Bifidobacteria are generally considered beneficial for human health and together with lactobacilli they are widely used in probiotic preparations and foods (Salminen, Deighton, Benno, & Gorbach, 1998). Several positive effects have been related to bifidobacteria. These include synthesis of vitamins, supplementation in digestion and absorption, inhibition of growth of exogenous organisms, and stimulation of the immune system (Mitsuoka, 1992). Analysis of human faecal samples have shown that bifidobacteria form one of the predominant culturable bacterial groups in the human colon where their numbers often reach the level of 10⁹–10¹⁰ cfu g⁻¹ (Finegold, Attebery, & Sutter, 1974; Mitsuoka, 1982; Finegold, Sutter, & Mathisen, 1983; Benno et al., 1989).

Bifidobacterial numbers in the human gut tend to decrease with age(Mitsuoka, 1992). To maintain a high level of bifidobacteria in the gut a two-fold strategy can be applied. Numbers of bifidobacteria can be increased either by continuous ingestion of bifidobacteria-containing preparations or foods, or food can be supplemented with substrates (bifidogenic factors or prebiotics) that specifically promote the growth of endogenous bifidobacteria in the gut ( Gomes & Malcata, 1999). The most thoroughly studied probiotic Bifidobacterium strain currently on the market is B. lactis Bb-12. In human studies B. lactis Bb-12 has shown efficacy in prevention of traveller's diarrhoea, treatment of viral diarrhoea (including rotavirus diarrhoea), modulation of intestinal flora, improvement of constipation, modulation of immune response, and alleviation of atopic dermatitis symptoms in children (Black et al., 1989; Marteau et al., 1990; Black, Einarsson, Lidbeck, Orrhage, & Nord, 1991; Alm, Ryd-Kjellen, Setterberg, & Blomquist, 1993; Link-Amster, Rochat, Saudan, Mignot, & Aeschlimann, 1994; Saavedra, Bauman, Oung, Perman, & Yolken, 1994; Schiffrin, Rochat, Link-Amster, Aeschlimann, & Donnet-Hughes, 1995; Fukushima, Kawata, Hara, Terada, & Mitsuoka, 1998; Kankaanpää, Sutas, Salminen, & Isolauri, 1998).

Prebiotics are compounds, usually carbohydrates, which are resistant to direct metabolism by the host and reach the colon where they are preferentially utilised by selected groups of beneficial bacteria (Ziemer & Gibson, 1998). Several prebiotics have been shown to promote growth of bifidobacteria in the colon ( Gibson, Beatty, Wang, & Cummings, 1995; Bouhnik and Bouhnik; Kleessen, Sykura, Zunft, & Blaut, 1997; Brighenti, Casiraghi, Canzi, & Ferrari, 1999; Kruse, Kleessen, & Blaut, 1999) and in a model system ( Kneifel, Rajal, & Kulbe, 2000). These include carbohydrates such as fructo-oligosaccharides (FOS), inulin and galacto-oligosaccharides (GOS). The majority of prebiotic and synbiotic (combination of prebiotic and probiotic) effects have been demonstrated with preparations of fructo-oligosaccharide nature ( Roberfroid, 1998; Roberfroid, van Loo, & Gibson, 1998). Pronouncedly fewer reports have been published on the specific effects of GOS in clinical studies. For this reason, this study was focused on this particular carbohydrate.

GOS, composed of lactose and galactose units, is found naturally in human milk and to a smaller extent in cows’ milk (Kunz & Rudloff, 1993; Saito, 1987). GOS is commercially produced from lactose by enzymatic reactions in which galactose residues are attached to lactose ( Zaraté & López-Leiva, 1990). Although some minor degradation of GOS may occur in the human small intestine, the main fraction of GOS is non-digestible and reaches the caecum and colon, where rapid fermentation takes place. GOS can be fermented by a number of intestinal bacteria including bifidobacteria, bacteroides, enterobacteria and lactobacilli ( Tanaka et al., 1983; Ishikawa et al., 1995). The few human feeding studies investigating the prebiotic potential of GOS suggest that GOS may have bifidogenic potential ( Ito et al., 1993; Bouhnik et al., 1997). In the present study, GOS-containing syrup and/or B. lactis Bb-12 were fed to healthy volunteers to study their effect on selected components of the faecal flora. Furthermore, the effect of GOS-containing syrup on colonisation and persistence of B. lactis Bb-12 in the gastrointestinal (GI) tract was evaluated.

2. MATERIALS AND METHODS

2.1. Selection criteria for bifidobacterial strains

Three commercially available bifidobacterial strains (B. lactis Bb-12, Bifidobacterium spp. 420, Bifidobacterium spp. H) were provided by European culture suppliers (Chr. Hansen, Horsholm, Denmark; Danisco Cultor, Niebuell, Germany, and SKW Biosystems, Boenen, Germany, respectively) as lyophilised powders. The strains were cultured in TPY broth (Scardovi, 1986) under anaerobic conditions for 48 h at 37°C and screened for their suitability to be used in the feeding trials. The selection criteria included technological properties (availability as a powder with high cell density, viable count stability in yoghurt), availability of scientific documentation about probiotic properties of the particular strain, and functional as well as metabolic characteristics (resistance against bile and acidity, utilisation of carbohydrates, including galacto-oligosaccharides).

GOS-containing syrup (preparation A in Table 1 and Fig. 2: 60% GOS, 20% lactose, 19% glucose, 1% galactose in 75% dry matter) was obtained commercially (kindly provided by Borculo Domo Ingredients, The Netherlands), GOS-containing preparation B (Table 1 and Fig. 2: 18% monosaccharides (galactose and glucose), 49% disaccharides (mainly lactose), 28% trisaccharides, 5% tetrasaccharides) was provided by Prof. K.D. Kulbe, Institute of Food Technology, University of Agricultural Science, Vienna, Austria, but originated from Dr. Patrick Adlerkreutz, Center for Chemistry and Chemical Engineering, Lund University, Sweden. Prof. Kulbe kindly provided all other sugars.

 

Table 1. Selection of bifidobacterial strains for incorporation into the test product
 

 

Fig. 2. Growth kinetics of probiotic Bifidobacterium lactis Bb-12 strain in TPY medium enriched with different carbohydrates at 1% (w/v). FOS A and B are fructo-oligosaccharide preparations, GOS A (syrup used in the present feeding trial) and B galacto-oligosaccharide preparations.

 

Bile and acidity resistance properties were evaluated based on culture tests (loop-streakings) on TPY medium containing 1.5% (w/v) agar–agar and different concentrations ranging from 0.1% to 0.5% (w/v) of bile salt mixture (Sigma B8756, Sigma Chemicals, St. Louis, USA) and on TPY agar of different pH values (3.5–5.0), under anaerobic conditions. Carbohydrate utilisation was monitored using the Bioscreen C system equipped with BioLink software package (Labsystem, Helsinki, Finland) which allows a simultaneous detection of optical densities of bacterial cultures in a chosen medium (TPY broth) supplemented with the different carbohydrates at 1% (w/v), in a microtiter format. To ensure anaerobic conditions throughout the screening, 1 mL of Oxyrase (Oxyrase Inc., Mansfield OH, USA) was added to 50 mL of TPY medium to remove oxygen. Furthermore, the microcavities of the test system were covered by two drops of sterile paraffin oil after inoculation. All tests were performed at 37°C. To assess the viable count stability, bifidobacterial counts in yoghurt were examined using the methodology described by Pacher and Kneifel (1996).

2.2. Products

B. lactis Bb-12 (selected for feeding trial based on characteristics introduced above) was applied as a freeze-dried powder containing, according to the supplier, 3×10¹⁰ cfu g⁻¹. The powder was divided and sealed in portion-sized (0.5 g) foil bags and the viability of the B. lactis Bb-12 strain in the bags was monitored twice a week throughout the study. The commercial galacto-oligosaccharide was used as GOS-containing syrup, which was divided into 10 g aliquots in plastic tubes. Due to the viscous nature of GOS-containing syrup its yield for each consumption occasion was estimated to be 9 g (containing 4.05 g GOS, 1.35 g lactose and 1.29 g glucose).

2.3. Subjects

The study population consisted of 30 healthy volunteers (27 females, 3 males), whose mean age was 32 years (range 22–47 years). Inclusion criteria included normal bowel function and no antimicrobial therapy for four weeks prior to the study. The volunteers retained their normal diet during the study with the exception that the consumption of probiotic products was forbidden. All volunteers gave their written informed consent to participate in the present study lasting for six weeks.

2.4. Study design

The feeding trial started with a two-week pre-feeding period during which the volunteers consumed a beaker of 125 mL plain low-fat, non-sugar yoghurt twice a day. This was followed by a two-week feeding period for which the volunteers were divided randomly to three groups each containing ten subjects. During the feeding-period (1) GOS-containing syrup (approx. 9 g), (2) B. lactis Bb-12 (0.5 g) or (3) GOS-containing syrup and B. lactis Bb-12 together (same amounts as before) were added to the yoghurt prior to twice-daily consumption. The daily dose for GOS was 8.1 g and for Bb-12 3×10¹⁰ cells. For practical reasons the volunteers added the prepared aliquots of GOS-containing syrup and/or B. lactis Bb-12 to the pre-packed yoghurts themselves prior consumption. Hence, volunteers knew which group they belonged to during the trial. However, the sample numbers were randomised and the groups were blinded for the sample handlers. The feeding trial was concluded with a two-week post-feeding period (normal diet without probiotic products).

Two volunteers from the GOS-containing syrup group interrupted the study: one started an antibiotic therapy and the other demonstrated gastrointestinal symptoms (pain from intestinal bloating) during the feeding period. Otherwise the products were well tolerated.

2.5. Microbiological analyses

Faecal samples were taken after the pre-feeding and feeding periods and twice during the post-feeding period. Samples were placed in plastic vials containing holes in the lid for gas exchange and immediately sealed in anaerobic bags (Anaerocult^® A mini, Merck, Darmstadt, Germany). The faecal samples were analysed within 0.5–5 h (mean 2.4 h) after defecation. Samples were weighed, diluted in peptone saline with cysteine (0.5 g L⁻¹) and anaerobic bacteria were cultured in an anaerobic chamber (H₂/CO₂/N₂; 10 : 10 : 80, MACS, Don Whitley Scientific Ltd., Shipley, West Yorkshire, UK). Bifidobacteria were enumerated on Beerens agar (Beerens, 1990) and Clostridium perfringens on Tryptose Sulfite Cycloserine agar (Merck) after 4 d of incubation in anaerobic jars (H₂/CO₂/N₂; 10 : 5 : 85, Anoxomat WS8000, Mart^® Microbiology, Lichtenvoorde, Holland), lactic acid bacteria on Rogosa agar (Oxoid, Basingstoke, Hampshire, UK) after 3 d of incubation in anaerobic jars (Anaerocult^® A, Merck) and coliforms on MacConkey agar (Difco Laboratories, Detroit, MI, USA) after 2 d of aerobic incubation. All plates were incubated at 37°C.

2.6. Identification of B. lactis Bb-12

Isolates from faecal samples from subjects consuming the probiotic during the feeding period (GOS-containing syrup+Bb-12 and Bb-12 groups) were subjected to detection of B. lactis Bb-12 by culture with a following randomly amplified polymorphic DNA (RAPD) genotyping step. Since B. lactis is more aerotolerant and more resistant to acidic conditions and elevated temperatures than most other human bifidobacterial strains (Gavini et al., 1991; Meile et al., 1997), detection of the strain from faecal samples was facilitated by an aerobic acid-treatment step prior to culture and by incubating the cultured plates at an elevated temperature. The acid-treatment and cultivation of B. lactis Bb-12 were performed as follows: A volume of 0.5 mL of faecal sample dilution 10⁻¹ was added to 4.5 mL 0.2  HCl/KCl buffer (pH 2). The solution was incubated aerobically at 37°C for 90 min, serially diluted, plated on Bifidobacterium agar (Sutter, Citron, Edelstein, & Finegold, 1985) and incubated at 44°C for 3 d in an anaerobic jar (Anaerocult A, Merck). Isolates resembling B. lactis Bb-12 by colony morphology were selected for RAPD analysis (up to 12 isolates/sample).

Template DNA for RAPD analysis was prepared from cultures grown anaerobically over-night in MRS+cysteine (0.5 g L⁻¹) broth (Oxoid) by mechanical lysis of the cells. Briefly, the cells were harvested from 1 mL of broth, re-suspended in 500 L of sterile water and mixed with 100 mg of glass-beads (acid-washed glass beads, 150–212 m, Sigma Chemical Co., St. Louis, MO, USA) followed by disruption of the cells by RiboLyser™ (Hybaid Ltd., Ashford, Middlesex, UK) at 6.5 m s⁻¹ speed for 30 s and removal of glass beads and undisrupted cell particles by centrifugation. RAPD was performed in a 50 L reaction volume consisting of 0.2 m dNTP (Finnzymes Ltd., Espoo, Finland), 0.4  random sequence primer, Dynazyme buffer (10 m Tris, 150 m KCl, 1.5 m MgCl₂, 0.1% Triton X-100, final conc.), 3 U DYNAzyme™ II DNA polymerase (Finnzymes Ltd.) and 5 L of template. Random primer OPA-2 (5′-TGCCGAGCTG-3′) was used for the fingerprinting of all isolates (232 isolates), and two additional primers, OPA-5 (5′-AGGGGTCTTG-3′) and OPA-13 (5′-CAGCACCCAC-3′) were used for the confirmation of identifications when necessary (90 isolates). The temperature profile in a thermocycler (Uno II, Biometra, Göttingen, Germany) was 35 cycles as follows: at 95°C for 1 min, at 32°C for 1 min, at 72°C for 2 min. The initial denaturation was at 95°C for 5 min and the final extension at 72°C for 10 min. Amplification products were analysed electrophoretically in 1% (w/v) agarose gel containing ethidium bromide (0.5 g mL⁻¹) and visualised under ultraviolet light.

The discriminatory power of the RAPD was validated with the strains B. adolescentis VTT E-981074^T, B. adolescentis VTT E-991436, B. animalis VTT E-96663^T, B. bifidum VTT E-97795^T, B. breve VTT E-981075^T, B. catenulatum VTT E-991437^T, B. denticolens VTT E-991434^T, B. dentium VTT E-991438^T, B. infantis VTT E-97796^T, B. inopinatum VTT E-991435^T, B. lactis VTT E-97847^T, B. longum VTT E-94505, B. longum VTT E-96664^T, B. longum VTT E-96702 and B. pseudocatenulatum VTT E-991439^T from VTT culture collection (Espoo, Finland), and B. bifidum Bb-11, B. infantis Bb-02 and B. lactis Bb-12 from Christian Hansen Ltd.

Primer OPA-2 (Fig. 1), OPA-5 and OPA-13 gave identical RAPD genotype to B. lactis VTT E-97847 and B. lactis Bb-12, but distinguished B. lactis Bb-12 from all the other bifidobacterial species. Hence, OPA-2 was considered discriminatory enough for the RAPD genotyping of the present isolates.

 

 

Fig. 1. RAPD profiles of Bifidobacterium strains. Lane 1: B. lactis Bb-12, lane 2: B. lactis VTT E-97847^T, lane 3: B. animalis VTT E-96663^T, lane 4: B. breve VTT E-981075^T, lane 5: B. infantis VTT E-97796^T, lane 6: B. longum VTT E-94505, lane 7: B. bifidum Bb-11, lane 8: B. adolescentis VTT E-981074^T, lane 9: B. catenulatum VTT E-991437^T, lane 10: B. pseudocatenulatum VTT E-991439^T, lane 11: B. dentium VTT E-991438^T, lane 12: B. denticolens VTT E-991434^T, lane 13: B. inopinatum VTT E-991435^T. A 1kb DNA ladder was used as a molecular weight standard.

 

2.7. Statistical analysis

Student's t-test was used to determinate the significant differences between the studied groups.

3. RESULTS

3.1. Probiotic strain and galacto-oligosaccharide selection

The results of the probiotic strain selection are shown in Table 1. In spite of comparable resistance and utilisation patterns of all three strains, Bb-12 was chosen, mainly due to two criteria, i.e. its extensive background of documentation regarding the probiotic properties and its higher viable count stability in the test product. The GOS-containing substrates were utilised moderately (GOS-containing syrup, preparation A) or only weakly (GOS-containing preparation B) by all strains. Fig. 2 shows the growth curves of B. lactis Bb-12 registered as optical densities of liquid bacterial cultures in TPY containing different carbohydrates.

3.2. The effect of the products on faecal bacteria

The results of the microbial analysis of faecal samples are shown in Fig. 3 and Table 2. Large inter-individual variations were observed in the numbers of bifidobacteria, lactic acid bacteria, C. perfringens and coliforms in GOS-containing syrup, GOS-containing syrup+Bb-12, and Bb-12 groups. Due to the overgrowth of B. lactis Bb-12 on Rogosa agar plates lactic acid bacterial numbers in the feeding-period samples of the two groups consuming the Bb-12 could not be determined.

 

Fig. 3. The changes in faecal bifidobacterial numbers of GOS-containing syrup (——×), Bb12 (– – – ) and GOS-containing syrup+Bb-12 (— — — ) groups. Lines represent the mean values and symbols represent the individual values. *P<0.05, **P<0.01.

 

Table 2. The mean numbers of lactic acid bacteria, C. perfringens and coliforms in faecal samples (log (cfu g⁻¹ wet weight), numbers in parenthesis indicate the minimum and maximum values)
 

 

Before the feeding period, the mean numbers of bifidobacteria were log₁₀ (cfu g⁻¹ wet weight) 9.2 in the GOS-containing syrup group, log₁₀ 8.7 in the Bb-12 group and log₁₀ 9.0 in the GOS-containing syrup+Bb-12 group. During the feeding period the mean numbers of bifidobacteria increased slightly in all study groups (Fig. 3). The increase was higher in the GOS-containing syrup+Bb-12 group (log₁₀ 9.6) than in the Bb-12 group (log₁₀ 9.1). These increases were statistically significant (P<0.01). In the GOS-containing syrup group the increase in bifidobacterial numbers (from log₁₀ 9.2 to 9.4) was not statistically significant. After one-week of the washout period bifidobacterial numbers were still elevated in the GOS-containing syrup+Bb-12 group (log₁₀ 9.5), but they decreased to the initial level after two-weeks of washout.

In the GOS-containing syrup group a slight decrease in the mean number of C. perfringens (P<0.05) was observed (Table 2). GOS-containing syrup consumption together with Bb-12 showed no effect on C. perfringens numbers. No statistically significant changes were detected in other bacterial numbers in different groups.

3.3. The effect of the products on B. lactis Bb-12

The numbers of isolates with B. lactis Bb-12 RAPD genotype were below the detection limit (log₁₀ 4) in all volunteers in the GOS-containing syrup+Bb-12 group before the feeding period, indicating that B. lactis Bb-12 was absent (or below the detection level) in the subjects’ faeces at this sampling occasion. Isolates with B. lactis Bb-12 RAPD genotypes were found in two volunteers’ samples in the Bb-12 group (log₁₀ 5.4 and 7.2) before the feeding period (Table 3).

 

 

Table 3. The numbers of B. lactis Bb-12 positive individuals and the mean numbers of Bb-12 in their faecal samples. Numbers in parenthesis indicate the minimum and maximum values
 

 

After the two-week feeding period isolates with B. lactis Bb-12 RAPD genotype were found in high numbers in both Bb-12 and GOS-containing syrup+Bb-12 groups (Table 3). The mean number of isolates with B. lactis Bb-12 RAPD genotype reached slightly higher level in the group consuming GOS-containing syrup in addition to B. lactis Bb-12 (log₁₀ 9.0) than in the group consuming B. lactis Bb-12 only (log₁₀ 8.6). This difference was not statistically significant. Isolates with B. lactis Bb-12 RAPD genotype represented 23% (range 9–100%) and 35% (range 3–71%) of the total faecal bifidobacterial numbers in Bb-12 and GOS-containing syrup+Bb-12 groups, respectively.

The numbers of isolates with B. lactis Bb-12 RAPD genotype were low after one- and two-week washout periods in both Bb-12 and GOS-containing syrup+Bb-12 groups (Table 3). After one-week of washout four of the 10 volunteers in both of the groups harboured isolates with B. lactis Bb-12 RAPD genotype, whereas after two-weeks of washout this genotype was detected in only two volunteers in each of the two groups. One of these volunteers harboured the B. lactis Bb-12 RAPD genotype already before the feeding period.

4. DISCUSSION

In the present study the effect of GOS-containing syrup (60% GOS) ingested alone (daily dose of GOS being 8.1 g) or together with B. lactis Bb-12 on selected components of the faecal flora of healthy volunteers was studied. GOS is utilised in vitro by several intestinal bacteria including bifidobacteria, and ingested GOS has occasionally been shown to increase the faecal bifidobacterial numbers (Ito et al., 1993; Bouhnik et al., 1997). In the present study B. lactis strain Bb-12 was shown to utilise GOS-containing syrup in vitro. This result is also in accordance with earlier findings (Kneifel et al., 2000), where a strong strain-dependent utilisation behaviour of probiotic bacterial candidates could be demonstrated. In addition, the Bb-12 strain exhibited optimum viable count stability and has a documented history of use as a probiotic microorganism. Therefore, it was chosen as a probiotic strain for the present synbiotic feeding trial. With the selection of a specific prebiotic–probiotic pair we wanted to study whether a specific prebiotic can enhance the survival and persistence of an ingested probiotic strain in the GI-tract.

A few human feeding trials have been performed with GOS of which three demonstrated an increase in faecal bifidobacterial numbers (Ito and Ito; Bouhnik et al., 1997) and two showed no effect ( Teuri, Korpela, Saxelin, Montonen, & Salminen, 1998; Alles et al., 1999) as a result of GOS consumption. Increase in the numbers of faecal bifidobacteria was shown when 10 g GOS was ingested daily by 20–32 year old subjects (initial level log₁₀ 8.6; increase log₁₀ 1.1; P<0.05; Bouhnik et al., 1997). Similarly, Ito et al. (1990) detected an increase in bifidobacterial numbers in 26–48 year old males with the daily consumption of 10 g GOS (initial level log₁₀ 9.8; increase log₁₀ 0.3; P<0.01) and even at a daily dose of 2.5 g GOS in 35–55 year old males (initial level log₁₀ 9.4; increase log₁₀ 0.3; P<0.05;Ito et al., 1993). Teuri et al. (1998) and Alles et al. (1999) did not find any statistically significant changes in bifidobacterial numbers when they studied the effect of daily doses of 7.5 and 15 g of GOS (initial bifidobacterial levels were log₁₀ 7.3 and 9.4, respectively). We observed a slight increase (from log₁₀ 9.2 to 9.4; the difference was not statistically significant) in bifidobacterial numbers with a daily dose of 8.1 g GOS. Thus, somewhat conflicting results have been obtained on the effect of GOS on faecal bifidobacterial numbers. This discrepancy may result from differences in the populations (sex, age), in the normal diets of the volunteers (Western vs. Japanese) and consequently in the normal faecal flora of the volunteers (e.g. varying bifidobacterial numbers and strain types), in sample handling, in cultivation techniques and in data handling. Furthermore, GOS, like many other prebiotic preparations (including inulin), is by no means specifically utilised by a certain bacterial group alone but instead is fairly widely used by a number of gastrointestinal bacteria (Tanaka et al., 1983; Ishikawa et al., 1995). This lack of specificity likely has abolished the postulated beneficial bifidogenic effect in some human feeding trials.

Most earlier B. lactis Bb-12 feeding trials have studied the effect of bifidobacterial ingestion on the total faecal bifidobacterial numbers only (numbers of specific exogenous bifidobacterial strain were not determined)(Link-Amster et al., 1994; Schiffrin et al., 1995). In the present study the daily ingestion of 3×10¹⁰ B. lactis Bb-12 cells increased the numbers of faecal bifidobacteria from log₁₀ 8.7 to 9.1. Earlier studies have observed higher increases in faecal bifidobacterial numbers than the present study with daily dose of approx. 10¹⁰ B. lactis Bb-12 cells (Link-Amster et al., 1994; Schiffrin et al., 1995). However, in these earlier studies the initial bifidobacterial levels were much lower than in the present study (log₁₀ 5.9; Link-Amster et al., 1994; log₁₀ 7.8; Schiffrin et al., 1995). A likely explanation for the different observations is that when initial bifidobacterial numbers are already high, it is difficult to further increase the size of the population by ingesting exogenous bifidobacterial cells. In the present study the total faecal bifidobacterial numbers decreased below or close to the detection level shortly after the feeding period ended. The same result, indicating transient colonisation of B. lactis Bb-12 in the GI-tract, was obtained in earlier studies as well (Link-Amster et al., 1994; Schiffrin et al., 1995).

The present study has shown that B. lactis Bb-12 strain survived well through the GI-tract. This was evident by increased numbers of faecal isolates with B. lactis Bb-12 RAPD genotype during probiotic feeding. B. lactis Bb-12 RAPD genotype was detected in the faecal samples of each of the 20 subjects consuming the probiotic strain (Bb-12 and GOS-containing syrup+Bb-12 groups). The earlier results of Fukushima et al. (1998) confirm our finding on good survival of Bb-12 in the human GI-tract. In the study of Fukushima et al. (1998) healthy children aged 15–31 months consumed at least 10⁹ of B. lactis Bb-12 for 20 days. B. lactis Bb-12, differentiated from other bifidobacterial strains by colonial and cellular morphology, was found in the faeces of five of seven children representing 27% (mean) of the total faecal bifidobacterial flora. In the present study B. lactis Bb-12 represented 23% or 35% (mean) of the total faecal bifidobacterial flora, depending on the study group.

Two of the 20 subjects consuming Bb-12 had a B. lactis Bb-12 RAPD genotype in their pre-feeding samples in the present study. This is likely explained by the fact that Bb-12 is commonly used in probiotic foods in Finland. In these two subjects a previous consumption of Bb-12-containing food had possibly led to a longer-term colonisation with the strain. Other possible explanations for the finding are that the two volunteers accidentally consumed Bb-12-containing products during the pre-feeding period or that the RAPD genotyping method applied for bifidobacterial fingerprinting was not discriminatory enough. However, the absence of B. lactis Bb-12 RAPD genotype from the pre-feeding samples of 18 out of 20 volunteers indicated that the genotyping method applied was generally discriminatory enough for the present purposes. Furthermore, when the genotyping was always preceded by a specific culture step selective for Bb-12, the likelihood that some of the isolates with B. lactis Bb-12 RAPD genotype actually represented some other strain becomes even smaller.

There are relatively few reports on synbiotic feeding trials on humans. Tanaka et al. (1993) failed to observe any change in faecal bifidobacterial numbers when GOS (daily dose 3 g or 10 g) was ingested simultaneously with an exogenous bifidobacterial strain (B. breve 4006). Bouhnik et al. (1996) made a similar conclusion with inulin (daily dose 18 g) and Bifidobacterium sp. (isolated from commercial fermented milk Ofilus, Yoplait, Paris, France). In the present study the ingestion of GOS-containing syrup had no significant effect on faecal B. lactis Bb-12 numbers, but total bifidobacterial numbers remained elevated after one-week washout period in the group consuming GOS-containing syrup and B. lactis Bb-12. B. lactis is known to be much more tolerant to many harsh conditions (including better acid and oxygen tolerance) than many other bifidobacterial strains(Meile et al., 1997; Crittenden et al., 2001). Hence, in the cases where a probiotic strain is a good survivor it may be difficult to show any additional value for the simultaneous prebiotic consumption.

In conclusion, the results show that B. lactis Bb-12 survives well in the human GI-tract. GOS-containing syrup (with a daily dose of 8.1 g), however, did not enhance the survival or persistence of the strain in the GI-tract.

ACKNOWLEDGEMENTS

Authors thank all the volunteers for their co-operation and Marja-Liisa Jalovaara and Niina Torttila for their skilful technical assistance. Support from Prof. K.D. Kulbe, co-ordinator of the EU project "Enzymatic lactose valorisation" (FAIR CT96-1048), is gratefully acknowledged.

REFERENCES

Alles, M.S., Hartemink, R., Meyboom, S., Harryvan, J.L., van Laere, K.M., Nagengast, F.M. and Hautvast, J.G., 1999. Effect of transgalacto-oligosaccharides on the composition of the human intestinal microflora and on putative risk markers for colon cancer. American Journal of Clinical Nutrition 69, pp. 980–991.

Alm, L., Ryd-Kjellen, E., Setterberg, G., & Blomquist, L. (1993). Effect of new fermented milk product ‘Cultura’ on constipation in geriatric patients. 1st lactic acid bacteria computer conference. May 1–June 30, 1994..

Beerens, H., 1990. An elective and selective isolation medium for Bifidobacterium spp. Letters in Applied Microbiology 11, pp. 155–157.

Benno, Y., Endo, K., Mizutani, T., Namba, Y., Komori, T. and Mitsuoka, T., 1989. Comparison of faecal microflora of elderly persons in rural and urban areas of Japan. Applied and Environmental Microbiology 55, pp. 1100–1105.

Black, F. T., Andersen, P. L., Orskov, J., Orskov, F., Gaarslev, K., & Laulund, S. (1989). Prophylactic efficacy of lactobacilli on traveller's diarrhoea. Travel Medicine, 333–335.

Black, F., Einarsson, K., Lidbeck, A., Orrhage, K. and Nord, C.E., 1991. Effect of lactic acid production bacteria on the human intestinal microflora during ampicillin treatment. Scandinavian Journal of Infections Diseases 23, pp. 247–254.

Bouhnik, Y., Flourié, B., Andrieux, C., Bisetti, N., Briet, F. and Rambaud, J.C., 1996. Effects of Bifidobacterium sp. fermented milk ingested with or without inulin on colonic bifidobacteria and enzymatic activities in healthy humans. European Journal of Clinical Nutrition 50, pp. 269–273.

Bouhnik, Y., Flourié, B., D’Agay-Abensour, L., Pochart, P., Gramet, G., Durand, M. and Rambaud, J.C., 1997. Administration of translacto-oligosaccharides increases faecal bifidobacteria and modifies colonic fermentation metabolism in healthy humans. Journal of Nutrition 127, pp. 444–448.

Bouhnik, Y., Vahedi, K., Achour, L., Attar, A., Salfati, J., Pochart, P., Marteau, P., Flourié, B., Bornet, F. and Rambaud, J.C., 1999. Short-chain fructo-oligosaccharides administration dose-dependently increases faecal bifidobacteria in healthy humans. Journal of Nutrition 129, pp. 113–116.

Brighenti, F., Casiraghi, M.C., Canzi, E. and Ferrari, A., 1999. Effect of consumption of a ready-to-eat breakfast cereal containing inulin on the intestinal milieu and blood lipids in healthy male volunteers. European Journal of Clinical Nutrition 53, pp. 726–733.

Crittenden, R.G., Morris, L.F., Harvey, M.L., Tran, L.T., Mitchell, H.L. and Playne, M.J., 2001. Selection of a Bifidobacterium strain to complement resistant starch in a synbiotic yoghurt. Journal of Applied Microbiology 90, pp. 268–278.

Finegold, S.M., Attebery, R. and Sutter, V., 1974. Effect of diet on human faecal flora: Comparison of Japanese and American diets. American Journal of Clinical Nutrition 27, pp. 1456–1469.

Finegold, S.M., Sutter, V.L. and Mathisen, G.E., 1983. Normal indigenous intestinal flora. In: Hentges, D.J., Editor, , 1983. Human intestinal microflora in health and disease, Academica Press, New York, pp. 3–31.

Fukushima, Y., Kawata, Y., Hara, H., Terada, A. and Mitsuoka, T., 1998. Effect of probiotic formula on intestinal immunoglobulin A production in healthy children. International Journal of Food Microbiology 42, pp. 39–44.

Gavini, F., Pourcher, A.-M., Neut, C., Monget, D., Romond, C., Oger, C. and Izard, D., 1991. Phenotypic differentiation of bifidobacteria of human and animal origins. International Journal of Systematic Bacteriology 41, pp. 548–557.

Gibson, G., Beatty, E., Wang, X. and Cummings, J., 1995. Selective stimulation of bifidobacterium in the colon by oligofructose and inulin. Gastroenterology 108, pp. 975–982.

Gomes, A. and Malcata, F., 1999. Bifidobacterium spp. and Lactobacillus acidophilus: Biological, biochemical, technology and therapeutical properties relevant for use as probiotics. Trends in Food Science & Technology 10, pp. 139–157.

Ishikawa, F., Takayama, H., Matsumoto, K., Ito, M., Chonan, O., Deguchi, Y., Kikuchi-Hayakawa, H. and Watanuki, M., 1995. Effects of 1–4 linked galacto-oligosaccharides on human faecal microflora. Bifidus 9, pp. 5–18.

Ito, M., Deguchi, Y., Matsumoto, K., Kimura, M., Onodera, N. and Yajima, T., 1993. Influence of galacto-oligosaccharides on human faecal microflora. Journal of Nutritional Science and Vitaminology 39, pp. 635–640.

Ito, M., Deguchi, Y., Miyamori, A., Matsumoto, K., Kikuchi, K., Matsumoto, K., Kobayashi, Y., Yajima, T. and Kan, T., 1990. Effects of administration of galacto-oligosaccharides on the human faecal microflora, stool weight and abdominal sensation. Microbial Ecology in Health and Disease 3, pp. 285–292.

Kankaanpää, P., Sutas, Y., Salminen, S. and Isolauri, E., 1998. Results on clinical demonstration of probiotics in children. In: Mattila-Sandholm, T. and Kauppila, T., Editors, 1998. Functional food research in Europe 3rd workshop FAIR CT96-1028, VTT symposium 187, Edita, Espoo, Finland, pp. 38–39.

Kleessen, B., Sykura, B., Zunft, H.J. and Blaut, M., 1997. Effects of inulin and lactose on faecal microflora, microbial activity and bowel habit in elderly constipated persons. American Journal of Clinical Nutrition 65, pp. 1397–1402.

Kneifel, W., Rajal, A. and Kulbe, K.D., 2000. In vitro growth behaviour of probiotic bacteria in culture media of prebiotic importance. Microbial Ecology in Health and Disease 12, pp. 27–34.

Kruse, H.P., Kleessen, B. and Blaut, M., 1999. Effects of inulin on faecal bifidobacteria in human subjects. British Journal of Nutrition 82, pp. 375–382.

Kunz, C. and Rudloff, S., 1993. Biological functions of oligosaccharides in human milk. Acta Paediatrica 82, pp. 903–912.

Link-Amster, H., Rochat, F., Saudan, K.Y., Mignot, O. and Aeschlimann, J.M., 1994. Modulation of a specific humoral immune response and changes in intestinal flora mediated through fermented milk intake. FEMS Immunology and Medical Microbiology 10, pp. 55–63.

Marteau, P., Pochart, P., Flourie, B., Pellier, P., Santos, L., Desjeux, J.F. and Rambaud, J.C., 1990. Effect of chronic ingestion of a fermented dairy product containing Lactobacillus acidophilus and Bifidobacterium bifidum on metabolic activities of the colonic flora in humans. American Journal of Clinical Nutrition 52, pp. 685–688.

Meile, L., Ludwig, W., Rueger, U., Gut, C., Kaufmann, P., Dasen, G., Wenger, S. and Teuber, M., 1997. Bifidobacterium lactis sp. Nov., a moderately oxygen tolerant species isolated from fermented milk. System and Applied Microbiology 20, pp. 57–64.

Mitsuoka, T., 1982. Recent Trends in Research on Intestinal Flora. Bifidobacteria Microflora 1, pp. 3–24.

Mitsuoka, T., 1992. Intestinal Flora and Aging. Nutrition Reviews 50, pp. 438–446.

Pacher, B. and Kneifel, W., 1996. Development of culture medium for the detection and enumeration of bifidobacteria in fermented milk products. International Dairy Journal 6, pp. 43–64.

Roberfroid, M.B., 1998. Prebiotics and synbiotics: Concepts and nutritional properties. British Journal of Nutrition 80, pp. S197–S202.

Roberfroid, M.B., van Loo, J.A.E. and Gibson, G., 1998. The bifidogenic nature of chicory inulin and its hydrolysis products. Journal of Nutrition 128, pp. 11–19.

Saavedra, J.M., Bauman, N.A., Oung, I., Perman, J.A. and Yolken, R.H., 1994. Feeding of Bifidobacterium bifidum and Streptococcus thermophilus to infants in hospital for prevention of diarrhea and shedding of rotavirus. Lancet 344, pp. 1046–1049.

Saito, T., 1987. Chemical structure of three neutral trisaccharides isolated in free form from bovine colostrum. Carbohydrate Research 165, pp. 43–51.

Salminen, S., Deighton, M.A., Benno, Y. and Gorbach, S., 1998. Lactic acid bacteria in Health and Disease. In: Salminen, S. and von Wright, A., Editors, 1998. Lactic acid bacteria, microbiology and functional aspects ((2nd ed.)),, Marcel Dekker, New York, USA, pp. 211–253.

Scardovi, V., 1986. Genus Bifidobacterium. In: Sneath, P.H., Mair, N.S., Sharpe, M.E. and Holt, J.G., Editors, 1986. Bergeys manual of systematic bacteriology Vol. 2, Williams and Wilkins, Baltimore, USA, pp. 1418–1434.

Schiffrin, E.J., Rochat, F., Link-Amster, H., Aeschlimann, J.M. and Donnet-Hughes, A., 1995. Immuno-modulation of human blood cells following the ingestion of lactic acid bacteria. Journal of Dairy Science 78, pp. 491–497.

Sutter, V., Citron, D., Edelstein, M., & Finegold, S. (1985). Wadsworth anaerobic bacteriology manual (4th ed.) (p. 109). USA: Star.

Tanaka, R., Takayama, H., Morotomi, M., Kuroshima, T., Ueyama, S., Matsumoto, K., Kuroda, A. and Mutai, M., 1983. Effects of administration of TOS and Bifidobacterium breve 4006 on the human faecal flora. Bifidobacteria Microflora 2, pp. 17–24.

Teuri, U., Korpela, L., Saxelin, M., Montonen, L. and Salminen, S., 1998. Increased fecal frequency and gastrointestinal symptoms following ingestion of galacto-oligosaccharide-containing yogurt. Journal of Nutritional Science and Vitaminology 44, pp. 465–471.

Zaraté, S. and López-Leiva, M.H., 1990. Oligosaccharide formation during enzymatic lactose hydrolysis: a literature review. Journal of Food Protection 53, pp. 262–268.

Ziemer, C. and Gibson, G., 1998. An overview of probiotics, prebiotics and synbiotics in the functional food concept: Perspectives and future strategies. International Dairy Journal 8, pp. 473–479.

(order Full Text from publisher)