrate utilisation

Example 18 Isolation and Characterisation of Triosephosphate isomerase from L. rhamnosus HNOOl

HNOOl triosephosphate isomerase AK5 was isolated by a series of experiments designed to identify HNOOl strain proteins that were up-regulated in response to physiological stresses encountered during industrial processes. Cells were subjected to heat or osmotic shock, proteins radiolabeled with [³⁵S]- methionine and [ S]-cysteine (Amersham, USA), and cell-free extracts from shocked and non-shocked HNOOl cultures compared by 2-D analysis and N- terminal sequencing as described for Example 15 (HNOOl phosphoenolpyruvate hydratase AK4).

A protein up-regulated by heat and osmotic shock was N-terminal sequenced and the polypeptide sequence is given in SEQ ID NO: 76. This sequence was used to search an HNOOl sequence database using the TBLASTN program (NCBI) and the corresponding polynucleotide and polypeptide sequences are given in SEQ ID NOS: 21 and 63 and shown in Figs. 39 and 40, respectively. Similarity searching using BLAST software revealed closest amino acid sequence similarity to triosephosphate isomerase sequences but with significant differences. Fig. 39 shows the nucleotide sequence containing L. rhamnosus strain HNOOl triosephosphate isomerase AK5 showing ATG initiation and translation stop codons (boxed).

Triosephosphate isomerase (EC 5.3.1.1) is involved in the glycolysis pathway, and catalyzes the isomerisation reaction of D-glyceraldehyde 3- phosphate to glycerone phosphate (bacterial enzyme: Fahey et al, Biochem. J.

124:77P, 1971). Applications for HNOOl triosephosphate isomerase AK5 include:

• methods of enhanced survival of industrial processes; • improved colonization of human intestinal environment; and

• altered metabolic characteristics through changes in carbohydrate utilisation

Example 19 Isolation and Characterisation of Fructose-bisphosphate Aldolase from L. rhamnosus HNOOl

HNOOl fructose-bisphosphate aldolase AM8 was isolated by a series of experiments designed to identify HNOOl strain proteins that were up-regulated in response to physiological stresses encountered during industrial processes. Cells were subjected to heat or osmotic shock, proteins radiolabeled with [³⁵S]- methionine and [³⁵S] -cysteine (Amersham, USA), and cell-free extracts from shocked and non-shocked HNOOl cultures compared by 2-D analysis and N- terminal sequencing as described for Example 15 (HNOOl phosphoenolpyruvate hydratase AK4). A protein upregulated by heat and osmotic shock was N-terminal sequenced and the amino acid is given in SEQ ID NO: 77. This was used to search an HNOOl sequence database using the TBLASTN program (NCBI) and and the corresponding polynucleotide and polypeptide sequences are given in SEQ ID NOS: 29 and 71 and shown in Figs. 74 and 75, respectively. Fructose-bisphosphate aldolase (EC 4.1.2.13) is involved in the glycolysis pathway, and catalyzes the elimination reaction of D-Fructose 1,6-bisphosphate to glycerone phosphate and D-glyceraldehyde 3 -phosphate (bacterial enzyme reviewed in: Ujita and Kimura, Methods in Enzymol. 90: 235-241, 1982). Applications for HNOOl fructose-bisphosphate aldolase AM8 include: • methods of enhanced survival of industrial processes;

• improved colonization of human intestinal environment; and

• altered metabolic characteristics through changes i carbohydrate utilisation

Example 20

Isolation and Characterisation Phosphoryl Carrier Protein HPR from L. rhamnosus HNOOl

HNOOl phosphoryl carrier protein HPR AA9 was isolated by a series of experiments designed to identify HNOOl strain proteins that were up-regulated in response to physiological stresses encountered during industrial processes. Cells were subjected to heat or osmotic shock, proteins radiolabeled with [³⁵S]- methionine and [ S]-cysteine (Amersham, USA), and cell-free extracts from shocked and non-shocked HNOOl cultures compared by 2-D analysis and N- terminal sequencing as described for Example 15 (HNOOl phosphoenolpyruvate hydratase AK4).

A protein upregulated by heat and osmotic shock was N-terminal sequenced and the determined amino acid sequence is given in SEQ ID NO: 78.

This sequence was used to search an HNOOl sequence database using the TBLASTN program (NCBI) and the corresponding polynucleotide and polypeptide sequences are given in SEQ ID NOS: 4 and 45 and shown in Figures 41 and 42, respectively. Similarity searching using BLAST software revealed closest amino acid sequence similarity to phosphoryl carrier protein HPR sequences but with significant differences.

Phosphoryl carrier protein HPR is involved in the phosphoenolpyruvate:carbohydrate phosphotransferase system (PTS) responsible for the uptalce and phosphorylation of a number of carbohydrates (De Reuse et al,

Gene 35:199-207, 1985; Gonzy-Treboul et al, Mol. Microbiol. 3:103-112, 1989). PTS is also involved in the regulation of various bacterial functions by various mechanisms, including catabolite repression, ihducer exclusion, and inducer expulsion (reviewed in Postma et al, Microbiol. Rev. 57:543-594, 1993; Reizer et al, Crit. Rev. Microbiol. 15:297-338, 1988; Saier et al, Microbiol.

142:217-230, 1996). Applications for HNOOl phosphoryl carrier protein HPR AA9 include:

• methods of enhanced survival of industrial processes; • improved colonization of human intestinal environment;

• altered metabolic characteristics through changes in carbohydrate utilization; and

• control of catabolite regulation.

Example 21

Isolation and Characterisation of Chaperone Protein dnaK from L. rhamnosus

HNOOl

HNOOl dnaK chaperone protein AM9 was isolated by a series of experiments designed to identify HNOOl strain proteins that were up-regulated in response to physiological stresses encountered during industrial processes. Cells were subjected to heat or osmotic shock, proteins radiolabeled with [³⁵S]- metliionine and [³⁵S]-cysteine (Amersham, USA), and cell-free extracts from shocked and non-shocked HNOOl cultures compared by 2-D analysis and N- terminal sequencing as described for Example 15 (HNOOl phosphoenolpyruvate hydratase AK4). A protein up-regulated by heat and osmotic shock was N-terminal sequenced and the determined amino acid sequence is given in SEQ ID NO: 79. This sequence was used to search an HNOOl sequence database using the TBLASTN program (NCBI) and the corresponding polynucleotide and polypeptide sequences are given in SEQ ID NOS: 30 and 72 and shown in Figs. 76 and 77, respectively. Similarity searching using BLAST software revealed closest amino acid sequence similarity to chaperone protein dnaK sequences but with significant differences.

Chaperone protein dnaK is a 70 kDa heat shock protein (HSP). DnaK chaperones act by binding and protecting exposed regions on unfolded or partially folded protein chains, and are involved in reactivating proteins that become aggregated after heat shock (reviewed in Lund, Adv. Microbial Physiol. 44:93- 140, 2001). Overexpression may contribute to plasmid instability (Lobacz and olslca, Ada Microbiol. Pol. 46:393-397, 1977). Applications for HNOOl chaperone protein dnaK AM9 include:

• methods of enhanced survival of industrial processes;

• improved colonization of human intestinal environment;

• altered protein translation characteristics; and

• methods to control plasmid stability.

Example 22 Isolation and Characterisation of Gly ceraldehvde-3 -phosphate Dehydrogenase fro J. rhamnosus HNOOl

HNOOl glyceraldehyde-3 -phosphate dehydrogenase AK7 was isolated by a series of experiments designed to identify HNOOl strain proteins that were upregulated in response to physiological stresses encountered during industrial processes. Cells were subjected to heat or osmotic shock, proteins radiolabeled with [³⁵S]-methionine and [³⁵S]-cysteine (Amersham, USA), and cell-free extracts from shocked and non-shocked HNOOl cultures compared by 2-D analysis and N- terminal sequencing as described for Example 15 (HNOOl phosphoenolpyruvate hydratase AK4).

A protein up-regulated by heat and osmotic shock was N-terminal sequenced and the determined amino acid sequence is given in SEQ ID NO: 80. This sequence was used to search an HNOOl sequence database using the TBLASTN program (NCBI) and the corresponding polynucleotide and polypeptide sequences are given in SEQ ID NOS: 23 and 65, and shown in Figs. 43 and 44, respectively. Similarity searching using BLAST software revealed the closest amino acid sequence similarity to glyceraldehyde-3 -phosphate dehydrogenase sequences but with significant differences. A second experiment was also performed to identify surface layer proteins extracted from Lactobacillus rhamnosus HNOOl strain. Surface layer proteins from were extracted using the method of Turner et αl., J Bacteriol. 179:3310- 3316, 1997. Briefly, 100 ml stationary phase HNOOl culture was pelleted by centrifugation, washed with an equal volume of 0.15M NaCl, resuspended in 1 ml of 5M LiCl and kept on ice for 15 min. The crude lysate was centrifuge^'d at

13,000 rpm using a microcentrifuge and analyzed by SDS-PAGE on a 12.5% gel. To facilitate better extraction of surface layer proteins, freeze-dried DR20 was extracted with 0.2% SDS and 5M LiCl₂ as described by Brennan et al, Infect. Imm. 52:840-845, 1986 and Toba et al, J. Imm. Methods 182:193-207, 1995. After 1-D electrophoresis according to standard laboratory methods, gels were blotted on a PVDF membrane using a Semi-dry blotting apparatus (Bio-Rad). A major surface protein with molecular weight between 30 and 46 kDa was excised and N- terminal sequencing performed using a protein sequencer (Applied BioSystems, Model 476A). The determined N-terminal sequence was identical to that obtained from the heat and osmotic shock experiments as described above.

Therefore, HNOOl gene AKT encodes glyceraldehyde-3 -phosphate dehydrogenase, which is up-regulated by shock and is a major cell surface protein.

Glyceraldehyde-3 -phosphate dehydrogenase (EC 1.2.1.12) is part of the glycolytic pathway and catalyzes the redox reaction of D-Glyceraldehyde 3- phosphate, phosphate and NAD⁺ to 3-phospho-D-glyceroyl phosphate and NADH (for bacterial enzyme see Amelunxen, Methods in Enzymol. 41:268-273, 1975;

DAlessio and Josse, J Biol. Chem. 246:4326-4333, 1971). The enzyme has also been found to be a major cell-surface component of several bacterial species including Saccharomyces cerevisiae (Delgado et al, Microbiol. 147:411-417, 2001), Candida albicans (Gil-Navarro et al, J. Bacteriol. 179: 4992-4999, 1997) and group A Streptococci (Pancholi and Fischetti, Proc. Natl. Acad. Sci. USA

90:8154-8158, 1993). Applications for HNOOl glyceraldehyde-3-phosphate dehydrogenase AK7 include:

• flavor and aroma enhancement; • enhanced survival of industrial processes;

• prolonged survival in storage;

• improved colonization of human intestinal environment;

• enhanced textural properties;

• enhanced adhesion to intestinal cell surfaces; and • altered metabolic characteristics

Example 23

Isolation and Characterisation of Transcription Regulator sorR from L. rhamnosus

HNOOl

The full-length polynucleotide sequence of a transcription regulator sorR, given in SEQ ID NO: 24 and shown in Fig. 45, was used to amplify the AL3 transcription regulator sorR gene from L. rhamnosus HNOOl DNA using standard

PCR methodology. The upstream and downstream primers were tagged with BamHI and Pstl restriction endonuclease recognition sequences to facilitate cloning. The polypeptide sequence of AL3 is given in SEQ ID NO: 66 and shown in Fig. 46.

Full-length HNOOl sorR transcription regulator AL3 was cloned into

BamHI and Pstl cut pFX3 vector (an in-house E. coli/Lactococcus lactis shuttle vector as used in Xu et al, FEMS Microbiol. Lett. 61:55-59, 1991), and transformed into competent E. coli DH5α cells according to standard laboratory methods. Positive transformants were selected, grown overnight, and the plasmid construct isolated using a QlAprep Spin Miniprep Kit (Qiagen). The pFX3 construct encoding the HNOOl sorR transcription regulator AL3 was digested using the restriction enzymes EcoRI and Nr l, which released a 500 bp internal AL3 fragment that was cloned into the pBΕryl vector cut with EcoRI and Smal. The 3.6 kb pBΕryl vector was constructed using the replicon and multiple cloning site (MCS) from the phagemid pBlueScript (pBS-SK+) (Stratagene, La Jolla CA, USA). The ampicillin resistance gene in pBS-SK+ was removed by digestion with i?cαl (Roche, Auckland, New Zealand) and the 1,953 bp fragment containing the ColΕl origin and multiple cloning site purified and treated with Klenow enzyme (Roche) to give a blunt-ended fragment. A gene encoding resistance to erythromycin (Εm) was isolated on a 1.6 kb fragment obtained after cutting pVA891 (Macrina et al, Gene 25:145-50, 1983) with CM and H dIII and treatment with Klenow to give blunt ends. The 1.6 kb Εm fragment was ligated to the 1,953 bp pBS-SK+ fragment, transformed into E. coli TGI (Gibson TJ, Studies on the Epstein-Barr virus genome. Ph.D. Thesis, University of

Cambridge, Cambridge, England, 1984), and plated on LB agar plates containing 200 μg/ml Em. Maintenance of α-complementation for blue/white colour selection of recombinant pBEryl clones was confirmed by growing E. coli colonies on agar plates containing IPTG /X-gal. The resulting pBΕryl construct encoding the ΗN001 sorR transcription regulator AL3 gene was transformed into competent ΗN001 cells and grown anaerobically for 48 hrs at 37 °C on MRS lactobacilli agar (Difco, Detroit MI) containing 2.5 μg/ml Εm. Εrythromycin-resistant ΗN001 were checked for integration of the plasmid construct into the sorR gene by PCR using vector- specific (T3 or T7) and AL3 internal fragment-specific primers. Colonies giving

PCR patterns consistent with the insertional inactivation of the endogenous ΗN001 sorR transcription regulator AL3 gene were assessed for sorbose auxotrophy.

Auxotrophy of selected ΗN001 mutants for metabolism of sorbose was tested by growing pure cultures (1% inoculum) overnight at 37 °C on MRS agar in the presence of 1% sorbose or 1% glucose, compared to wild-type HNOOl and undefined mutant HNOOl (Em-resistant cultures with intact sorR transcription regulator AL3 gene, the result of a random integration event) cultures.

Results in Table 15 indicate that the AL3^~ HNOOl mutant strain failed to utilize sorbose as a carbon source in contrast to wild type HNOOl and undefined mutant HNOOl strain. This result was confirmed by growing pure cultures (1% inoculum) overnight at 37 °C in liquid MRS broth with 1%> sorbose or 1% glucose and measuring absorbance at 600 nm. Again, results showed a clear difference in growth between the AL3^~ mutant strain, and the wild-type and undefined mutant HNOOl strains containing intact the AL3 gene. Thus, the AL3 gene is required for sorbose metabolism in HN001 , and encodes the sorR transcriptional regulator.

Table 15. Results of assessment of sorbose auxotrophy.

+: growth; -: no growth

The sorR transcriptional regulator is required for the transcription of the sorbose operon, so regulating the utilization of L-sorbose as a carbon source, and its expression is induced by sorbose (Yebra et al, J. Bacteriol. 182:155-163, 2000; Sprenger and Lengeler, Mol. Gen. Genetics 209:352-359, 1987). Applications for the HNOOl sorR transcriptional regulator include: • Reagents for the control or modification of metabolic processes; and • Construction of sorbose-inducible HNOOl expression vectors using the sorR gene promoter Example 24

Isolation and Characterisation of Formamidopyrimidine-DNA-Glycosylase from

L. rhamnosus HNOOl

The full-length polynucleotide sequence of formamidopyrimidine-DNA- glycosylase (fpg) from L. rhamnosus strain HNOOl, given in SEQ ID NO: 25 and shown in Fig. 47 with ATG initiation and translation stop codons (boxed), was used to amplify the AL4 fpg gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and Sail restriction endonuclease recognition sequences to facilitate cloning.

AL4 was then cloned into the EcoRI and Sail sites of the pKK223-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain DH5α competent cells according to standard laboratory protocols. The polypeptide sequence of AL4 is given in SΕQ ID NO: 67 and shown in Fig. 48. Expression of the fpg AL4 protein was confirmed by SDS-PAGE analysis.

AL4 fpg activity was assayed according to the previously published methods (Duwat et al, Microbiol 141:411-417, 1995; Zhang et al, Nucleic Acids Res. 26:4669-4675, 1998) that examined the ability of fpg to suppress the spontaneous mutator phenotype of fpg or mutY mutants of E. coli. The E. coli strain CSH117 (Miller, in: A short course in Bacterial Genetics, Cold Spring

Harbor Press, Cold Spring Harbor, NY, 1992) that contained a mutated mutY gene was obtained from the E. coli Genetic Stock Centre (Yale University, USA) and transformed with the ρKK223-3 construct encodmg the HNOOl fpg AL4 gene according to standard laboratory methods. Positive transformants were selected according to ampicillin resistance, and used to innoculate 7 ml LB broth cultures containing 100 μg/ml ampicillin and incubated aerobically at 37 °C with shaking. Cultures containing pKK223-3 constructs encoding AL4 or empty pKK223-3 vector were grown to similar OD at 600 nm, serially diluted, and plated in triplicate on LB plates with and without 100 μg/ml rifampicin (Sigma). Plates were mcubated overnight at 37 °C and colonies counted. Results are shown in Table 16 as mean plate counts from three independent experiments.

Results in Table 16 indicate there was a significant difference in the frequency of mutations leading to rifampicin resistance in E. coli CSH117 transformed with pKK223-3 encoding HNOOl fpg AL4 and empty pKK22-3 vector (p < 0.001 by paired Student's t-test (1 -tailed)). Becuase the presence of AL4 suppressed the spontaneous mutation rate, it was concluded that AL4 encoded the HNOOl fpg protein.

Table 16. Spontaneous mutagenesis in E. coli CHS117 expressing the

HNOOl fpg AL4 gene.

The fpg protein (EC 3.2.2.23) is a DNA glycosylase/AP lyase that removes oxidized purine residues present in DNA, including the highly mutagenic

C8-oxo-guanine (8-oxoG) generated in DNA by active oxygen during metabolism (Laval et al, Mutation Res. 233:73-79, 1990; Boiteux et al, EMBO J. 6: 3177- 3183, 1987). The fpg protein exhibits three catalytic activities in vitro (Olga et al, J. Biol. Chem. 275:9924-9929, 2000): a DNA glycosylase that excises modified nucleotide bases (Laval et al, Mutation Res. 402:93-102, 1998), an AP lyase that incises DNA at abasic sites by an elimination mechanism, and a deoxyribophosphodiesterase that removes 5'-terminal deoxyribose phosphate residues. Applications for the HNOOl φg AL4 protein include: • Reagents or techniques to improve the survival of HNOOl in aerobic conditions;

• Enhanced survival of HNOOl in industrial processes; and

• Enhanced survival in the intestinal environment

Example 25

Isolation and Characterisation of Acetoin Dehydrogenase from L. rhamnosus

HNOOl

The full-length polynucleotide sequence of acetoin dehydrogenase from L. rhainnosis strain HNOOl, given in SEQ ID NO: 32 and shown in Fig. 49 with ATG initiation and translation stop codons (boxed), was used to amplify the API acetoin dehydrogenase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and Sail restriction endonuclease recognition sequences to facilitate cloning.

API was then cloned into the EcoRI and SaR sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain K12

XL-lBlue competent cells according to standard laboratory protocols. The polypeptide sequence of acetoin dehydrogenase API is given in SEQ ID NO: 74 and shown in Fig. 50. The acetoin dehydrogenase API protein was expressed as a fusion protein with glutathione S-transferase (GST), and purified using Glutathione Sepharose 4B resin (Pharmacia Biotech) according to the manufacturer's instructions. An aliquot of purified API -GST fusion protein was confirmed by SDS-PAGE analysis.

Acetoin dehydrogenase activity was assayed according to published methods (Rattray et al, Int. Dairy J. 10:781-789, 2000) with some modifications. Briefly, acetoin dehydrogenase activity was measured spectrophotometrically by monitoring the change in absorbance of the cofactor NADH at 340 nm. Aliquots of the purified AP4-GST fusion protein solution were added to reaction mixtures containing 50 mM 2[N-morpholino]ethanesulphonic acid (MES, Sigma) buffer pH 5:5 at 30 °C and the reactions started by the addition of 0.5 mM NADH and 37 mM diacetyl (Sigma) in a total volume of 1 ml. The change in optical density at

340 mn was measured, and rates of NADH utilization measured as an indicator of acetoin dehydrogenase activity. Enzyme activity was calculated as the amount of protein required to convert 1 μmol diacetyl and NADH to acetoin and NAD⁺ per minute at pH 5.5 at 30 °C. Enzyme activity of API-GST fusion protein was compared to that of an irrelevant GST-fusion protein, GST protein and elution buffer only.

Results in Fig. 51 and Table 17 indicate significant background utilization of NADH in the reactions. Similar rates were observed for elution buffer, GST protein and irrelevant fusion protein, indicating that the GST fusion protein did not exhibit acetoin reductase activity. Nonetheless, presence of the API -GST fusion protein gave significantly greater acetoin dehydrogenase activity than background, indicating that HNOOl API protein encodes acetoin dehydrogenase. Figure 51 shows the results of an acetoin reductase assay as measured by oxidation of NADH co-factor by OD at 340 nm in the presence of acetoin substrate. •, elution buffer only; ■, purified irrelevant GST-fusion protein; A, purified GST protein; ♦, purified API -GST fusion protein.

Table 17. Acetoin reductase activity of API GST-fusion protein compared to elution buffer, GST protein, and irrelevant GST-fusion protein controls.

Acetoin dehydrogenase (EC 1.1.1.5) catalyzes the reduction of diacetyl to acetoin, and acetoin to 2,3-butanediol as part of the pyruvate to 2,3-butanediol pathway (reviewed in Sarmiento and Burgos, Methods in Enzymol. 89:516-523,

1982). Diacetyl is an important flavor component in a variety of dairy products including butter, buttermilk, sour cream, fermented cream and cheese. Like its metabolites or related compounds acetoin, acetaldehyde and 2,3-butanediol, diacetyl plays a role in flavor when present in trace amounts (reviewed in

Escamilla-Hurtado et al, Rev. Latinoamerican Microbiol. 38:129-37, 1996). A mixture of all these compounds is produced during lactic acid fermentation, and particular proportions of these compounds lead to characteristic flavors in dairy products. Applications for HNOOl acetoin dehydrogenase API include:

• Methods to modulate the production of important flavor compounds;

• Techniques to modify pyruvate metabolic pathways;

• Industrial production of flavor compounds; and • Methods to control diacetyl levels in dairy products

Example 26

Isolation and Characterisation of Aflatoxin Bi Aldehyde Reductase from L. rhamnosus HNOOl

The full-length polynucleotide sequence of aflatoxin Bi aldehyde reductase from L. rhamnosus strain HNOOl, given in SEQ ID NO: 15 and shown with ATG initiation and translation stop codons (boxed) in Fig. 52, was used to amplify the AIT aflatoxin B_t aldehyde reductase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and SaR restriction endonuclease recognition sequences to facilitate cloning.

AIT was then cloned into the EcoRI and Sail sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain DH5α competent cells according to standard laboratory protocols. The polypeptide sequence of aflatoxin Bj aldehyde reductase AI7 is given in SEQ ID NO: 57 and shown in Fig. 53. The aflatoxin Bi aldehyde reductase AI7 protein was expressed as a fusion protein with glutathione S-transferase (GST) and purified using Glutathione Sepharose 4B resin (Pharmacia Biotech) according to the manufacturer's instructions. An aliquot of the purified AI7 protein was checked by SDS-PAGE analysis. AI7 activity was assayed by the previously published method of Ellis and

Hayes, Biochem. J. 312:535-541, 1995 with some modifications. Briefly, the aldehyde- and ketone reducing activity of aflatoxin B aldehyde reductase was assayed using 4-nitrobenzyl alcohol as substrate and NADPH as a cofactor. Enzyme activity was assessed spectrophotometrically by monitoring the utilization of NADPH at an OD of 340 nm. Reaction volumes of 1 ml containing

100 M sodium phosphate pH 6.6 and 0.2 mM NADPH were prepared, aliquots of purified AI7 protein added, and the changes in OD measured. Enzyme activity was compared between reactions containing AI7-GST fusion protein, irrelevant GST-fusion protein, elution buffer used during protein purification and water only. Enzyme activity was calculated as μmol NADP used/min ml.

Results in Fig. 54 and Table 18 indicate that HNOOl AI7 protein fused with GST exhibited significant aldehyde reductase activity, while the irrelevant GST-fusion protein, as well as the water and elution buffer controls, showed no activity whatsoever. Therefore, the aflatoxin B_\ aldehyde reductase activity of the AP4-GST fusion protein was due to the AI7 moiety rather than the GST. Also, the results showed increased rate of substrate utilization proportional to the amount of AI7 protein added, indicating that aflatoxin Bj aldehyde reductase activity of AI7 was dose dependent. Therefore, AI7 encodes HNOOl aflatoxin Bi aldehyde reductase. Fig. 54 shows the experimental results of aflatoxin Bi aldehyde reductase assay according to oxidation of the NADPH co-factor in the presence of acetoin substrate. X, water only; +, Sepharose column elution buffer only; •, irrelevant GST-fusion protein; ■, 10 μl purified AP4-GST fusion protein; A 20 μl purified AP4-GST fusion protein.

Table 18. aflatoxin Bi aldehyde reductase activity of AI7 GST-fusion protein compared to elution buffer, water and irrelevant GST-fusion protein controls.

/12506

104

Aflatoxin Bi aldehyde reductase metabolizes the carcinogen aflatoxin Bi (AFBi) by converting the protein-binding dialdehyde form of AFBj -dihydrodiol to the non-binding di-alcohol metabolite, and is associated with AFBj -resistance in animal studies (Ellis et al, Proc. Natl. Acad. Sci. USA 90:10350-10354, 1993; Hayes et al, Cancer Res. 53:3887-3894, 1993). The enzyme is also active against other substrates including a particular class of Icetone (Icetone groups on adjacent carbon atoms, eg. 9,10-phenanthrenequinone), as well as aromatic and aliphatic aldehydes (Ellis and Hayes, Biochem. J. 312:535-541, 1995). Applications for

HNOOl aflatoxin Bi aldehyde reductase AI7 include:

• Anti-carcinogenic or chemoprotectant reagents;

• Probiotic bacterial strains with anti-cancer effects;

• Research tools for cancer research;

• Enhanced flavor or aroma characteristics; • Removal of undesirable flavors; and

• Description and application of novel metabolic pathways

Example 27 Isolation and Characterisation of 6-Phospho-β-galactosidase from L. rhamnosus

HNOOl

The full-length polynucleotide sequence of 6-phospho-β-galactosidase, given in SEQ ID NO: 31 and shown with the translation stop codon (boxed) in Fig. 78, was used to amplify the A05 6-phospho-β-galactosidase gene from L. rhamnosus HNOOl DNA using standard PCR methodology. The upstream and downstream primers were tagged with EcoRI and Sail restriction endonuclease recognition sequences to facilitate cloning.

A05 was then cloned into the EcoRI and Sail sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain DH5 competent cells according to standard laboratory protocols. The polypeptide sequence of aflatoxin B\ aldehyde reductase AO5 is given in SEQ ID NO: 73 and shown in Fig. 79. The 6-phospho-β-galactosidase AO5 protein was expressed as a fusion protein with glutathione S-transferase (GST) and purified using Glutathione Sepharose 4B resin (Pharmacia Biotech) according to the manufacturer's instructions. An aliquot of the purified AO5 protein was checked by SDS-PAGE analysis.

A05 activity was assayed using standard laboratory methods as follows. Briefly, crude cell lysates were prepared by resuspending a 10 ml overnight culture of E. coli DH5α cells in 1 ml lysis buffer (50 mM potassium phosphate pH 7.8, 400 mM NaCl, 100 mM KCl, 10 % glycerol, 0.5% Triton X-100, 10 mM imidazole). Cells were sonicated and spun to sediment cell debris according to standard laboratory methods. Aliquots of 50 μl of cell lysate were added to 900 μl reaction buffer (100 mM KH₂PO₄ pH 7.0, 2 mM MgCl₂) and 50 μl substrate O- nitrophenyl β-D-glycopyranoside (ONPG) (Sigma). Utilization of OPNG was measured spectrophotometrically by monitoring change in absorbance at 420 nm and enzyme activity calculated as μmol OPNG used/min/ml.

6-Phospho-β-galactosidase enzyme activity was compared in crude lysates from E. coli DH5α transformed with pGEX-6P-3 encoding A05, pGEX-6P-3 encoding an irrelevant protein, and lysis buffer only. Experimental results in Fig. 55 and Table 19 indicate that while reactions containing crude lysates from cells transformed with an irrelevant GST-fusion protein or lysis buffer only exhibited little or no enzyme activity, crude lysate from E. coli expressing AO5-GST fusion protein showed significant enzyme activity. Fig. 55 shows the experimental determination of 6-Phospho-β- galactosidase enzyme activity as measured by substrate utilisation using crude lysates of strains transformed with pGex-6P-3 encoding A05 (♦), pGex-6P-3 encoding an irrelevant protein (■), or using lysis buffer only (X). Table 19. 6-Phospho-β-galactosidase enzyme activity in crude cell lysates

Enzyme activity was also measured in increasing amounts of crude cell lysates to assess dose-dependency. Results shown in Fig. 56 and Table 20 indicate that increasing amounts of cell lysates from cells expressing the AO5-

GST fusion protein led to proportional increases in 6-phospho-β-galactosidase enzyme activity. Therefore, A05 encodes HNOOl 6-phospho-β-galactosidase. Fig. 56 shows 6-Phospho-β-galactosidase enzyme activity as measured experimentally by substrate utilisation using increasing amounts of crude lysate from strains transformed with pGex-6P-3 encoding A05-GST fusion protem. ♦,

50 μl lysate; ■, 100 μl lysate; A, 200 μl lysate; •, 200 μl lysis buffer only.

Table 20. 6-Phospho-β-galactosidase enzyme activity in increasing amounts of crude cell lysates.

6-Phospho-β-galactosidase (EC 3.2.1.85) catalyzes the hydrolysis of O- glycosyl bonds of 6-phospho-beta-D-galactosides to give alcohols and 6-phospho- D-galactose, and is involved in lactose utilization (Hengstenberg and Morse, Methods in Enzymol. 42:491-494, 1975). Applications for HNOOl 6-phospho-β- galactosidase A05 include:

• flavor and aroma enhancement; • nutritional enhancement;

• altered bacterial metabolic/growth characteristics; and

• removal of bitter or undesirable flavors

Example 28 Isolation and Characterisation of Aromatic Aminotransferase from L. rhamnosus

HNOOl

The full-length polynucleotide sequence of aromatic aminotransferase of L.rhamnosus strain HNOOl, given in SEQ ID NO: 11 and shown in Fig. 57 with ATG initiation and translation stop codons (boxed), was used to amplify the AHT aromatic aminotransferase gene from L. rhamnosus HNOOl DNA using standard

PCR methodology. The upstream and downstream primers were tagged with EcoRI and Sail restriction endonuclease recognition sequences to facilitate cloning.

AHT was then cloned into the EcoRI and SaR sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into E. coli strain DH5α competent cells according to standard laboratory protocols. The polypeptide sequence of aflatoxin Bj aldehyde reductase AI7 is given in SEQ ID NO: 53 and shown in Fig. 58. The aflatoxin Bi aldehyde reductase AI7 protein was expressed as a fusion protem with glutathione S-transferase (GST) and purified using Glutathione Sepharose 4B resin (Pharmacia Biotech) according to the manufacturer's instructions. An aliquot of the purified AI7 protein was checked by SDS-PAGE analysis.

Aromatic aminotransferase activity was assayed according to previously published methods (Yvon et al, Appl. Environ. Microbiol. 63:414-419, 1997) with modifications. The assay is composed of two parts: the first is an aminotransferase reaction using the aromatic amino acid phenylalanine as substrate and results in the production of glutamate from α-ketoglutarate. The second part of the assay is the colorimetric determination of the glutamate. For the phenylalanine transamination, 250 μl reaction mixtures containing 70 mM 02/12506

108

Tris-HCl pH 8.0, 3 mM L-phenylalanine, 10 mM α-ketoglutarate and 0.05 μM pyridoxal 5' phosphate were incubated with purified proteins or elution buffer at 37 °C for 15 min. Aliquots of 20 μl were then taken and glutamate levels determined by adding to a reaction mixture containing 65 mM Tris pH 9.0, 1.3 mM EDTA, 40 mM hydrazine, 19.5 mM NAD⁺, 65 mM ADP, with and without 2.4 U glutamate dehydrogenase in a total volume of 250 μl in the wells of a microtitre plate. Reactions were incubated at 37 °C for 40 min and absorbance at 340 nm measured using a plate reader (Molecular Devices, Sunnyvale CA). Enzyme activity of the purified AH7-His-Thio fusion protein was compared a purified irrelevant His-Thio-fusion protein, elution buffer used to elute the purified proteins from the Ni-NTA columns and water only and results are shown in Table 21. Glutamate concentrations were calculated using a standard curve, and assays on all samples and standards were performed in triplicate. Enzyme activities were calculated as μmol glutamate produced/min/ml and specific activities calculated using protein concentrations obtained using the BCA protein assay kit (Pierce) according to the manufacturer's instructions. Results indicate that while the irrelevant fusion protein, elution buffer and water resulted in little glutamate production, AP5 fusion protein exhibited significant aminotransferase activity using phenylalanine as substrate. Therefore, HNOOl AP5 encodes an aromatic amino acid transferase.

Table 21. Aromatic amino acid transferase activity in HNOOl AH7 purified protein as measured by glutamate production.

12506

109

Aromatic amino acid transferase (EC 2.6.1.57) catalyzes the transfer of amino groups between an aromatic amino acid and α-ketoglutarate to its aromatic oxo-acid and L-glutamate (Mavrides and Orr, J. Biol. Chem. 250:4128-4133, 1975). The products of enzymatic amino acid degradation play a major role in cheese flavor development. Degradation products from aromatic amino acids have both positive and negative impacts on cheese flavor (Dunn and Lindsay, J Dairy

Sci. 68:2859-2874, 1985; Engels et al, Int. Dairy J. 7:225-263, 1997). Therefore, the applications of HNOOl aromatic amino acid aminotransferase AP5 include: • flavor and aroma enhancement;

• removal of off-flavors;

• altered levels of biogenic amines; and

• altered metabolic characteristics.

Example 29

Isolation and Characterisation of Acetate Kinase from L. rhamnosus HNOOl

The full-length polynucleotide sequence of acetate kinase, given in SEQ ID NO: 33 and shown in Fig. 59 with ATG initiation and translation stop codons (boxed), was used to amplify the AP5 acetate kinase gene from L. rhamnosus strain. The upstream and downstream primers were tagged with EcoRI and Sail restriction endonuclease recognition sequences to facilitate cloning.

AP5 was then cloned into the EcoRI and Sa l sites of the pGΕX-6P-3 expression vector (Pharmacia Biotech) and transformed into the E. coli strain IC12 XL-lBlue competent cells according to standard laboratory protocols. The polypeptide sequence of the acetate kinase AP5 polypeptide is given in SEQ ID

NO: 75 and shown in Fig. 60 and was expressed as a fusion protein with glutathione S-transferase (GST) and purified using Glutathione Sepharose 4B resin (Pharmacia Biotech) according to the manufacturer's instructions. An aliquot of the purified AP5 protein was checked by SDS-PAGE analysis. 02/12506

110

AP5 activity was assayed based on a published method for analysis of the related carbamate kinase (Crow and Thomas, J Bacteriol. 150:1024-1032, 1982) with modifications. Briefly, the assay uses a couple reaction such that acetyl phosphate and ADP is converted to CO₂, NH₃ and ATP in the presence of acetate kinase. The produced ATP is then combined with glucose by the enzyme hexolcinase to give glucose-6-phosphate, which in turn is reduced by glucose-6- phosphate dehydrogenase using the NADP⁺ cofactor. Because the hexolcinase glucose-6-phosphate dehydrogenase enzymes are provided in excess, acetate kinase activity can be assessed spectrophotometrically by monitoring NADPH production at an OD of 340 nm. Reaction mixtures of 730 μl 200 mM Tris-HCL pH 7.9, 73 μl 200 mM acetyl phosphate, 36.5 μl 200 mM ADP, 36.5 μl 200 mM

MgCl₂, 73 μl 500 mM glucose, 7 μl 100 mM NADP⁺ and 7 μl hexolcinase glucose-6-phosphate dehydrogenase were prepared and allowed to equilibrate at 37 °C. Purified AP5-GST fusion protein and sterile milliQ water was added to a final volume of 1 ml, and changes in OD at 340 nm measured. Enzyme activity was compared between purified AP5-GST fusion protein, irrelevant fusion protein, and elution buffer used to elute the purified proteins off the Sepharose column and the results is shown in Table 22. Enzyme activities were calculated as μmol NAPDH produced/min/ml, and specific activities calculated using protein concentrations obtained using the BCA protein assay kit (Pierce) according to the manufacturer's instructions. Results in Table 22 indicate that while elution buffer and irrelevant GST-fusion protein showed little or no enzyme activity, the AP5- GST fusion protein exhibited significant activity. Therefore, AP5 encodes HNOOl acetate kinase.

Table 22. Acetate kinase activity of HNOOl protein AP5.

2/12506

111

Acetyl kinase (EC 2.7.2.1) catalyzes the phosphotransfer between ADP and acetyl phosphate to give ATP and acetate (Nishimura and Griffith, Methods in En ymol. 71:311-316, 1981). Acetate, a flavor compound in its own right, can give ammonia and carbon dioxide, both of which have important flavor and texture impacts in cheese (Fox et al, Crit. Rev. Food Sci. Nutr. 29:237-53,

1990). Applications for HNOOl acetate kinase AP5 include:

• flavor and aroma enhancement;

• removal of off-flavors; • altered texture characteristics; and

• altered metabolic characteristics

Example 30 Isolation and Characterisation of Basic Surface Protein from L. rhamnosus HNOOl

The full-length polynucleotide sequence of basic surface protein from L. rhamnosus strain HNOOl, given in SEQ ID NO: 6 and shown in Fig. 61 with ATG initiation and translation stop codons (boxed), was used to amplify the AC9 basic surface protein gene, but excluding the predicted N-terminal Type II signal sequence. The primer sequences used are given in SEQ ID NOS: 34 and 35 and were tagged with EcoRI and BamHI restriction endonuclease recognition sequences, respectively, to facilitate cloning. AC9 sequence was then amplified from HNOOl strain genomic DNA, purified, cloned into EcoRI 5αmHI-cut pGΕX- 6P-3 expression vector, and transformed into E. coli DH5α cells according to standard laboratory methods. The polypeptide sequence of basic surface protein

AC9 is given in SΕQ ID NO: 47 and shown in Fig. 62. The basic surface protein AC9 was expressed as a fusion protein with glutathione S-transferase (GST), bound to Glutathione Sepharose 4B resin (Pharmacia Biotech), and PreScission 2506

112

protease used to cleave off the basic surface protein AC9 protein, according to the manufacturer's instructions. An aliquot of the purified AC9 protein was checked by SDS-PAGE analysis.

Purified AC9 protein (14 μg) was labeled by radio-iodination with 0.1 mCi iodine- 125 (Amersham Pharmacia) using IODO-BEADS iodination reagent (Pierce) following the manufacturer's instructions. The radio-iodinated protein was separated from unincorporated iodine- 125 and excess sodium iodide- 125 using a PD-10 desalting column (Amersham-Pharmacia) according to the manufacturer's instructions, except that the elution was performed in phosphate buffered saline in twelve 500 μl aliquots. Radioactivity in eluted fractions was quantitated on a Bioscan Quick Count QC-4000/XER Benchtop Radioisotope

Counter (Bioscan, Inc.) and fractions containing the first peak of radioactivity (corresponding to labeled AC9 protein) were pooled and bovine serum albumin added to a final concentration of 10 mg/ml.

To analyze the binding of polypeptide AC9 to proteins associated with intestinal surface proteins Icnown to act as ligands for bacterial adhesins, different intestinal protein ligands were dot blotted onto a nitrocellulose membrane using a Convertible Filtration Manifold System (Life Technologies) following the manufacturer's instructions. Duplicate dots of approximately 1 μg of type I collagen from calf skin, type IV collagen from human placenta, fibronectin from human plasma, laminin from the basement membrane of Engelbreth-Holm-Swarm mouse sarcoma and type III mucin partially purified from porcine stomach and bovine serum albumin included as a negative control (all proteins were obtained from Sigma) were blotted. The blot was incubated at room temperature on an orbital shaker in 10 ml phosphate buffered saline, pH 7.4, containing 0.1% Tween 20 and 5 mg/ml bovine serum albumin for 1 hour. Radio-iodinated AC9 protein was then added to a final concentration of approximately 500 ng/ml, and incubated at room temperature for a further hour. The blot was washed three times in approximately 40 ml phosphate buffered saline, pH 7.4, containing 0.1% Tween 20 at room temperature for 10 minutes, then autoradiographed against X- ray film at -80 C overnight. The autoradiograph was developed and the resulting /12506

113

image digitized with a FluorS Multilmager (BioRad). Binding by AC9 protein to the intestinal protein ligands was quantitated using Bio-Rad Quantity One software by measuring the density of the signal on the autoradiograph resulting from radiolabeled AC9 protein binding to the different ligands and subtracting the background density of blank film. To quantitate relative amounts of protein ligands blotted, blots were stained with Ponceau S using standard procedures

(Ausubel et al, Current Protocols in Molecular Biology, John Wiley & Sons, 2001), and quantitated as for the autoradiograhs. The density corresponding to AC9 protein binding to individual ligands was divided by the average density of Ponceau S staining of the ligand to give the relative AC9 bound to each ligand. Binding of iodinated AC9 (1.50 xlO⁷ dpm) was compared to binding of iodinated mucus adhesion promoting (mapA) protein of Lactobacillus reuteri (GenBank accession number AJ293860) as a positive control, and iodinated irrelevant HNOOl protein (7.00 X lO⁶ dpm) as a negative control.

Results in Table 23 indicate that while the irrelevant HNOOl protein did not bind to any of the intestinal adhesin ligands, both the AC9 protein and the positive control protein mapA showed significant binding to mucin. Therefore, AC9 encodes the HNOOl basic surface protein.

Table 23. Density of autoradiographic signals from AC9 basic surface protein binding to dot blots of intestinal proteins, compared to a positive control (mapA) and negative control (irrelevant HNOOl protein). Results represent mean of relative density of two dots.

The basic surface protein of Lactobacillus fermentum is a surface-bound molecule that belongs to a family of ATP-binding cassette (ABC) receptor solute binding proteins (Turner et al, J. Bacteriol. 179:3310-3316, 1997; Tarn et al, 506

114

Microbiol. Rev. 57:320-346, 1993). Basic surface protein has also been shown to be involved in cysteine uptalce (Turner et al, J. Bacteriol. 181.2192-2198, 1999) and has been used as an attachment site for immunodominant proteins in the development of new vaccine strategies (Turner et al, Infect. Imm. 67:5486-5489, 1999). Applications for HNOOl basic surface protein AC9 include:

• enhanced adhesion to intestinal surface and cell lines;

• enhanced survival in intestinal environment;

• altered metabolic characteristics;

• altered flavor or aroma characteristics; • enhanced probiotic effects;

• reagents to block or modify adherence of bacteria to mucosal surfaces; and

• development of vaccine carriers,

Example 31

Isolation and Characterisation of Outer Membrane Protein A from L. rhamnosus

HNOOl

The full-length polynucleotide sequence of outer membrane protein A from L. rhamnosus strain HNOOl, given in SEQ ID NO: 27 and shown in Fig. 63 with ATG initiation and translation stop codons (boxed) was used to amplify the

N-terminal region of AL8 outer membrane protein A gene. The primer sequences are given in SEQ ID NOS: 36 and 37, respectively, and were tagged with BamHI and Xhόl restriction endonuclease recognition sequences, respectively, to facilitate cloning. AL8 sequence was then amplified from HNOOl strain genomic DNA, purified, cloned into BamHI/XhoI-cut pGEX-6P-3 expression vector, and transformed into E. coli DH5α cells according to standard laboratory methods. The polypeptide sequence of outer membrane protein A AL8 is given in SΕQ ID NO: 69 and shown in Fig. 64. The outer membrane protein A AL8 was expressed as a fusion protein with glutathione S-transferase (GST), bound to Glutathione Sepharose 4B resin (Pharmacia Biotech), and PreScission protease used to cleave off the outer membrane protein A ALS protein, according to the manufacturer's instructions. An aliquot of the purified AL8 protein was checked by SDS-PAGE analysis.

Purified AL8 protein (20 μg) was then labeled by radio-iodination with 0.1 mCi iodine-125 (Amersham Pharmacia) using IODO-BEADS iodination reagent (Pierce) following the manufacturer's instructions. Radio-iodinated protein was separated from unincorporated iodine-125 and excess sodium iodide-125 using a PD-10 desalting column (Amersham-Pharmacia) according to the manufacturer's instructions, except that the elution was performed in phosphate buffered saline, in twelve 500 μl aliquots. Radioactivity in eluted fractions was quantitated on a Bioscan Quick Count QC-4000/XER Benchtop Radioisotope Counter (Bioscan,

Inc.) and fractions containing the first peak of radioactivity (corresponding to labeled AL8 protein) were pooled and bovine serum albumin added to a final concentration of 10 mg/ml.

To analyze AL8 protein binding to proteins associated with intestinal surface proteins known to act as ligands for bacterial adhesins, different intestinal protein ligands were dot blotted onto a nitrocellulose membrane using a Convertible Filtration Manifold System (Life Technologies) following the manufacturer's instructions. Duplicate dots of approximately 1 μg of type I collagen from calf skin, type IV collagen from human placenta, fibronectin from human plasma, laminin from the basement membrane of Engelbreth-Holm-Swarm mouse sarcoma and type III mucin partially purified from porcine stomach and bovine serum albumin included as a negative control (all proteins were obtained from Sigma) were blotted. The blot was incubated at room temperature on an orbital shaker in 10 ml phosphate buffered saline, pH 7.4, containing 0.1 % Tween 20 and 5 mg/ml bovine serum albumin for 1 hour. Radio-iodinated AL8 protein was then added to a final concentration of approximately 500 ng/ml, and incubated at room temperature for a further hour. The blot was then washed three times in approximately 40 ml phosphate buffered saline, pH 7.4, containing 0.1% Tween 20 at room temperature for 10 minutes, then autoradiographed against X- ray film at -80 C overnight. The autoradiograph was developed and the resulting image digitised with a FluorS Multilmager (BioRad). Binding by AL8 protein to the intestinal protein ligands was quantitated using Bio-Rad Quantity One software by measuring the density of the signal on the autoradiograph resulting from radio labelled AL8 protein binding to the different ligands and subtracting the background density of blank film. To quantitate relative amounts of protein ligands blotted, blots were stained with Ponceau S using standard procedures

(Ausubel et ah, Current Protocols in Molecular Biology, John Wiley & Sons, 2001), and quantitated as for the autoradiograhs. The density corresponding to AL8 protein binding to individual ligands was divided by the average density of Ponceau S staining of the ligand to give the relative AL8 bound to each ligand. Binding of iodinated AL8 (3.2 x 10⁷ DPM) was compared to binding of iodinated mucus adhesion promoting (mapA) protein of Lactobacillus reuteri (GenBank accession number AJ293860) (6.6 x 10 dpm) as a positive control, and iodinated irrelevant HNOOl protein (7.0 x 10⁶ DPM) as a negative control.

Results in Table 24 indicated that while the irrelevant HNOOl protein did not bind to any of the intestinal adhesin ligands, both the AL8 protein and the positive control protein map A showed significant binding to mucin. Therefore, AL8 encodes the HNOOl outer membrane protein A.

Table 24. Density of autoradiographic signals from AL8 outer membrane protein A binding to dot blots of intestinal proteins, compared to a positive control (mapA) and negative control (irrelevant HNOOl protein). Results represent mean of relative density of two dots.

The outer membrane protein A of Rickettsia spp. is a 190 lcDa surface bound molecule required for the adhesion of Rickettsia to host cells (Li and Walker, Microbial Path. 179:3310-3316, 1998). Rickettsial outer membrane protein A is also an immunodominant protein and has been used for the serotyping of riclcettsial strains (Philip et ah, J. Imm. 121:1961-1968, 1978). Applications for HNOOl outer membrane protein AL8 include:

• enhanced adhesion to intestinal surface and cell lines; • enhanced survival in intestinal environment;

• altered texture characteristics;

• enhanced probiotic effects;

• reagents to block or modify adherence of bacteria to mucosal surfaces; and • development of vaccine carriers

Example 32 Isolation and Characterisation of Extracellular Matrix Binding Protein from L. rhamnosus HNOOl

The full-length polynucleotide sequence of extracellular matrix binding protein, AM4, from L. rhamnosus strain HNOOl, given in SEQ ID NO: 28 and shown in Fig. 65, was used to amplify the N-terminal region of AM4 extracellular matrix binding protein gene. The primer sequences used are given in SEQ ID NOS: 38 and 39, respectively, and were tagged with EcoRI and Notl restriction endonuclease recognition sequences, respectively, to facilitate cloning. AM4 sequence was then amplified from HΝ001 strain genomic D A, purified, cloned into EcoRI/Notl-cut pGΕX-6P-3 expression vector, and transformed into E. coli DH5α cells according to standard laboratory methods. The polypeptide sequence of extracellular matrix binding protein AM4 is given in SEQ ID NO: 70 and shown in Fig. 66. The extracellular matrix binding protein AM4 was expressed as a fusion protein with glutathione S-transferase (GST) and purified using Glutathione Sepharose 4B resin (Pharmacia Biotech), according to the manufacturer's instructions. An aliquot of the purified AM4-GST fusion protein was checked by SDS-PAGE analysis. Purified AM4 protein (10 μg) was labeled by radio-iodination with 0.1 mCi iodine-125 (Amersham Pharmacia) using IODO-BEADS iodination reagent (Pierce) following the manufacturer's instructions. Radio-iodinated protein was separated from unincorporated iodine-125 and excess sodium iodide-125 using a PD-10 desalting column (Amersham-Pharmacia) according to the manufacturer's instructions, except that the elution was performed in phosphate buffered saline in twelve 500 μl aliquots. Radioactivity in eluted fractions was quantitated on a Bioscan Quick Count QC-4000/XER Benchtop Radioisotope Counter (Bioscan, Inc.) and fractions containing the first peak of radioactivity (corresponding to labeled AM4 protein) were pooled and bovine serum albumin added to a final concentration of 10 mg/ml.

To analyze binding of the AM4 protein proteins associated with intestinal surface proteins known to act as ligands for bacterial adhesins, different intestinal protein ligands were dot blotted onto a nitrocellulose membrane using a Convertible Filtration Manifold System (Life Technologies) following the manufacturer's instructions. Duplicate dots of approximately 1 μg of type I collagen from calf skin, type IV collagen from human placenta, fibronectin from human plasma, laminin from the basement membrane of Engelbreth-Holm-Swarm mouse sarcoma and type III mucin partially purified from porcine stomach and bovine serum albumin included as a negative control (all proteins were obtained from Sigma) were blotted. The blot was incubated at room temperature on an orbital shaker in 10 ml phosphate buffered saline, pH 7.4, containing 0.1 % Tween 20 and 5 mg/ml bovine serum albumin for 1 hour. Radio-iodinated AM4 protein was then added to a final concentration of approximately 500 ng/ml, and incubated at room temperature for a further hour. The blot was then washed three times in approximately 40 ml phosphate buffered saline, pH 7.4, containing 0.1 %

Tween 20 at room temperature for 10 minutes, then autoradiographed against X- ray film at -80 C overnight. The autoradiograph was developed and the resulting image digitised with a FluorS Multilmager (BioRad). Binding by AM4 protein to the intestinal protein ligands was quantitated using Bio-Rad Quantity One software by measuring the density of the signal on the autoradiograph resulting from radiolabeled AM4 protein binding to the different ligands and subtracting the background density of blank film. To quantitate relative amounts of protein ligands blotted, blots were stained with Ponceau S using standard procedures (Ausubel et ah, Current Protocols in Molecular Biology, John Wiley & Sons, 2001), and quantitated as for the autoradiograhs. The density corresponding to AM4 protein binding to individual ligands was divided by the average density of

Ponceau S staining of the ligand to give the relative AM4 bound to each ligand. Binding of iodinated AM4-GST fusion protein (3.3 xlO⁷ DPM) was compared to binding of iodinated mucus adhesion promoting (mapA) protein of Lactobacillus reuteri (GenBank accession number AJ293860) (6.6 x 10⁶ DPM) as a positive control, and iodinated irrelevant HNOOl protein (7.0 x 10⁶ dpm) as a negative control.

Results in Table 25 indicate that while the irrelevant HNOOl protein did not bind to any of the intestinal adhesin ligands, the AM4 fusion protein gave a very similar binding pattern to the positive control protein mapA, with significant binding to mucin and collagen types I and IV. Therefore, AM4 encodes the

HNOOl extracellular matrix binding protein.

Table 25. Density of autoradiographic signals from AM4-GST fusion protein to dot blots of intestinal proteins, compared to a positive control

(mapA) and negative control (irrelevant HNOOl protein). Results represent mean of relative density of two dots.

The extracellular matrix binding protein is a surface bound molecule required for the adhesion of Streptococcus spp. to the extracellular matrix, exposed during tissue injury (Manganelli and van de Rijn, Infect. Imm. 67:50-56, 1999). Applications for HNOOl extracellular matrix binding protein AM4 include:

• enhanced adhesion to intestinal surface and cell lines;

• enhanced survival in intestinal environment; • altered texture characteristics;

• enhanced probiotic effects;

• reagents to block or modify adherence of bacteria to surfaces; and

• development of vaccine carriers

Example 33

Isolation and Characterisation of High-Molecular- Weight Adhesion Protein from

L. rhamnosus HNOOl

The full-length polynucleotide sequence of high-molecular-weight adhesion protein, AL7, from L. rhamnosus strain HNOOl given in SEQ ID NO: 26 and shown in Fig. 67 with ATG initiation and translation stop codons (boxed), was used to amplify the N-terminal region of ALT high-molecular- weight adhesion protein gene. The primer sequences used are given in SEQ ID NOS: 40 and 41, respectively, and were tagged with BamHI and EcoRI restriction endonuclease recognition sequences, respectively, to facilitate cloning. ALT sequence was then amplified from HNOOl strain genomic DNA, purified, cloned into _5αmHI/EcoRI-cut pGΕX-6P-3 expression vector, and transformed into E. coli DH5α cells according to standard laboratory methods. The polypeptide sequence of high-molecular- weight adhesion protein AL7 is given in SEQ ID NO:

68 and shown in Fig. 68. The high-molecular-weight adhesion protein AL7 was expressed as a fusion protein with glutathione S-transferase (GST) and expression was checked by SDS-PAGE analysis.

Lysates of DH5α clones containing pGEX-6P-3 expressing AL7-GST fusion protein, lysates of DH5α clones containing pGEX-6P-3 expressing irrelevant HNOOl GST-fusion protein, and crude cell wall cytoplasmic HNOOl protein preparations (prepared by standard laboratory methods) were separated by SDS-PAGE. Proteins were blotted onto nitrocellulose membranes using a Trans-Blot SD Semi-Dry Electrophoretic Transfer Cell (Bio-Rad) according to the manufacturer's instructions. The nitrocellulose blot was then blocked overnight at 4 C in phosphate buffered saline, pH 7.4, 0.1%) Tween 20 (PBS-T), containing 5% non-fat dried milk. Rabbit anti-sera raised against HNOOl cell wall proteins (supplied by Dr. Paul O'Toole, Institute of Molecular Biosciences, Massey University, Palmerston North, New Zealand) were diluted 1:5000 in PBS-T, 5%o non-fat dried milk and incubated with the blot for 1 hr at room temperature. The blot was washed three times for 15 min each in PBS-T and incubated at room temperature in 50 ml PBS-T, 5% non-fat dried milk containing a 1 :3000 dilution of a horseradish peroxidase-labeled antibody against rabbit Ig (Amersham Pharmacia) for 20 min. The blot was washed six times in PBS-T at room temperature for 15 min each, and binding visualized using the ECL Western blotting detection system (Amersham Pharmacia) according to the manufacturer's instructions.

Results of the Western blot revealed that the anti-sera detected a number of proteins from HNOOl raised against the HNOOl cell wall preparations. While several of these proteins were found in both the cell wall and cytoplasmic preparations of HNOOl, these proteins consisted of bands of approximately 66 lcDa and less. In addition, a number of high molecular weight protein bands were detected in the HNOOl cell wall protein preparations that were not present in the HNOOl cytoplasmic protein preparations. These bands ranged from approx. 130 kDa to approx. 220 lcDa or greater. Therefore the cell wall antisera specifically detected several large cell wall proteins from HNOOl. Of the E. coli extracts, the only signal came from the lysate of the DH5α clone containing pGEX-6P-3 expressing the N-terminal region of AL7. This strong band was approximately 97 kDa, the same size as the AL7-GST fusion protein. Lysates from E. coli clones expressing unrelated proteins showed no cross-reactivity with the HNOOl cell wall anti-sera. This data indicates that ALT encodes a high- molecular- weight adhesion protein at the cell surface. The high-molecular-weight adhesion protein is a homologue of the surface-bound molecule of Haemophilus influenzae shown to be involved in adhesion to human cell lines (Barenkamp and St Geme, Mol. Microbiol. 19: 1215- 1223, 1996; St Geme et al, Proc. Natl. Acad. Sci. USA 90:2875-2879, 1993). Applications for HNOOl high-molecular- weight adhesion protein AL7 include:

• enhanced adhesion to intestinal surfaces and cell lines;

• enhanced survival in intestinal environment;

• altered texture characteristics;

• enhanced probiotic effects; • reagents to block or modify adherence of bacteria to surfaces; and

• development of vaccine carriers.

Example 34 Isolation and Characterisation of Periplasmic Binding Protein 1 (PEBl) from L. rhamnosus HNOOl

The full-length polynucleotide sequence of a periplasmic binding protein 1 (PEBl), AJ4, from L. rhamnosus strain HNOOl, given in SEQ ID NO: 16 and shown in Fig. 69 with ATG initiation and translation stop codons (boxed), was used to amplify the AJ4 PEBl gene from HNOOl strain genomic DNA by PCR according to standard laboratory methods. Primers were tagged with BamHI and EcoRI to facilitate cloning. AJ4 PCR products were purified, cloned into _9αmHI/EcoRI-cut pGΕX-6P-3 expression vector, and transformed into E. coli DH5α cells according to standard laboratory methods. The polypeptide sequence of PEBl AJ4 is given in SEQ ID NO: 58 and shown in Fig. 70. The PEBl AJ4 was expressed as a fusion protein with glutathione S-transferase transferase (GST), bound to Glutathione Sepharose 4B resin (Pharmacia Biotech), and PreScission protease used to cleave off the PEBl AJ4 protein, according to the manufacturer's instructions. An aliquot of the purified AJ4 protein was checked by SDS-PAGE analysis. Purified AJ4 protein (10 μg) was then labeled by radio-iodination with 0.1 mCi iodine-125 (Amersham Pharmacia) using IODO-BEADS iodination reagent (Pierce) following the manufacturer's instructions. Radio-iodinated protein was separated from unincorporated iodine-125 and excess sodium iodide-125 using a PD-10 desalting column (Amersham-Pharmacia) according to the manufacturer's instructions, except that the elution was performed in phosphate buffered saline in twelve 500 μl aliquots. Radioactivity in eluted fractions was quantitated on a Bioscan Quick Count QC-4000/XER Benchtop Radioisotope Counter (Bioscan, Inc.) and fractions containing the first peak of radioactivity (corresponding to labeled AJ4 protein) were pooled and bovine serum albumin added to a final concentration of 10 mg/ml.

To analyze the binding of the AJ4 protein proteins associated with intestinal surface proteins Icnown to act as ligands for bacterial adhesins, different intestinal protein ligands were dot blotted onto a nitrocellulose membrane using a Convertible Filtration Manifold System (Life Technologies) following the manufacturer's instructions. Duplicate dots of approximately 1 μg of type I collagen from calf skin, type IV collagen from human placenta, fibronectin from human plasma, laminin from the basement membrane of Engelbreth-Holm-Swarm mouse sarcoma and type III mucin partially purified from porcine stomach and bovine serum albumin included as a negative control (all proteins were obtained from Sigma) were blotted. The blot was incubated at room temperature on an orbital shaker in 10 ml phosphate buffered saline, pH 7.4, containing 0.1% Tween 20 and 5 mg/ml bovine serum albumin for 1 hour. Radio-iodinated AJ4 protein was then added to a final concentration of approximately 500 ng/ml, and incubated at room temperature for a further hour. The blot was washed three times in approximately 40 ml phosphate buffered saline, pH 7.4, containing 0.1%

Tween 20 at room temperature for 10 minutes, and autoradiogfaphed against X- ray film at -80 C overnight. The autoradiograph was developed and the resulting image digitized with a FluorS Multilmager (BioRad). Binding by AJ4 protein to the intestinal protein ligands was quantitated using Bio-Rad Quantity One software by measuring the density of the signal on the autoradiograph resulting from radio-labeled AJ4 protein binding to the different ligands and subtracting the background density of blank film. To quantitate relative amounts of protein ligands blotted, blots were stained with Ponceau S using standard procedures (Ausubel et al, Current Protocols in Molecular Biology, John Wiley & Sons, 2001), and quantitated as for the autoradiograhs. The density corresponding to AJ4 protein binding to individual ligands was divided by the average density of

Ponceau S staining of the ligand to give the relative AJ4 bound to each ligand. Binding of iodinated AJ4 protein (2.6 x 10^δ DPM) was compared to binding of iodinated mucus adhesion promoting (mapA) protein of Lactobacillus reuteri (GenBank accession number AJ293860) (1.3 x 10⁶ dpm) as a positive control, and iodinated irrelevant HNOOl protein (1.4 x 10⁶ DPM) as a negative control.

Results shown in Fig. 71 demonstrate that while the irrelevant HNOOl protein showed no significant binding to the intestinal proteins, AJ4 and the positive control protein mapA showed significant binding to mucin. AJ4 also showed some binding to laminin, fibronectin, and collagen type IV. Therefore, AJ4 encodes the HNOOl PEBl. Fig. 71 shows the relative density of autoradiographic signals from AJ4 protein (grey bars) to dot blots of intestinal proteins, compared to a positive control (mapA, white bars) and negative control (irrelevant HNOOl protein, black bars). Results for each dot (duplicates) are shown. The PEBl is a surface-bound molecule required for the adhesion of

Campylobacter spp. to intestinal epithelial cells and is required for effective colonization of the gut environment (Pei et ah, Infect. Imm. 66:938-943, 1998; Pei and Blaser, J. Biol. Chem. 268:18717-18725, 1993). Applications for HNOOl PEBl AJ4 include:

enhanced adhesion to intestinal surface and cell lines; enhanced survival in intestinal environment; altered texture characteristics; enhanced probiotic effects; • reagents to block or modify adherence of bacteria to surfaces; and • development of vaccine carriers.

SEQ ID NOS: 1-83 are set out in the attached Sequence Listing. The codes for nucleotide sequences used in the attached Sequence Listing, including the symbol "n," conform to WIPO Standard ST.25 (1998), Appendix 2, Table 1.

While in the foregoing specification this invention has been described in relation to certain preferred embodiments, and many details have been set forth for purposes of illustration, it will be apparent to those slcilled in the art that the invention is susceptible to additional embodiments and that certain of the details described herein may be varied considerably without departing from the basic principles of the invention.

Claims (1)

1. We claim:

   1. An isolated polynucleotide comprising a nucleotide sequence present in Lactobacillus rhamnosus strain HNOOl that encodes a polypeptide having at least one of the following activities: enzyme activity; anti-infection activity; lactose digestion modulating activity; immune system modulating activity; amino acid, lipid or carbohydrate metabolic activity; flavor, texture or aroma modulating activity; multistress resistance and survival activity; antigenic activity; adhesion activity; and regulatory activity.

   2. An isolated polynucleotide of claim 1, comprising a nucleotide sequence selected from the group consisting of: (1) the sequences recited in SEQ ID

   NOS: 1-33; and (2) sequences comprising a nucleotide sequence producing an Expectation ("E") value of 0.01 or less when compared to a sequence of (1) above using the BLASTN algorithm version 2.04 set to the default parameters described in the specification, above.

   3. An isolated polynucleotide of claim 1 comprising a nucleotide sequence having at least 75% identical nucleotides to a compare sequence selected from the nucleotide sequences recited in SEQ ID NOS: 1-33, the percentage identical nucleotides being determined by aligning the sequence and the compare . sequences using the BLASTN algorithm version 2.04 set at default parameters, identifying the number of identical nucleotides over aligned portions of the sequence and the compare sequences, dividing the number of identical nucleotides by the total number of nucleic acids of the compare sequence, and multiplying by 100 to determine the percentage identical nucleotides.

   4. An isolated polynucleotide of claim 1 comprising a nucleotide sequence that hybridizes to a polynucleotide comprising a sequence recited in SEQ ID NOS: 1-33 under stringent hybridization conditions;

   5. An isolated polynucleotide comprising a nucleotide sequence selected from the group consisting of: (1) complements of the sequences recited in SEQ ID NOS: 1-33; (2) reverse complements of the sequences recited in SEQ ID NOS: 1-33; and (3) reverse sequences of the sequences recited in SEQ ID NOS: 1-33

   6. An isolated polynucleotide comprising a nucleotide sequence selected from the group consisting of: (1) sequences comprising a nucleotide sequence that is a 200-mer of a sequence of claim 3; (2) sequences comprising a nucleotide sequence that is a 100-mer of a sequence of claim 3; and (3) sequences comprising a nucleotide sequence that is a 40-mer of a sequence recited in claim 3.

   7. An isolated polynucleotide of claim 1 comprising a nucleotide sequence that differs from a nucleotide sequence recited in SEQ ID NOS: 1-33 as a result of conservative substitutions.

   8. An isolated polynucleotide of claim 1 comprising a nucleotide sequence that differs from a nucleotide sequence recited in SEQ ID NOS: 1-33 as a result of deletions and/or insertions totaling less than 10%) of the total sequence length.

   9. An isolated polynucleotide of claim 1 comprising a nucleotide sequence that differs from a nucleotide sequence recited in SEQ ID NOS: 1-33 as a result of substitutions, insertions, and/or deletions totaling less than 15% of the total sequence length.

   10. An isolated polypeptide encoded by an isolated polynucleotide of any of claims 1-9.

   11. An isolated polypeptide comprising an amino acid sequence selected from the group consisting of : (1) the sequences recited in SEQ ID NOS: 42-75;

   (2) sequences producing an producing an Expectation ("E") value of 0.01 or less when compared to a sequence recited in (1) above using the

   BLASTP algorithm version 2.0.11 set to default parameters; (3) sequences comprising an amino acid sequence having at least 75% identical amino acid residues with a compare sequence selected from the amino acid sequences recited in (1) and (2) above, the percentage identical amino acids being determined by aligning the sequence and the compare sequences using the BLASTP algorithm version 2.0.411 set at default parameters, identifying the number of identical amino acids over aligned portions of the sequence and the compare sequences^", dividing the number of identical amino acids by the total number of amino acids of the compare sequence, and multiplying by 100 to determine the percentage identical amino acids; (4) sequences differing by codon alterations that reflect the degeneracy of the genetic code; and (5) functionally similar sequences differing only by conservative amino acid substitutions.

   12. A fusion protein comprising at least one polypeptide according to claim 11.

   13. A kit comprising a plurality of oligonucleotide probes or primers comprising at least 10 contiguous residues complementary to 10 contiguous residues of a nucleotide sequence recited in claim 1.

   14. A genetic construct comprising a polynucleotide of claim 1.

   15. The genetic construct of claim 14, wherein said polynucleotide encodes a polypeptide that modifies the flavor, aroma, texture and health-related benefits of milk-derived products selected from the group consisting of: peptidase (pepO), esterase (AA7), glyceraldehyde 3 -phosphate dehydrogenase (AK7), acetoin dehydrogenase, 6-phospho galactosidase, aromatic aminotransferase, acetyl kinase (AP5), malic enzyme (AA5), malate dehydrogenase (AG3), malY/Pat B pyridoxal 5 'phosphate aminotransferase, histidinol-phosphate aminotransferase (AG2), dihydropicolinate synthase, dihydropicolinate reductase, aspartate aminotransferase, mal Y aminotransferase, cystathione lyase, serine dehydratase, and aflatoxin Bl aldehyde reductase.

   16. The genetic construct of claim 14, wherein said polynucleotide encodes a polypeptide that increases the survivability of a microbe used in the manufacture of dairy products and probiotic supplements, wherein said polypeptide is selected from the group consisting of: formamidopyrimidine-DNA-glycosylase (fpg, AL4);^" basic surface protein (AC9); outer membrane protem A (AL8); extracellular matrix binding protein (AM4); high molecular weight adhesion protein (AL7); periplasmic binding protein(PEBl, AJ4); autoaggregation protein (AG5); phosphoenolpyruvate hydrolase, tagatose bisphosphate aldolase, phosphoglycerate kinase, triose phosphate isomerase, fructose-bis phosphate aldolase, phosphoryl carrier protein HPR AA9, tagatose bisphosphate aldolase, and dnaK chaperone protein (AM9).

   17. A transgenic cell comprising a genetic construct according to any of claims 14-16.

   18. A genetic construct comprising, in the 5 '-3' direction:

   (a) a gene promoter sequence;

   (b) a polynucleotide sequence comprising at least one of the following: (1) a polynucleotide coding for at least a functional portion of a polypeptide encoded by a nucleotide sequence described in claim 1; and (2) a polynucleotide comprising a non-coding region of a gene coding ^"for an polypeptide encoded by a nucleotide sequence ^• selected from the group consisting of sequences recited in claim 1 ; and

   (c) a gene termination sequence.

   19. The genetic construct of claim 14-16 or 18 wherein the polynucleotide is in a sense orientation.

   20. The genetic construct of claim 14-16 or 18 wherein the polynucleotide is in an antisense orientation.

   21. The genetic construct of claim 18, wherein the gene promoter sequence and gene termination sequences are functional in a prokaryote or eucaryote.

   22. A method for modulating the polynucleotide content or composition of an organism comprising transforming the organism with a genetic construct of claim 14-16 or 18.

   23. A method of identifying an organism or reproductive material or an extract therefrom as having a specific origin, the method comprising detecting in the genetic complement of the organism, material or extract the presence or absence of a polynucleotide identifier representative of said origin, the polynucleotide identifier comprising a sequence recited in SEQ ID NOS: 1-33.

   24. The method of claim 23 wherein the organism is a bacterial cell or a yeast cell.

   25. The method of claim 23 wherein the presence or absence of the polynucleotide identifier is detected by isolating DNA from the organism or material and contacting the isolated DNA with at least one oligonucleotide probe specific for the polynucleotide identifier.

   26. The method of claim 23 wherein the isolated DNA is contacted with a plurality of oligonucleotide probes in an array format.

   27. A method for improving the properties of microbes used in the manufacture of milk-derived products and probiotic supplements, which comprises modulating the polynucleotide content or composition of said microbes by transforming said microbes with one or more polynucleotide sequences selected from the group consisting of:

   (a) Lactobacillus rhamnosus strain HNOOl sequences encoding polypeptides that modify the flavor, aroma, texture and health-related benefits of milk-derived products; and

   (b) Lactobacillus rhamnosus strain HNOOl sequences encoding polypeptides that increase the survivability of said microbes in dairy product manufacturing processes.

   28. The method of claim 27, wherein said polypeptides are selected from the group consisting of: peptidase (pepO), esterase (AA7), glyceraldehyde 3- phosphate dehydrogenase (AK7), acetoin dehydrogenase, 6-phospho - galactosidase, aromatic aminotransferase, acetyl kinase (AP5), malic enzyme (AA5), malate dehydrogenase (AG3), malY/Pat B pyridoxal 5 'phosphate aminotransferase, histidinol-phosphate aminotransferase (AG2), dihydropicolinate synthase, dihydropicolinate reductase, aspartate aminotransferase, mal Y aminotransferase, cystathione lyase, serine dehydratase, aflatoxin Bl aldehyde reductase, formamidopyrimidine- DNA-glycosylase (fpg, AL4); basic surface protein (AC9);outer membrane protein A (AL8); extracellular matrix binding protein (AM4); high molecular weight adhesion protein (AL7); periplasmic binding protein (PEBl, AJ4); autoaggregation protein (AG5); phosphoenolpyruvate hydratase, tagatose bisphosphate aldolase, phosphoglycerate kinase, triose phosphate isomerase, fructose-bis phosphate aldolase, phosphoryl carrier protein HPR AA9, and dnaK chaperone protein (AM9).

   9. A method for modifying the flavor, aroma, texture and/or nutritional and health benefits of milk-derived products, which comprises adding one or more polypeptides to the milk being processed, wherein said polypeptides are selected from the group consistmg of Lactobacillus rhamnosus strain HNOOl peptidase (pepO), esterase (AA7), glyceraldehyde 3 -phosphate dehydrogenase (AK7), acetoin dehydrogenase, 6-phospho galactosidase, aromatic aminotransferase, acetyl kinase (AP5), malic enzyme (AA5), malate dehydrogenase (AG3), malY/Pat B pyridoxal 5 'phosphate aminotransferase, histidinol-phosphate aminotransferase (AG2), dihydropicolinate synthase, dihydropicolinate reductase, aspartate aminotransferase, mal Y aminotransferase, cystathione lyase, serine dehydratase, aflatoxin Bl aldehyde reductase, and tagatose bisphosphate aldolase.

   30. A therapeutic composition effective for treating or preventing a gastrointestinal condition or disorder in a mammal caused by the presence of pathogenic microbes in the gastrointestinal tract or by the absence of normal intestinal microbes in the intestinal tract, wherein said composition comprises one or more species or strains of probiotic microbes that are non-pathogenic to mammalian organisms, wherein said microbes comprise one or more expressible polynucleotide sequences derived from

   Lactobacillus rhamnosus HNOOl that encode polypeptides selected from the group consisting of autoaggregation protein (AG5), glyceraldehyde 3- phosphate dehydrogenase (AK7) basic surface protein (AC9), outer membrane protein A (AL8), extracellular matrix binding protein (AM4), high molecular weight adhesion protein (AL7), and periplasmic binding protein 1 (PEBl, AJ4).

   31. The therapeutic composition of claim 30, wherein said microbes are selected from a lactic acid-producing species of Bacillus, Lactobacillus, Sporolactobacillus, or Bifidiobacterium.

   32. A transgenic microbial population which comprises an expressible polynucleotide sequence isolated from Lactobacillus rhamnosus HNOOl that encodes aflatoxin _\ aldehyde reductase (AI7), wherein said composition is effective in detoxifying carcinogens, including aflatoxin, and wherein said microbial population is nonpathogenic and can be administered to a mammal for use as anticarcinogenic agent .

   33. A genetic construct comprising a polynucleotide sequence comprising at least one of the following: (1) a polynucleotide coding for at least a functional portion of a

   Lactobacillus rhamnosus HNOOl sor R transcriptional regulator polypeptide ; and (2) a polynucleotide comprising a promoter region of said Lactobacillus rhamnosis HNOOl sor R transcriptional regulator gene.

   34. A genetic construct comprising in the 5 '-3 ' direction,

   (a) an inducible gene promoter;

   (b) a Lactobacillus rhamnosus HNOOl polynucleotide encoding acetoin dehydrogenase; and (c) a gene termination sequence.

   35. A transgenic organism comprising a genetic construct of claim 34.

   36. The transgenic organism of claim 35, selected from lactic-acid producing bacteria and brewer's yeast.

   37. A method for modulating the level of diacetyl in a food or beverage, comprising: measuring the level of diacetyl produced during the manufacture of a food or beverage compared with a predetermined level, and adding an effective amount of a transgenic organism of claim 36 to said food or beverage to reduce the level of diacetyl to the predetermined level.

   38. An isolated polynucleotide comprising a nucleotide sequence present in Lactobacillus rhamnosus strain HNOOl that encodes a polypeptide capable of modifying the flavor, aroma, texture and/or nutritional and health benefits of milk-derived products, wherein said polypeptides are selected from the group consisting of Lactobacillus rhamnosus strain HNOOl peptidase (pepO), esterase (AA7), glyceraldehyde 3 -phosphate dehydrogenase (AK7), acetoin dehydrogenase, 6-phospho galactosidase, aromatic aminotransferase, acetyl kinase (AP5), malic enzyme (AA5), malate dehydrogenase (AG3), malY/Pat B pyridoxal 5 'phosphate aminotransferase, histidinol-phosphate aminotransferase (AG2), dihydropicolinate synthase, dihydropicolinate reductase, aspartate aminotransferase, mal Y aminotransferase, cystathione lyase, serine dehydratase, aflatoxin Bl aldehyde reductase, and tagatose bisphosphate aldolase.

   39. An isolated polynucleotide comprising a nucleotide sequence present in Lactobacillus rhamnosus strain HNOOl that encodes a polypeptide that increases the survivability of microbes in dairy product manufacturing processes, wherein said polypeptides are selected from the group consisting of malic enzyme (AA5), malate dehydrogenase (AG3), glyceraldehyde 3 -phosphate dehydrogenase (AK7), dihydropicolinate synthase, dihydropicolinate reductase, aspartate aminotransferase, serine dehydratase, formamidopyrimidine-DNA-glycosylase (fpg, AL4), basic surface protein (AC9), outer membrane protein A (AL8), extracellular matrix binding protein (AM4), high molecular weight adhesion protein (AL7), periplasmic binding protein(PEBl, AJ4), autoaggregation protein (AG5), phosphoenolpyruvate hydratase, tagatose bisphosphate aldolase, phosphoglycerate kinase, triose phosphate isomerase, fructose-bis phosphate aldolase, phosphoryl carrier protein HPR AA9, and dnaK chaperone protem (AM9).

   40. A transgenic microbial population for use in cheese ripening, which comprises an expressible polynucleotide sequence isolated from Lactobacillus rhamnosus HNOOl that encodes peptidase (pepO).