The inuence of a variety of prebiotic sugars (galacto-, manno-, fructooligosaccharides, lactulose and others) and basic carbohydrates on the growth of a selection of probiotic bidobacteria (9 strains), Lactobacillus acidophilus (8 strains) and other lactobacilli (9 strains) was investigated by applying in vitro methodologies based on optical density measurement. It has been shown that some of the bacteria could markedly utilize the substrates but with pronounced variation, depending on the individual nature of the strains. Besides their capability to grow in galacto- and fructooligosacharide containing media, a distinct growth in lactulose-based substrates was evident for most of the strains tested. Results presented can be used for selecting probiotic strains and prebiotic sugars to form synbiotic formulations.

Key words: probiotic bacteria, prebiotic carbohydrates, growth.

INTRODUCTION

Bidogenic carbohydrates and sugar preparations have become an important item within the area of functional food ingredients. These compounds are primarily designed to exert benecial effects to humans by positively inuenc­ing the composition of the intestinal microora (1, 2). Such products have recently been called prebiotics and turned out to be an interesting alternative to probiotics that basically are administered as live bacteria. There is scien­tically proven evidence that some prebiotic compounds are capable of promoting the growth of bidobacteria and also of lactobacilli in the colon, since they can pass the upper intestinal region without being hydrolysed (1, 3, 4).

Today, prebiotic sugars are manufactured at large scale from different natural sources and by different technolo­gies. Most of the substances available at the market are inulin-based and made from various plant material such as chicory, topinambur, artichoke etc. They consist of glucose molecules linked with b(21) fructose oligomers of vary­ing chain length. Another category are the whey-based products which contain galacto-oligosaccharides resem­bling sugar components usually occuring naturally in breast milk. Moreover, a variety of other carbohydrate preparations produced by diverse biochemical or enzymat­ical techniques have been described (46). According to these authors, the array of oligosaccharides can be grouped into twelve categories, ranging from the galacto­oligosaccharide type, over lactulose, fructo-oligosaccha­rides, malto-oligosaccharides up to the xylo-oligo­saccharides. Frequently, prebiotics have also been de­scribed assolubleber ' materials whichin addition to their biological signicancepossess several other advan­tages such as reduced caloric value, reduced cariogenity and textural effects exerted on the food matrix applied in. During the last years, many efforts have been made to develop multiple functional foods or preparations contain­ing pro- and prebiotics. By combining these two compo­nents in so-called synbiotic products, a twofold positive effect on the gut microora can be expected (1, 7).

A variety of probiotic bacterial strains has been incorpo­rated in different kinds of synbiotics, in combination with prebiotic sugars. Because of these reasons, it was of inter­est to consider the inuence of prebiotics on the growth of bacteria with intestinal importance. Following this target, this study deals with the in vitro investigation of the inuence of various prebiotic sugars (commercially avail­able sugars and research samples of oligosaccharidic and polysaccharidic type) andfor comparison purposes-basic carbohydrates on the growth behaviour of a selection of probiotic bidobacterial and lactobacilli strains.

MATERIALS AND METHODS

Bacteria, prebiotic carbohydrates and sugar derivatives

Bacterial test strains were obtained from different Eu­ropean culture suppliers and strain collections (Table I). Although most of these test strains have been applied in probiotic products, their specic documentation on such properties was not available for all of them. Cultures were maintained in appropriate basal media. For culturing the lactobacilli, MRS broth (8) was prepared with the following ingredients (g per l): peptone from casein (10.0), meat extract (8.0), yeast extraxt (4.0), glucose (20.0), Tween 80 (1.0), K₂HPO₄ (2.0), (NH₄)₂ citrate (2.0), sodium acetate (5.0), MgSO₄ (0.2), MnSO₄ (0.02), pH6.5. Bidobacte­ria were cultured in TPY medium (9) which was composed as follows (concentrations in g per l): tryptone (10.0), peptone (5.0), yeast extract (2.5), glucose (5.0), Tween 80 (1.0), K₂HPO₄ (2.0), MgCl₂ (0.5), ZnSO₄. 7H₂O (0.25), CaCl2 (0.15), FeCl3. 6H2O (0.003), cysteine HCl (0.5), pH6.5. Each of the components used for media prepa­ration originated from one single lot, in order to facilitate comparable growth conditions throughout the experiments.

All microorganisms were cultured under anaerobic con­ditions (80% nitrogen, 10% hydrogen, 10% carbon dioxide) provided by an anaerobic incubator (WTC Binder, Tutt­lingen, Germany). Before inoculation of the test media containing the different carbohydrates and sugar prepara­tions (Table II), bacteria were activated for 24 h at 37°C in their basal media containing glucose. These suspensions were used as bacterial stock cultures for growth intensity measurements.

Table I
Test strains used in this study

Measurement of bacterial growth intensity

With each of the carbohydrates and prebiotic substrates a series of six bacteriological tubes containing 10 ml MRS (lactobacilli) or TPY (bidobacteria) broth supplemented with 1% (wv) carbohydrate substrate was prepared for each strain, with the exception of sodium alginate which was applied only at 0.1% (wv) because of its natural turbidity interfering with the optical readings. Since no sufcient information was available regarding the specic carbon contents of some of the test samples, sugars were added based on simplied weight-per-volume basis. The tubes were inoculated with 100 ml of a 24 h old bacterial stock culture giving an inoculation strength of approxi­mately 10⁴ to 10⁵ per total test tube volume. In preceding trials it was found that most test strains developed their maximum growth activities during theirrst 24 h of incu­bation, followed by some stationary phase. Because of this experience this incubation time was chosen throughout the experiments. After incubation, the growth intensities of the test strains were examined based on optical density (OD) measurements using a Hitachi U 2000 photometer set at 600 nm and equipped with 1 cm cuvettes. For this pur­pose, the incubated tubes were agitated for 10 s on a Vortex shaker to re-suspend sedimented cells, followed by transferring 1 ml of this suspension into the optical cu­vette. Arithmetical mean values were calculated from all single OD readings and compared with the OD values obtained with the cultures grown in a basal medium containing glucose as a reference (100%).

In order to determine the growth activity by considering the growth kinetics of the bacteria and also to verify the results from the above described test series, a selection of

bacterial strains was monitored for their growth behaviour in media with various carbohydrate sources (same compo­sition as described above) using the automatic BioScreen^R C system (Labsystems, Helsinki, Finland) connected with a

PC equipped with a BioLink^R software package (Labsys­tems). To maintain anaerobic conditions in the microtiter cavities, 1 ml of Oxyrase^R (Oxyrase Inc., Manseld OH, USA) enzyme solution was added to 50 ml of substrate in order to remove dissolved oxygen, before dispensing it automaticly into the microtubes. In addition, all microvials were covered by adding two drops of sterile parafne oil, after bacterial inoculation. The temperature of the microt­iter plates was maintained at 37°C. The growth induction period (duration of time until reaching the exponential multiplication phase of the bacteria) as well as the slope of the growth curves at their period of maximum growth velocity were determined graphically.

Table II
Carbohydrates and prebiotic preparations used in this study

RESULTS

Growth behaviour of bidobacteria in media containing prebiotics

Nine bidobacterial strains (numbers 19) were compared for their growth activities in TPY medium supplemented with different basic and prebiotic sugars. The single values of the optical density measurements exhibited a low varia­tion yielding a coefcient of variation of B10%. Because of this observation and also in order to facilitate the interpretation of the graphics only the arithmetical mean values of the relative growth intensity patterns are pre­sented. As shown in Fig. 1, fructose was utilized markedly only by four strains (nos. 2, 7, 8, 9), whereas the other disaccharidic sugars led to pronounced growth (partly more than 100%) of eight bidobacterial strains. Galac­tooligosaccharide preparations and the mannooligosaccha­ride sample gave heterogeneous patterns, ranging from weak sugar utilisation by strains 1, 3, 4, 5 and 6, up to higher growth responses (e.g., nos. 7 to 9) which were quite in accordance with those obtained with glucose. With the exception of one strain and the guar gum-based pow­der, the fructooligosaccharides varied from giving moder­ate to marked growth activity, obviously also depending on the brand of sugar product applied. Furthermore, it was evident that lactulose enabled distinct growth of some of the bidobacteria (e.g., nos. 2, 4, 7, 8, 9), while the inulin-containing media were more or less acceptable sub­strates for two strains (nos. 2 and 9). The growth intensi­ties obtained with the other carbohydrate compounds were less pronounced, however, we observed response values of even \50% to lactobionic acid with a few bidobacterial strains.

Growth behaviour of lactobacilli in media containing prebiotics

In Fig. 2 the growth patterns of all Lactobacillus acidophilus strains (nos. 1017) are shown. While glucose and fructose were utilised quite well by these bacteria, with some strains decites were observed regarding sucrose (no. 15) and lactose (no. 12). While the Elixor syrup representing an industrially manufactured galactooligosaccharide sugar yielded pronounced utilisation patterns with all of these strains, the galactooligosaccharide powder from a Japanese manufacturer gave lower growth values with two L. acidophilus strains (nos. 11 and 15). The individual growth responses to the mannooligosaccharide varied con­siderably among the bacteria, exhibiting the highest re­sponse with strain no. 13 (mean value of approximately 87%). The results of guar-gum based product were not comparable to those obtained with the fructooligosaccha­ride preparations. In particular, strain no. 11 grew pronouncedly (140%) on a substrate containing Actilight. Under similar conditions, high growth responses (116%, 121%) were also shown with strains no. 13 and 14, respec­tively. The Raftilose powder yielded enhanced bacterial growth intensities with two, and the FOS research sample with three strains. As already experienced with some bidobacteria, again lactulose was well utilised by some L. acidophilus strains. The inulin-containing substrates gave a pronounced response only with two strains. Neither lacti­tol, nor sodium alginate and lactobionic acid showed marked patterns with these test strains.

Fig. 1. In vitro growth response patterns of strains of Bidobac­terium (for explanation of strain codes see Table I) in media containing carbohydrates of prebiotic importance (basic sugars served as reference substrates, growth intensity in glucose medium100%; for abbreviations see Table II; Lact. Acid . . . Lactobionic Acid).

Fig. 2. In vitro growth response patterns of strains of Lactobacil­lus acidophilus (for explanation of strain codes see Table I) in media containing carbohydrates of prebiotic importance (basic sugars served as reference substrates, growth intensity in glucose medium100%; for abbreviations see Table II; Lact. Acid . . . Lactobionic Acid).

Fig. 3. In vitro growth response patterns of other lactobacilli strains (L. rhamnosus, L. casei, L. paracasei, L. reuteri, L. gasseri; for explanation of strain codes see Table I) in media containing carbohydrates of prebiotic importance (basic sugars served as reference substrates, growth intensity in glucose medium100%; for abbreviations see Table II; Lact. Acid . . . Lactobionic Acid).

Fig. 4. Growth kinetics of probiotic strain no. 4 (for explanation of strain code see Table I) in TPY medium with selected prebiotic oligosaccharides and glucose (reference) as monitored by the BioScreen^R C analyzer. The induction time (period until onset of the exponential growth phase, example shown for sugar E      t_E) was calculated as a measure for the duration of the lag phase; the slope of the exponential phase of the growth curve (lines indicated along the graphs) was calculated as a measure for growth intensity.

In Fig. 3 the growth patterns of lactobacilli (nos. 18 to 26) representing members of L. rhamnosus, casei, paraca­sei, reuteri and gasseri are shown. The mono- and disac­charide-containing media resulted in heterogeneous growth patterns of the lactobacilli. L. reuteri strains (nos. 24, 25) and also L. gasseri (no. 26) only produced a weak turbidity in the fructose broth. Also with sucrose, some of the test strains (nos. 9, 20, 22, 23) did not grow well. With the exception of strain no. 22, the galactooligosaccharide preparations were utilised, and the mannooligosaccharide gave high responses with one L. reuteri strain (no. 25). In general, the growth on fructooligosaccharides was less pronounced, except with some marked utilisation patterns with Actilight. Strain no. 21 exhibited a pronounced tur­bidity with all the oligofructose samples tested and with inulin. Interestingly, lactulose enabled good growth prop­erties to several strains. Lactitol resulted in relatively high turbidities with strains no. 18, 19, 20 and 23. Poor growth rates were observed with the alginateand d lactobionic acid-based substrates.

Growth kinetics of probiotics

In Fig. 4 the growth curves obtained with one Bidobac­terium strain (no. 4) in TPY media supplemented with a selection of prebiotic sugars are shown exemplarily. The development of turbidity was followed continuously in the BioScreen C system by optical density measurements. The growth properties of the strain were in fair agreement with the above described results based on static assessment after 24 h. The growth curves were evaluated for their induction periods (time until onset of exponential multiplication) and the slope calculated for the exponential phase. Table III shows the data calculated exemplarily for one Bidobac­terium and one Lactobacillus strain. From these data it is evident that the induction period indicating the lag phase of the bidobacterial strain in the media containing the two oligofructose products was comparable to that ob­tained with glucose. Even the exponential growth intensi­ties were in fair accordance. When inulin was used as a carbohydrate source distinctly longer lag phases were ob­served than with the other sugars. The bidobacterial growth curve of galactooligosaccharide product almost gave the same shape as depicted for oligofructose, albeit with a lower slope during exponential bacterial multiplication.

Synbiotic combinations

Taking into consideration the above described results, an attempt was made to summarize the results in order to illustrate the variety of responses and to enable a simplied selection of possible synbiotic mixtures consisting of car­bohydrates and bacterial strains which can be stimulated or utilized markedly by these sugars. Only the marked growth intensity results were used as selection criteria (Table IV).

DISCUSSION

Several authors have studied the positive effects induced by the

administration of oligosaccharides at different dose levels. In most studies a signicant increase of the colonic bidobacterial microora caused by long-term application of prebiotic sugars has been demonstrated in vivo and in vitro (3, 1012). In addition to the shift in proportions of the intestinal micropopulation, various secondary effects initiated by metabolic activities of the microora have been observed (7, 1315). Although certain bacterial groups with benecial attributes can be stimulated by prebiotic carbohydrates, differing results have been re­ported about the possible capability of other microbial representatives of the gastrointestinal tract to utilize them as well (16, 17). Furthermore, dose-response side effects and disadvantages to the human organism have been shown with prebiotic sugars (2, 18, 19). These experiences have on one hand contributed to a better understanding of the complex microbial ecosystem of the gut, but may, on the other hand, also restrict the basic meaning of a pre­biotic, according to its denition (1).

In the light of this situation, this study was undertaken to examine whether probiotic bacteria are able to utilize oligosaccharides which are claimed to be of prebiotic importance. In accordance with many reports, this in vitro screening has demonstrated in general that bidobacteria are capable of utilizing galacto-, manno, and fruc­tooligosacharides, however with differing intensities and levels. These differences may be due to the individual nature of the strains. It should also be mentioned that strains of the same species and originating from one producer yielded different utilization patterns. The bidobacterial utilization patterns of other carbohydrate compounds (as shown in the lower quarter of the corre­spondinggure) such as inulin, lactulose, lactitol etc. are very heterogeneous, and lactulose caused a distinct growth intensity with a few particular strains of Bidobacterium. While alginate preparations have been shown to increase the bidobacterial populations (20) in the human intestine, the present in vitro screening gave less promising results. However, it has to be admitted that due to its natural turbidity this compound was not applied at a concentra­tion comparable with that of the other carbohydrates. Therefore, only moderate growth of some of the bidobac­teria was observed.

The Lactobacillus acidophilus strains also grew well in substrates containing the most commonly used prebiotic sugar preparations, although differences in the growth response values were more pronounced regarding the oligofructose samples applied. Among the other sugars and derivatives, lactulose gave the most distinct growth levels. Similar observations with bidobacteria and lacto­bacilli have also been reported earlier (21, 22). The results with the other lactobacilli have indicated some preferences rather for the galacto- than for the fructooligosaccharides, albeit with one positive exception, strain no. 21. Interest­ingly, members of Lactobacillus rhamnosus, casei, paracasei and reuteri were also utilizing lactulose pronouncedly.

Synbiotics follow the strategy to exploit a twofold po­tential by combining pro- with prebiotics. However, this system is only sensible if both componentst together, which can be considered by examining the growth of the particular bacterial strain exposed to a substrate with the prebiotic sugar to be applied. This study has shown that it would be useful to take into account the specic utilization patterns of a probiotic bacterial strain before incorporat­ing it into a product containing a prebiotic sugar. Based on these results, a better selection of suitable sugars and bacteria for formulating synbiotics has been made feasible. The ability of a probiotic strain to compete successfully with other microorganisms present in the colonic ecosys­tem depends on several factors. Although its capabiltiy to utilize in vitro certain carbohydrates gives some indication whether the probiotic strain may succeed or not under given conditions, a final conclusion can only be given based on in vitro experiments.

Table III

Growth kinetic data calculated for two bacterial strains no. 4 and 14 in media with selected prebiotics monitored in the BioScreen C analyzer (for explanation of strain codes see Table I)

ACKNOWLEDGEMENTS

This study was supported by the EU under FAIR project CT96-1048 "Enzymatic lactose valorization". The provision of cultures

and sugar preparations by the above mentioned suppliers and companies is gratefully acknowledged.

REFERENCES

1. Gibson GR, Roberfroid MB. Dietary modulation of the human colonic microbiota introducing the concept of prebi­otics. J Nutr 1995; 125: 1401 12.
2. Tomomatsu H. Health effects of oligosaccharides. Food Technology 1994; 48: 61 5.
3. Gibson GR, Collins, MD. Concept of balanced colonic mi­crobiota, prebiotics, and synbiotics. In: Hanson, LA, Yolken, RH, eds. Probiotics, other Nutritional Factors, and Intestinal Micro ora. Nestle Nutrition Services, Philadelphia, Lippin­cott-Raven, 1999: 139 156.
4. Playne, MJ, Crittenden, RG. Commercially available oligosaccharides. In: Bulletin of the IDF no. 313, Brussels, 1996: 10 22.
5. Matsumoto, K, Kobayashi, Y, Ueyama, S, Watanabe, T, Tanaka, R, Kan, T., Kuroda, A, Sumihara, Y. Galac­tooligosaccharides In: Nakakuki, T. ed. Oligosaccharides. Production, properties and applications. Japanese Technology Reviews, Vol. 3, no. 2, Gordon and Breach Science Publish­ers, Switzerland, Australia, 1993: 90 106.
6. Crittenden, RG. Prebiotics. In: Tannock, GW, ed. Probi­otics a Critical Review. Norfolk, UK, Horizon Scienti c Press, 1999: 141 156.
7. Ziemer CJ, Gibson GR. An overview of probiotics, prebiotics and synbiotics in the functional food concept: perspectives and future strategies. Int Dairy J 1998; 8: 473 9.
8. DeMan JD, Rogosa M, Sharpe ME. A medium for the cultivation of lactobacilli. J Appl Bacteriol 1960; 23: 130 5.
9. Scardovi, V. Genus Bi dobacterium. In: Bergey 's s Manual of Systematic Bacteriology Vol 2, Sneath, P, Mair, N, Sharpe, E, Holt, JG, eds. New York, Williams and Wilkins, 1986: 1418 1434.
10. Gibson GR, Beatty ER, Wang X, Cummings JH. Selective stimulation of bi dobacteria in the human colon by oligofruc­tose and inulin. Gasteroenterology 1995; 108: 975 82.
11. Yamazaki H, Dilawri N. Measurement of growth of bi dobacteria on inulofructosaccharides. Lett Appl Microbiol 1990; 10: 229 32.
12. Tanaka, R, Matsumoto, K. Recent progress on prebiotics in Japan, including galacto-oligosaccharides. In: Bulletin of the IDF no. 336, Brussels, 1998: 21 27.
13. Sghir A, Chow JM, Mackie RI. Continuous culture selection of bi dobacteria and lactobacilli from human faecal samples using fructooligosacchride as selective substrate. J Appl Mi­crobiol 1998; 85: 769 77.
14. Bouhnik, Y, Flourie , B, D Agay-Abensour, Pochart, P, Gramet, G, Durand, M, Rambaud, J-C. Administration of transgalacto-oligosaccharides increases fecal bi dobacteria and modi es colonic fermentation metabolism in healthy humans. J Nutr 1997; 127: 444 448.
15. Hidaka H, Eida T, Takizawa T, Tokunaga T, Tashiro Y. Effects of fructooligosacchrides on intestinal flora and human health. Bi dobacteria Micro ora 1986; 5: 37 50.
16. Hartemink R, Quataert MCJ, van Laare KMJ, Nout MJR, Rombouts FM. Degradation and fermentation of fructo-oligosaccharides by oral streptococci. J Appl Bacteriol 1995; 79: 551 7.
17. Hartemink R, van Laere KMJ, Rombouts FM. Growth of enterobacteria on fructo-oligosacharides. J Appl Microbiol 1997; 83: 367 74.
18. Alles MS, Hautvast JGA, Nagengast FM, Hartemink R, van Laere KMJ, Jansen JBM. Fate of fructo-oligosaccharides in the human intestine. Brit J Nutr 1996; 76: 211 21.
19. Ito M, Deguchi Y, Miyamori A, Matsumoto K, Kikuchi H, Kobayashi Y, Yajima T, Kann T. Effects of administration of galactooligosaccharides on the human faecal micro ora, stool weight and abdominal sensation. Microb Ecol Health Dis 1990; 3: 285 92.
20. Terada A, Hara H, Mitsuoka T. Effect of dietary alginate on the faecal microbiota and faecal metabolic activity in humans. Microb Ecol Health Dis 1995; 8: 259 66.
21. Modler WH. Bi dogenic factors sources, s, metabolism and applications. Int Dairy J 1994; 4: 383 407.
22. Salminen, S, Salminen, E. Lactulose, lactic acid bacteria, intestinal microecology and mucosal protection. Scand J Gas­troenterol 1997; 32 Suppl 222: 45 48.

(order Full Text from publisher)