|
|
||||||||
Laboratory of Dairy Science, Research Group of Animal Product Science, Division of Bioresources and Product Science, Graduate School of Agriculture, Hokkaido University, Sapporo-shi 060-8589, Japan
Corresponding author: H. Kumura; e-mail: kumura{at}anim.agr.hokudai.ac.jp.
| ABSTRACT |
|---|
|
|
|---|
Key Words: dairy yeast probiotics Caco-2 cell
Abbreviation key: YPD = yeast-peptone-dextrose
| INTRODUCTION |
|---|
|
|
|---|
Lactobacillus and Bifidobacterium strains have been extensively studied as probiotic agents (Borriello et al., 2003; Tuohy et al., 2003) because they are normal inhabitants of the intestinal tracts of humans and other vertebrates. In addition, these strains have been frequently used to produce fermented dairy products. Even if a strain of interest belongs to the same bacterial species, its probiotic effect may vary according to the properties of individual strain. Consequently, preferable strains are selected using an in vitro model system that reflects specific effects. However, before screening, strains should be checked for acid and bile tolerance to ensure survival during passage through the gastrointestinal tract. In addition, adhesion to the intestinal mucosa is considered important for exclusion of pathogens and undesirable bacteria. Furthermore, growth of the probiotic microorganism should also lead to acceptable flavor of the functional food.
Some fermented milks such as kefir and koumiss contain lactic acid bacteria and lactose-fermenting yeasts (Seiler, 2003). Yeasts can also be found in some traditional cheeses. Debaryomyces hansenii, Kluyveromyces lactis, and Yarrowia lipolytica have frequently been found (De Boer and Kuik 1987; Lopez-Diaz et al., 1995) as predominant species, although they have not been adopted for deliberate use. It is not unusual to find a yeast count of 105 to 107 cells per gram of cheese (Fleet, 1990), with beneficial effects such as interaction between starter cultures, production of aroma components, and inhibitory effects against spoilage microorganisms (Jakobsen and Narvhus, 1996). However, studies of yeasts from a probiotic standpoint have conducted in a limited manner using other species (Guslandi et al., 2003; Mansour-Ghanaei et al., 2003).
In this study, yeasts isolated from commercial blue-veined cheese and kefir were tested as potential probiotics. The ability of the yeasts to adhere to human intestinal cells was observed using the enterocyte-like Caco-2 cell culture system, and tolerance to acid and bile was evaluated. Factors responsible for adhesion of yeast strains to Caco-2 cells were also discussed.
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
Adhesion Assay
The adhesion assay was performed by the method of Chauvière et al. (1992). Caco-2 cells (passages 51 to 71) were seeded at a concentration of 4 x 104/cm2 on plastic coverslips placed in 24-well tissue culture plates. The culture medium was changed every other day. The Caco-2 monolayers at postconfluence after 10 d were washed 3 times with PBS. Yeast strains, randomly selected from each species identified, were cultured in YPD medium at 27°C for 2 d. Subsequently, they were harvested and washed twice with basal medium by centrifugation. After cell density was counted in the suspension dispersed in the basal medium (Packard and Ginn, 1985), yeast cells were seeded at a concentration of 106 in 24-well culture plates and incubated at 37°C for 90 min in 5% CO2-95% air. After incubation, the monolayers were washed 4 times with PBS, fixed with methanol, and stained with Giemsa solution followed by microscopic examination under oil immersion. Each adhesion assay was conducted in triplicate with cells from 3 successive passages. For each glass coverslip monolayer, the number of adherent yeast and Caco-2 cells was counted in 20 random microscopic areas. Adhesion ability was expressed as the number of yeast cells adhering to 100 Caco-2 cells.
Effects of Low pH and Bile on Viability and Growth of Yeast Strains
Effect of exposure to low pH was qualitatively evaluated by transferring one portion of the subcultured yeast medium to 0.5% NaCl containing 1 M HCl, pH 2.0 (10 mL), and incubating at 37°C for 3 h. Subsequently, a portion of the solution was inoculated into YPD medium and incubated at 27°C for 72 h. As an index of the growth, turbidity was visually monitored. Saline was used in place of NaCl/HCl as a control.
Growth of the strains in the presence of bile was examined using YPD medium containing 0.1% cholic acid (Wako Pure Chemical Co., Osaka, Japan). Incubation was carried out at 27 or 37°C for 24 h. If necessary, anaerobic incubation was performed using YPD medium containing 0, 0.3, or 0.5% cholic acid, with the surface of the medium in the test tube covered by liquid paraffin oil.
Proteinase Treatments
The influence of proteinase treatment on adhesion ability of K. lactis S1 and S25 and K. lodderae K2 was evaluated by the method of Sarem et al. (1996), with some modifications. The strain cultured in YPD medium at 27°C for 2 d was harvested and washed 3 times with PBS. The cells were dispersed in 0.05 M Tris-HCl buffer (pH 7.5) in the presence or absence of 0.1% trypsin (Sigma Chemical Co.) or proteinase K (Invitrogen, Tokyo, Japan). When the cells were treated with 0.1% pepsin (Sigma Chemical Co.), the buffer system was replaced by 0.05 M glycine-HCl buffer (pH 2.5). After incubation at 37°C for 1 h, cells were harvested, washed 3 times with PBS, and suspended in basal medium. During the enzyme treatments, cell density was approximately 107/mL. Then, yeast cells were counted and seeded at a concentration of 106 in 24-well culture plates to perform the adhesion assay. Dual treatment, consisting of incubation with pepsin followed by tryptic digestion, was also performed; the reactions were carried out at 37°C for 1 h for each enzyme.
Immunological Analysis
Effect of yeast strains on proinflammatory cytokine of IL-8 by Caco-2 cells was evaluated by the method of Morita et al. (2002). The Caco-2 monolayers at postconfluence after 10 d were incubated with K. lactis S1 or S25, whose viable cell count was 106 in 24-well culture plates as described above (adhesion assay). After incubation at 37°C for 24 h, the culture medium was recovered to quantify IL-8 using a commercial ELISA kit (Quantikine, R&D Systems, Inc., MN). For comparison, culture medium containing LPS (0.1 mg/mL) prepared from E. coli O111:B4 (Sigma Chemical Co.) was also investigated.
Statistical Analysis
Significant differences among means were analyzed by Tukey-Kramers test. A value of P < 0.05 was considered significant. All statistical calculations were carried out with JMP software (SAS Institute, Inc., Cary, NC).
| RESULTS |
|---|
|
|
|---|
Figure 1
shows the number of yeast cells adhesive to 100 Caco-2 cells. The genus Kluyveromyces possessed adhesion ability to Caco-2 cells; this was particularly evident for K. lactis (Figure 2
). Debaryomyces hansenii showed adhesion ability comparable to that of K. marxianus and K. lodderae; however, the adhesion ability of D. occidentalis was much lower. Other yeasts, including S. cerevisiae, Y. lipolytica, and C. humilis, were poorly adhesive to Caco-2 cells.
|
|
|
The component located on the surface of microorganisms could affect mucosal immune modulation. It has been reported that proteins expressed on the surface of the cell walls in lactobacilli are responsible for bacterial adhesion to intestinal epithelial cells (Vidgrén et al., 1992; Toba et al., 1995; Roos et al., 1996; Sarem et al. 1996; Sillanpää et al., 2000; Åvall-Jääskeläinen et al., 2003; Fernández et al., 2003). Consequently, correlation between adhesion ability and expression of cell surface proteins of the yeast strains was evaluated by pretreatment of the yeast cells with proteinases, including intestinal enzymes such as pepsin and trypsin. Among 5 strains of K. lactis, strains S1 and S25 were selected because significant differences in adhesion ability were found. Kluyveromyces lodderae was investigated for comparison.
Table 3
shows the percentage of yeast cells adherent to Caco-2 cells after each treatment. Apparently, adhesion ability was partially impaired when the 3 strains were incubated with pepsin; however, the value was comparable to the pepsin-free buffer solution (pH 2.5) used as a control. The results suggested that the reduction of adhesion ability was due to low pH rather than the peptic effect. The adhesion ability of the 3 strains was maintained following trypsin treatment; however, significant reduction due to proteinase K was recognized in K. lactis S1 and S25, but not in K. lodderae. Dual treatment with pepsin and trypsin led to an insignificant decrease of adhesion ability compared with the control. In general, adhesion ability was maintained at more than 50%.
|
|
| DISCUSSION |
|---|
|
|
|---|
Saccharomyces cerevisiae has been widely applied in industry, and beneficial effects such as promotion of iron absorption (Mai et al., 2002) and improvement of intestinal conditions (Takasaki and Saitoh, 1997) have been reported. Our study demonstrated that S. cerevisiae showed negligible adhesion ability; however, it could survive acidic conditions and grow in the presence of bile.
Among the yeasts tested in this study, K. lactis was the most adhesive to Caco-2 cells, although a significant difference was observed depending on the strain. When the same parameter was adopted, some strains of bifid-obacterium showed a value of >350 (Crociani et al., 1995), and Lactobacillus rhamnosus GG gave the highest value of >1500 (Lee et al., 2000; Bernardeau et al., 2001). However, these were interesting results because values less than 250 were frequently obtained in most adhesive lactobacilli and bifidobacteria (Chauvière et al., 1992; Crociani et al., 1995; Lee et al., 2000; Gopal et al., 2001). In the case of K. lactis, the value was apparently intermediate; however, it should be considered that a single yeast cell needs a larger space to adhere to the surface of an intestinal cell than a bacterium does because of the size difference between bacteria and yeasts.
Kluyveromyces lactis proliferated under anaerobic conditions, and showed acid and bile tolerance; however, growth was moderate at 37°C. Thus, we concluded that K. lactis as a probiotic agent could be expected to enhance immune modulation rather than improve the intestinal flora. Thus, we examined the component expressed on the surface of the cell wall that related to adhesion ability.
Proteins responsible for adhesion have been characterized and cloned (Vidgrén et al., 1992; Toba et al., 1995; Roos et al., 1996; Mukai et al., 1997; Sillanpää et al., 2000; Åvall-Jääskeläinen et al., 2003). However, the involvement of a proteinaceous factor in adhesion seems to depend on the strain; the adhesion ability of Lactobacillus gasseri and L. delbrueckii ssp. lactis to Caco-2 cells was impaired when the bacterial cells were treated with trypsin, whereas no reduction was observed in the case of L. acidophilus (Sarem et al. 1996; Fernández et al., 2003). In our study, proteinase K caused negligible reduction of adhesion ability for K. lodderae and significant reduction of adhesion ability for 2 strains of K. lactis. The results implied involvement of a proteinaceous factor for the adhesion in K. lactis; however, the function was unaffected by single or dual treatment with pepsin and trypsin. Consequently, it can be concluded that a protein-mediated adhesion system had a minor role in these yeast cells and that the adhesion properties would be maintained in the presence of intestinal proteinases.
For K. lactis, exposure to acidic conditions produced a more significant effect than proteinase treatment did with respect to subsequent adhesion ability. Because incubation of K. lactis in an acidic buffer led to more than 50% reduction of viability (data not shown), decrease of adhesion activity after incubation in acidic control buffer (Table 3
) could be ascribed to decreased living cells.
Treatment with sodium metaperiodate led to reduction of adhesion ability to Caco-2 cells of some probiotic strains including L. gasseri and L. acidophilus (Fernández et al., 2003), which implied that carbohydrates influence the adherence of the strain to intestinal epithelial cells. Consequently, we tried to examine adhesion of K. lodderae and 2 strains of K. lactis that had been incubated in 1% sodium metaperiodate solution. However, the treatment resulted in considerable reduction of the cell viability, probably due to the sensitivity of the cell wall of the yeast to oxidation. Thus, the contribution of carbohydrates in adherence of the yeast strains to Caco-2 cells remains unclear.
Finally, production of IL-8 by Caco-2 cells was investigated after their exposure to the yeast strains. Interleukin-8 is well known as the representative proinflammatory cytokine and its synthesis by enterocytes can be induced in response to bacterial enteric pathogens. Even some strains of bifidobacteria were demonstrated to be able to promote IL-8 secretion by Caco-2 cells (Morita et al., 2002); however, the 2 strains of K. lactis tested in this study showed little effect on its production.
Further studies are required to consider the physiological effects of yeasts from dairy products as probiotic microorganisms, in particular, the mucosal immunological modulation induced by oral administration of the yeast. Among the strains used in this study, K. lactis was of interest for further investigation for this purpose. Although this species is a lactose-fermenting organism and could cause a blowing problem such as expansion of the container by the production of CO2 gas under certain delivery conditions, it might be possible to incorporate it into some dairy products such as mold-type cheeses and frozen desserts.
| ACKNOWLEDGEMENTS |
|---|
|
|
|---|
We are very grateful to Hidemasa Motoshima, Yotsuba Dairy Industry, for the gift of yeast strains isolated from kefir, and we thank Koutaroh Ishikawa for technical assistance. We express our appreciation to Satoshi Ishizuka for his valuable suggestions for statistical analysis.
Received for publication May 6, 2004. Accepted for publication August 15, 2004.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. S. Martins, A. C. P. Rodrigues, F. C. P. Tiago, F. J. Penna, C. A. Rosa, R. M. E. Arantes, R. M. D. Nardi, M. J. Neves, and J. R. Nicoli Saccharomyces cerevisiae strain 905 reduces the translocation of Salmonella enterica serotype Typhimurium and stimulates the immune system in gnotobiotic and conventional mice J. Med. Microbiol., March 1, 2007; 56(3): 352 - 359. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |