Original Article | Published:

Dose–response study of probiotic bacteria Bifidobacterium animalis subsp lactis BB-12 and Lactobacillus paracasei subsp paracasei CRL-341 in healthy young adults

Abstract

Objective:

This study was performed to investigate the dose–response effects of supplementation with Bifidobacterium animalis subsp lactis (BB-12) and Lactobacillus paracasei subsp paracasei (CRL-431) on blood lipids, recovery from feces and bowel habits. Changes of the fecal microflora was analyzed in the 1010 CFU/day probiotic and placebo group.

Design:

The study was designed as a randomized, placebo-controlled, double-blinded, parallel dose–response study.

Subjects:

Healthy young adults (18–40 years) were recruited by advertising in local newspapers. Of the 75 persons enrolled, 71 (46 women, 25 men, mean age 25.6 years (range 18–40 years)) completed the study.

Intervention:

The volunteers were randomly assigned into five groups receiving either placebo or a mixture of the two probiotics in the concentration of 108, 109, 1010 or 1011 CFU/day in 2 weeks run-in period, 3 weeks intervention and 2 weeks wash-out. Diary reporting bowel habits and well being (abdominal bloating, flatulence and headache) was kept for all 7 weeks and blood lipids, fecal recovery of BB-12 and CRL-431, as well as fecal microflora was tested before, immediately and 2 weeks after intervention.

Results:

The fecal recovery of BB-12 increased significantly (P<0.001) with increasing dose. In the group receiving 1011 CFU/day BB-12 was recovered from 13 out of 15 volunteers. CRL-431 was not recovered in any of the fecal samples. Supplementation with probiotics did not change the fecal bacterial composition. A significant linear increase in fecal consistency (looser stool) with increasing probiotic dose (P=0.018) was observed. No overall dose–response effect was found on the blood lipids. High doses of probiotics were well tolerated.

Conclusion:

A dose-related recovery of BB-12 from feces was observed.

Sponsorship:

The study was sponsored by Chr. Hansen A/S, Hoersholm, Denmark.

Introduction

It is generally accepted that the composition of the intestinal microflora influences the health and well being of humans. With the many publications showing beneficial effects of probiotic bacteria, there has been an increasing interest in the mechanism behind. Most clinical studies have tested only one dose of probiotics, ranging from 4 × 108 CFU/day for determination of gastrointestinal colonization and immune modulation (Valeur et al., 2004) up to 1011 CFU/day in infant formula given to allergic children (Isolauri et al., 2000). In a meta-analysis of the therapeutic effect of lactobacilli on acute diarrhea a dose–response effect was suggested (Van Niel et al., 2002). The importance of the dose was emphasized by the joint working group of FAO/WHO (2002) who recommend to define probiotics as ‘live microorganisms which when administered in adequate amounts confer a health benefit on the host.’ However, the adequate amount needs to be defined for each strain and product independently. To our knowledge no other study has investigated the dose–response effects of supplementation with more than two doses of probiotic bacteria. Fecal recovery of two different doses of Lactobacillus GG has been reported showing difficulties recovering from the low dose (1.6 × 108 CFU/day) whereas all the volunteers in the higher dose group (1.2 × 10(10) CFU) had detectable numbers of Lactobacillus GG in their feces during the test period using colony morphology and metabolism as identification principle (Saxelin et al., 1995).

Evidence exists that probiotics are effective in treatment and prevention of acute infectious diarrhea in children (Szajewska and Mrukowicz, 2001). This was recently demonstrated with Bifidobacterium animalis subsp lactis BB-12 as well as L. reuteri ATCC 55730 (Weizman et al., 2005). It has also been suggested that probiotics may alleviate clinical symptoms of pediatric atopic diseases (Isolauri et al., 2000; Kalliomäki et al., 2001, 2003; Rosenfeldt et al., 2003; Rosenfeldt et al., 2004) and modulate immune functions (Schiffrin et al., 1995; Fukushima et al., 1998; Blum and Schiffrin, 2003; Hart et al., 2004; Valeur et al., 2004). Concerning the effect on blood lipids contradictory results has been reported. Where some studies demonstrate an effect of probiotics on blood lipids in both normocholesterolemic (Agerholm-Larsen et al., 2000b; Kiessling et al., 2002) and in hypercholesterolemic (Anderson and Gilliland, 1999; Agerholm-Larsen et al., 2000b; Kiessling et al., 2002; Xiao et al., 2003) individuals, others report no effect (Lewis and Burmeister, 2005).

BB-12 has been used in clinical trials either alone (Fukushima et al., 1997; Fukushima et al., 1998; Isolauri et al., 2000; Kirjavainen et al., 2002) or together with other bacteria such as Lactobacillus acidophilus LA-5 (Alm et al., 1993; Laake et al., 1999) or Streptococcus thermophilus (Langhendries et al., 1995; Chouraqui et al., 2004; Saavedra et al., 2004). The latter combination exerted prophylactic activity against infectious rotavirus diarrhea in children (Saavedra et al., 1994; Chouraqui et al., 2004). Furthermore, long-term consumption seems to be well tolerated in infants and a significantly lower frequency of colic or abdominal irritability was reported in two different doses of probiotic intake compared to the placebo group (Saavedra et al., 2004). Supplementation with BB-12 has shown to alleviate clinical symptoms of atopic dermatitis in children (Isolauri et al., 2000) and to increase mucosal immunoglobulin A level (Fukushima et al., 1998). Previously supplementation with BB-12 either alone or in combination with other probiotics, was found to change the ratio between lactic acid bacteria and enterobactericeae in favor of the former (Langhendries et al., 1995; Fukushima et al., 1997, 1998; Laake et al., 1999; Kirjavainen et al., 2002), and the bowel habits in children as well as in elderly seem to be changed towards a more desirable pattern with softer and more bowel movements (Alm et al., 1993; Saavedra et al., 1998). Concerning the effect on blood lipids in vitro BB-12 showed the ability to deconjugate bile salt and thereby remove cholesterol from a culture medium (Klaver and van der Meer, 1993). Recently, BB-12 has shown the ability to reduce total cholesterol as well as very low-density lipoprotein+low-density lipoprotein (LDL) cholesterol in rats fed a cholesterol-enriched diet (Abd El-Gawad et al., 2005).

Lactobacillus paracasei subsp paracasei (CRL-431) in combination with L. acidophilus has shown significant effects in the prevention (Gonzalez et al., 1990) and treatment of diarrhea in children (Gonzalez et al., 1995; Gaon et al., 2002) and in adults with bacterial overgrowth-related chronic diarrhea (Gaon et al., 2002). In an in vitro study CRL-431 inhibited the growth of enteropathogenic bacteria, suggesting an efficacy in preventing and treating intestinal infections (Gonzalez et al., 1993).

In the present study, we have combined two probiotic cultures, BB-12 and CRL-431, in doses ranging from 108 to 1011 CFU/day in order to investigate the dose–response effects on blood lipids, composition of the microflora and tolerance in healthy young adults. Furthermore, the recovery of BB-12 and CRL-431 from fecal samples was determined. The effects on various immune parameters were also studied and are to be published elsewhere (Christensen et al., 2006).

Materials and methods

Study design

The study was conducted at the Department of Human Nutrition, the Royal Veterinary and Agricultural University, Copenhagen, as a randomized, controlled double-blind, parallel 7 weeks dose–response trial, with a 2 weeks run-in period, 3 weeks of intervention and 2 weeks wash-out. Study participants were randomized into five groups receiving daily two capsules of either placebo or probiotic doses of approximately 108, 109, 1010 or 1011 CFU/day, respectively. The probiotics and placebo was dispensed in a cellulose capsule size 1. Placebo consisted of dextrose whereas BB-12 and CRL-431 were mixed in approximately equal amounts in the desired dose. Volunteers were instructed to take capsules in connection with a meal. The study design with the schedule for blood and fecal sampling is shown in Figure 1. At enrollment the participants received written and verbal information about the objectives of the study, the inclusion and exclusion criteria and the importance of avoiding fermented milk products and supplements with probiotics during the 7 weeks trial period was emphasized. Otherwise, they were instructed to live and eat as they used to. Furthermore, they were explained how to fill in the diary and how to collect, store and deliver the fecal samples. At all four visits the participants were asked about changes in their well being and whether they had been ill or had taken any kind of medication since the last visit.

Figure 1
figure1

Design of the study with run-in, intervention and wash-out phase along with visits and sample collection. B indicates delivery of blood sample and F of fecal sample.

The protocol was approved by the Municipal Ethical Committee of Copenhagen and Frederiksberg, Denmark, journal number (KF) 02-093/02 and by the Danish Medicines Agency, journal number 2612-2178.

Participants

The study cohort comprised 75 healthy volunteers recruited by advertising in local papers. The volunteers gave informed written consent before the study started. Exclusion criteria were gastrointestinal, endocrine or immune system diseases including atopic diseases, colostomia patients, pregnancy or lactation, treatment with corticosteroids, antibiotics (within the last 8 weeks), antacids (within the last 4 weeks), immune suppressive drugs, any drug potentially affecting the gastrointestinal function and flora. The sample size was based on data from Gill and Rutherfurd (2001). With a relevant difference in the granulocyte activity of 20% between probiotic treatment and placebo and a strength of 90% 11 volunteers were needed in each treatment group. Fifteen volunteers were included in each group in order include potential to drop-outs.

Randomization

The 75 volunteers were randomly divided into five groups of 15 persons each. The participant numbers in the placebo group and the 1010 CFU/day group were marked with a star in order to be identified by the staff without disclosing who were in the two groups, as it was planned to make more analyses on samples from these two groups. The randomization list and the packing of medication was performed by a person without any practical involvement in the study.

Diary

Everyday throughout the 7 weeks study period the participants filled in a diary describing their bowel habits including the frequency of defecation and the fecal consistency according to the illustrations from the Bristol scale which have seven levels going from ‘hard nuts’ (level 1) to ‘watery’ feces (level 7) (Heaton and Thompson, 1999). At the end of each week they reported if they had suffered from abdominal bloating, had observed changes in flatulence or had experienced problems with headache during the past week.

Blood samples

At all four visits a blood sample was drawn from the antecubital vein after 10 min rest in a supine position. The participants had been fasting for a minimum of 10 h. The blood was analyzed for total cholesterol, high-density lipoprotein (HDL) cholesterol and triacylglycerol. LDL cholesterol was calculated according to the equation (Friedewald et al., 1972): LDL cholesterol (mmol/l)=total cholesterol (mmol/l)−HDL cholesterol (mmol/l)−(triacylglycerol/5 (mmol/l)). The blood for the total cholesterol, HDL cholesterol and triacylglycerol analyses was collected in tubes without anticoagulants, centrifuged for 3000g at 20°C for 15 min and the resulting serum was stored at −80°C. For the analysis of total cholesterol and triacylglycerol the cholesterol CHOD-PAP (no. 2016630) and the triacylglycerol GPO-PAP (no. 2016648) methods, respectively, were used, and for the measurement of HDL cholesterol the HDL-C plus second generation (no. 3030024) were used. All three kits were purchased from Roche Diagnostics, Mannheim, Germany. The analyses were performed using a Cobas Mira Plus autoanalyzer (Roche, Basel, Schwitzerland). Blood samples were analyzed at the Department of Human Nutrition at the Royal Veterinary and Agricultural University.

Content of probiotics in capsules

Initially, the content of live bacteria in the capsules was examined. For enumeration of BB-12 appropriate dilutions were spread on freshly made de Man, Rogosa, Sharpe (MRS, Oxoid cm 361, Hampskin, UK) agar plates with 4 mg/ml tetracycline (Sigma T-3383, Sigma-Aldrich, Steinheim, Germany) and 0.05% (w/v) cysteinhydrochloride (Merck KGaA 102830, Darmstadt, Germany) and incubated anaerobically (anaerogen, Oxoid) for 3 days at 37°C. For enumeration of CRL-431, dilutions were spread on MRS agar plates with 50 μg/ml vancomycin (Sigma v-2002) and incubated aerobically for 3 days at 37°C. Total count was determined on MRS agar plates incubated anaerobically for 3 days at 37°C. The results are given in Table 1.

Table 1 Daily dose of probiotic bacteria in each group measured with selective media at start of study

Fecal samples

At visits 2, 3 and 4 the participants delivered one fresh (<24 h) fecal sample divided into two plastic containers. At home the two containers were stored in a cool-bag containing two frozen chill elements and transported to the department, where they were stored at 5°C for up to 3 h before further analysis. Only fecal samples from the placebo and the 1010 CFU/day group were used for the analyses of the bacterial profile. The analyses were performed at Chr. Hansen A/S.

Recovery of BB-12 and CRL-431

Approximately 3 g of feces were weighed out in stomacher filter bags and 10-fold diluted in peptone water with 0,05% (w/v) cysteinehydrochloride followed by 3 min of stomaching (1 min high velocity and 2 min medium velocity). After mixing, 10-fold dilutions were made to appropriate concentrations and the samples were spread on freshly made MRS agar plates with 50 μg/ml vancomycin for selective isolation of CRL-431 and incubated for 3 days aerobically at 37°C. For selective isolation of BB-12 the samples were spread on MRS plates with 4 μg/ml tetracycline and 0.05% (w/v) cysteinehydrochloride and incubated anaerobically (anaerogen, Oxoid) for 3 days at 37°C (Lim et al., 1993; Charteris et al., 1998). After incubation 10 solitary colonies were randomly picked from plates with 20–200 colonies and further spread on MRS agar and incubated anaerobically for 3 days at 37°C. The morphology of the isolated colonies were noticed for later concentration estimation of recovered bacteria. Pure CRL-431-like colonies were aerobically grown in MRS broth with 10 mM threonine (Sigma t-8375) overnight at 37°C and further identified as described below. Pure BB-12-like colonies were incubated under anaerobic conditions in MRS with 0.05% (w/v) cysteine hydrochloride and frozen at −18°C with 15% (w/v) glycerol, and used for further verification by fluorescent whole cell hybridization.

Pulsed field gel electrophoresis

Pulsed field gel electrophoresis (PFGE) was performed according to: Hung and Bandziulis (Promega Notes number 24, page 1–3: Megabase DNA Analysis: Chromosomal DNA Preparation, Restriction, and Pulsed-Field Electrophoresis 1990). NotI was used as restriction enzyme for CRL-431-like colonies and SpeI for the verification of BB-12 findings in visit 2 samples and the placebo group. The electrophoresis parameters were 5.7 V/cm, switch times 2–30 s with a linear ramp factor and 24 h run time. PFGE fingerprints of all samples were directly compared by visual inspection to reference PFGE fingerprints of CRL-431, respectively, BB-12, which were included in all gels.

Spiking of CRL-431 in feces

Recovery of CRL-431 from feces was tested in vitro. Fecal samples were artificially spiked with CRL-431 at levels between 105 and 1011 CFU/g feces and incubated anaerobically as well as aerobically for 0, 2, 6 and 24 h at 4°C. After incubation, CRL-431 was recovered from feces as described above.

Fluorescent whole cell hybridization of BB-12-like colonies

Whole cell hybridization was performed as published previously (Beimfohr et al., 1993) with some modifications. The fixation step was omitted and instead the cell material from purified isolates from the semiselective media was directly smeared onto microscope slides (Marienfeld) and dried at 37°C for 30 min. The cell smears were covered with a 1 mg/ml lysozyme solution in 100 mM Tris/HCl, 50 mM EDTA (pH 8.0) and incubated at 37°C for 30 min. The B. animalis subsp lactis-specific probe, Bila-4 (Brockmann et al., in preparation), was labeled with Cy3. Eub338 (Stahl et al., 1989) was used as universal probe and labeled with fluorescein.

Bacterial spectrum

Fecal samples from the 1010 CFU/day probiotic group and the placebo group were tested in appropriate dilutions on specific agars for the number of enterobacteriaceae (violet red bile agar, aerobic 37°C overnight), enterococci (Slanetz & Bartley Oxoid CM377, aerobic incubation 2 days at 37°C), clostridia (TSC Oxoid CM587+SR88, anaerobic incubation 2 days at 37°C), bifidobacteria (BIM-25; (Munoa and Pares, 1988) anaerobic incubation 5 days at 37°C) lactobacilli (Rogosa agar Oxoid CM627; anaerobic incubation 3 days at 37°C) anaerobic and aerobic mesophilic bacteria (reinforced clostridial agar with calf blood, hemin and vitamin K, Oxoid CM151, anaerobic/aerobic incubation, respectively, 3–5 days at 37°C) and bacteroides (Bacteroides Bile Esculine agar, Quelab, Canada, anaerobic incubation 2 days at 37°C).

Statistical analyses

All statistical analyses were performed using SPSS 12.0 (SPSS®, Chicago, IL, USA). Normal distributed variables were described as mean±s.d. To assess whether there was a dose–response effect of probiotics, linear regression analyses were performed with probiotic dose coded as 0 (placebo), 1 (108), 2 (109), 3 (1010) and 4 (1011 CFU/day). Using this strategy, a dose–response effect was defined as a slope different from unity of the line between these points. The regression coefficient thus reflects the theoretic change in the dependent variable per 10 times increase of probiotic dose covering from 0 to 1011 CFU/day. A successive analysis strategy was used; first step, as mentioned above was a linear regression analysis using probiotic dose in a linear scale. If this test was significant, we added to the model the different probiotic doses as a categorical variable, whereby the deviation from the estimated line was estimated for each point. Independence and normal distribution of residuals were assessed in each model by plotting standardized residuals against predicted values. Level of significance was defined as P<0.05, and 95% confidence intervals were reported with the estimates. As the difference of baseline values rather reflected intra-individual variation than effects caused by changed behaviour in the run-in period, it was decided to include both baseline values of the dependent variable (visit 1 before the run-in period, and visit 2, immediately before intervention) as covariates in all analyses.

Results

Baseline characteristics of the participants

Of the 75 persons enrolled, 71 (46 women, 25 men, mean age 25.6 years (range 18–40 years)) completed the study. Two dropped out due to lack of time (one from the placebo and one from the 1010 CFU/day group), one developed hay fever 1 week into the run-in phase (1010 CFU/day group) and one developed diarrhea before the first visit (108CFU/day group). Baseline characteristics of the participants who completed the study are shown for each group in Table 2. Despite randomization the proportion of males was lower in the 1011 CFU/day group than in the other groups. The baseline values of blood lipids refer to the blood sample taken after 2 weeks run-in period just before the intervention started. At baseline the females had significantly higher HDL cholesterol (1.76 versus 1.43 mmol/l; P<0.001) and HDL:LDL ratio (0.95 versus 0.73 mmol/l; P=0.039) compared to males. The total cholesterol concentration was also higher among females (4.36 versus 4.01 mmol/l; P=0.076). Serum triacylglycerol (P=0.80) and LDL cholesterol (P=0.75) did not differ between females and males. All participants were normocholesterolemic.

Table 2 Baseline concentrations of blood lipids, sex distribution and age in the placebo group and the four groups receiving increasing doses of probiotics

Recovery of BB-12 and CRL-431

BB-12 was recovered from feces in a dose-dependent manner. When the dose of BB12 ingested was treated as a linear, continuous variable, the chance of a positive fecal recovery increased significantly by a factor 2.4 (OR (95% CI)=2.4 (1.5; 3.7)) per 10 times higher dose ingested (Table 3). In the 1011 CFU/day dose group, BB-12 was recovered in 13 out of 15 fecal samples after intervention with an average concentration of 8 × 107 CFU/g feces.

Table 3 Recovery of BB-12 from fecal samples

Surprisingly, BB-12 was also recovered in feces from some of the participants of the placebo group as well as before administration in the active groups. This was the case in six of the samples from different participants in the placebo group (one after run-in, two after intervention and three after wash-out) and in two of the run-in samples in the 1011 CFU/day group. After breaking the code the positive findings in these groups were confirmed by PFGE fingerprinting of the isolates. BB-12 was also recovered from two samples in the 1011 CFU/day group and one in the 109 CFU/day group after wash-out. CRL-431 was not recovered from any of the fecal samples. However, when recovering from artificially contaminated fecal samples, CRL-431 was found in all samples for spiking concentrations of 107 CFU/g or higher, and in half of the samples for a concentration 105 CFU/g.

Bacterial profile

No significant changes in the fecal microflora between the 1010 CFU/day group and the placebo group could be detected between visits 2 and 3 (Table 4).

Table 4 Baseline bacterial counts and changes in the bacterial profile during the intervention period (visits 2 and 3) for the placebo group and the 1010 CFU/day probiotic group

Blood lipids

There was no dose–response effect on any of the blood lipid parameters (Table 5). Each of the analyses was adjusted for age, sex and for the two baseline concentrations (before the run-in period and before the intervention). One of the participants had eaten a fatty meal short before the third blood sample and was excluded from the calculations. Serum HDL increased insignificantly with increasing probiotic dose (B (95% CI)=0.025 (−0.008; 0.059)), but it did not influence the HDL:LDL ratio (Figure 2). The estimate corresponds to a 0.025 mmol/l increase in HDL cholesterol per 10 times increase in probiotic dose. There were no effects of probiotics on blood lipids at the end of the wash-out period as assessed by linear regression (data not shown). BB-12 was isolated in the fecal samples from three participants in the placebo group and from two participants from the 1011 CFU/day group after the run-in period. Excluding these individuals from the analyses did not make any changes in the conclusions (data not shown). Using recovery of BB-12 as the predictor of blood lipid concentrations in linear regression analyses revealed no coherence between blood lipid concentration and the recovery of BB-12 (data not shown).

Table 5 Regression coefficient (95% CI) for dose–response analyses of the effect of increasing doses of probiotics on blood lipids (N=70)
Figure 2
figure2

The difference in blood concentration of LDL cholesterol, HDL cholesterol and HDL:LDL ratio from baseline to after intervention (baseline being represented by the mean concentration of the blood samples at day −14 and day 0). Bars represent 95% CI.

Daily records

For every 10 times increase in probiotic dose the frequency of defecation was 0.05 times higher in the last week of the intervention period (B (95%CI)=0.056 (−0.014;0.13)) when adjusted for age, sex and defecation frequency in the run-in period (P=0.12) (Figure 3). Mean (s.d.) frequency of defecation in the run-in period was 1.46 (0.42) times/day. On a weekly average, five of the 71 (7%) participants reported more than two defecations per day in the run-in period, while 14 of the 71 (19%) reported more than two defecations per day during the intervention. The highest reported weekly average in the run-in period was 2.5 defecations per day and after the intervention 3.8 defecations per day.

Figure 3
figure3

The frequency of defecation in the run-in period (dark gray bar) and in the intervention period (light grey bar). Linear trend in the intervention period, controlled for run-in values, P=0.12. Error bars denote the 95% CI.

There was a statistically significant linear increase in fecal consistency with increasing probiotic dose (B (95% CI)=0.11 (0.02;0.21); P=0.018, when adjusted for age, sex, and consistency in the run-in period (Figure 4). A decrease in consistency reflects a more solid stool, meaning that the stools of the groups receiving probiotics were looser than that of the placebo group in the intervention period. The mean values of consistency before and after the intervention were at the level of 3–4, which is considered as normal. In the run-in period four persons reported a weekly average in fecal consistency at level 2, two persons at level 5 and none at level 1, 6 and 7. After the intervention eight persons reported a weekly average in fecal consistency at level 2, three persons at level 5 and none at level 1, 6 and 7.

Figure 4
figure4

Fecal consistency in the run-in period (dark gray bar) and in the intervention period (light gray bar). Linear trend in the intervention period controlled for run-in values, P=0.018. Error bars denote the 95% CI.

Reported potential side effects were high with 68% complaining of flatulence, 37% complaining of abdominal bloating and 22% complaining of headache in the run-in period without any obvious patterns of change during intervention or wash-out. However, with the small number of participants in each group the statistical power to identify changes is low for such subjective parameters.

Discussion

Of the two probiotics used in the present study only BB-12 was recovered from the fecal samples. After consumption, BB-12 demonstrated not only an ability to survive in the human gastrointestinal environment but also a dose-dependent increase in the fecal recovery, and was generally recovered in concentrations ranging from 1 × 104 up to 8 × 108 CFU/g feces. In the low-dose groups, BB-12 was probably overgrown by other bifidobacteria on the agar plate and was not found by the random selection of colonies used here. BB-12 was surprisingly recovered in the placebo group in fecal samples from six different individuals distributed in the run-in, the intervention and the wash-out period in spite of the instructions to abstain from fermented milk products or probiotic supplements during the entire study period. The recovery of BB-12 was verified by PFGE fingerprinting. A questionnaire asking deeper into eating habits during the study period was sent to the participants after the intervention, but this did however not unveil the reason for these findings. As BB-12 is widespread in dairy products the participants could have ingested products containing BB-12 without realizing it. Another possibility is that BB-12 may colonize for a longer period in some individuals who had consumed food products with BB-12 before the trial, and therefore could be isolated even at the end of the run-in period. Owing to the limited recoveries in the wash-out period of the treatment groups this seems however unlikely.

Although CRL-431 originally was isolated from feces of a child (Gonzalez et al., 1993), CRL-431 was not recovered from feces in this study. Either the bacteria died during passage of the intestinal tract or did not survive in concentrations sufficiently high to be detected among the other bacteria. The detection method used here and survival in feces was tested by spiking CRL-431 to fecal samples in concentrations ranging from 105 to 1011 CFU/g feces. To simulate storage after defecation for up to 24 h, they were counted after 0, 2, 6 and 24 h. In a concentration of 107 CFU/g, CRL-431 was detected in all samples, whereas at a lower concentration (105 CFU/g), CRL-431 was only detected in half of the samples. Here, 105 CFU/g was found to be the lowest detectable concentrations using the combined selective isolation and PFGE technique. However, the high-dose group were approximately 1011 CFU/day, so either CRL-431 died or at least were reduced 5 decades or more. In contrast to this, CRL-431 has previously been recovered from human fecal samples by their specific colony morphology (Gaon et al., 2002) without confirming this by any molecular method though and the present study questions this recovery.

No significant changes in microflora was found after the intervention in the two groups tested. As it was not practically possible to receive the fecal samples immediately after ‘delivery’ we had to decide how to store the samples. Freezing them immediately after ‘delivery’ would probably change the fecal composition and kill some bacteria and cold storage up to 24 h would probably also changes the bacterial composition to some degree; however, the method used in this study was expected to be the best and therefore chosen. Chosing the media to calculate the specific bacterial groups upon, created another double-edged decision since selective media are known to inhibit the wanted bacteria to some degree too. In this way, various selective media underestimates some bifidobacteria species and fails to recover others (Apajalahti et al., 2003). However, a significant increase in bifidobacteria could only be seen in doses where fecal recovery reaches the concentrations of endogenous bifidobacteria and in this case the background concentration of bifidobacteria (Table 4) are determined to be more than 1 log higher than the average fecal recovery of BB-12 for the volunteers receiving 1010 CFU/day (results not presented). Alternatively molecular methods such as polymerase chain reaction (PCR) in combination with denaturing gradient gel electrophoresis (DGGE) could have been used to study changes in microflora. PCR-DGGE has been shown to be useful in demonstrating changes that occur in the composition of the fecal microflora (Tannock, 2002).

Results from several in vitro studies indicate that some strains of lactobacilli and bifidobacteria are able to reduce the concentration of cholesterol in a culture medium when grown under anaerobic conditions in the presence of bile acid (Tahri et al., 1996; Brashears et al., 1998; Pereira and Gibson, 2002). Likewise strains of B. animalis isolated from feces have shown to possess bile salt hydrolase activity (Tanaka et al., 1999). This has not been tested specifically for BB-12 but in vivo the hypocholesterolemic effect of bifidobacteria has recently been demonstrated in a study where rats were fed a cholesterol-rich diet along with BB-12 or Bifidobacterium longum BB-46-supplemented milk yoghurt as well as soy yoghurt (Abd El-Gawad et al., 2005). Clinical studies in humans are, however, ambiguous regarding the effects of probiotics on the concentrations of the blood lipids, showing both effect and absence there of. Underlining the importance of clinical trials, a recent study by Lewis and Burmeister (2005) have shown that even though in vitro studies with freeze-dried L. acidophilus showed a reduction in cholesterol, no reduction was seen in vivo. A study by Kiessling et al. (2002) with healthy normo- and hypercholesterolemic women (19–56 years) consuming yoghurt supplemented with L. acidophilus 145 and B. longum 913 for 6 months, showed a significant increase in HDL cholesterol in both the normo- and hypercholesterolemic women. When a cholesterol-lowering effect has been shown it is, however, more obvious in hypercholesterolemic individuals (Anderson and Gilliland, 1999; Kiessling et al., 2002; Xiao et al., 2003). The present study could not demonstrate any significant changes in the lipid profile in healthy young normocholesterolemic individuals after 3 weeks intervention.

The incidence of coronary heart disease (CHD) is directly correlated to LDL cholesterol and inversely correlated to HDL cholesterol (Dean et al., 2004). The protective role of HDL is supported by descriptions such as 2–4% increase in risk of CHD for every 0.0259 mmol/l decrease in HDL-cholesterol level (Dean et al., 2004). In the present dose–response study, a 10 times increase in probiotic dose caused an increase in HDL cholesterol of 0.025 mmol/l. The overall difference between the placebo group and the 1010 CFU/day group resulted in 0.075 mmol/l (4.6%) increase in HDL cholesterol and a decrease in LDL cholesterol of 0.021 mmol/l (0.9%) in the intervention period. Even though no overall significant differences in the HDL:LDL ratio was found, it cannot completely be rejected that the increase in HDL and decrease in LDL from baseline to the end of the intervention in the 1010 CFU/day group, could have a clinical prophylactic effect in healthy, normocholesterolemic individuals or people with elevated cholesterol levels. In general, the duration of studies showing effect on blood lipids seem to be of longer duration than the present (Agerholm-Larsen et al., 2000a).

Some studies indicate, that probiotics can influence the bowel habits of healthy individuals (Alm et al., 1993; Marteau et al., 2002; Xiao et al., 2003) and not only of people suffering from diarrhea. In the present study, healthy individuals in the placebo group reported slightly more solid stools compared to the probiotic groups. The present study also revealed a tendency to an increase in the frequency of defecation with increasing dose of probiotics. These effects on the bowel function have been reported elsewhere too, for example in a study with healthy individuals receiving a milk product fermented with B. longum, where one of 16 persons in the placebo group and five of 16 persons in the supplemented group experienced increased fecal frequency in the intervention period (Xiao et al., 2003). Another study showed that elderly constipated individuals experienced a significant improvement of bowel movements after consumption of fermented milk (P<0.05) supplemented with BB-12 and L. acidophilus LA-5 (Alm et al., 1993).

In general, the increasing doses of probiotic bacteria were well tolerated and appeared not to cause any major differencies in the risk of abdominal bloating, flatulence or headache in the intervention groups and no volunteers reported adverse side effects during the intervention.

In conclusion, BB-12 was recovered in feces in a dose-dependent manner, while CRL-431 was not recovered. The increasing dose of probiotics was well tolerated and did not seem to cause any adverse side effects. Increasing dose of probiotics showed an effect on bowel habits, with the observed changes before and after the intervention within a normal range. Short-term supplementation of increasing doses of BB-12 and CRL-431 in healthy, young normocholesterolemic persons had no conclusive influence on the blood lipid profile.

Additional information

Guarantor: KF Michaelsen.

Contributors: DCE, CNL and KFM wrote the protocol. DCE performed the study. CNL did the microbiological analysis of fecal samples and study product. EB performed the fluorescent whole cell hybridization of BB-12 like colonies. MB performed the PFGE of CRL-431 like colonies. PK performed the statistical analysis. CNL and SN wrote the first draft of the paper and all contributors participated in the revision and final approval of the paper.

References

  1. Abd El-Gawad IA, El-Sayed EM, Hafez SA, El-Zeini HM, Saleh FA (2005). The hypocholesterolaemic effects of milk yoghurt and soy-yoghurt containing bifidobacteria in rats fed on a cholesterol enriched diet. Int Dairy J 15, 37–44.

  2. Agerholm-Larsen L, Bell ML, Grunwald GK, Astrup A (2000a). The effect of a probiotic milk product on plasma cholesterol: a meta-analysis of short-term intervention studies. Eur J Clin Nutr 54, 856–860.

  3. Agerholm-Larsen L, Raben A, Haulrik N, Hansen AS, Manders M, Astrup A (2000b). Effects of 8 week intake of probiotic milk products on risk factors for cardiovascular diseases. Eur J Clin Nutr 54, 288–297.

  4. Alm L, Ryd-Kjellen E, Setterberg G, Blomquist L (1993). Effect of a new fermented milk product ‘CULTURA’ on constipation in geriatric patients. First Lactic Acid Bacteria Computer Conference Proceedings. Horizon Scientific Press, Norfolk, UK.

  5. Anderson JW, Gilliland SE (1999). Effects on fermented milk (yogurt) containing Lactobacillus acidophilus L1 on serum cholesterol in hypercholesterolemic humans. J Am Coll Nutr 18, 43–50.

  6. Apajalahti JHA, Kettunen A, Nurminen PH, Jatila H, Holben WE (2003). Selective plating underestimates abundance and shows differential recovery of bifidobacterial species from human feces. Appl Environ Microbiol 69, 5731–5735.

  7. Beimfohr C, Krause A, Amann R, Ludwig W, Schleifer K-H (1993). In situ identification of lactococci, enterococci and streptococci. Syst Appl Microbiol 16, 450–456.

  8. Blum S, Schiffrin EJ (2003). Intestinal microflora and homeostasis of the mucosal immune response: implications for probiotic bacteria? Curr Issues Intestinal Microbiol 4, 53–60.

  9. Brashears MM, Gilliland SE, Buck LM (1998). Bile salt deconjugation and cholesterol removal from media by Lactobacillus casei. J Dairy Sci 81, 2103–2110.

  10. Charteris WP, Kelly PM, Morelli L, Collins JK (1998). Antibiotic susceptibility of potentially probiotic Bifidobacterium isolates from the human gastrointestinal tract. Lett Appl Microbiol 26, 333–337.

  11. Chouraqui J-P, Van Egroo L-D, Fichot M-C (2004). Acidified milk formula supplemented with Bifidobacterium lactis: impact on infant diarrhea in residential care settings. J Ped Gastroent Nutr 38, 288–292.

  12. Christensen HR, Larsen CN, Kæstel P, Rosholm L, Sternberg C, Michaelsen KF et al. (2006). Immunomodulatory potential of supplementation with probiotics: A dose-response study in healthy young adults. FEMS Immunol Med Microbiol, in press.

  13. Dean BB, Borenstein JE, Henning JM, Knight K, Merz NB (2004). Can change in high-density lipoprotein cholesterol levels reduce cardiovascular risk? Am Heart J 147, 966–976.

  14. FAO/WHO (2002). Guidelines for the evaluation of probiotics in food. Joint working group report on drafting guidelines for the evaluation of probiotics in food. London, Ontario, Canada April 30 and May 1.

  15. Friedewald WT, Levy RI, Fredrickson DS (1972). Estimation of the concentration of low-density-lipoprotein cholesterol in plasma, without use of the preparative ultra-centrifuge. Clin Chem 18, 499–509.

  16. Fukushima Y, Li S-T, Hara H, Terada A, Mitsuoka T (1997). Effect of follow-up formula containing bifidobacteria on fecal flora and fecal metabolites in healthy children. Biosci Microflora 16, 65–72.

  17. Fukushima Y, Kawata Y, Hara H, Terada A, Mitsuoka T (1998). Effect of a probiotic formula on intestinal immunoglobulin A production in healthy children. Int J Food Microbiol 42, 39–44.

  18. Gaon D, Garmendia C, Murrielo NO, de Cucco Games A, Cerchio A, Quintas R et al. (2002). Effect of Lactobacillus strains (L. casei and L. acidophilus Cerela) on bacterial overgrowth-related chronic diarrhea. Medicina 62, 159–163.

  19. Gill HS, Rutherfurd KJ (2001). Probiotic supplementation to enhance natural immunity in the elderly: effects of a newly characterized immunostimulatory strain Lactobacillus rhamnosus HN001 (DR20(TM)) on leucocyte phagocytosis. Nutr Res 21, 183–189.

  20. Gonzalez S, Albarracin G, Locascio de Ruiz Pesce M, Male M, Apella MC, Pesce de Ruiz Holgado A et al. (1990). Prevention of infantile diarrhea by fermented milk. Microbiologie-Aliments-Nutrition 8, 349–354.

  21. Gonzalez SN, Apella MC, Romero NC, Nader De Macias ME, Oliver G (1993). Inhibition of enteropathogens by lactobacilli strains used in fermented milk. J Food Prot 56, 773–776.

  22. Gonzalez S, Cardozo R, Apella M, Oliver G (1995). Biotherapeutic role of fermented milk. Biotherapy 8, 129–134.

  23. Hart AL, Lammers K, Brigidi P, Vitali B, Rizzello F, Gionchetti P et al. (2004). Modulation of human dendritic cell phenotype by probiotic bacteria. Gut 53, 1602–1609.

  24. Heaton KW, Thompson WG (1999). Diagnosis. In: Grant Thompson W, Heaton KW (eds). Irritable Bowel Syndrome. Health press: Oxford, 27.

  25. Isolauri E, Arvola T, Sütas Y, Moilanen E, Salminen S (2000). Probiotics in the management of atopic eczema. Clin Exp Allergy 30, 1604–1610.

  26. Kalliomäki M, Salminen S, Arvillommi H, Kero P, Koskinen P, Isolauri E (2001). Probiotics in primary prevention of atopic disease: a randomised placebo-controlled trial. Lancet 357, 1076–1079.

  27. Kalliomäki M, Salminen S, Poussa T, Arvillommi H, Isolauri E (2003). Probiotics and prevention of atopic disease: 4-year follow-up of a randomised placebo-controlled trial. Lancet 361, 1869–1871.

  28. Kiessling G, Schneider J, Jahreis G (2002). Long term consumption of fermented diary products over 6 month increases HDL cholesterol. Eur J Clin Nutr 56, 843–849.

  29. Kirjavainen PV, Arvola T, Salminen S, Isolauri E (2002). Aberrant composition of gut microbiota of allergic infants: a target of bifidobacterial therapy at weaning? Gut 51, 51–55.

  30. Klaver F, van der Meer R (1993). The assumed assimilation of cholesterol by lactobacilli and Bifidobacterium bifidum is due to their bile salt-deconjugating cativity. Appl Environ Microbiol 59, 1120–1124.

  31. Laake KO, Bjorneklett A, Bakka A, Midtvedt T, Norin KE, Eide TJ et al. (1999). Influence of fermented milk on clinical state, fecal bacterial counts and biochemical characteristics in patients with ileal- pouch- anal-anastomosis. Microb Ecol Health Dis 11, 211–217.

  32. Langhendries JP, Detry J, Van Hees J, Lamboray JM, Darimont J, Mozin MJ et al. (1995). Effect of a fermented infant formula containing viable bifidobacteria on the fecal flora composition and pH of healthy full-term infants. J Pediatr Gastroenterol Nutr 21, 177–181.

  33. Lewis SJ, Burmeister S (2005). A double-blind placebo-controlled study of the effects of Lactobacillus acidophilus on plasma lipids. Eur J Clin Nutr 59, 776–780.

  34. Lim KS, Huh CS, Baek YJ (1993). Antimicrobial susceptibility of bifidobacteria. J Dairy Sci 76, 2168–2174.

  35. Marteau P, Cuillerier E, Meance S, Gerhardt MF, Myara A, Bouvier M et al. (2002). Bifidobacterium animalis strain DN-173 010 shortens the colonic transit time in healthy women: a double-blind, randomized, controlled study. Aliment Pharmacol Ther 16, 587–593.

  36. Munoa FJ, Pares R (1988). Selective medium for isolation and enumeration of Bifidobacterium spp. Appl Environ Microbiol 54, 1715–1718.

  37. Pereira DI, Gibson GR (2002). Cholesterol assimilation by lactic acid bacteria and bifidobacteria isolated from the human gut. Appl Environ Microbiol 68, 4689–4693.

  38. Rosenfeldt V, Benfeldt E, Nielsen SD, Michaelsen KF, Jeppesen D, Valerius NH et al. (2003). Effects of probiotic Lactobacillus strains in children with atopic dermatitis. J Allergy Clin Immunol 111, 389–395.

  39. Rosenfeldt V, Benfeldt E, Valerius N, Pærregaard A, Michaelsen KF (2004). Effects of probiotics on gastrointestinal symptoms and small intestinal permeability in children with atopic dermatitis. J Pediatr 145, 612–616.

  40. Saavedra J, Abi-Hanna A, Moore N, Yolken R (1998). Effect of long term consumption of infant formulas with bifidobacteria and S. thermophilus on stool patterns and diaper rash in infants. J Pediatr Gastroenterol Nutr 27, 483.

  41. Saavedra JM, Abi-Hanna A, Moore N, Yolken RH (2004). Long-term consumption of infant formulas containing live probiotic bacteria: tolerance and safety. Am J Clin Nutr 79, 261–267.

  42. Saavedra JM, Bauman NA, Oung I, Perman JA, Yolken RH (1994). Feeding of Bifidobacterium bifidum and Streptococcus thermophilus to infants in hospital for prevention of diarrhea and shedding of rotavirus. Lancet 344, 1046–1049.

  43. Saxelin M, Pessi T, Salminen S (1995). Fecal recovery following oral administration of Lactobacillus Strain GG (ATCC 53103) in gelatine capsules to healthy volunteers. Int J Food Microbiol 25, 199–203.

  44. Schiffrin EJ, Rochat F, Link-Amster H, Aeschlimann JM, Donnet-Hughes A (1995). Immunomodulation of human blood cells following the ingestion of lactic acid bacteria. J Dairy Sci 78, 491–497.

  45. Stahl DA, Devereux R, Amann RI, Flesher B, Lin C, Stromley J (1989). Ribosomal RNA based studies of natural microbial diversity and ecology. In: Hattori T, Ishida Y, Maruyama Y, Morita R, Uchida A (eds). Recent Advances in Microbial Ecology. Japan Scientific Societies Press: Tokyo, Japan, pp 669–673.

  46. Szajewska H, Mrukowicz J (2001). Probiotics in the treatment and prevention of acute infectious diarrhea in infants and children: A systematic rewiev of published, double-blind, placebo-controlled trials. J Pediatr Gastroenterol Nutr 33, S17–S25.

  47. Tahri K, Grill PJ, Schneider F (1996). Bifidobacteria strain behavior towards cholesterol:coprecipitation with bile salts and assimilation. Curr Microbiol 33, 187–193.

  48. Tanaka H, Doesburg K, Iwaski T, Mierau I (1999). Screening of lactic acid bacteria for bile salt hydrolase activity. J Dairy Sci 82, 2530–2535.

  49. Tannock GW (2002). Analysis of the intestinal microflora using molecular methods. Eur J Clin Nutr 56 (Suppl 4), S44–S49.

  50. Valeur N, Engel P, Carbajal N, Connolly E, Ladefoged K (2004). Colonization and immunomodulation by Lactobacillus reuteri ATCC55730 in the human gastrointestinal tract. Appl Environ Microbiol 70, 1176–1181.

  51. Van Niel CW, Feudtner C, Garrison MM, Christakis DA (2002). Lactobacillus therapy for acute infectious diarrhea in children: A meta-analysis. Pediatrics 109, 678–684.

  52. Weizman Z, Asli G, Alsheikh A (2005). Effect of a probiotic infant formula on infections in child care centers: comparison of two probiotic agents. Pediatrics 115, 5–9.

  53. Xiao JZ, Kondo S, Takahashi N, Miyaji K, Oshida K, Hiramatsu A et al. (2003). Effects of milk products fermented by Bifidobacterium longum on blood lipids in rats and healthy adult male volunteers. J Dairy Sci 86, 2452–2461.

Download references

Acknowledgements

The skillful technical assistance of Abdallah Albayasli, Helle Schack Andersen, Karin Schlichter, Rita Lohmann and Susanne Stoeving Jensen is highly appreciated. The study was financed by Chr. Hansen A/S.

Author information

Correspondence to K F Michaelsen.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • dose–response study
  • probiotics
  • cholesterol
  • recovery from feces
  • constipation

Further reading