Interaction of bifidobacteria with the gut and their influence in the immune function

Gabriela Perdigón^1,2, Mónica Locascio¹, Marta Medici¹, Aida Pesce de Ruiz Holgado^1,2 and Guillermo Oliver¹

1. CERELA. Chacabuco 145, 4000 Tucumán, Argentina.
2. Instituto de Microbiología, Facultad de Bioquímica, Química y Farmacia. Universidad Nacional de Tucumán, Argentina.

Address correspondence to: Dra. Gabriela Perdigón. CERELA, Chacabuco 145, (4000) Tucumán, ARGENTINA. Te/Fax: (+54-381) 431 0465. E-mail: perdigon@cerela.org.ar

Key words: Bifidobacteria, Homologous and Heterologous strains, Immune System.

ABSTRACT: Bifidobacteria are predominant in the lumen of the large intestine and confer various health benefits on the host. They are also used in the preparation of new fermented milks (bioyogurts) or added to conventional yogurt to generate probiotic effects. The colonization of the gut by bacteria tends to be host specific due partly to the way in which bacteria adhere to the intestinal wall. Using a homologous strain of Bifidobacterium animalis in an experimental mouse model, we analyzed by immunofluorescence labelled-bacteria and transmission electronic microscopy the importance of this bacterial interaction with epithelial an immune cells associated to the gut, and the effect of feeding of B. animalis in the immune response. It was able to adhere and interact with both small and large intestine. In spite of this interaction with the gut, no modifications in the immune state (secretory or systemic response) were observed.
A heterologous strain of Bifidobacterium adolescentis from human faeces, was neither uncapable of binding to the intestine, nor influence the immune system activation, when it was administered during 2, 5 or 7 consecutive days; we believe that using a homologous strain, oral tolerance is developed even when the microorganism interacts with the immune cells associated with the intestine. However, we cannot ignore the beneficial effect of these microorganisms, especially in the prevention of intestinal infections. We think that this property exerted by bifidobacteria is more related to other mechanisms such as competitive inhibition, acid production or others, than enhancement of the immune state.

Introduction

Bifidobacteria are important species of the intestinal tract and they are associated with a healthy status in humans (Bartram et al., 1994). These microorganisms have been found to be dominant in the large intestine, especially in the proximal colon, whereas lactobacilli are also found in the distal end of the small intestine
(Kimura et al., 1997; Mc Cartney et al., 1996; Tannock, 1999).
Milks fermented by bifidobacteria have been developed in many areas of the world, and research suggests several potential probiotic advantages compared to products containing lactic acid bacteria alone (Zavaglia et al., 1998; Yildirim and Johnson, 1998; Holzapfel et al., 2001; Gomes and Xavier Malcata, 1999). The potential probiotic advantages of bifidobacteria are related in particular to the antimicrobial effects (Gibson and Wang, 1994; Misra and Kuila, 1995) immunomodulation (Link-Amster et al., 1994; Schiffrin et al., 1995) alleviation of constipation (Seki et al., 1978) reduction of risk of cancer (Abdelali et al., 1995; Grill et al., 1995a; Moore and Moore, 1995) and modulation of the intestinal microflora. These finding suggest that bifidobacteria may help maintain healthy functioning of the intestinal tract, because they act directly, through their antimicrobial activity, or indirectly by immunomodulation via intestinal cells or modification of the function of the normal microflora in order to exert a probiotic effect. Bifidobacteria must reach the gut in large numbers. It has been recommended that a minimun number of 10⁶-10⁷ viable microorganisms/g in the product should be ingested (Bouhnik, 1993). Other studies have shown that cholesterol can be removed by bifidobacteria from growth medium through both, assimilation and precipitation with conjugated bile salts (Grill et al., 1995b; Tahri et al., 1997). In spite of the numerous properties attributed to these microorganisms the functions of endogenous bifidobacteria in the colon have not been completely elucidated. Until now the selection of strains of bifidobacteria to be included in fermented milks has been based on the survival, in the product and through the intestinal tract after consumption. However, this point is a technological challenge because bifidobacteria tend to lose viability in fermented milks, and they do not survive well in a pH less than 4.5 or in aerobic conditions. Thus scientific interest is turning now toward their functions as essential criteria of selection, in order to ensure their beneficial effects on health of the consumer. It is necessary to establish a basic knowledge concerning the stability of bifidobacteria probiotic strains and their effects on eucaryotic-cells or on the microbial community inhabiting the large intestine. It is also important to know if exogenous bifidobacteria can become part of the microflora or not, or if the bacterial colonization by these bacteria is dependent on a dynamic bifidobacterial population.
The present work aims to compare homologous and heterologous species of bifidobacteria with respect to their interaction with the gut and their influence on the immune function, when these bifidobacteria strains are orally administered.

Material and Methods

Animals

BALB/c mice aged 6 to 8 week, weighing 25 to 30 g purchased from a closed colony of the Institute CERELA, Tucumán, were used in our experiments. Each experimental group consisted of 3 _ 5 mice.

Bifidobacterium strains

The strains used were Bifidobacterium animalis CRL 1247 isolated from mouse and Bifidobacterium adolescentis CRL 1243 isolated from faeces of infants. The bifidobacteria were cultured in MRS broth (Oxoid) supplemented with glucose (1% w/v) and lactose (1%w/v) for 8 h at 37ºC under anaerobic conditions.

Fluorescent labelling of bacteria

Bacteria used were labelled with fluorescein isothiocyanate (FITC) prior to introduction into the mouse intestine. To label the bacteria, the inoculum (10⁷ cells) was pelleted and the bacterial pellet was resuspended (dilution 1/10) in PBS with FITC (100 mg per ml in PBS) and incubated for 1 h at 37ºC in the dark. The labelled bacteria were washed four times with PBS to remove unincorporated FITC. The final pellet was resuspended in PBS to achieve 10⁷ cells. Then, 0.2 ml of each bacterial suspension was administered by intubation to different groups of mice. Animals were sacrificed by cervical dislocation at 60 min intervals. The small and large intestine from 3 animals were removed and processed for histological examination (Saint-Marie, 1962).

Preparation of samples for electron microscopy

Groups of 3 mice that had received by intubation 0.2 ml of different unlabelled suspensions (10⁷ cells) of each strains, were sacrificed by cervical dislocation 30 min after Lactic Acid Bacteria (LAB) administration. The ileum near Peyer's patches in the small intestine and large intestine were removed. The intestinal contents were eliminated by washing with 1 ml of PBS. Tissues were fixed in formaldehyde (40%) and glutaraldehyde (10%). Specimens were then washed in PBS and fixed in 1% osmium tetroxide, dehydrated in ethanol, cleared in propylene oxide and finally embedded in a low viscosity medium. The sections from small and large intestine were stained with saturated uranyl acetate in 50% ethanol and 4% of citrate. Sections were examined by transmission electron microscopy. The sections from small and large intestine were photographed and the micrographs were produced at 4,900 x or 12,000 x magnification. Specimens from the control group of mice, which had not been dosed with bacteria, were processed in the same way. Two histological slices from each animal and each group were analyzed.

Feeding procedure

Mice were administered with 1x10⁷ of each bifidobacteria/day/mouse. The microorganisms were suspended in 5 ml of non-fat milk (NFM 10%) and were administered at 20% v/v in the drinking water for 2, 5 and 7 consecutive days. A control group received NFM 10% in the same conditions as the test group. All animals were fed ad libitum with a conventional balanced diet.

Histological preparation for determination of IgA producing cells

At the end of each period of feeding, 4 mice were sacrificed and samples from the ileum near Peyer's patches of small intestine and large intestine were removed. The intestinal fluids were washed out with 1 ml 0.01 M, phosphate buffered saline solution, pH = 7.2. Tissues were placed in ethanol in order to be processed by the Saint Marie's technique (1962). Once fixed, dehydrated and embedded in paraffin at 56ºC, they were cut in 4 mm serial paraffin sections and used to perform the immunofluorescence test.

Immunofluorescence test

The number of IgA producing cells were determined by direct immunofluorescence test. It was performed using the monospecific antibody (a chain specific) conjugated with fluorescein isothiocyanate (FITC) (Sigma, St. Louis, MO, USA). Histological samples from the small and large intestine were incubated with 0.2 ml of IgA antibody at 1/100 dilution for 30 min at room temperature. Then, they were washed three times with 0.01 M_phosphate buffered saline pH = 7.2. Three histological slices from each animal and for each period of feeding were analyzed. The number of positive cells were determined in 10 fields (magnification x 100).

Phagocytosis test

At the end of each feeding period with B. animalis or B. adolescentis, 5 mice of each group and periods of feeding were sacrificed by cervical dislocation and a non-specific immune response was performed by "ex vivo"; phagocytosis test using peritoneal macrophages according Perdigón et al., 1986. Peritoneal cells from mice of different group and period of feeding were recovered. Macrophages were suspended at a concentration of 10⁶ cells/ml in RPMI 1640 medium (Gibco). Then, 0.2 ml of Candida albicans suspension at a concentration of 5 x 10⁶ microorganisms/ml opsonized with mouse autologous serum was added to 0.2 ml of each macrophage suspension and incubated for 15 min at 37ºC. The phagocytosis rate was measured by counting 100 cells using a Zeiss optical microscope.

Statistical Analysis

Data were expressed as the mean (M) of n independent experiments + standard errors of the mean (SEM). Student's test was used to calculate the statistical significance of the results.

Results

Visualization of fluorescent Lactic acid bacteria in the gut

When we examined the histological slices from tissues of animals that received B. animalis (homologous strain), we observed fluorescent bacteria in the immune cells associated with the Peyer's patches of small intestine (Fig. 1 a and b) and in the large intestine (Fig. 2a and b).

a

b
Figure 1. a: Photograph of Peyer's patches from control animals that received non-fluorescent bacteria. X 40 b: Histological slice of Peyer's patch tissue from animals that received labelled B. animalis (10⁷ cells) by oral intubation. Numerous FITC labelled bacteria are observed in Peyer's patch. X 40.

a

b
Figure 2. a: Photograph of large intestine from control animals. b: Photograph of large intestine from animals which were given labelled B. animalis. X 40.

In the case of B. adolescentis (heterologous strain) small and large intestine samples did not show fluorescence. See Fig. 3a and b.

a

b
Figure 3. a: Histological slice of small intestine tissue from mice that received labelled B. adolescentis (10⁷ cells) by oral intubation. Numerous FITC labelled bacteria are in the intestinal lumen. X 40. b: Photograph of large intestine tissue from animals which were given labelled B. adolescentis. X 40.

Transmission electron microscopy studies

In these studies we observed that B. animalis was able to interact with the epithelial cell of small intestine (Fig. 4 a and b), and it internalized into the epithelial cells of large intestine. Fig. 5 a, b and c. B. adolescentis only showed lysosomal activation of the epithelial cells in small and large intestine (Fig. 6).

a

b
Figure 4. a: Transmission micrograph of control mouse epithelial cells from small intestine. Magnification x 4,900. b: Electron micrograph of epithelial cell from small intestine following oral inoculation of B. animalis. Intense lysosomal activity can be seen. X 7,800

a

b

c
Figure 5. a: Micrograph of control mouse epithelial cell from large intestine. X 4,900 b: Transmission electron micrograph of bacterium in the process of internalization to the large intestine after oral administration of B. animalis. X 56,600. c: Transmission electron micrograph of epithelial cell from large intestine after administration of B. animalis. Bacteria can be seen inside the epithelial cell. X 12,800

Figure 6. Micrograph of epithelial cell from small intestine after B. adolescentis administration. Lysosomal activity is observed. Magnification x 4,900

Determination of IgA⁺ B cells in lamina propria of small and large intestine

When we analyzed the effect of feeding of bifidobacteria in the IgA producing cells on lamina propria of the small intestine we observed that both strain B. adolescentis and B. animalis did not induce a significant increase in any period of administration compared with the control values. As regard the number of IgA⁺ cells present in the large intestine, we found that B. animalis induced a significant increase for a period of 5 and 7 days of administration in relation to the control, while B. adolescentis only increased these cells after 7 days of administration (Table 1).

Determination of peritoneal macrophage activation

We observed that either B. animalis or B. adolescentis did not enhance the phagocytic activity of peritoneal macrophage after the different periods of feeding assayed. The values obtained were less than the controls (Table 2).

Discussion

The gastrointestinal flora is without doubt the most complex part of normal mammalian microflora. Its composition differs among animal species, in individuals within the same species, and also during the life of the same individual. It plays an important role into protecting the mucosal surface from pathogens as well as the non-pathogens, avoid their attachment and entry into the intestinal mucosa.
Normal microflora has a marked effect on the mucosal immune system (Moreau and Gaboriau-Routhiau, 2000). The immune system of the gut is part of a specialized defense system for mucosal membranes. There is an immune response to the intestinal flora with presence of serum antibodies against numerous structures on the many different bacteria normally present in the gut. It has been shown that secretory IgA in the intestinal lumen prevents enteropathogen infections, and absorption of allergenic food proteins and carcinogens (Kagnoff, 1993; Mestecky and Mc Ghee, 1987).
In previous work we demonstrated that some lactic acid bacteria can increase mucosal immunity at different levels of the intestinal tract (Perdigón et al., 1995; Perdigón et al., 1999a, Perdigón et al., 1999b) and we also determined the pathways of internalization of different LAB with the gut (Perdigón et al., 2000). It has been previously suggested that bifidobacteria are able to enhance the immune function by increasing the cytokine production in macrophages stimulated by them (Marin et al., 1997), favouring the IgA synthesis (Yasui et al., 1992), the IgA response to dietary antigens (Takahashi et al., 1998) or the proliferation of Peyer's patch immune cells (Yasui and Ohwaki, 1991).
An accurate understanding of the mechanisms whereby probiotics exert beneficial effects on the host require studies of the interaction between probiotic strains and the intestinal microflora of the host. To demonstrate an immune function it is necessary to study the interaction of probiotic strain with immune cells associated with the intestinal mucosa. Mucosa epithelial cells also plays and important role in barrier defense and as cytokine producers (Kagnoff et al., 1996). These cells can respond to probiotics. It was also demonstrated that an individual bacteria can modify the host environment and stimulate for example fucosylation of glycolipids in the cells of small intestine (Gordon et al., 1997; Umesaki et al., 1996).
In the present work we demonstrated by fluorescent labelling techniques the importance of host specificity of bifidobacteria in determining the interaction with the immune cells associated with the gut. Where as B. animalis (homologous strain) showed interaction with immune cells of Peyer's patches and with the large intestine (Fig. 1b and Fig. 2b), B. adolescentis (heterologous strain) was incapable of binding to either the small or the large intestine (Fig. 3 a and b). However, by transmission electron microscopy (TEM) this latter
strain showed an increase in the epithelial cell activation in the small and large intestine with a lisosomal activation of the enterocytes, indicating this bacterial antigen could interact with these cells (Fig. 6).
Bifidobacterium animalis by TEM was able to make contact with the epithelial cells of the small intestine and to induce epithelial activation (Fig. 4b). It has been described that pathogens (bacteria, virus and parasites) can cross the mucosal barrier using different routes: transcellular route, paracellular route across cell for the tight junctions and via M cells (Hershberg and Mayer, 2000). We believe that B. animalis non-pathogenic bacteria was able to internalize into epithelial cells because of the host specificity (Fig. 5 b and c). B. adolescentis did not have this capacity because it was of human origen. Even though we could demonstrate that B. animalis interacted with the immune cells of Peyer patches, we could not demonstrate that antigen uptake was through M cells. We believe that the internalization observed was through the epithelial cells associated to follicles (FAE). The nature of the mucosal immune response induced will depend in part on the pathway of antigen internalization to the Peyer's patches. If via M cells it will induce a protective humoral immune response (Neutra and Kraehenbuhl, 1992), if via enterocytes it may or may not interact with dendritic cells or underlying macrophages favouring the immune response, or the clearence of the antigen (Eckman et al., 1995). We observed that B. animalis enters to the intestine through epithelial cells; this is also possible because it has been reported that the wall of the colon provides a special ecological site for attachment and proliferation of bifidobacteria (Matsuki et al., 1999). Some strains such as Bifidobacterium infantis secrete polysaccharides which can initiate adhesion to the epithelial cells of the intestine (Shah, 1997).
These previous findings would justify the results obtained on the number of IgA⁺ cells associated with the small and large intestine (Table 1), where for B. adolescentis this IgA cell was only increased in large intestine, and for B. animalis, even when we observed fluorescent bacteria in the immune cells of Peyer's patches, we did not find an increase in the IgA⁺ cells in the small intestine. We cannot exclude that this homologous strain could induce oral tolerance (Elson and Zivory, 1996). We suggested that the increase of these cells in the large intestine might due to an activation of epithelial cells with IL6 or other cytokine release, which are able to induce "in situ" clonal expansion the IgA⁺ cells (Kagnoff et al., 1996). Phagocytosis by peritoneal macrophages was not increased for any of the bifidobacteria strains assayed (Table 2). B. animalis would have only a local effect and may induce oral tolerance to T cells associated with the gut which would not be stimulated to produce cytokines. These biological messengers are keys for the immune response induction. The diminution of the immune response observed for B. adolescentis would be due to the lack of adhesion to the gut wall.
This study may form the basis for possible application of bifidobacteria as probiotic, where the species specificity, at least for colonization by bifidobacteria, is the fundamental importance in the selection of strains for probiotic use. We think that the beneficial effects for human health attributed to the bifidobacteria especially as anti diarrhea or antitumor activity, would be mediated by mechanisms other than those involving the immune system, for example: competitive inhibition, increase of acidity, affecting the activity of gut microflora or its composition, or may also affect the metabolism of the gut microorganisms.

Acknowledgement

This work was partly supported by grant from the Consejo Nacional de Investigaciones Cientificas y Técnicas (CONICET) PIP 5011 and the Consejo de Investigaciones de la Universidad Nacional de Tucumán (CIUNT) Nº26 D/127, Argentina.

References

1. Abdelali H, Cassand P, Saussotte B, Koch-Bocabeille B, Narbonne J (1995). Antimutagenicity of components of dairy products. Mutations Res. 331: 133-141.         [ Links ]
2. Bartram H, Scheppach W, Gerlach S, Ruckdeschel G, Kelber E, Kasper H (1994). Does yogurt enriched with Bifidobacterium longum affect colonic microbiology and fecal metabolites in healthy subjects? Am J Clin Nutr 59: 428-432.         [ Links ]
3. Bouhnik Y (1993). Survie et effects chez l'homme des bactéries ingéreés dans les laits fermentés. Lait. 73: 241-247.         [ Links ]
4. Eckman L, Kagnoff M, Fierrer J (1995). Intestinal epithelial cells as watchdogs for the natural immune system. Trends Microbiol. 3: 118-120.         [ Links ]
5. Elson CH, Zivory J (1996). Oral tolerance: Acommentary. In: Essentials of Mucosal Immunology. Kagnoff M. and Kiyono H., Ed. Academic Press, Inc., San Diego, CA. pp. 543-554.         [ Links ]
6. Gibson G, Wang X (1994). Regulatory effects of bifidobacteria on the growth of other colonic bacteria. J Appl Bacteriol. 77: 412-420.         [ Links ]
7. Gomes AMP, Xavier Malcata F (1999). Bifidobacterium ssp. and Lactobacillus acidophilus: biological, biochemical, technological and therapeutical properties relevant for use as probiotics. Trends Food Sc. Techn. 10: 139-157.         [ Links ]
8. Gordon J, Hooper V, Bry L, Midtved T, Falk P (1997). Interaction between epithelial cells and bacteria, normal and pathogenic response. Science 276: 965-967.         [ Links ]
9. Grill J, Crociani J, Ballongue J (1995a). Effects of bifidobacteria on nitrites and nitrosamines. Lett Appl Microbiol 20: 328-330.         [ Links ]
10. Grill J, Maniginot-Dürc C, Schneider F, Ballongue J (1995b). Bifidobacteria and probiotic effect: action of Bifidobacterium species on conjugated bile salts. Curr Microbiol 31: 23-27.         [ Links ]
11. Hershberg R, Mayer L (2000). Antigen processing and presentation by intestinal epithelial cells-polarity and complexity. Immunol Today 21: 123-128.         [ Links ]
12. Holzapfel W, Haberer P, Geisen R, Bjorkroth J, Schillinger V (2001). Taxonomy and important features of probiotic microorganisms in food and nutrition. American J Clin Nutr 73: 365-373.         [ Links ]
13. Kagnoff M (1993). Immunology of the intestinal tract. Gastroenterology 105: 1275-1280.         [ Links ]
14. Kagnoff M, Lars E, Yang S, Huang G, Jung H, Reed S, Fierer J (1996). Intestinal epithelial cells: an integral component of the mucosal immune system. In: Essentials of Mucosal Immunology. Kagnoff M., Kiyono H., Ed. Academic Press, Inc, San Diego, California. pp. 63-70.         [ Links ]
15. Kimura K, Mc Cartney A, Mc Connell M, Tannock G (1997). Analysis of fecal populations of bifidobacteria and lactobacilli and investigation of the immunologic response of their human host to the predominat strains. App Environ Microbiol 63: 3394-3398.         [ Links ]
16. Link-Amster H, Rochat F, Saudan K, Mignot O, Aeschlimann J (1994). Modulation of a specific humoral immune response and changes in intestinal flora mediated through fermented milk intake. FEMS Immunol Med Microbiol 10: 55-64.         [ Links ]
17. Marin M, Lee J, Murtha J, Ustunol Z, Pestka J (1997). Differential cytokine production in clonal macrophage and T cell lines cultured with bifidobacteria. J Dairy Sc 80: 2713-2720.         [ Links ]
18. Matzuki T, Watanabe K, Tanaka R, Fukuda M, Ayaizu H (1999). Distribution of bifidobacterial species in human intestinal microflora examined with 165rRNA gene targeted species-specific primers. App Environ Microbiol 65: 4506-4512.         [ Links ]
19. Mc Cartney A, Wenzhi W, Tannock G (1996). Molecular analysis of the composition of the bifidobacteria and lactobacilli microflora of humans. App Environ Microbiol 62: 4608-4613.         [ Links ]
20. Mestecky J, Mc Ghee J (1987). Immunoglobulin A (IgA): molecular and cellular interactions involved in IgA biosynthesis and immune response. Adv Immunol 40: 153-245.         [ Links ]
21. Misra A, Kuila R (1995). Antimicrobial substances from Bifidobacterium bifidum. Indian J Dairy Sc 48: 612-614.         [ Links ]
22. Moreau MC, Gaboriau-Routhiau V (2000). Influence of resident intestinal microflora on the development and functions of the intestinal-associated lymphoid tissue. In: Probiotics 3. Immunomodulation by the gut microflora and probiotics. Fuller R., Perdigón G., Ed. Kluwer Academic Publishers. pp. 69-104.         [ Links ]
23. Moore W, Moore L (1995). Intestinal flora of populations that have a high risk of colon cancer. App Environ Microbiol 61: 3202-3207.         [ Links ]
24. Neutra M, Kraehenbuhl J (1992). Transepithelial transport and mucosal defense. It the role of M cells. Trends Cell Biol 2: 134-138.         [ Links ]
25. Perdigón G, Alvarez S, Gobbato N, De Budeguer M, P De Ruiz Holgado A (1995). Comparative effect of the adjuvant capacity of Lactobacillus casei and LPS on the intestinal secretory antibody response and resistance to Salmonella typhimurium infection in mice. J Agric Immunol 7: 283-294.         [ Links ]
26. Perdigón G, Alvarez S, Medina M, Vintiñi E (1999a). Influence of the oral administration of lactic acid bacteria on IgA producing cells associated to bronchus. Int J Immunother Pharmacol 12: 97-102.         [ Links ]
27. Perdigón G, Alvarez S, Vintiñi E, Medina M, Medici M (1999b). Study of the possible mechanisms involved in the mucosal immune system activation by lactic acid bacteria. J Dairy Sci 82: 1108-1114.         [ Links ]
28. Perdigón G, Medina M, Vintiñi E, Valdez JC (2000). Intestinal pathway of internalisation of lactic acid bacteria and gut mucosal immunostimulation. Int J Immunopath Pharmacol 13: 141-150.         [ Links ]
29. Perdigón G, Nader de Macias ME, Alvarez S, Oliver G, Pesce de Ruiz Holgado A (1986). Effect of perorally administered lactobacilli on macrophage activation in mice. Infect Immun 53: 104-410.         [ Links ]
30. Saint-Marie G (1962). A paraffin embedding technique for studies employing immuno-fluorescence. J Histochem Cytochem10: 250-256.         [ Links ]
31. Schiffrin E, Rochat F, Link-Amster H, Aeschlimann J, Donnet-Hughes A (1995). Immunomodulation of human blood cells following the ingestion of lactic acid bacteria. J Dairy Sci 78: 491-497.         [ Links ]
32. Seki M, Igarashi M, Fukuda Y, Simamura S, Kawashima T, Ogasa K (1978). The effect of Bifidobacterium cultured milk on the regularity among aged group. Nutrit Food 4: 379-387.         [ Links ]
33. Shah N (1997). Bifidobacteria: characteristics and potential for application in fermented milk products. Milchwissenschaft 52: 16-20.         [ Links ]
34. Tahri K, Grill J, Schneider H (1997). Involvement of trihydroxyconjugated bile salts in cholesterol assimilation by bifidobacteria. Curr Microbiol 34: 79-84.         [ Links ]
35. Takahashi T, Nakagawa E, Nara T, Yajima T, Kuwata T (1998). Effect of orally ingested Bifidobacterium longum on the mucosal IgA response to mice dietary antigens. Biosci Biotechnol Biochem 62: 10-15.         [ Links ]
36. Tannock G (1999). Microecology of lactobacilli and bifidobacteria inhibiting the digestive tract: Essential knowledge for successful probiotic research. In: Probiotics, other Nutritional Factors and Intestinal Microflora. Hanson H., Yolben R. Ed., Nestlé, Nutrition Workshop series, Nestec Ltd Vevey, Lippincott. Raven Publishers. Philadelpia. pp. 17-28.         [ Links ]
37. Umesaki Y, Okada Y, Matsumoto S, Inaoka A, Setoyana H (1996). Segmented filamentous bacteria are indigenous intestinal bacteria that activate intraepithelial lymphocytes and induce MHC Class II molecules and fucosyl asialo GM1 glycolipids on the small intestine epithelial cells in the ex-germ free mouse. Microbiol Immunol 39: 555-562.         [ Links ]
38. Yasui H, Nagaoka N, Mike A, Hayakawa K, Ohwaki M (1992). Detection of Bifidobacterium strains that induce large quantities of IgA. Microbiol Ecol in Health Dis 5: 155-162.         [ Links ]
39. Yasui H, Ohwaki M (1991). Enhancement of immune response in Peyer's patch cells cultured with Bifidobacterium breve. J Dairy Sci 74: 1187-1195.         [ Links ]
40. Yildirim Z, Johnson M (1998). Characterization and antimicrobial spectrum of Bifidocin B a bacteriocin produced by Bifidobacterium bifidum NCFB 1454. J Food Prot 61: 47-51.         [ Links ]
41. Zavaglia A, Kociukinski G, Perez P, de Antoni G (1998). Isolation and characterization of Bifidobacterium strains for probiotics formulation. J Food Prot 61: 865-873.         [ Links ]

Received on April 24, 2002.
Accepted on December 2, 2002.