immunomodulation by treatment with lactobacillus casei strain shirota

International Journal of Food Microbiology 41 (1998) 133–140

Immunomodulation by treatment with Lactobacillus casei strainShirota

*Takeshi Matsuzaki

Yakult Central Institute for Microbiological Research, 1796 Yaho, Kunitachi-shi, Tokyo 186, Japan

Accepted 6 March 1998

Abstract

Lactobacillus casei strain Shirota (LcS) has been shown to have potent anti-tumour and anti-metastatic effects ontransplantable tumour cells and to suppress chemically-induced carcinogenesis in rodents. In particular, intrapleural (i.pl.)administration of LcS into tumour-bearing mice has been shown to effectively inhibit the growth of tumour cells in thethoracic cavity and to significantly prolong survival time. Also, i.pl. administration of LcS has been shown to induce theproduction of several cytokines, such as IFN-g, IL-1b and TNF-a, in the thoracic cavity of mice, resulting in the inhibitionof tumour growth and increased survival. On the other hand, oral administration of LcS has been shown to inhibit the growthof implantable tumour cells in rodents, and to restore the decreased mitogenic response of tumour-bearing mice.Administration of LcS has also been shown to inhibit chemically-induced bladder cancer in rodents. These findings suggestthat treatment with LcS has the potential to ameliorate or prevent a variety of diseases through modulation of the host’simmune system, specifically cellular immune responses. 1998 Elsevier Science B.V.

Keywords: Anti-tumour activity; Immunomodulation; Lactobacillus casei strain Shirota; Oral administration

1. Introduction LAB which has been selected for its specific bio-logical activity in man and animals. LcS is widely

In recent years, lactic acid bacteria (LAB) have used in the production of probiotic dairy productsbecome of major economic importance to the food and is also used as a food ingredient both in Japanindustry and their role in maintaining gut health has and in Europe. Numerous papers document thealso been receiving increased research attention. In safety and functional characteristics of the strain.particular, the immune stimulating effect of LAB has Since the strain is not directly cytotoxic to tumourbeen investigated using several research models. cells in vitro, it is postulated that the mechanisms ofLactobacillus casei strain Shirota (LcS) is a strain of action appear to be mediated by augmentation of the

host’s immune system. This paper reviews some ofthe recent studies which examined the immune

*Tel.: 1 81 425 77 8960; fax: 1 81 425 77 3020. modulating effect of LcS.

0168-1605/98/$19.00 1998 Elsevier Science B.V. All rights reserved.PI I : S0168-1605( 98 )00046-4

134 T. Matsuzaki / International Journal of Food Microbiology 41 (1998) 133 –140

2. Anti-tumour and anti-metastatic effect of various transplantable murine tumours, we alsointravenous or intralesional injections of LcS examined the effect of i.v. or i.l. injections of LcS on

a highly metastatic variant of B16 melanoma inAnti-tumour and anti-metastatic effects of in- C57BL/6 mice (Matsuzaki et al., 1987a, 1988a). It

travenous (i.v.) or intralesional (i.l.) injections of LcS was shown that the treatment with LcS significantlywere examined using several transplantable tumours. reduced the formation of lung metastases (Table 1).It was reported that LcS (YIT No. 9018) was among The anti-metastatic effect of i.l. injections of LcSthe bacteria which possessed strong anti-tumour was also examined against lymph node metastasesactivity (Yokokura et al., 1981). It was also reported using the line-10 hepatoma tumour system in guineathat LcS showed strong anti-tumor activity against pigs (Matsuzaki et al., 1985). It was shown thatseveral tumours and that the anti-tumour activity of significantly more guinea pigs in the LcS treatedLcS was reduced by treatment with carrageenan, groups survived and were free from tumours andwhich is known to reduce macrophage activity (Kato lymph node metastases when compared with those inet al., 1981). It was also observed that the treatment the control group (Table 2). Also, a delayed typewith LcS augmented the cytolytic activity of skin reaction was observed following intra-dermalperitoneal macrophages. These results suggest that re-inoculation of line-10 tumour cells in guinea pigsthe anti-tumour activity of LcS might be macro- in which the tumour had been cured by the i.l.phage-dependent. injection of LcS, and tumour growth at the re-

As LcS showed a strong anti-tumour effect in inoculated site was completely rejected. These ob-servations indicate that the treatment with LcS waseffective in inducing a systemic tumour immunity in

Table 1 strain-2 guinea pigs.Anti-metastatic effect of LcS on a highly metastatic variant of B16

Since i.v. or i.l. injection of LcS effectivelymelanoma in C57BL/6 miceinhibited the formation of lung metastases, we

21 21Dose (mg kg day ) No. of metastases (mean6SD) examined the activity of the alveolar macrophagesExp. 1 Exp. 2 from the local site which might be involved in the

activation. As indicated in Fig. 1, we found that theSaline 170636 117638LcS (2) 76611 80619 cytolytic activity of the alveolar macrophages was

* *LcS (4) 47610 55630 augmented and reached a peak 1–3 days after LcS** **LcS (10) 28616 38620 injection. We also found that the natural killer (NK)

4B16-BL6 cells (5 3 10 ) were inoculated intravenously (i.v.) into activity of the axially lymph node cells was aug-C57BL/6 mice (n 5 7) on day 0, and LcS was injected i.v. four mented for 7 days after the LcS injection and that thetimes on days 7, 10, 13 and 16. The number of pulmonary

activity reached a peak on day 3 (Fig. 2). It wasmetastases was counted on day 21 after tumour inoculation.reported that LcS augmented NK activity of spleen* **Statistical significance compared with control: P , 0.05, P ,

0.01. cells (Kato et al., 1984). These results suggest that

Table 2Anti-metastatic effect of LcS on line-10 hepatoma in strain-2 guinea pigs

Dose of LcS Result after 73 days

No. of survivors No. of tumour-free animals No. of animals with lymph node metastases

Control 0 /6 0 /6 6/6*400 mg 3 1 3/6 3/6 3/6**400 mg 3 4 6/7 5/7 2/7

1 mg 3 1 2/7 2/7 5/7***1 mg 3 4 5/6 5/6 1/6

6Line-10 tumour cells (5 3 10 ) were inoculated intradermally in strain-2 guinea pigs. LcS was injected intralesionally when the tumourswere 9–10 mm in diameter. The control group was injected with phosphate buffered saline (PBS).

* ** ***Statistical significance compared with control: P , 0.05, P , 0.01, P , 0.001.

T. Matsuzaki / International Journal of Food Microbiology 41 (1998) 133 –140 135

the augmentation of cytolytic activity of the alveolarmacrophages and NK activity of the axially lymphnode cells were involved in the inhibition of lungand lymph node metastases.

Following these experiments we investigated theeffect of priming (pre-sensitizing the animals withLcS before inoculation with the primary tumours) onthe anti-metastatic effect of LcS against metastatictumour (Matsuzaki et al., 1990). LcS was injectedsubcutaneously (s.c.) 3 weeks and 1 week beforeLewis lung carcinoma (3LL) inoculation. It wasdemonstrated that priming significantly augmented

Fig. 1. Alveolar macrophage-mediated cytotoxic activity against the anti-metastatic effect of LcS to the extent thatB16-BL6. LcS (250 mg/mouse) was injected i.v. into C57BL/6 metastases-free lungs were observed (Table 3). Inmice (n 5 20) and alveolar macrophages were collected. Cytolytic

addition, when the LcS-induced primed peritoneal51activity was measured by Cr-release assay.exudated cells were subsequently stimulated withLcS in vitro, we found that the number of CD4 1 Tcells in the LcS-treated group was increased com-pared with that of the control group (Fig. 3). The

Fig. 2. Natural killer (NK) cell activity of axillary lymph node Fig. 3. Flow cytometric analysis of T cells from mice primed with6cells of C57BL/6 mice. LcS (250 mg/mouse) was injected into LcS. Peritoneal exudated cells (1 3 10 ) from normal or LcS-

the left front footpads of C57BL/6 mice (n 5 20) and the axillary primed mice injected with LcS were treated with PE-conjugatedlymph node cells were collected. NK cell activity was measured anti-CD4 or FITC-conjugated anti-CD8 antibodies and analysed

51by Cr-release assay. with a fluorescent activated cell-sorter.

Table 3aAnti-metastatic effect of LcS by priming with LcS in mice

Group No. of pulmonary metastases (median) Weight of primary tumour (g, mean6SD)

Control 34, 18, 13, 13, 11 (13) 7.6161.26*LcS 13, 5, 4, 3, 2 (4) 3.9161.53**Priming 1 LcS 5, 2, 0, 0, 0 (0) 1.0060.53

aC57BL/6 mice were primed with LcS (100 mg/mouse) subcutaneously (s.c.) 3 and 1 weeks before tumour inoculation. 3LL cells5(5 3 10 /mouse) were inoculated s.c. on day 0. LcS (250 mg/mouse) was injected intralesionally on days 3, 6, 9, 12 and 15, and the number

of pulmonary metastases and the weight of the primary tumours were determined on day 21.* **Statistical significance compared with control: P , 0.01, P , 0.001.


induced LcS-reactive CD4 1 T cells have the po- or anti-metastatic activity of LcS depends on thetential to produce several cytokines. Therefore, we production of anti-tumour cytokine, e.g. TNF-a, andthen investigated the production of several cytokines subsequently a series of down-stream immune re-in the peritoneal cavity. As indicated in Fig. 4, we actions.found increases in cytokines, such as IL-1b, IL-2,IFN-g and TNF-a. These results clearly indicatedthat the treatment with LcS augmented the host’s 3. Anti-tumour effect of intrapleural injection ofimmune responses and induced cytokine production LcSwhich in turn contributed to an augmentation of theanti-tumour immunity against transplantable Reports have mentioned that some bacteria, suchtumours. Thus, it was concluded that the anti-tumour as BCG, OK432, or a preparation of Streptococcus

pyogenes, have been used in the treatment of lungcancer patients with malignant pleurisy (Uchida andMicksche, 1983; Bakker et al., 1986; Luh et al.,1992). Subsequently, a system for the evaluation ofthe anti-tumour effect of bacterial preparationsagainst pleural effusion was developed and used tocompare the anti-tumour effects of several bacterialpreparations (Matsuzaki and Yokokura, 1987b; Mat-suzaki et al., 1988b). BALB/c mice were inoculatedwith Meth A tumour i.pl.. Then, LcS was injectedi.pl. and the subsequent survival was monitored. Thei.pl. injection of LcS into tumour-bearing mice wasshown to effectively inhibit the growth of tumourcells in the thoracic cavity and to significantlyprolong the survival of the mice in comparison withother bacterial preparations (Table 4). It was alsoshown that i.pl. injection of LcS stimulated NK cellactivity of the thoracic exudated cells and augmentedthe phagocytic activity of the thoracic macrophageagainst sheep red blood cells. We found that theeffect of LcS on the survival of Meth A-bearing micewas in part dependent on T cells since there was noeffect in either nude mice or SCID mice (Fig. 5)(Matsuzaki et al., 1996). We also demonstrated thatan anti-CD4 monoclonal antibody did not affect thesurvival of the mice and the effect of LcS was in partdiminished by treatment with an anti-CD8 mono-Fig. 4. Interleukin-1b (IL-1b), IL-2, interferon-g (IFN-g) and

tumour necrosis factor-a (TNF-a) production in LcS-primed mice. clonal antibody. These results show that CD8 1 TLcS (100 mg/mouse) was injected subcutaneously (s.c.) in the cells may play an important role in the action of theright groin of C57BL/6 mice (n 5 5) 3 weeks and 1 week before anti-tumour effect of LcS. This result is supported bytumour inoculation. After Lewis lung carcinoma were inoculated

the observation that LcS had no anti-tumour effect ins.c. in the left groin on day 0, LcS (250 mg/mouse) was injectedT cell deficient mice. We examined the production ofintraperitoneally on day 3, and the peritoneal fluid was collected

on day 6. IL-1b, IFN-g and TNF-a were analysed in the peritone- several cytokines, especially IFN-g, IL-1b, TNF-a]

al fluid. IL-1b and IL-2 were measured by the incorporation of and nitric oxide, by thoracic exudated cells induced3H-thymidine (TdR) in thymocytes of C3H/HeJ mice and IL-2 by LcS in a time-course study. IFN-g is a typicaldependent CTLL-2 cells, respectively. IFN-g and TNF-a were

type 1 helper T cell cytokine, or Th1 cytokine andmeasured by ELISA. The activity of TNF-a was expressed as thethe augmentation of Th1 cytokines enhances theinhibition rate (% cytostasis) against TNF-a-dependent L929

cells. immune responses. As indicated in Fig. 6, LcS


Table 4Anti-tumour effect of intrapleural administration of LcS in BALB/c mice

aBacterial preparation Dose (mg) Survival days (mean6SD) T/C (%)

Control 10.861.5 100**LcS 500 18.264.0 169***250 20.063.6 185***100 27.060.7 250*BCG 500 14.263.1 131

250 13.864.2 128**100 15.561.1 144

C. parvum 500 11.262.3 104250 9.661.2 89

*100 13.061.1 120**OK 432 250 15.762.2 145*100 15.763.6 145

5BALB/c mice (n 5 10) were inoculated with Meth A tumour (1 3 10 /mouse) intrapleurally (i.pl.) on day 0, and several bacterialpreparations were injected i.pl. on days 1, 3, 5, 7 and 9.aT/C (%) 5 (mean survival days of treated group/mean survival of control group) 3 100.

* ** ***Statistical significance compared with control: P , 0.05, P , 0.01, P , 0.001.

induced a continuous production of IFN-g, IL-1b,TNF-a and nitric oxide at an elevated level in thethoracic cavity. We also found an increase in IL-12production of the macrophages (data not shown).IL-12 is a cytokine which stimulates the productionof IFN-g, and plays an important role in the induc-tion of cellular immunity. With regard to othercytokines, or Th2 cytokines, IL-4, IL-6 and IL-10levels were found to be only slightly elevated (Fig.7) (Matsuzaki et al., 1996). Moreover, it was demon-strated using RT-PCR that cytokine gene expressionof the thoracic macrophages was strongly induced inthe group with LcS when compared with the controlgroup which received phosphate buffered saline (Fig.8). Taken together, these findings clearly suggest thatthe i.pl. treatment with LcS augments host immuneresponses such as cytokine production and killingactivity of tumour cells, resulting in the exhibition ofstrong anti-tumour effects. In clinical trials, it wasdemonstrated that i.pl. administration of LcS sig-nificantly prolonged the survival of lung cancerpatients with malignant pleural effusions in a phaseIII randomized controlled study in Japan (Masuno etal., 1991). Furthermore, it was reported that theLcS-treated patients were found to have higher

Fig. 5. Anti-tumour effect of LcS on BALB/c, nude and SCID Quality of Life (QOL) scores by an evaluation which5mice. Meth A cells (1 3 10 cells /mouse) were inoculated intrap-included health parameters such as chest pain, in-leurally (i.pl.) into mice on day 0. LcS (100 mg/mouse) wassomnia, nausea /vomiting, chest discomfort andinjected i.pl. on days 3 and 6, and the survival was monitored.

Numbers in parentheses indicate tumour-free mice / tested mice. anorexia (Masuno et al., 1994). Studies to date


Fig. 6. LcS induced changes in IFN-g, IL-1b, TNF-a and nitric oxide (NO ) production in the thoracic cavity of BALB/c mice. BALB/cx5mice (n 5 3) were inoculated with Meth A cells (1 3 10 ) intrapleurally (i.pl.) on day 0. LcS (100 mg/mouse) was injected i.pl. on days 3

and 6, and measurements were made after the final injection.

demonstrate the potential of various LcS preparations of orally administered LcS on the growth of sec-in counteracting tumour formation. Therefore, we ondary tumours was investigated after surgical re-expect that LcS, particularly LC 9018, a heat-killed moval of primary tumours (Kato et al., 1992, 1994).preparation of LcS, will be used in trials not only in In this model, five days after inoculation of Colon 26Japan but throughout the world. tumour cells in the abdomen of the mice, the primary

tumour was surgically removed. At this stage, thetumour-bearing period had been too short to establishtumour immunity. Then, the animals were inoculated

4. Effect of oral administration of LcS with Colon 26 into the hind footpad and the growthof the tumours was monitored. LcS was orally

Recently, the potential health benefits of consum- administered for 7 consecutive days after the surgicaling probiotic dairy products have received increased removal of the primary tumours to examine theattention. It was reported that the growth of both effect of LcS administration on tumour immunity. Asimplantable and chemically-induced tumour cells are a result of this experiment, tumours grew graduallyinhibited in rodents consuming LcS (Tomita et al., in the control group because tumour immunity was1994; Yokokura, 1994). Oral administration of LcS not sufficiently established, while tumour growth into Meth A-bearing mice inhibited tumour growth in a the LcS-treated group was significantly inhibiteddose-dependent manner. It was shown that oral compared with that in the control. These resultsadministration of LcS also inhibited methylcholan- indicate that oral administration of LcS modulatesthrene-induced carcinogenesis in C3H/HeN mice the immune responses in the process of the establish-and bladder carcinogenesis induced by BBN, a ment of tumour immunity and suggest that LcScarcinogen, in rats. These results clearly suggest that enhances tumour-specific immunity. It was alsooral administration of LcS is effective in the preven- shown that the reduced mitogenic response of cellstion of carcinogenesis in rodents. To determine collected from tumour-bearing mice was restored bywhether immune responses were involved, the effect oral administration of LcS.


Fig. 8. Cytokine gene expression induced by LcS. Meth A cells5(1 3 10 cells /mouse) were inoculated intrapleurally (i.pl.) into

mice on day 0. LcS (100 g/mouse) was injected i.pl. on days 3and 6. The thoracic exudated cells were collected and the gene

Fig. 7. IL-4, IL-6 and IL-10 production in the thoracic cavity of expression was measured by RT-PCR.BALB/c mice. LcS (100 mg/mouse) was injected intrapleurally(i.pl.) into BALB/c mice (n 5 3) on days 3 and 6, and thecytokines in the thoracic cavity were examined over time.

5. Conclusion

LcS shows potent anti-tumour and anti-metastaticOral administration of LcS has been shown to effects in rodents and, in particular, i.pl. injection of

suppress the recurrence of superficial bladder cancer LcS effectively inhibits tumour growth in thein a double-blind clinical trial (Aso et al., 1992, thoracic cavity not only in mice but also in humans,1995). In addition, the mitogenic response of the resulting in a prolonged survival period. It is sug-peripheral blood lymphocytes (PBL) of cancer pa- gested that these activities are dependent on thetients has been shown to increase after oral adminis- stimulation of the production of several cytokinestration of LcS (Asano, 1989). Oral administration of such as IFN-g, IL-1b and TNF-a. Moreover, oralLcS has also been shown to augment the ratio of administration of LcS suppressed tumour growth andT-helper and NK cells in PBL of colon cancer carcinogenesis in rodents, and exhibited a preventivepatients. Furthermore, LcS administration has been effect on the recurrence of superficial bladder cancerdemonstrated to reduce suppressor / inducer and sup- in humans. In particular, LcS promoted the pro-pressor T cells (Sawamura et al., 1994). These duction of several different cytokines which arereports suggest that oral administration of LcS has a involved in the regulation of host cellular immunedefinite immune modulating effect in several human responses.diseases. Taken together, these results lead us to conclude


Masuno, T., Kishimoto, S., Ogura, T., Honma, T., Niitani, H.,that LcS has the potential to ameliorate or prevent aFukuoka, M., Ogawa, N., 1994. Control of carcinomatousvariety of diseases through modulation of the host’spleural effusion with LC 9018 and quality of life in lung

immune system, specifically cellular immune re- cancer patients. Biotherapy 8, 847–856.sponses. Matsuzaki, T., Yokokura, T., Azuma, I., 1985. Anti-tumour

activity of Lactobacillus casei on Lewis lung carcinoma andline-10 hepatoma in syngeneic mice and guinea pigs. CancerImmunol. Immunother. 20, 18–22.Acknowledgements

Matsuzaki, T., Yokokura, T., Azuma, I., 1987. Anti-metastaticeffect of Lactobacillus casei YIT9018 (LC 9018) on a highlyThe author gratefully acknowledges Dr. Colettemetastatic variant of B16 melanoma in C57BL/6J mice.

Shortt for editing and helpful discussion of the Cancer Immunol. Immunother. 24, 99–105.manuscript. Matsuzaki, T., Yokokura, T., 1987. Inhibition of tumour metastasis

of Lewis lung carcinoma in C57BL/6 mice by intrapleuraladministration of Lactobacillus casei. Cancer Immunol. Im-munother. 25, 100–104.References

Matsuzaki, T., Yokokura, T., Mutai, M., 1988. The role of lymphnode cells in the inhibition of metastasis by subcutaneousAsano, M., 1989. Lactobacillus casei, immunology and cancer. In:injection of Lactobacillus casei in mice. Med. Microbiol.The Story of Lactobacillus casei and Cancer. Bunshodo,Immunol. 177, 245–253.Tokyo.

Matsuzaki, T., Yokokura, T., Mutai, M., 1988. Anti-tumour effectAso, Y., Akaza, H., the BLP Study Group, 1992. Prophylacticof intrapleural administration of Lactobacillus casei. Cancereffects of a Lactobacillus casei preparation on the recurrenceImmunol. Immunother. 26, 209–214.of superficial bladder cancer. Urol. Int. 49, 125–129.

Matsuzaki, T., Shimizu, Y., Yokokura, T., 1990. Augmentation ofAso, Y., Akaza, H., Kotake, T., Tsukamoto, T., Imai, K., Naito, S.,anti-metastatic effect on Lewis lung carcinoma (3LL) inthe BLP Study Group, 1995. Preventive effect of a Lac-C57BL/6 mice by priming with Lactobacillus casei. Med.tobacillus casei preparation on the recurrence of superficialMicrobiol. Immunol. 179, 161–168.bladder cancer in a double-blind trial. Eur. Urol. 27, 104–109.

Bakker, W., Nijhuis-Heddes, J.M.A., van der Velde, E.A., 1986. Matsuzaki, T., Hashimoto, S., Yokokura, T., 1996. Effects onPost-operative intrapleural BCG in lung cancer: A 5-year anti-tumour activity and cytokine production in the thoracicfollow-up report. Cancer Immunol. Immunother. 22, 155–159. cavity by intrapleural administration of Lactobacillus casei in

Kato, I., Kobayashi, S., Yokokura, T., Mutai, M., 1981. Anti- tumour-bearing mice. Med. Microbiol. Immunol. 185, 157–tumour activity of Lactobacillus casei in mice. Gann 72, 161.517–523. Sawamura, A., Yamaguchi, Y., Toge, T., Nagata, N., Ikeda, H.,

Kato, I., Yokokura, T., Mutai, M., 1984. Augmentation of mouse Nakanishi, K., Asakura, A., 1994. The enhancing effect of oralnatural killer cell activity by Lactobacillus casei and its Lactobacillus casei on the immunologic activity of colonsurface antigens. Microbiol. Immunol. 28, 209–217. cancer patients. Biotherapy 8, 1567–1572.

Kato, I., Endo, K., Yokokura, T., 1992. Induction of anti-tumourTomita, K., Akaza, H., Nomoto, K., Yokokura, T., Matsushita, H.,

activity with oral administration of Lactobacillus casei inHonma, Y., Aso, Y., 1994. Influence of Lactobacillus casei on

mice. Biotherapy 6, 1121–1126.rat bladder carcinogenesis. Nihon Hinyoukika Gakkai 85,

Kato, I., Endo, K., Yokokura, T., 1994. Effects of oral administra-655–663.

tion of Lactobacillus casei on anti-tumour responses inducedUchida, A., Micksche, M., 1983. Randomized controlled study ofby tumour resection in mice. Int. J. Immunopharmacol. 16,

OK-432 in the treatment of cancerous pleurisy. Jpn. J. Cancer29–36.Chemother. 10, 1497–1503.Luh, K.T., Yang, P.C., Kuo, S.H., Chang, D.B., Yu, C.J., Lee, L.N.,

Yokokura, T., Kato, I., Mutai, M., 1981. Antitumour effect of1992. Comparison of OK-432 and mitomycin C pleurodesisLactobacillus casei (LC 9018). In: Mitsuoka, T. (Ed.), Intesti-for malignant pleural effusion caused by lung cancer. Cancernal Flora and Carcinogenesis. Gakkai-Syuppan Center, Tokyo,69, 674–679.pp. 72–88.Masuno, T., Kishimoto, S., Ogura, T., Honma, T., Niitani, H.,

Fukuoka, M., Ogawa, N., 1991. A comparative trial of LC Yokokura, T., 1994. Anti-tumour and immuno-stimulating ac-9018 plus doxorubicin and doxorubicin alone for the treatment tivities of orally administered Lactobacillus casei. In: Theof malignant pleural effusion secondary to lung cancer. Cancer Third International Symposium on Intestinal Flora. Intestinal68, 1495–1500. Flora and Human Health, 72–88.

immunomodulation by treatment with lactobacillus casei strain shirota

Documents