In a further aspect the present invention provides a method for increasing infectious disease resistance in a host which method comprises administering a supplement or food of the invention to the host. The host may be a human or other animal. It will be understood that this aspect of the invention also embraces a method for the manufacture of a food or supplement of the invention for use in a method for increasing infectious disease resistance in a host which method comprises administering a supplement or food of the invention to the host.
Features of Probiotics
A probiotic which needs to be viable to exert its probiotic effect should possess at least the following four characteristics:
2) it should remain stable and viable for long periods under storage and field conditions,
3) it should have the ability to survive (not necessarily grow) in the intestine,
4) it must produce a beneficial effect in the host animal.
All microorganisms can be divided into three groups in terms of their relative safety: non-pathogenic, opportunistic pathogens and pathogenic. The first step in the selection of microbial strains for probiotic use is that it must be representative of microorganisms that are 'Generally Recognized As Safe' (GRAS) microorganisms (Havenaar et al. , 1992). In the
dairy industry, the currently used probiotic bacteria include lactic acid bacteria (LAB), bifidobacteria and yeasts. The use of Lactobacillus in foods has a long history and most strains are considered GRAS microorganisms. This also true for dairy propionibacteria.
Safety is an important requirement for probiotics. There are increasing demands to extend the range of foods containing probiotic organisms from dairy foods to infant formulae, baby foods, fruit juice-based products, cereal-based products and pharmaceuticals. New and more specific strains of probiotic bacteria are being selected. However, these novel probiotic organisms may not share the safety of traditional strains. Their safety should be carefully assessed before their use in food.
Three approaches can be used to assess the safety of a probiotic strain: studies on the intrinsic properties of the strain; studies on the pharmacokinetics of the strain; and studies searching for interactions between the strain and the host. Models and methods recommended to test the safety of probiotic bacteria include (1) determination of the intrinsic properties of bacteria and strains selected for probiotic use, for example, adhesion factors, antibiotic resistance, plasmid transfer, enzyme profile; (2) assessment of the effects of the metabolic products of the bacteria; (3) assessment of the acute and sub acute toxicity of ingestion of extremely large amounts of the bacteria ; (4) estimation of the in vitro infective properties of probiotic bacteria using cell lines and human intestinal mucus degradation; (5) assessment of infectivity in animal models, for example, immunocompromised animals or lethally irradiated animals; (6) determination of the efficacy of ingested probiotic bacteria as measured by dose- response (minimum and maximum dose required, consequent health effects), assessment of the effect of massive probiotic doses on the composition of human intestinal microflora; (7) assessment of the side-effects in human volunteer studies and in clinical studies of various disease-specific states; (8) epidemiological surveillance of people ingesting large amounts of newly introduced probiotic bacteria for infections; (9) extra attention to genetically modified strains and strains derived from animals.
Assessment of a novel probiotic can be performed by in vitro methods, animal models and human subjects. In vitro studies provide indirect measure of the potential for a test organism to invade intestinal cells and to damage the intestinal mucus. Most strains of lactic acid bacteria have shown no invasive properties towards a human gut epithelial cell line, Caco-2. There is no report on the potential for propionibacteria to invade intestinal cells and to damage the intestinal mucus and no toxicity data is available relating to the ingestion of large quantities of dairy propionibacteria.
An important factor modifying toxicity is microbial metabolism in the gut. Profound toxicological and carcinogenic consequences may result from changes in microbial enzyme activity in animal gut. It is generally known that colonic microflora can generate mutagens, carcinogens and tumour promoters from dietary and endogenously produced precursors. The
bacterial enzymes involved in these harmful processes include azoreductase, nitroreductase, nitrate reductase, P-glucuronidase and p-glycosidase. Some probiotic strains of Lactobacillus and Bifidobacteriunz sp. have been shown to reduce 0-glucuronidase and p-glycosidase activities in the lower gut of rats.
Six isolated Propio7zibacterium spp. strains from human faeces have been found to have (3-glucuronidase activity (Nanno et al. , 1986), whereas two strains of P. freudenreichii and three strains of P. acidipropionici in another study have been shown to lower the (3- glucuronidase activity in mice, but had no inducing effects on the activities of azoreductase, nitroreductase, nitrate reductase, and (3-glycosidase of the intestinal microflora (Perez-Chaia et al. , 1999).
During the development of probiotics, it is very important to investigate the viability of the microorganisms during processing and storage (Havenaar et al. , 1992). The processing and storage methods for a probiotic are determined by its application and the administration methods.
High correlation has been found between the results of in vitro and in vivo studies of gastrointestinal tract transit tolerance of probiotics (Havenaar et al. , 1992). Therefore, selection of probiotic strains with gastrointestinal tract tolerance can be based on in vitro experiments (Havenaar et al. , 1992). Several in vitro methods have been developed to select gastrointestinal transit tolerant probiotic strains. The most widely used methods include the use of HCl-acidified distilled water, broth and buffers (Chou and Weimer, 1999, Chung et al., 1999, Clark et al. , 1993, Clark and Martin, 1994, Wang et al. , 1999).
Another in vitro method, which simulates the human upper gastrointestinal transit conditions, was developed by Charteris et al (Charteris et al. , 1998). The simulated gastric juice (pH2.0) contained pepsin (0.3% w/v) and sodium chloride (0. 5% w/v) while the simulated small intestinal juice (pH8.0) contained pancreatin (0.1% w/v) and sodium chloride (0. 5% w/v).
Selection of bile-resistant bacteria can be made by culturing on selective agar medium with various levels of bile (Gilliland et al. , 1984, Ibrahim and Bezkorovainy, 1993, Clark and Martin, 1994, Chung et al. , 1999). Because Oxgall (Oxoid) is readily available and is used extensively in selective media for human enteric pathogens, its use has been adopted widely to select bile-tolerant probiotic strains (Clark and Martin, 1994).
The ability to survive and colonize the intestine can be both species and strain- dependent (Wang et al. , 1999).
Adhesion to intestinal surfaces is regarded as the first, step for colonization and immune stimulation for probiotics (Havenaar et al., 1992). Since it is very difficult to study bacterial adhesion in vivo, selection of strains with the capacity to adhere to gastrointestinal
cells is based on in vitro tests (Tuomola et al. , 1999, Sarem et al. , 1996, Lehto and Salminen, 1997, Crociani et al., 1995, Mayra-Makinen et al. , 1983). One of the models used is the human intestinal Caco-2 epithelial cell line. The Caco-2 cell line was originally isolated from a human colon adenocarcinoma (Pinto et al., 1983). This cell line spontaneously differentiates under standard in vitro culture conditions and the differentiated cells express characteristics of mature enterocytes (Pinto et al. , 1983).
In vitro methods are suitable for assessment of some of the probiotic bacteria selection criteria. Nevertheless, regardless of the results of in vitro tests, it is still difficult to predict the actual conditions in vivo (Havenaar et al. , 1992). The adherence and colonization in vivo is affected by different animal species, the specificity of microorganism species and strains, and the different food or feed consumed (Havenaar et al. , 1992). Adherence ill vitro is no guarantee for adherence in vivo and subsequent colonization but is usually predictive of in vivo efficacy.
The results reported here provide evidence that Propionibacterium jensenii 702 will adhere and colonise in vivo. Confirmation is provided by in vivo rat studies. Animal models using Balb/c mice have been used to determine the survival through gastrointestinal tract of strains of Bifidobacterizim spp, which have shown resistance to acid and bile in in vitro tests (Wang et al. , 1999). In this study, the mice were oro-gastrically fed with a strain of Bifidobacteriurrz, and the viable cell numbers of this strain in faeces were determined. The recovery rate of the strains of Bifidobacterium spp. in faeces was found to be only 4.3% (Wang et al. , 1999).
The survival and colonisation of some Propionibacterium strains in the gastrointestinal tract has been examined through in vitro tests, animal studies and human trials (Perez-Chaia et al. , 1995, Mantere-Alhonen, 1983, Bougle et al. , 1999). A strain of P. freudenreichii has been reported to be able to survive during in vitro gastric digestion at pH4.8 (Mantere-Alhonen, 1983), while in in vivo tests, strain P. acidipropionici CRL 1198 has shown the ability to survive in the gut of BALB/c male albino mice (Perez-Chaia et al.,
1995). In a human trial, it was also found that part of the ingested propionibacteria were able to survive the digestive transit (Bougle et al., 1999). In the present study, the gastrointestinal resistance of selected Propionibacterium strains has been tested under both in vitro and in vivo conditions.
A Propionibacterium strain, P. freudenreichii ssp. shennanii JS, has been shown to be able to adhere to human gut epithelial Caco-2 cell line in vitro (Lehto and Salminen, 1997). The adhesion rate was 12.2 % of the number of added bacteria. The results also show that the adhesion of P. fi-eudenreichii ssp. shermanii JS was significantly reduced by previously adhered Lactobacillus rhairayaosus LC-705. This may indicate that bacteria may compete for adhesion sites
(Lehto and Salminen, 1997). In the present study, the adhesive abilities of selected Propionibacterium strains have been tested using a subclone of the Caco-2 cell line, c2bbel.
In in vivo tests, strain P. acidipropionici CRL 1198 has shown the ability to establish in the gut of BALB/c male albino mice (Perez-Chaia et al. , 1995). The numbers of fed P. acidopropionici CRL 1198 remain stable in the mouse gut contents and mouse gut walls one week after the cessation of the diet. In contrast, in a human trial, propionibacteria were found to be able to adhere but unable to colonise the digestive tract (Bougle et al. , 1999). The numbers of fecal propionibacteria returned to the level observed prior to the supplementation within a few days of stopping the supplement.
Because Propinibacterium strains utilise lactate, they can be isolated, from food and environmental samples, using Yeast Extract Lactate medium (YELA), which contains sodium lactate, casein peptone, yeast extract and agar (Harrigan, 1998a, Fessler et al. , 1998, Britz and Riedel, 1994, Cummins and Johnson, 1986).
Primary identification of isolated strains to the genus Propionibacterium is based on microscopic examination for morphological and staining characteristics, cultural characteristics and simple biochemical tests. Propionibacterium is separated from related bacterial species by positive Gram stain, negative acid fast stain, no endospore formation, positive catalase, positive lactate fermentation, and irregular shaped rods (Harrigan, 1998b).
Conventional methods of identifying Propionibacterium strains to species level include carbohydrate fermentation tests, nitrate reduction test, -hemolysis test and morphological analysis of cells and colonies. Different selected biochemical characteristics of Propioiiibacteri, uyjz species are shown in Table 1 (Cummins and Johnson, 1986, Holt et al., 1997).
Table 1 Characteristics differentiating the species of Proponibacteriurn a
Characteristeristics P. acidipropionici P.acne P.avidum P.freudenreichii P.granulosum P.jensenii P.lymphophilum P.thoenii
To gray cream or pink gray pink White to red- brown -EIemolysis-d (+)- (-)-~ + Nitrate reduction + d+----d- a Symbols : +, 90% or more of strains are positive; (+), 80-89% of strains are positive ; d+, 40-90% of strains are positive ; d, 21-79% of strains are positive; d-, 10-40% of strains are positive; (-), 11-20% of strains are positive; -, 90% or more of strains are negative.
Conventional methods are mostly based on phenotypic characteristics, however, species differentiation is not always reproducible due to variation in specific phenotypic characteristics (Cummins and Johnson, 1986, Britz and Riedel, 1994). Therefore, alternative classification techniques, such as protein profile studies and genetic analysis, are used to identify the genus and species more accurately, more sensitively and more rapidly (Jones and Krieg, 1986, Jones, 1986).
The protein profile method is based on the principle that closely related organisms should have similar or identical kinds of cellular proteins (Jones and Krieg, 1986). A strain is considered to belong to a particular species when there is more than 70% identity of the protein profile with the type strain (Fessler et al. , 1999). Whole cells, cellular membrane fractions and water-soluble protein fractions are separated by polyacrylamide gel electrophoresis (PAGE) and stained gels are used to distinguish related from unrelated organisms (Jones and Krieg, 1986). Sodium dodecyl sulfate-PAGE (SDS-PAGE) of soluble proteins has been used to distinguish different species of Propionibacterium. It has been reported that the SDS-PAGE profile of cell free soluble protein can differentiate the four species of dairy propionibacteria (Fessler et al. , 1999, Baer, 1987, Riedel and Britz, 1992).
These results also show the correlation between a SDS-PAGE protein profile identification method and a genetic method (Fessler et al. , 1999).
There are several genetic analysis methods applied to the identification of dairy Propioyaibacteriura species, including DNA fingerprinting by pulsed-field gel electrophoresis and various PCR methods (Gautier et al. , 1996, Meile et al. , 1999, Fessler et al. , 1998, Riedel et al. , 1998). The details of PCR methods and primers used in Propionibacterium identification are summarised respectively in Table 2 and Table 3 (Meile et al. , 1999, Fessler et al. , 1999, Rossi et al., 1998, Fessler et al., 1998, Riedel et al. , 1998, Riedel et al., 1994, Rossi et al. , 1999) Table 2 PCR Methods used to identify Propionibacterium
Method Primer name Identification specificity Target Endonucleases
RAPD-OPL-01 Strains within P. species Genome DNA N/A
PCR OPL-02 Strains within P. species Genome DNA
SK-2 Strains within P. species Genome DNA
DF4 Strains within P. species Genome DNA
OPL-05 Four dairy P. species Genome DNA
PCR & A-B Four dairy P. species 23s rRNA MspI
RFLP 16sPl-16sP4 Four dairy P. species 16s rDNA AluI, HaeIII
16sP3-16sP4 Four dairy P. species 16s rDNA HpaII
MPCR gdl-bakllw Genus Propionibacterimn 16s rDNA N/A
gdl-bak4 Genus Propionibacterium 16s rDNA
SpecificP PB1-PB2 Dair P. & P. acne 16s rDNA
CR PF-PB2 P. freudenreichii
PJ-PB2 P. jensenii
PA-PB2 P. acidipropionici
PT3-PB2 P. tlaoenii & P. acne
Table 3 Primers used in identifying Propionibacterium
Primer name (SEQ ID No) Primer sequence OPL-01 (SEQ ID No : 1) 5'-GGCATGACCT-3' OPL-02 (SEQ ID No : 2) 5'-TGGGCGTCAA-3' SK-2 (SEQ ID No : 3) 5'-GCCGCCGCCGCC-3' DF4 (SEQ ID No : 4) 5'-CGCCGCCGTCGC-3' OPL-05 (SEQ ID No : 5) 5'-ACGCAGGCAC-3' A (SEQ ID No : 6) 5'-MADGCGTAGNCGAWGG-3' B (SEQ ID No : 7) 5'-GTGWCGGTTTNBGGTA-3' 16sPl (SEQ ID No : 8) 5'-GGGTGACCGGCCACA-3' 16sP3 (SEQID No : 9) 5'-AAGGTGGGGATGAGC-3' 16sP4 (SEQ ID No : 10) 5'-TCGGGTGTTACCGAC-3' Bak4 (SEQID No : 11) 5'-AGGAGGTGATCCARCCGCA-3' Gdl (SEQ ID No : 12) 5'-TGCTTTCGAT ACGGGTTGAC-3' Bakllw (SEQ ID No : 13) 5'-AGTTTGATCMTGGCTCAG-3' PB 1 (SEQ ID No : 14) 5'-AGTGGCGAAGGCGGTTCTCTGGA-3' PB2 (SEQ ID No : 15) 5'-TGGGGTCGAGTTGCAGACCCCAAT-3' PF (SEQ ID No : 16) 5'-CTTTCATCCATGACGAAGCGCAAG-3' PJ (SEQ ID No : 17) 5'-GACGAAGTGCCTATCGGGGTG-3' PA (SEQ ID No : 18) 5'-GACGAAGGCATTCTTTTAGGGTGT-3' PT3 (SEQ ID No : 19) 5'-GGACAAAAGGCCTTTCGGGGTTT-3' B=C, G, T; D=A, G, T; M=A, C; N=A, G, C, T; R=A, G; W=A, T
Different PCR methods have different capabilities. A multiplex-PCR (MPCR) using two sets of primers, namely, gdl-bakl lw and gdl-bak4, can differentiate Propionibacterium from related genera but can not differentiate species of Propionibacterium (Meile et al., 1999). RAPD-PCR method using primer OPL-05 can differentiate four dairy Propionibacteriunz, while other RAPD-PCR methods using primers OPL-01, OPL-02, SK2,
DF-4 give better results in dividing different strains into different groups within single species (Fessler et al. , 1999, Rossi et al., 1998). The use of both RFLP and PCR (primer A and B) targeting 23s rDNA and restriction endonuclease MspI can differentiate dairy Propioyaibacterium species from related species (Fessler et al. , 1998). The four dairy Propionibacterium species can be separated using other combinations of PCR and restriction fragment length polymorphisms (RFLP) which target 16s rDNA. These methods use primer sets 16sPl-16sP4, 16sP3-16sP4 in combination with the endonucleases Aluni, HaeI, HpaII (Riedel et al., 1998, Riedel et al. , 1994). Recently, a rapid, genus-specific and species- specific PCR targeting the genes encoding 16s rRNA was developed to detect the dairy Propionibacteria in environmental samples (Rossi et al. , 1999). In this method, the primers PB1-PB2, PF-PB2, PJ-PB2, PA-PB2, and PT3-PB2 were designed specifically for dairy propionibacteria and P. acne ; P. freudenreichii, P. jensenii ; P. acidipropionici ; and P. thoenii respectively.
The human mouth and intestine provide suitable habitats for numerous bacterial genera. Propionibacteria are found as normal oral flora (Sutter, 1984) and have also been found in the human colon (Allison et al. , 1989, Macfarlane et al. , 1986). Propionibacteria are observed to be present from 10 4 3 cfuJg to 1012 cfu/g in faeces (Finegold et al. , 1983, Macfarlane, 1986).
ABBREVIATIONS a Alpha AIDS Auto-Immune Deficiency Syndrome APC Antigen Presenting Cell p Beta BCG M. bovis Bacille Calmette-Guerin CD Cluster of Differentiation CD1 Cluster of Differentiation one CD4+ T-cells Cluster of Differentiation four positive T cells CD8+ T-cells Cluster of Differentiation eight positive T cells CFU Colony Forming Units Con A Concanavalin A CPM Counts Per Minute CT Cholera Toxin CTB Cholera Toxin subunit B CTL Cytotoxic T Lymphocytes 8 Delta Da Dalton DN ap T cells Double negative alpha beta T cells DNA Deoxyribose nucleic acid DOTS Directly Observed Therapy Short Course DTH Delayed-type hypersensitivity ELISA Enzyme-Linked Immunosorbent assay et al. and others y Gamma g gram GALT Gut-Associated Lymphoid Tissue
HIV Human Immunodeficiency Virus IFN Interferon Ig Immunoglobulin IL Interleukin kDa Kilodalton L Litre LAM Lipoarabinomannans LJ Lowenstein-Jensen (medium) M. Mycobacterium M Molar mA Milli Ampere MALT Mucosal Associated Lymphoid Tissue MDR-TB Multi-drug resistant tuberculosis mg milligram MHC Major Histocompatibility Complex Min minute ml millilitre mM Millimolar MSM Modified Sauton's Medium MWCO Molecular Weight Cut-Off NIAID National Institute of Allergy and Infectious Diseases NK Natural Killer nm nanometre PBS Phosphate Buffered Saline pg picagram PMSF Phenyl Methyl Sulfonyl Fluoride RPMI Roswell Park Memorial Institute (medium) SA-HRP Streptavidin Horse-Radish Peroxidase SDS-PAGE Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis sec seconds SLA Sodium Lactate Agar SLB Sodium Lactate Broth STCF Short Term Culture Filtrate Th T helper cells TAP1 Transporter Associated with Antigen Processing 1 TB Tuberculosis TCR T cell receptor TNF Tumour Necrosis Factor TNF-a alpha-tumour necrosis factor IlCi microCurie 119 microgram u. l microlitre UV ultra violet (light) V Volts WHO World Health Organisation WTB Whole Tuberculosis cells (sonicated) Brief description of the drawings
Figure 1 shows genus-specific amplification with primer set PB 1-PB2. Lane 1 : PCR DNA Marker (FN-1, Biotech), Lane 2,12, 13: 702, Lane 3: 801, Lane 4: 901, Lane 5: 1001, Lane 6: P. freudenreichii CSCC 2200, Lane 7: P. freudenreichii CSCC 2201; Lane 8: P. freudenreichii CSCC 2207, Lane 9: P. acidopropionici ATCC 25562, Lane 10, 201al Lane
11: P. jensenii NCFB572, Lane 14, 201b, Lane 15: P. thoenii ACM 365, Lane 16: P. freudenreichii CSCC 2207, Lane 17: P. acidopropionici ATCC 25562
Figure 2 shows species-specific amplification with primer set PF-PB2. Lane 1: PCR DNA Marker (FN-1, Biotech), Lane 2: 201al, Lane 3: 201b, Lane 4: 702, Lane 5: 801, Lane 6: 901, Lane 7,8 : 1001, Lane 9: P. freudenreichii CSCC 2200; Lane 10: P. freudenreichii CSCC 2201, Lane 11 : P. freudenreichii CSCC 2206, Lane 12: P. freudenreichii CSCC 2207, Lane 13: P. freudenreichii CSCC 2216
Figure 3 shows species-specific amplification with primer set PJ-PB2 Lane 1: PCR DNA Marker (FN-1, Biotech), Lane 2: Negative control, Lane 3-5: 702; Lane 6: P. acidopropionici ATCC 25562, Lane 7,8 : Pfreudenreicliii CSCC 2206, Lane 9: P. freudenreiclaii CSCC 2207; Lane 10: Pjensenii NCFB571; Lane 11 : Pjensenii NCFB572
Figure 4 shows species-specific amplification with primer set PT3-PB2 Lane 1: PCR DNA Marker (FN-1, Biotech), Lane 2,3 : P. thoenii ACM 365, Lane 4,5 : 702
Figure 5 shows electrophoretic profiles obtained for tested strains of Propionibacterium spp. by RAPD-PCR with primer OPL-05. Lane 1: PCR DNA Marker (Bio-Rad AmpliSize Molecular Ruler, 50-2000bp ladder); Lane 2: 1001 ; Lane 3: 901; Lane 4: 801 ; Lane 5: 201b; Lane 6: P. freudenreichii CSCC 2206; Lane 7: 201al ; Lane 8: P. acidopropionicii ATCC 25562; Lane 9: P. acidopropionicii 341
Figure 6 shows SDS-PAGE analysis of whole cell-water-soluble proteins of an isolated strain 702 and reference strains. Lane 1,10 : SDS-PAGE Protein Marker (Bio-Rad Precision Protein Standard, Broad Range, Unstained), Lane 2: Lactobacillus acidophilus (MJLA1), Lane 3: P.freudenreichii CSCC 2207, Lane 4: P. acidopropionici ATCC 25562, Lane 5 and 6: P. thoenii ACM 365, Lane 7 to 9: Strain 702
Figure 7 shows SDS-PAGE analysis of whole cell water-soluble proteins of an isolated strain 702 and reference strains. Lane 1: SDS-PAGE Protein Marker (Bio-Rad Precision Protein Standard, Broad Range, Unstained), Lane 2: Lactobacillus acidophilus MJLA1, Lane 3: P. freudenreichii CSCC 2207, Lane 4: P. acidopropionici ATCC 25562, Lane 5: P. thoenii ACM 365, Lane 6,7 : P. jensenii NCFB 572, Lane 8-9: 702
Figure 8 shows SDS-PAGE analysis of whole cell water-soluble proteins of six isolated strains and five Propionibacterium reference strains. Lane 1; SDS-PAGE Protein Marker (Bio-Rad SDS-PAGE Molecular Standards, low range); Lane 2: P. freuradenreichii CSCC2200; Lane 3: P. fretcfzdettreicltii CSCC2201; Lane 4: P. freunderzreichii CSCC2206; Lane 5: P. freudenreichEi CSCC2207; Lane 6: Pjreuiidenreichii CSCC2216; Lane 7 strain