Lactulose, lactitol and lactobionic acid are compounds that can be produced from lactose (or whey) and which, unlike lactose, are not absorbed in the small intestine of lactose-absorbing subjects. Thus, all these compounds have potential to function as prebiotics, substrates that promote the growth of beneficial microbes in the large intestine (Kontula, 1999).

The molecular formula of lactulose is similar to that of lactose (disaccharide of galactose and glucose), the only difference being that in lactulose the glucose residue of lactose is isomerised to fructose(Strohmaier, 1998). Lactulose is used in various types of food products (infant formula, baby food, confectionary, soft drink, milk products) and also pharmaceutically to improve hepatic encephalopathy and constipation ( Mizota, 1996; Strohmaier, 1998). The growth-promoting effect of lactulose for Bifidobacterium was found almost 50 years ago (Strohmaier, 1998). In studies performed in the 1960s and in the 1970s the bifidogenicity of lactulose was shown in formula fed infants (for a review see Strohmaier, 1998). More recently the ability of lactulose to support the growth of intestinal bacteria has been more widely studied. In addition to Bifidobacterium some Clostridium, Lactobacillus and Peptostreptococcus species are able to metabolise lactulose extensively in vitro (Hoffman, Mossel, Korus, & van De Kamer, 1964; Sako, Matsumoto, & Tanaka, 1999; Kneifel, Rajal, & Kulbe, 2000). Furthermore, representatives of several other genera e.g. Bacteroides and Enterococcus can utilise lactulose to a lesser extent (Sahota, Bramley, & Menzies, 1982; Sako et al., 1999).

Beside lactulose, lactitol, a sugar alcohol consisting of galactose and sorbitol, is also used as a sugar substitute in foods such as sugar-free confections, chocolates, chewing gums, no-sugar added baked goods and ice creams (Kummel & Brokx, 2001), and medically to improve hepatic encephalopathy and constipation( Riggio et al., 1990a; Petticrew, Watt, & Sheldon, 1997). Lactitol and lactulose both inhibit detrimental ammonia production and cause a favourable fall in pH in the faeces of healthy subjects ( Kitler et al., 1992). Similarly to lactulose also lactitol is fermented in vitro by several important intestinal bacteria including representatives of Bacteroides, Clostridium, Lactobacillus, Enterococcus and Bifidobacterium species (Kitler et al., 1992; Kontula, Suihko, von Wright, & Mattila-Sandholm, 1999; Kneifel et al., 2000; Kontula, Suortti, Tenkanen, Mattila-Sandholm, & von Wright 2000; Kummel & Brokx, 2001). However, lactulose seems to be a better substrate for intestinal bacteria compared to lactitol since more bacterial species and even strains within species are able to utilise it ( Kitler et al., 1992; Kneifel et al., 2000).

The main applications of lactobionic acid include the use as an ingredient of solutions stabilising organs during transport for transplantation, and the use of calcium lactobionate as calcium supplement in the pharmaceuticals. Other potential industrial applications for lactobionic acid include e.g. the use as a raw material for the synthesis of surface-active substances (detergents). In the food industry several possible applications of lactobionic acid are under investigation including reduction of souring and ripening time in cheese and yoghurt production, promotion of stable gel structures, elimination of bitterness and flavour enhancement, improvement in the taste perception of sourness, and preservation of aroma freshness and protection against oxidation of partially hydrogenated vegetable fats. Furthermore, calcium lactobionate seems to contribute to the improvement of crispness (Gerling, 1998). The data on the utilisation of lactobionic acid by intestinal bacteria are very scarce. In in vitro trials some species/strains of Bifidobacterium and Lactobacillus, or their cell-free extracts, have shown activity against lactobionic acid (Harju, 1993; Kneifel et al., 2000).

The fact that lactose derivatives are not selectively enhancing the growth of only beneficial (probiotic) bacteria such as Lactobacillus spp. and Bifidobacterium spp. in vitro, does not rule out the possibility that they still may show prebiotic potential in vivo. Lack of specificity in growth enhancement in vitro is a common phenomenon among prebiotics. However, in vivo, in the complex microbial ecosystem of the human gut, they may act more selectively (Van Loo et al., 1999). The effect of lactulose on the faecal flora has been investigated in several studies, both in healthy subjects and in subjects with hepatic encephalopathy, cirrhosis or constipation, whereas the effect of lactitol on the faecal flora has been scantly reported, and no reports are available on the effects of lactobionic acid on the gut microbiota ( Vince, Zeegen, Drinkwater, O’Grady, & Dawson, 1974; Riggio et al., 1990b, Kitler et al., 1992; Terada, Hara, Kataoka, & Mitsuoka, 1992; Ballongue & Schumann, 1997; Mangin et al., 2002).

The aim of the present study was to investigate the ability of intestinal Lactobacillus rhamnosus, L. salivarius and L. paracasei strains, selected on the basis of their good lactose derivative utilisation compared to other intestinal Lactobacillus isolates, to utilise lactose derivatives in different growth conditions and to determine how lactose derivatives effect the stability and different characteristics of the potentially probiotic strains.

2. MATERIALS AND METHODS

2.1. Bacterial strains

Four Lactobacillus strains isolated from human faecal or biopsy samples using rye pentosan, lactitol or lactobionic acid enrichment were included into the study (Table 1) (Kontula, 1999). The strains L. rhamnosus VTT E-97800 and VTT E-97948, L. salivarius VTT E-981006, and L. paracasei VTT E-97949, originating from the VTT culture collection, were chosen to the present study on the basis of their good utilisation of lactose derivatives compared to other intestinal Lactobacillus isolates. In previous studies L. rhamnosus strains VTT E-97800 and VTT E-97948 grew on lactulose and lactitol well (E-97948 grew also on lactobionic acid moderately), L. salivarius VTT E-981006 grew on lactulose well and on lactitol and lactobionic acid moderately, and L. paracasei VTT E-97949 grew on lactulose and lactitol well and on lactobionic acid moderately (Kontula, 1999).

 

 

Table 1. Lactobacillus strains used in the study
 

 

Lactobacillus strains were maintained at −70°C and initially grown on MRS (Oxoid) agar plates at 37°C for 2 d in anaerobic jars (H₂/CO₂/N₂; 10:5:85, Anoxomat WS8000, Mart^® Microbiology, Lichtenvoorde, Holland). Growth tests were initiated by growing Lactobacillus strains anaerobically in MRS broth at 37°C overnight and, unless otherwise stated, all further incubations were also performed in anaerobic conditions at 37°C.

2.2. Growth media

To study the ability of Lactobacillus strains to utilise lactose derivatives carbohydrate-free MRS was used as a basal growth medium. The basal medium had the following components (gL⁻¹): peptone from casein (10.0) (Merck, Germany), yeast nitrogen base amino acids (5.0) (Difco, USA), Na-acetate×3H₂O (5.0) (Riedel-de Haën, Germany), K₂HPO₄×3H₂O (2.0) (Merck, Germany), (NH4)₃C₆H₅O₇×2H₂O (2.0) (BDH, England), MgSO₄×7H₂O (0.2) (Merck, Germany), MnSO₄×4H₂O (0.05 g) (Merck, Germany) and Tween 80 (1 mL) (Fluka, Switzerland). The pH was adjusted to 6.2 and the medium sterilised at 121°C for 15 min.

Stock solutions (10%, unless otherwise stated) of lactulose (ICN Biomedicals Inc., USA), lactitol (Suomen Xyrofin Oy, Finland), lactobionic acid (Aldrich, Germany), glucose (BDH, England), galactose (Merck, Germany), and lactose (BDH, England) were prepared in ion-exchanged water and filter-sterilised. Sterile substrate solutions were added into basal MRS-medium to obtain final carbohydrate concentrations of 0.5%, 1% or 2%. The pH of the supplemented MRS was checked and adjusted if necessary (in the case of lactobionic acid to pH 6.5 to ensure that it would remain mainly in the dissociated form throuhgout the incubation).

2.3. Growth on lactose derivatives

Initial testing for the growth of Lactobacillus strains on lactose derivatives was performed as follows: Basal MRS supplemented with different concentrations of carbohydrates (0.5%, 1% or 2%) was divided into test tubes (5 mL in each) and inoculated with 50 L of the overnight Lactobacillus culture. Glucose was used as a positive control and water as a negative control. Test tubes were incubated at 37°C for 2 d after which the pH and turbidity (Multiskan, Labsystems, Finland) of the growth media were measured.

The utilisation of carbohydrates by Lactobacillus strains was more thoroughly studied using the turbidometer Bioscreen C system equipped with BioLink software package (Labsystems, Finland) which allows a simultaneous detection of optical densities of several bacterial cultures in a microtiter format. Bacterial suspensions (30 L from an overnight MRS broth culture diluted 1:10) was added into the wells of honeycomb plates (Labsystems, Finland) containing 270 L of the basal MRS medium supplemented with carbohydrates (0.5%, 1% or 2%). All runs were performed aerobically (Bioscreen C system does not enable the creation of anaerobic conditions) at 37°C for 2 d and the turbidity of the growth medium measured at 10 min intervals. The turbidometer automatically plotted the bacterial growth curve and calculated the corresponding growth area. The results were calculated as percentages of growth area. All measurement were performed in triplicate parallels and performed twice.

2.4. Acid and bile tolerance

Initial screening for acid tolerance (to test the suitable pH for the assays) was performed with the modified method of Jin, Ho, Abdullah, and Jalaludin (1998): Lactobacillus strains were grown in 10 mL of MRS broth overnight anaerobically at 37°C. The bacterial cells were centrifuged (3000 rpm, 10 min), washed with 5 mL of PBS buffer (0.007  Na-phosphate in 0.85% NaCl, pH 7.2) and resuspended in 1 mL of PBS buffer (cell suspension for acid and bile tolerance tests). A portion of 0.2 mL of cell suspension was mixed with 2 mL of PBS buffer (pH 2.0, 3.0, and 7.2). The cell suspensions were incubated at 37°C in water bath and samples (0.5 mL) were taken after 1, 2, and 3 h. Samples were diluted in tenfold steps in PS broth (Maximal Recovery Diluent, IDG, UK) and 10 L samples of dilutions −3 to −7 were inoculated as spots on MRS agar plates (two spots per dilution). Plates were incubated anaerobically at 37°C overnight and Lactobacillus microcolonies were enumerated from the spots.

After the initial screening the test conditions were modified (pH 2.5 and sampling time of 2 h were used in the acid tolerance tests) and the effect of lactulose and lactitol on acid and bile tolerance was tested as follows: Lactobacillus strains were grown in 10 mL of basal MRS broth supplemented with 1% (final conc.) of glucose, lactulose or lactitol at 37°C overnight. The bacterial cells were centrifuged, washed and resuspended in PBS buffer as above (cell suspension for acid and bile tolerance tests). For the acid tolerance test 0.2 mL of cell suspension was mixed with 2 mL of PBS buffer (pH 2.5 and 7.2), and for the bile tolerance test 0.2 mL of cell suspension was mixed with 2 mL of freshly prepared PBS buffer with 1.5% bile extract (Sigma B-8631) (pH 7.2). The cell suspensions were incubated at 37°C in water bath and samples (0.5 mL) were taken from pH 7.2 suspensions after 0, 2 and 3 h (controls for acid and bile tests), from pH 2.5 suspensions after 2 h and from bile suspensions after 3 h. PBS buffer (pH 7.2) without any additives was used as a control. Lactobacillus numbers in samples were determined as above. Acid and bile tolerance tests were performed twice or thrice (depending on the strain).

2.5. Antimicrobial activity

The effect of lactulose on the antimicrobial activity of Lactobacillus strains was tested using the following indicator strains: Escherichia coli 50 VTT E-94564 (ATCC 11775), Staphylococcus aureus ATCC 6538, Salmonella enterica serovar Typhimurium VTT E-012041 (SH5014), Candida albicans NCPF 3179, and C. krusei NCPF 3848. E. coli and S. aureus were grown on nutrient agar or broth (Difco, USA) and yeasts on yeast maltose agar or broth (Difco, USA). For the antimicrobial activity testing the indicator strains were grown in appropriate broth (see above) aerobically overnight at 37°C. Microbial numbers in the cultures were determined by culture and dilution −2 (in PS) was further used in the activity testing.

Lactobacillus supernatants for testing were obtained as follows: Test strains were grown in glucose- or lactulose-supplemented (final conc. 1%) basal MRS broth at 37°C for 2 d. Bacterial cells were centrifuged (3000 rpm, 10 min) and supernatants recovered for the analysis. The pH of the supernatants was measured and the pH of the control (basal MRS) was adjusted to same level (pH 3.8–4.0) using lactic acid (control I) and HCl (control II).

The antimicrobial activity was tested using Bioscreen C equipment (see above). 240 L of appropriate growth medium for each indicator strain was added into the wells of honeycomb plates in triplicate. Thirty microlitre of an indicator strain culture (see above) and 30 L of Lactobacillus supernatant (from glucose- or lactulose-grown cultures), lactic acid control, HCl control, or growth medium were added into the wells. The run was performed aerobically at 37°C for 1 d and the turbidity of the growth medium measured at 10 min intervals. Two parallel runs were performed. In addition, similar run was performed with supernatants and control with pH adjustment to 6.0 with NaOH. Antimicrobial activity was measured by comparing the areas of the growth curves from wells with (A) and without (B) Lactobacillus supernatant. Inhibition percentage (C) was calculated as follows: 100%−(A/B×100%)=C%.

2.6. Stability during cold storage

The effect of lactulose on the stability of Lactobacillus strains during cold-storage was studied in three ways: (1) freshly grown Lactobacillus cells were added into lactulose-supplemented skim milk and the inoculated milk samples were kept at 4°C, (2) freeze-dried Lactobacillus cells were added into lactulose-supplemented skim milk and the inoculated milk samples were kept at 4°C, (3) Lactobacillus strains were first grown in lactulose-supplemented skim milk and the fermented milk samples were then kept at 4°C. As a control, skim milk without lactulose supplementation was used in all experiments. Pasteurised skim milk (Valio, Finland) packed in 1 L cartons was used in all the experiments. For the stability studies (1 and 2) the milk was further sterilised at 115°C for 15 min, whereas for the fermentation+stability study the milk was only heat-treated (95°C for 10 min). Each test was performed twice.

The stability of fresh Lactobacillus cells in skim milk supplemented with lactulose was studied at 4°C for 22 d. The strains were grown in 40 mL of MRS broth anaerobically at 37°C for 1 d (starting from 1% inoculum). Cells were centrifuged (4000 rpm, 10 min), resuspended into 15 mL of H₂O, centrifuged (4000 rpm, 15 min) and resuspended into 8.5 mL of skim milk. Four milliliters of cell-suspension was transferred into test tube containing either 16 mL of skim milk (control) or 14 mL of skim milk supplemented with 2 mL of 10% lactulose (final conc. 1%). Test tubes were kept at 4°C and bacterial numbers enumerated after 0, 8, 15, and 22 d storage. The spot technique described earlier (see acid and bile tolerance) was used for the culture.

The stability of freeze-dried Lactobacillus cells in lactulose-supplemented skim milk was studied as follows: The content of ampoules containing freeze-dried bacterial cells (prepared at VTT culture collection) was suspended into 36 mL of skim milk. The suspension was divided into two test tubes, and the other was supplemented with lactulose (final conc. 1%). The tubes were kept at 4°C and bacterial numbers enumerated as above after 0, 10, 17 and 24 d storage.

Fermentation combined to stability studies was performed as follows: Lactobacillus strains were grown in MRS broth overnight. Eighteen milliliters of heat-treated milk in test tubes was inoculated with 180 L of the overnight culture. The milk was supplemented either with 1 mL of 20% lactulose (final conc. 1%) or with 1 mL of water (control). The tubes were incubated at 42°C aerobically for 20 h and transferred into 4°C for 22 d. Bacterial numbers were enumerated as above before and after fermentation (time points 0 and 20 h), and three times during the cold storage (7, 15 and 22 d).

3. RESULTS

3.1. Growth on lactose derivatives

The results of the anaerobic growth experiments performed in test tubes are shown in Fig. 1. At the substrate concentration of 0.5% the differences in the growth of L. rhamnosus strains VTT E-97800 and VTT E-97948 on different carbohydrates were small. At this lowest substrate concentration L. paracasei VTT E-97949 and L. salivarius VTT E-981006 could not utilise lactitol and lactitol/lactobionic acid, respectively, as efficiently as the other carbohydrates. With increased substrate levels (1% and 2%) growth of all four strains was poorer on lactitol and lactobionic acid than on glucose, galactose, lactose and lactulose. At these substrate levels the two L. rhamnosus strains utilised glucose, galactose and lactose about equally (and lactulose slightly less), whereas L. paracasei did not show clear preference among the four carbohydrates, and L. salivarius utilised lactulose the best. The final pHs of cells suspension grown on glucose varied between 3.6 and 4.4, on galactose between 3.7 and 4.5, on lactose between 3.7 and 4.4, on lactulose between 3.8 and 4.4, on lactitol between 4.3 and 5.0, and on lactobionic acid between 4.5 and 6.0 (Fig. 1). Since the pK_a value for lactobionic acid is 3.6, the substrate remained mainly in the dissociated form throughout the growth period.

 

 

Fig. 1. The growth of Lactobacillus strains in test tubes (anaerobic growth) on glucose, galactose, lactose, lactulose, lactitol and lactobionic acid at 0.5% (black columns), 1% (gray columns) and 2% (white columns) substrate concentrations after 2 d.

 

The results of the kinetic growth experiments (aerobic growth) performed in Bioscreen are shown in Fig. 2. At the substrate concentration of 0.5% L. rhamnosus VTT E-97800 reached the same final cell density (OD 1.2) when growing on glucose, galactose, lactose, lactulose, and lactitol, however the growth kinetics of the strain was different when growing on different substrates. When growing on glucose and galactose the stationary phase was obtained within 13 h, on lactose or lactulose within 20 h and on lactitol after 25 h. Lactobionic acid was poorly utilised aerobically. At the substrate concentration of 1% similar differences in growth kinetics were observed. When growing on lactobionic acid the strain reached the same final cell density as with other substrates (OD appr. 1.5), but the growth rate was slower (stationary phase was reached after 36 h, whereas for other carbohydrates the time varied between 16 and 24 h). At the substrate concentration of 2% glucose and galactose were the best (fastest) utilised substrates (stationary phase after 16 h), whereas lactulose, lactose, lactobionic acid and lactitol were well utilised but the growth rate was slower than with glucose and galactose (stationary phase after 22–32 h) (Fig. 2a). The results of growth studies with L. rhamnosus VTT E-97948 strain were similar to those of L. rhamnosus VTT E-97800 (data not shown).

 

 
   

Fig. 2. The growth of Lactobacillus strains on glucose, galactose, lactose, lactulose, lactitol and lactobionic acid at 0.5, 1 and 2% substrate concentrations during kinetic growth experiments (aerobic growth) performed in Bioscreen.

 

At the substrate concentration of 0.5% L. paracasei VTT E-97949 reached the final optical density of 1.0 when growing on lactose and lactulose. When growing on glucose, galactose, lactitol and lactobionic acid the final optical densities varied between 0.6 and 0.7. At substrate concentrations of 1% and 2% lactulose and lactose were still the best utilised substrates, but the difference to other substrates was smaller compared to the experiment with the 0.5% substrate concentration (Fig. 2b). At the substrate concentration of 0.5% L. salivarius VTT E-981006 utilised lactose better than other substrates (final OD 0.7). At 1% and 2% substrate concentrations lactulose was the best utilised substrate (final ODs 1.3 and 1.6, respectively) and also lactose was well utilised compared to glucose (Fig. 2c).

3.2. Effect of lactulose and lactitol on the acid and bile tolerance

In the initial screening all Lactobacillus strain survived well at pH 3.0, whereas only L. salivarius VTT E-981006 was able to survive at pH 2.0 (data not shown). Therefore, an intermediate pH of 2.5 was used in studies where the effect of carbohydrate on acid stability was investigated. Furthermore, since the strains utilised lactulose and lactitol better than lactobionic acid, only the effect of the first two lactose derivatives was studied. In general lactulose or lactitol did not show clear effect on the acid or bile tolerance of the strains, except that L. salivarius E-981006 seemed to tolerate bile better when grown on lactulose than on glucose or lactitol (a two log-value difference in the means of the results of the tests performed three times) (Fig. 3).

 

 

Fig. 3. Acid and bile tolerance of the Lactobacillus strains after growth with glucose, lactulose or lactitol.

 

3.3. Effect of lactulose on the antimicrobial activity

Since lactulose was the favoured substrate of the Lactobacillus strains antimicrobial testing and cold storage stability testing were performed only with this lactose derivative. Lactulose did not have an effect to the antimicrobial activity of the studied Lactobacillus strains. Lactobacillus supernatants (pH appr. 4.0) inhibited the growth of E. coli, S. aureus and S. enterica serovar Typhimurium strains almost completely (98%, 95–96%, and 96%, respectively). Lactic acid and hydrochloric acid (pH appr. 4.0) inhibited the growth as follows: E. coli: 98% inhibition for lactic acid and 96% for hydrochloric acid; S. aureus: 96% for lactic acid and 69% for hydrochloric acid; S. enterica: 96% for both acids. No antimicrobial activity was observed when the pH of Lactobacillus supernatants was adjusted to 6.0. All supernatants at pH 6.0 increased the growth of E. coli, S. aureus and S. enterica up to 34%.

The antimicrobial activity of the Lactobacillus strains against C. krusei and C. albicans was studied only with pH 4.0 supernatants. The growth medium filtrates of Lactobacillus strains did not prevent the growth of the studied yeast strains, in fact they enhanced the growth (up to 55%).

3.4. Effect of lactulose on the stability during cold storage

The results of the stability studies are shown in Table 2. Lactulose did not have an effect on the stability of the Lactobacillus strains when the stability of fresh bacterial cells in skim milk and in skim milk supplemented with 1% lactulose was studied at 4°C. The viability of the test strains remained constant during 22 d storage period in skim milk with or without lactulose supplementation. When the experiment was performed with freeze-dried cells instead of fresh ones, lactulose once again did not have a significant effect on the strain stability. In these test conditions the viability of the L. rhamnosus VTT E-97800 and VTT E-97948 strains remained fairly constant during the 24 d storage period, whereas the culturable numbers of L. paracasei VTT E-97949 increased during the 24 d storage period (probably due to splitting of cell-chains). The viability of the L. salivarius VTT E-981006 strain remained constant during the 24 d storage in lactulose supplemented skim milk, whereas in plain skim milk the viability decreased over one log-value after 17 d.

 

 

Table 2. The stability of fresh, freeze-dried or fermented Lactobacillus cells in skim milk with or without lactulose supplementation at +4°C
 

 

When Lactobacillus strains were grown in plain skim milk or in skim milk supplemented with 1% lactulose (incubation at 42°C for 20 h aerobically) lactulose did not effect the stability of the strains during the 22 d storage at 4°C. The viability of the L. rhamnosus VTT E-97800, L. rhamnosus VTT E-97948 and L. paracasei VTT E-97949 strains remained constant during 22 d storage period in fermented skim milk and in fermented lactulose skim milk. The viability of the L. salivarius VTT E-981006 strain decreased 1.5 log-values during the 22 d storage period both in fermented skim milk and in fermented lactulose skim milk.

4. DISCUSSION

In the present study the ability of lactulose, lactitol and lactobionic acid to improve the technological and functional properties of putative probiotic Lactobacillus strains, representing species L. rhamnosus, L. paracasei and L. salivarius, was studied. The strains for the study were selected on the basis that they utilised lactose derivatives well compared to several other intestinal Lactobacillus isolates (Kontula, 1999). Lactulose, lactitol and lactobionic acid are unabsorbable lactose derivatives with prebiotic potential ( Kontula, 1999). All three sugar derivatives can be utilised in varying extent by different Lactobacillus and Bifidobacterium species/strains (Sahota et al., 1982; Smart, Pillidge, & Garman, 1993; Kneifel et al., 2000).

In the present study lactulose was the favoured lactose derivative when Lactobacillus strains were grown anaerobically at substrate concentrations 1% or 2%. This finding is in an agreement with the results of Kneifel et al. (2000). L. rhamnosus strains VTT E-97800 and VTT E-97948 and L. paracasei strain VTT E-97949 utilised lactobionic acid anaerobically either equally or even better than lactitol at the concentrations of 1% and 2%, whereas L. salivarius strain VTT E-981006 grew poorly on lactobionic acid at all substrate concentrations.

In kinetic growth experiments (growth in aerobic conditions) a similar trend in substrate preference was noted. Glucose and galactose were the fastest utilised carbohydrates by the two L. rhamnosus strains, followed by lactulose and lactose. Growth on lactobionic acid started more slowly, but on the higher substrate concentrations the final OD reached (after 48 h) was as high as with other substrates tested. The two L. rhamnosus strains were originally isolated using different enrichment carbohydrates (lactitol and rye pentosans), but regardless of this fact they behaved almost identically in the kinetic growth experiments. The preferred substrates for the L. paracasei strain were lactose and lactulose, especially at lower substrate concentrations. For the L. salivarius strain the preferred substrates were lactose, glucose and lactulose. Unlike the other tested strains L. salivarius never utilised lactobionic acid well, regardless of the substrate concentration. L. salivarius VTT E-981006 was originally isolated using lactobionic acid enrichment (Kontula et al., 2000), but clearly this substrate is not preferred by the strain. Data on the final cell densities (anaerobic growth) and kinetic growth experiments (aerobic growth) indicates that increasing the substrate concentration from 0.5% to 1% enhanced the growth of the Lactobacillus strains, whereas no difference between substrate concentrations 1% and 2% was observed.

In tests where the effect of lactulose and lactitol on the bile and acid tolerance of the Lactobacillus strains were studied the general trend was that these lactose derivatives did not improve the properties of the strains. However, there was one exception, which warrants further studies: L. salivarius VTT E-981006 cells grown on lactulose seemed to tolerate 1.5% bile better (difference of 2-log values) than the cells grown on glucose or lactitol. Since the test was performed only three times no statistical analysis was performed and the result still remains presumptive.

Lactose derivative lactulose did not have an effect on the antimicrobial activity of the Lactobacillus strains. Lactobacillus growth medium filtrates showed antimicrobial activity against E. coli, S. aureus, S. enterica serovar Typhimurium at pH 4.0 but not at pH 6.0 (most likely due to lactic acid effect). Growth medium filtrates (at pH 4.0) slightly enhanced the growth of Candida yeasts. The phenomenon that Lactobacillus growth medium filtrates lose their antimicrobial activities at higher pH values is well-known (see e.g. Silva, Jacobus, Deneke, & Gorbach, 1987). Also the growth-enhancement of yeasts by Lactobacillus has been noted before (Daeschel, Andersson, & Fleming, 1987).

Lactulose was also tested for its ability to improve the stability of the strain during cold storage (appr. 3 weeks) at 4°C. The storage tests were performed in three ways to mimic the situation of how probiotics can be formulated into foods (in this study skim milk). They can be added as fresh cells, as freeze-dried cells or they can be grown in the product. Since lactulose was well utilised by all test strains, only 1% lactulose supplementation in skim milk was tested. Heat-treatment as such results in the formation of lactulose in milk, but in low levels (appr. 0.1%). In general additional lactulose did not seem to have an effect on the cold-storage stability of the Lactobacillus strains. The exception was once again L. salivarius VTT E-981006. When freeze-dried cells were used L. salivarius seemed to survive better in lactulose supplemented skim milk than in plain skim-milk after 17 d. This test was performed twice and therefore also this finding warrants further confirmation.

In conclusion, lactose derivatives did not largely effect the technological or functional properties of L. rhamnosus, L. paracasei, and L. salivarius. However, we got preliminary indication that lactulose might improve the bile tolerance and cold-storage stability of L. salivarius. In our hands L. salivarius strains have proved less robust in many test conditions compared to especially L. rhamnosus. Our results may therefore suggest that specific prebiotics may benefit less robust probiotic strains, whereas for robust strains further enhancement of their performance with a prebiotic may be difficult. Shin, Lee, Petska, and Ustunol (2000) were able to show that adding large quantities (5%) of galactooligosaccharide (GOS) and especially fructooligosaccharide (FOS) to growth medium (skim milk) enhanced the survival of two Bifidobacterium strains during 4 weeks cold-storage. Since we studied both different prebiotic preparations and strains from a different genus (Lactobacillus, the strains of which are generally more robust technologically than Bifidobacterium strains) our results are not comparable to the results of Shin et al. (2000). However, the possibility that by using larger quantities of lactose derivatives (1% in the present study) a more pronounced effect on the properties the tested Lactobacillus strains could have been detected, cannot be ruled out.

Present results furthermore indicate that finding synbiotic (combination of prebiotic and probiotic) pairs where the prebiotic would benefit the specific probiotic strain, e.g. during production and formulation into foods, is not a simple task. However, although no in vitro benefit is seen, in vivo the synbiotic pair may act differently; e.g. although lactulose did not benefit L. rhamnosus VTT E-97800 strain in the present in vitro study, we have found that in vivo lactulose prolonged the persistence of this strain in the human GI-tract after discontinuation of the probiotic feeding (Kontula et al., 2002). We have also performed another synbiotic feeding trial with a galactooligosaccharide preparation and Bifidobacterium lactis strain Bb-12 (Alander et al., 2001). However, in this study the prebiotic did not benefit the probiotic, which gives further support to our finding that developing synbiotic pairs where the prebiotic enhances the performance of the probiotic is not easily accomplished. Although the concept of prebiotic benefiting the probiotic strain is not included into the definition of synbiotic, this kind of capacity would naturally give added value for the synbiotic product, and is therefore worth pursuing for.

ACKNOWLEDGEMENTS

Authors thank Päivi Lepistö and Niina Torttila for their skilful technical assistance. The work was partly funded by the EU project "Enzymatic lactose valorisation" (FAIR CT96-1048).

REFERENCES

Alander et al. (2001). M. Alander, J. Mättö, W. Kneifel, M. Johansson, B. Kögler, R. Crittenden, T. Mattila-Sandholm and M. Saarela , Effect of galacto-oligosaccharide supplementation on human faecal microflora and on survival and persistence of Bifidobacterium lactis Bb-12 in the gastrointestinal tract. International Dairy Journal 11 (2001), pp. 817-825.

Ballongue and Schumann (1997). J. Ballongue, C. Schumann and P. Quignon , Effects of lactulose and lactitol on colonic microflora and enzymatic activity. Scandinavian Journal of Gastroenterology 32 Suppl 222 (1997), pp. 41-44.

Daeschel, Andersson, & Fleming (1987). M.A. Daeschel, R.E. Andersson and H.P. Fleming , Microbial ecology of fermenting plant materials. FEMS Microbiology Reviews 46 (1987), pp. 357-367.

Gerling (1998). Gerling, K.-G. (1998). Large scale production of lactobionic acid-use and new applications. International Dairy Federation, 9804, 251-256 (Special issue).

Harju (1993). M. Harju , Production and properties of lactulose, lactitol and lactobionic acid. Bulletin of the International Dairy Federation 289 (1993), pp. 27-30.

Hoffman, Mossel, Korus and van De Kame (1964). K. Hoffman, D.A.A. Mossel, W. Korus and J.H. van De Kamer , Untersuchungen über die Wirkungsweise der Lactulose (-Galactosido-Fructose) im Darm. Klinische Wochenschrift 42 (1964), pp. 126-130.

Jin, Ho, Abdullah, and Jalaludin (1998). L.Z. Jin, Y.W. Ho, N. Abdullah and S. Jalaludin , Acid and bile tolerance of Lactobacillus isolated from chicken intestine. Letters in Applied Microbiology 27 (1998), pp. 183-185.

Kitler et al. (1992). M.E. Kitler, M. Luginbuhl, O. Lang, F. Wuhl, A. Wyss and G. Lebek , Lactitol and lactulose: An in vivo and in vitro comparison of their effects on the human intestinal flora. Drug Investigation 4 (1992), pp. 73-82.

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

Kontula (1999). P. Kontula , In vitro and in vivo characterization of potential probiotic lactic acid bacteria and prebiotic carbohydrates. Finnish Journal of Dairy Science 54 (1999), pp. 1-84.

Kontula et al. (2002). P. Kontula, L. Nollet, M. Saarela, T. Vilpponen-Salmela, W. Verstraete, T. Mattila-Sandholm and A. von Wright , The effect of lactulose on the survival of Lactobacillus rhamnosus in the simulator of the human intestinal microbial ecosystem (SHIME) and in vivo. Microbial Ecology in Health and Disease 14 (2002), pp. 90-96.

Kontula, Suortti, Tenkanen, M., Mattila-Sandholm and von Wrigh (2000). P. Kontula, M.-L. Suihko, T. Suortti, M. Tenkanen, T. Mattila-Sandholm and A. von Wright , The isolation of lactic acid bacteria from human colonic biopsies after enrichment with lactose derivatives and rye arabinoxylo-oligosaccharides. Journal of Food Microbiology 17 (2000), pp. 13-22.

Kontula, Suihko, von Wright & Mattila-Sandholm (1999). P. Kontula, M.-L. Suihko, A. von Wright and T. Mattila-Sandholm , The effect of lactose derivatives on intestinal lactic acid bacteria. Journal of Dairy Science 82 (1999), pp. 249-256.

Kummel & Brokx (2001). K.F. Kummel and S. Brokx , Lactitol as a functional prebiotic. Cereal Foods World 46 (2001), pp. 424-429.

Mangin et al. (2002). I. Mangin, Y. Bouhnik, A. Suau, V. Rochet, L. Raskine, P. Crenn, F. Dyard, J.-C. Rambaud and J. Dore , Molecular analysis of intestinal microbiota composition to evaluate the effect of PEG and lactulose laxatives in humans. Microbial Ecology in Health and Disease 14 (2002), pp. 54-62.

Mizota (1996). Mizota, T. (1996). Lactulose as a growth promoting factor for Bifidobacterium and its physiological aspects. International Dairy Federation, Bulletin no. 313, 43-48.

Petticrew, Watt, & Sheldon (1997). Petticrew, M., Watt, I., & Sheldon, T. (1997). Systemic review of the effectiveness of laxatives in the elderly. Health Technology Assessesment, 1 (13):i-iv, 1-52.

Riggio et al. (1990a). O. Riggio, G. Balducci, F. Ariosto, M. Merli, S. Tremiterra, V. Ziparo and L. Capocaccia , Lactitol in the treatment of chronic hepatic encephalopathy-a randomised cross-over comparison with lactulose. Hepatogastroenterology 37 (1990), pp. 524-527.

Riggio et al. (1990b). O. Riggio, M. Varriale, G.P. Testore, R. Di Rosa, E. Di Rosa, M. Merli, A. Romiti, C. Candiani and L. Capocaccia , Effect of lactitol and lactulose admistration on the fecal flora in cirrhotic patients. Journal of Clinical Gastroeneterology 12 (1990), pp. 433-436.

Sahota, Bramley, & Menzies (1982). S.S. Sahota, P.M. Bramley and I.S. Menzies , The fermentation of lactulose by colonic bacteria. Journal General Microbiology 128 (1982), pp. 319-325.

Sako, Matsumoto, & Tanaka (1999). T. Sako, K. Matsumoto and R. Tanaka , Recent progress on research and applications of non-digestible galacto-oligosaccharides. International Dairy Journal 9 (1999), pp. 69-80.

Shin, Lee, Petska, and Ustunol (2000). H.-S. Shin, J.-H. Lee, J.J. Petska and Z. Ustunol , Growth and viability of commercial Bifidobacterium spp. in skim milk containing oligosaccharides and inulin. Journal of Food Science 65 (2000), pp. 884-887.

Silva, Jacobus, Deneke, & Gorbach (1987). M. Silva, N.V. Jacobus, C. Deneke and S.L. Gorbach , Antimicrobial substance from a human Lactobacillus strain. Antimicrobial Agents and Chemotherapy 31 (1987), pp. 1231-1233.

Smart, Pillidge, & Garman (1993). J.B. Smart, C.J. Pillidge and J.H. Garman , Growth of lactic acid bacteria and bifidobacteria on lactose and lactose-related mono-, di- and trisaccharides and correlation with distribution of -galactosidase and phospho--galactosidase. Journal of Dairy Research 60 (1993), pp. 557-568.

Strohmaier (1998). Strohmaier, W. (1998). Lactulose: Status of health-related applications. International Dairy Federation, Bulletin no. 9804, 262-271.

Terada, Hara, Kataoka, & Mitsuoka (1992). A. Terada, H. Hara, M. Kataoka and T. Mitsuoka , Effect of lactulose on the composition and metabolic activity of the human faecal flora. Microbial Ecology in Health and Disease 5 (1992), pp. 43-50.

Van Loo et al (1999). J. Van Loo, J. Cummings, N. Delzenne, H. Englyst, A. Franck, M. Hopkins, N. Kok, G. Mcfarlane, D. Newton, M. Quigley, M. Roberfroid, T. van Vlient and E. van den Heuvel , Functional food properteis of non-digestable oligosaccharides: A consensus report from the endo project (DGXII AIRII-CT-94-1095). British Journal of Nutrition 81 (1999), pp. 121-132.

Vince, Zeegen, Drinkwater, O'Grady and Dawso (1974). A. Vince, R. Zeegen, J.E. Drinkwater, F. O'Grady and A.M. Dawson , The effect of lactulose on the faecal flora of patients with hepatic encephalopathy. Journal of Medical Microbiology 7 (1974), pp. 163-168.

(order Full Text from publisher)