A new synbiotic, Lactobacillus casei subsp. casei together with dextran, reduces murine and human allergic reaction

Authors


  • Editor: Alex van Belkum

Correspondence: Tomohiko Ogawa, Department of Oral Microbiology, Asahi University School of Dentistry, 1851-1 Hozumi, Mizuho, Gifu 501-0296, Japan. Tel./fax: +81 58 329 1421; e-mail: tomo527@dent.asahi-u.ac.jp

Abstract

We studied the development of atopic dermatitis-like skin lesions in NC/Nga mice and the allergic symptoms and blood patterns of healthy volunteers during the cedar (Cryptomeria japonica) pollen season in Japan following oral administration of a new synbiotic, Lactobacillus casei subsp. casei together with dextran. The combination of L. casei subsp. casei and dextran significantly decreased clinical skin severity scores and total immunoglobulin E levels in sera of NC/Nga mice that had developed picryl chloride-induced and Dermatophagoides pteronyssinus crude extract-swabbed atopic dermatitis-like skin lesions. During the most common Japanese cedar pollen season, synbiotic L. casei subsp. casei and dextran in humans led to no significant changes in total nasal and ocular symptom scores, in the levels of cedar pollen-specific immunoglobulin E, interferon-γ and thymus and activation regulated chemokine or in the number of eosinophils in sera, whereas the placebo group showed a tendency for increased levels of cedar pollen-specific immunoglobulin E, thymus and activation regulated chemokine and number of eosinophils, and a decrease in interferon-γ levels. Thus, the oral administration of synbiotic L. casei subsp. casei together with dextran appears to be an effective supplement for the prevention and treatment of allergic reactions.

Introduction

Atopic dermatitis (AD) is a very common skin disease characterized by pruritic and eczematous skin lesions, which appears in patients with a personal or family history of atopic disorders (Uehara & Kimura, 1993). Elevated serum immunoglobulin E (IgE) levels are also a characteristic feature in many AD patients. The incidence of AD is increasing, particularly in industrialized countries, and studies have suggested that 10–15% of the general population is affected by AD during childhood (Coleman et al., 1997; Leung, 1997). It has also been reported that environmental factors, such as mite antigens, air pollution and mental stress, may contribute to the development and progression of AD (Scalabrin et al., 1999; Schafer et al., 1999; Schmid-Ott et al., 2001). Of these factors, mite antigens have been emphasized because of their high potency as allergens.

The most common cause of seasonal allergic rhinitis in Japan is Japanese cedar (JC) pollen, and the number of patients affected has been reported to be approximately 10–30% of the Japanese population (Ohashi et al., 1998). Pollinosis is well recognized as an allergy and is characterized by sneezing, itchy and watery eyes, a runny nose and a burning sensation in the palate and throat. These clinical conditions are caused, at least in part, by lipid mediators and other toxic proteins from both eosinophils and mast cells (Dolovich et al., 1989; Fokkens et al., 1997). Several types of cytokine mediate the facilitation of the functions of these cells, such as endothelial adhesion and degranulation, as do C–C chemokines, which are mainly produced by T-helper type 2 (Th2) CD4+ T cells (Bradding et al., 1995; Sim et al., 1995). Ohashi et al. (1997) also demonstrated that a significant increase in specific IgE and interleukin-4 (IL-4), and a significant decrease in interferon-γ (IFN-γ) were observed during the pollen season in a nonimmunotherapy group of patients with seasonal allergic rhinitis caused by JC pollen. Further, plasma levels of thymus and activation regulated chemokine (TARC), a member of the C–C chemokine family, in patients with either allergic rhinitis or AD have been demonstrated to be significantly higher than those in healthy subjects (Terada et al., 2001).

It is well known that colonic microflora play an important role in health (Rastall, 2004). Accordingly, there is great interest in the use of prebiotic supplements as functional food constituents to control the composition of colonic microflora in order to improve health (Gibson & Roberfroid, 1995; Roberfroid, 2000). Prebiotic supplements stimulate the growth and colonization of probiotic bacteria. Therefore, ingested nonpathogenic microorganisms, such as Lactobacillus and Bifidobacterium species, are beneficial to health. A synbiotic is a supplement that contains both a prebiotic and a probiotic that work together to improve the friendly flora of the human intestine (Bengmark, 2003). Recently, we have shown that Lactobacillus casei subsp. casei JCM 1134T has a specific ability to utilize dextran, as oral administration of the bacterium in conjunction with dextran effectively enhanced the humoral immune responses to bovine serum albumin (BSA) in BALB/c mice (Ogawa et al., 2005). It has also been demonstrated that the oral administration of Bifidobacterium bifidum BGN4 and L. casei 911 suppresses the allergic responses in an ovalbumin-induced allergy mouse model (Kim et al., 2005). In addition, probiotic bacteria have been reported to prevent and treat inflammatory bowel disease in a number of previous studies (Shanahan, 2000). In the present study, we examined whether oral administration of L. casei subsp. casei together with dextran suppressed AD-like skin lesions in an allergy mouse model (mice were sensitized and challenged with picryl chloride and Dermatophagoides pteronyssinus crude extract). Further, we investigated the effect of oral administration of L. casei subsp. casei together with dextran on various allergic factors in blood samples of healthy volunteers during the most common JC pollen allergic season.

Materials and methods

Bacteria

Lactobacillus casei subsp. casei strain JCM 1134T was obtained from the Japan Collection of Microorganisms (Riken Biosource Centre, Saitama, Japan) and aerobically cultured in Lactobacilli-MRS broth (Difco Laboratories, Detroit, MI) at 37°C for 24 h. Bacterial cells were collected by centrifugation, washed three times with saline and lyophilized.

Mice

Six-week-old male NC/Nga mice were purchased from Charles River Japan (Yokohama, Japan) and given MF pellet chow (Oriental Yeast Co. Ltd., Tokyo, Japan) prior to the experiments. The animals received humane care in accordance with our institutional guidelines and the legal requirements of Japan.

Experimental diet

Lactobacillus casei subsp. casei showed the ability to use dextran, which has a molecular weight of 10 000 (Meito Sangyo Co. Ltd., Nagoya, Japan); the other intestinal bacteria tested were not found to metabolize dextran (Ogawa et al., 2005). Food-grade dextran that had passed a safety test was used in the following experiments. NC/Nga mice were given MF powdered chow (control diet; CD), MF powdered chow with dextran (average molecular weight, 10 000) (dextran diet; DD), lyophilized L. casei subsp. casei in MF powdered chow (L. casei subsp. casei diet; LD) or lyophilized L. casei subsp. casei in MF powdered chow with dextran (L. casei subsp. casei and dextran diet; LDD) during the experimental period.

Lactobacillus species in small intestinal contents from NC/Nga mice

NC/Nga mice were divided into four groups of six each and given CD, DD (500 μg of dextran per day), LD (1 × 107 CFU of lyophilized L. casei subsp. casei per day) or LDD (1 × 107 CFU of lyophilized L. casei subsp. casei and 500 μg of dextran per day). On day 56, the contents were taken from their small intestines. A 10-fold aliquot of each sample was prepared and 100 μL of each suspension was smeared onto Lactobacilli-MRS agar (BD Biosciences, Mountain View, CA), a selective enrichment medium containing 0.5% (weight in volume, w/v) dextran, and then incubated aerobically at 37°C for 24 h. Following incubation, colonies of dextran-fermentable Lactobacillus species were identified by periodic acid Schiff (PAS) staining and then counted (Shailasree et al., 2004). Briefly, 7.5% polyacrylamide gels containing 0.075% dextran T2000 (Pharmacia Fine Chemicals, Uppsala, Sweden) were equilibrated in 40 mM acetic acid buffer (pH 6.2) and overlaid on a colony plate, which was incubated overnight at 37°C. PAS stain was then applied, yielding a clear zone in which dextran was digested.

Development of AD-like skin lesions using picryl chloride

Dermatitis was induced in mice using 2,4,6-trinitro-1-chlorobenzene (picryl chloride), as described previously (Taniguchi et al., 2003). Briefly, picryl chloride was obtained from Tokyo Kasei Chemical Co. Ltd. (Tokyo, Japan) and recrystallized with ethanol. It was then used to induce dermatitis according to standard instructions provided by Charles River Japan. The mice were shaved to remove hair from the abdominal and ear areas the day before sensitization. On day 0, the abdomen, chest and foot pad of each mouse were sensitized with 150 μL of 5% picryl chloride dissolved in an ethanol and acetone mixture (4 : 1). On day 4, both ears were challenged with 30 μL of 0.8% picryl chloride dissolved in olive oil. On days 5–56, 0.8% picryl chloride solution was applied to both ears once a day.

Development of AD-like skin lesions using mite antigen

Mite-induced dermatitis was produced according to a previously published method with some modifications (Sasakawa et al., 2001). Briefly, D. pteronyssinus mites (LSL Co. Ltd., Tokyo, Japan) were suspended in phosphate-buffered saline (PBS; Sigma Chemical Co., St Louis, MO), crushed by supersonic waves and centrifuged at 1500 g for 15 min at 4°C. The supernatant was filtered using a 0.8 μm filter (0.8 μm syringe filter; Nalge Nunc International K.K., Tokyo, Japan) and the protein concentration was measured using a BCA Protein Assay Reagent Kit (Pierce Biotechnology, Inc., Rockford, IL). The supernatant (1 mg mL−1 of protein) was used as D. pteronyssinus mite antigen (DMA). The mice were shaved to remove hair from the whole back. The back and ears were then swabbed with 150 and 20 μL of 4% sodium dodecyl sulphate (SDS), respectively, to destroy the skin barrier function. After the SDS had dried, the back and ears were again swabbed with 150 and 20 μL of DMA, respectively. The treatment protocol with SDS and DMA was carried out once a day for 8 weeks.

Inhibitory effect of the synbiotic on the development of AD-like skin lesions

NC/Nga mice that received the MF diet were divided into four groups of six mice each. Starting from 7 days before sensitization, the mice were fed CD, DD, LD or LDD. The first sensitization with picryl chloride or DMA was applied on day 0. The body weights of the mice, severity of the AD-like lesions and total IgE levels were measured once a week or every 2 weeks during the experimental period, as described below.

Evaluation of the severity of the AD-like skin lesions

Ear lesion status was scored on the basis of previously published reports (Matsuda et al., 1997; Matsuoka et al., 2003). The total clinical skin severity score for AD-like lesions was defined as the sum of individual scores, graded as 0 (none), 1 (mild), 2 (moderate) and 3 (severe), for each of the following five signs and symptoms: itching, erythema/haemorrhage, oedema, excoriation/erosion and scaling/dryness.

Measurement of murine total IgE

Serum specimens were obtained from mice and stored at −20°C until quantitative analysis. The concentration of total IgE in the specimens was measured using a mouse IgE enzyme-linked immunosorbent assay (ELISA) kit (Shibayagi Co. Ltd., Gunma, Japan).

Human subjects

We examined 14 healthy volunteers (12 males and two females), aged 25–62 years. None had taken antibiotics, antiallergic drugs, supplemental fibre or vitamin supplements during the 4-week period before the experiment began. All subjects were informed about the details of the study and each signed an informed consent form approved by the Ethics Committee of Asahi University (reference number 15007). The volunteers were randomly divided into two parallel groups: the synbiotic group (six males and one female, aged 40.71±10.73 years) and the placebo group (six males and one female, aged 40.14±16.50 years). Each group volunteer ingested 0.5 g of triturated powder containing lyophilized L. casei subsp. casei (5 × 109 CFU) together with 1 g of dextran once per day, or 1.5 g of triturated powder once per day, as a control, for 28 days. Blood specimens were obtained on days 0, 14 and 28.

Pollen counts

Pollen counts in the atmosphere were obtained from the data acquisition system of the Ministry of the Environment of Japan (http://kafun.nies.go.jp/). The levels were measured during the 4-week period from 8 March to 5 April 2005 in the experimental area of Aichi Environmental Research Centre, Aichi, Japan. The values are presented as the particle number per cubic metre per day.

Symptom assessments

Nasal and ocular symptoms, described in subject diary cards, were classified according to a previously reported method (Ito et al., 1997): sneezing, nasal discharge, nasal blockage, nasal itching, itchy eyes and watery eyes were scored as 0 (none), 1 (mild), 2 (moderate), 3 (severe) or 4 (violent). The total symptom score was determined from the sum of the individual scores for each of the six symptoms.

Measurement of blood specimens

Blood specimens from the subjects were examined to determine the levels of IgE specific for JC pollen and TARC. The number of eosinophils was counted at Mitsubishi Kagaku Bio-Clinical Laboratories Inc. (Tokyo, Japan). IFN-γ levels in the serum samples were measured using a human IFN-γ ELISA kit (Bender MedSystems GmbH, Vienna, Austria).

Statistical analysis

Skin severity scores and the number of Lactobacillus species colonies obtained from each group of NC/Nga mice were compared using Steel's test. The total IgE level in each group of NC/Nga mice was subjected to a Dunnett's multiple comparison post hoc test. For the human study, the Wilcoxon signed rank test was used for paired data analysis to compare with the values on day 0 in each group. Comparisons between the synbiotic and placebo groups were made using a Mann–Whitney U-test. Statistical significance was set at P<0.01 and P<0.05.

Results and discussion

Effects of feeding Lactobacillus casei subsp. casei together with dextran on the development of AD-like skin lesions in picryl chloride-treated NC/Nga mice

NC/Nga mice develop AD-like skin lesions when kept in conventional surroundings, but not when under specific–pathogen–free (SPF) conditions, and also develop AD-like skin lesions by repeated applications of picryl chloride (Matsuda et al., 1997). In agreement with previous findings, the clinical severity of AD-like skin lesions of CD-fed NC/Nga mice increased gradually as a function of the number of challenges with picryl chloride (Fig. 1a). To examine the effects of LDD feeding on the development of AD-like skin lesions, picryl chloride-treated NC/Nga mice were given 500 μg per day of dextran and the indicated cells of Lactobacillus casei subsp. casei. As shown in Fig. 1b, feeding L. casei subsp. casei together with dextran effectively decreased the development of AD-like skin lesions in a bacterial cell-dependent manner. When 1 × 107 CFU of L. casei subsp. casei per day was given, an increased amount of dextran also resulted in a significant decrease in total clinical skin severity scores (Fig. 1c). Similarly, increased IgE levels were observed in CD-fed NC/Nga mice (Fig. 2a), whereas feeding L. casei subsp. casei together with dextran effectively decreased these levels (Figs 2b and c) in a bacterial cell- and dextran dose-dependent manner.

Figure 1.

 Effects of the oral administration of Lactobacillus casei subsp. casei together with dextran on the clinical skin severity scores of picryl chloride-induced atopic dermatitis (AD)-like skin lesions in NC/Nga mice. (a) Clinical skin severity scores were monitored once a week starting from 7 days before sensitization with 5% picryl chloride, as described in ‘Materials and methods’. Values represent the mean±SEM of the total score of six mice in each group. (b) The mice were fed the indicated cells of L. casei subsp. casei together with 500 μg of dextran from day −7, with the first sensitization with 5% picryl chloride occurring on day 0. Clinical skin severity scores were monitored on days 0, 28 and 56. Values represent the mean±SEM of the total score of six mice in each group. (c) The mice were fed 1 × 107 CFU of L. casei subsp. casei together with the indicated doses of dextran from day −7, with the first sensitization with 5% picryl chloride occurring on day 0. Clinical skin severity scores were monitored on days 0, 28 and 56. Values represent the mean±SEM of the total score of six mice in each group. The mean values were significantly different from none on days 0, 28 and 56, respectively (*P<0.05).

Figure 2.

 Effects of the oral administration of Lactobacillus casei subsp. casei together with dextran on the total immunoglobulin E (IgE) levels in serum of picryl chloride-sensitized NC/Nga mice. (a) Total IgE levels in serum were monitored on day −7 and then once every 2 weeks starting from day 0, as described in ‘Materials and methods’. Values represent the mean±SEM of the total score of six mice in each group. (b) The mice were fed the indicated cells of L. casei subsp. casei together with 500 μg of dextran from day −7, with the first sensitization with 5% picryl chloride occurring on day 0. Total IgE levels in serum were monitored on days 0, 28 and 56. Values represent the mean±SEM of the total score of six mice in each group. (c) The mice were fed 1 × 107 CFU of L. casei subsp. casei together with the indicated doses of dextran from day −7, with the first sensitization with 5% picryl chloride occurring on day 0. Total IgE levels in serum were monitored on days 0, 28 and 56. Values represent the mean±SEM of the total score of six mice in each group. The mean values were significantly different from none on days 0, 28 and 56, respectively (**P<0.01, *P<0.05).

Effects of feeding Lactobacillus casei subsp. casei together with dextran on the development of DMA-induced AD-like skin lesions in NC/Nga mice

House dust mites are a primary source of allergens in house dust in most temperate humid areas of the world, and mite allergens sensitize and induce asthma, perennial rhinitis and AD in a large number of patients with allergic disease (Arlian & Platts-Mills, 2001). Allergens derived from the dust mite species D. pteronyssinus and Dermatophagoides farinae have been recognized as the most important allergens in allergic disease worldwide (Platts-Mills et al., 2000). Exposure to mite allergens, shown by a positive skin test and an increased level of specific IgE in serum, is related to both reduced lung function and augmented airway hyperresponsiveness, indicating that hypersensitivity to mites is a significant independent factor for the development of asthma (Van der Heide et al., 1998). We examined whether feeding of L. casei subsp. casei together with dextran would abrogate the development of DMA-induced AD-like skin lesions in NC/Nga mice. We swabbed the back and ear areas with DMA and observed the clinical skin severity. CD-fed NC/Nga mice showed a significant increase in total clinical skin severity scores and IgE levels (Figs 3a and 4a). In contrast, mice fed LD (1 × 107 CFU of L. casei subsp. casei per day) or DD (500 μg of dextran per day) showed a decrease in the total clinical skin severity score, and those given LDD (1 × 107 CFU of L. casei subsp. casei and 500 μg of dextran per day) had significantly reduced scores on days 28 and 56 (Fig. 3b). A decrease in total IgE levels was also found by feeding LDD, and to a lesser extent LD and DD, at 56 days after the first sensitization (Fig. 4b). Further, NC/Nga mice given LDD showed the largest number of dextran-utilizable Lactobacillus species in their small intestinal contents (Fig. 5). These results clearly indicate that orally administered dextran-fermentable L. casei subsp. casei can control IgE levels, leading to a suppression of DMA-induced AD-like skin lesions in NC/Nga mice.

Figure 3.

 Effects of the oral administration of Lactobacillus casei subsp. casei together with dextran on the clinical skin severity scores of Dermatophagoides pteronyssinus mite antigen (DMA)-induced atopic dermatitis (AD)-like skin lesions in NC/Nga mice. (a) Clinical skin severity scores were monitored once a week starting from 7 days before sensitization with 5% DMA (1 mg mL−1 of protein), as described in ‘Materials and methods’. Values represent the mean±SEM of the total score of six mice in each group. (b) The mice were fed control diet (CD), dextran diet (DD), L. casei subsp. casei diet (LD) or L. casei subsp. casei and dextran diet (LDD) from day −7, with the first sensitization with DMA occurring on day 0. Clinical skin severity scores were monitored on days 0, 28 and 56. Values represent the mean±SEM of the total score of six mice in each group. The mean values were significantly different from CD group on days 0, 28 and 56, respectively (*P<0.05).

Figure 4.

 Effects of the oral administration of Lactobacillus casei subsp. casei together with dextran on the total immunoglobulin E (IgE) levels in serum of Dermatophagoides pteronyssinus mite antigen (DMA)-sensitized NC/Nga mice. (a) Total IgE levels in serum were monitored on day −7 and then once every 2 weeks starting from day 0, as described in ‘Materials and methods’. Values represent the mean±SEM of the total score of six mice for each group. (b) The mice were fed control diet (CD), dextran diet (DD), L. casei subsp. casei diet (LD) or L. casei subsp. casei and dextran diet (LDD) from day −7, with the first sensitization with DMA occurring on day 0. Total IgE levels in serum were monitored on days 0, 28 and 56. Values represent the mean±SEM of the total score of six mice in each group. The mean values were significantly different from CD group on days 0, 28 and 56, respectively (**P<0.01, *P<0.05).

Figure 5.

Lactobacillus species in NC/Nga mice small intestinal contents determined by a culture method. NC/Nga mice were fed control diet (CD), dextran diet (DD), L. casei subsp. casei diet (LD) or L. casei subsp. casei and dextran diet (LDD). On day 56, the contents were taken from their small intestines and smeared onto Lactobacilli-MRS agar. After 24 h of incubation, dextran-fermentable Lactobacillus species colonies were identified by periodic acid Schiff (PAS) staining, as described in ‘Materials and methods’. Bacteria in 10 culture plates from each group were counted and the results are expressed as the mean±SEM. The mean values were significantly different for the groups fed CD (*P<0.05) or LD (P<0.05).

Effects of the oral administration of Lactobacillus casei subsp. casei together with dextran on healthy subjects during the JC pollen allergy season

Allergic disease is characterized by an increase in allergen-specific IgE levels (Terada et al., 2001). Allergen exposure activates mast cells, and, consequently, various chemical mediators and cytokines capable of inducing the recruitment of allergy-related inflammatory cells, such as eosinophils, neutrophils and Th2 cells, are released (Bradding et al., 1993). An overproduction of Th2-derived cytokines, such as IL-4, IL-5 and IL-13, is also associated with allergic diseases (Christodoupoulos et al., 2000), as is the serum TARC level (Kakinuma et al., 2001). The TARC receptor CCR4 is selectively expressed on Th2 cells and TARC is related to the induction of a Th2-type response (Sugawara et al., 2002). In the present study, elevated IgE levels, increased numbers of eosinophils and enhancement of TARC production were observed in the sera of healthy subjects following natural exposure to JC pollen. It has been reported previously that the ingestion of probiotic Lactobacillus species reduces the incidence of atopic diseases (Kalliomäki et al., 2003) and alleviates allergic symptoms, whereas no significant changes in blood parameters between test and placebo groups have been shown (Rosenfeldt et al., 2003; Ishida et al., 2005).

Japanese cedar pollinosis is a relatively common disorder in Japan that presents serious discomfort between February and April of each year. We examined the effects of the oral administration of L. casei subsp. casei together with dextran on healthy volunteers during this season. Pollen counts were recorded during the 4 weeks from 8 March to 5 April 2005 in the experimental area of Aichi Environmental Research Centre, Aichi, Japan (Fig. 6a), and dispersion was found to be increased on day 13 during the study period. The alterations in the total symptom scores in the two groups corresponded to the alterations in the pollen count during the pollen season: the first peak in the mean total symptom score in both groups was observed when the pollen count showed marked pollination in the environment; however, the mean total symptom score in the synbiotic group was lower than that in the placebo group during the study period (Fig. 6b). Comparison between the synbiotic and placebo groups did not show a significant difference in the total symptom scores by analysis with the Mann–Whitney U-test.

Figure 6.

 Effects of oral administration of Lactobacillus casei subsp. casei with dextran on allergic reactions to pollen during the experimental period in the area of the Aichi Environmental Research Centre, Aichi, Japan: pollen counts (a) and the estimation of symptoms (b). Values in (b) represent the mean±SEM of the total score of seven subjects in each group, according to the estimation of nasal and ocular symptoms described in ‘Materials and methods’. The mean values were significantly different from the values on day 0 (*P<0.05). Comparison between the synbiotic and placebo groups did not show significant differences by analysis with a Mann–Whitney U-test.

The group of human subjects fed 5 × 109 CFU of L. casei subsp. casei and 1 g of dextran for 28 days showed no significant changes in the levels of cedar pollen-specific IgE, IFN-γ, TARC and the number of eosinophils, whereas the placebo group given 1.5 g of control product showed significant changes, including increased levels of cedar pollen-specific IgE and TARC, and increased numbers of eosinophils, as well as a decrease in IFN-γ levels, at several time points relative to the values on day 0 (Figs 7a–h). However, no statistically significant differences were found between the synbiotic and placebo groups with regard to blood parameters by analysis with a Mann–Whitney U-test. No significant differences in the total symptom scores or blood parameters in humans were found between the synbiotic and placebo groups in our study. One of the reasons for this may be that the number of subjects studies was too small. Further, it is possible that a longer period of synbiotic supplementation may be required to obtain significant changes in the total symptom scores or blood parameters.

Figure 7.

 Effects of the oral administration of Lactobacillus casei subsp. casei with dextran on blood patterns in human subjects. Synbiotic (a, c, e, g) and placebo (b, d, f, h) groups were given 1.5 g of the test product containing 5 × 109 CFU of L. casei subsp. casei with 1 g of dextran per day and 1.5 g of triturated powder per day, respectively, for 28 days. Blood specimens were taken on days 0, 14 and 28, and the levels of immunoglobulin E (IgE) specific for Japanese cedar (JC) pollen (a, b), interferon-γ (IFN-γ (c, d) and thymus and activation regulated chemokine (TARC) (e, f), and the number of eosinophils (g, h), were analysed. Values represent the mean±SEM of the total level of seven subjects in each group. The mean values were significantly different from the values on day 0 (*P<0.05). Comparison between the synbiotic and placebo groups with regard to the blood parameters did not show significant differences by analysis with a Mann–Whitney U-test.

In conclusion, the present results indicate that the oral administration of a new synbiotic, probiotic L. casei subsp. casei together with its prebiotic dextran, is effective for the control of allergic reactions in humans and mice by modulating the serum levels of IgE and cytokines, and the number of eosinophils. We also found that this synbiotic supplement attenuated allergic symptoms.

Acknowledgement

We thank Mr Mark Benton for critical reading of the manuscript.

Ancillary