Persistence of probiotic strains in the gastrointestinal tract when administered as capsules, yoghurt, or cheese

International Journal of Food Microbiology 144 (2010) 293–300

Contents lists available at ScienceDirect

International Journal of Food Microbiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j f o o d m i c r o

Persistence of probiotic strains in the gastrointestinal tract when administered as capsules, yoghurt, or cheese Maija Saxelin a, Anna Lassig b, Heli Karjalainen a, Soile Tynkkynen a, Anu Surakka a, Heikki Vapaatalo c, Salme Järvenpää d, Riitta Korpela a,c,1, Marja Mutanen b, Katja Hatakka a,⁎ a

Valio Ltd, Research and Development, P.O. Box 30, FI-00039 Valio, Helsinki, Finland University of Helsinki, Applied Chemistry and Microbiology, Nutrition, P.O. Box 66, FI-00014 University of Helsinki, Helsinki, Finland University of Helsinki, Institute of Biomedicine, Pharmacology, P.O. Box 63, FI-00014 University of Helsinki, Helsinki, Finland d MedCare Foundation, Hämeentie 1, FI-44100 Äänekoski, Finland b c

a r t i c l e

i n f o

Article history: Received 10 November 2009 Received in revised form 6 October 2010 Accepted 17 October 2010 Keywords: Probiotic bacteria Food matrix Persistence in gastrointestinal tract Lactobacillus Bifidobacterium Propionibacterium

a b s t r a c t Most clinical studies of probiotics use freeze-dried, powdered bacteria or bacteria packed in capsules. However, probiotics are commercially available in various food matrices, which may affect their persistence in the gastrointestinal tract. The objective of the study was to compare oral and faecal recovery during and after administration of a combination of Lactobacillus rhamnosus GG and LC705, Propionibacterium freudenreichii subsp. shermanii JS, and Bifidobacterium animalis subsp. lactis Bb12 as capsules, yoghurt, or cheese. This randomized, parallel-group, open-label trial (n = 36) included a 4-week run-in, 2-week intervention, and 3week follow-up period. Participants consumed 1010 cfu/day of probiotic combination and provided saliva and faecal samples before, during, and after the intervention. Strain-specific real-time PCR was used to quantify the strains. L. rhamnosus GG was the only probiotic strain regularly recovered in saliva samples. During the intervention period it was recovered in the saliva of 88% of the volunteers at least once. No difference was found between the yoghurt and cheese groups. At the end of the intervention, L. rhamnosus GG and LC705 counts were high in faecal samples of all product groups (8.08 and 8.67 log10 genome copies/g, respectively). There was no matrix effect on strain quantity in faeces or the recovery time after ceasing the intervention. For P. freudenreichii subsp. shermanii JS and B. animalis subsp. lactis Bb12, a matrix effect was found at the end of the intervention (P b 0.01 and P b 0.001, respectively) and in the recovery time during follow-up (P b 0.05 for both). Yoghurt yielded the highest faecal quantity of JS and Bb12 strains (8.01 and 9.89 log10 genome copies/g, respectively). The results showed that the administration matrix did not influence the faecal quantity of lactobacilli, but affected faecal counts of propionibacteria and bifidobacteria that were lower when consumed in cheese. Thus, the consumption of probiotics in yoghurt matrix is highly suitable for studying potential health benefits and capsules provide a comparable means of administration when the viability of the strain in the capsule product is confirmed. © 2010 Elsevier B.V. All rights reserved.

1. Introduction One of the main preconditions for a bacterial strain to be called probiotic is its ability to survive in the gastrointestinal environment, although the importance of viability for the beneficial effects of probiotics is not well defined since inactivated and dead cells also have immunological and health-promoting effects (Ghadimi et al., 2008; Lievin-Le Moal et al., 2007; Lopez et al., 2008). The gastrointestinal microbiota together with the host's defence mecha-

⁎ Corresponding author. Tel.: + 358 50 384 0992; fax: + 358 10 381 3019. E-mail addresses: [email protected] (M. Saxelin), riitta.korpela@helsinki.fi (R. Korpela), katja.hatakka@valio.fi (K. Hatakka). 1 Present address: University of Helsinki, Institute of Biomedicine, Pharmacology, Finland. 0168-1605/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ijfoodmicro.2010.10.009

nisms resists colonization with invading bacteria (Lievin-Le Moal and Servin, 2006). Faecal recovery during the oral administration of a probiotic strain is a standard method to show survival in the gastrointestinal tract. Furthermore, the recovery of adherent probiotic strains from the intestinal gut mucosa confirms their persistence (Alander et al., 1999; Mättö et al., 2006). The persistence of the mucosal colonization of live Lactobacillus rhamnosus GG was about one week, but two of seven volunteers still carried the strain two weeks after discontinuing consumption (Alander et al., 1999). A dose response study with freeze-dried L. rhamnosus GG powder indicated that consumption of 1010 bacteria per day is needed to detect the strain in stool samples (Saxelin et al., 1991). However, a later clinical study showed that the strain was recovered in stool samples after consumption of a much lower dose (108) when consumed in liquid milk (Hatakka et al., 2001). Also, fermented milk seemed to protect

294

M. Saxelin et al. / International Journal of Food Microbiology 144 (2010) 293–300

the strain in the gastrointestinal environment (Saxelin et al., 1993). This is in agreement with in vitro studies, which showed that milk proteins and a cheese matrix enhance the survival of bacterial strains in acidic conditions (Sharp et al., 2008). Similarly, the faecal recovery of a Lactobacillus plantarum strain in gastrointestinal tract was higher and more frequent when 6 × 109 cfu was administered in fermented sausage than in freeze-dried powder (Klingberg and Budde, 2006). However, the gastrointestinal survival of Bifidobacterium animalis DN173010 was equally good when a high dose (approximately 1011 cfu) was administered in fermented milk or in freeze-dried powder (Rochet et al., 2008). The enumeration method of probiotic bacteria may have a significant effect on the survival analysis. Lahtinen et al. (2006) showed that real-time PCR and FISH (fluorescent in situ hybridization) gave higher numbers of bifidobacteria than plate counting since part of the cells were not culturable. Real-time PCR method has been used for direct detection and quantification of bacterial species from DNA isolated from faeces (Haarman and Knol, 2006; Matsuki et al., 2004; Requena et al., 2002; Rinttilä et al., 2004). Moreover, a couple of strain-specific real-time PCR assays have been developed, e.g. for the strains L. rhamnosus GG (Ahlroos and Tynkkynen, 2009), L. casei Shirota (Fujimoto et al., 2008), and Lactococcus lactis subsp. cremoris FC (Maruo et al., 2006), which improve the detection of a specific probiotic strain amongst other intestinal bacteria. The combination of L. rhamnosus GG, L. rhamnosus LC705, Propionibacterium freudenreichii subsp. shermanii JS, and B. animalis subsp. lactis Bb12 has shown to relieve gastrointestinal discomfort (Kajander et al., 2008; Saxelin et al., 2010). Since many of the clinical intervention studies are conducted with a probiotic ingredient as such, the effect of the administration matrix warrants further evaluation. In the present study, we investigated the gastrointestinal survival and the persistence of colonization of these four probiotic strains when administered as capsules, yoghurt, or cheese. 2. Subjects and methods 2.1. Subjects The intervention was conducted in Helsinki, Finland, between April and June, 2006. Healthy adult volunteers were recruited, mostly from the personnel of Valio Ltd, University Pharmacy Ltd, and FujitsuSiemens Ltd, Helsinki, Finland, via e-mail advertisements and posters. Inclusion criteria were healthy persons (chronic diseases controlled with proper medications were allowed, if not mentioned in the exclusion criteria), aged 18 to 60 years, and acceptance of the study protocol. Exclusion criteria were allergy to milk, pregnancy, lactation, diagnosed chronic intestinal diseases, medications associated with intestinal diseases (e.g. lactase enzyme, constipation or diarrhoea medications), alcohol abuse (more than 21 servings per week), drug abuse, or being a prisoner, disabled, participating in another clinical study, and unwillingness to follow the study protocol. Antibiotics were not allowed two months before the intervention. Background information was collected from the volunteers through screening interviews, and included questions regarding general health, medications, and manner of living. Based on previous experience (Alander et al., 1999; Saxelin et al., 1995; Yli-Knuuttila et al., 2006), the recruitment goal was 10–15 persons per group. Altogether, 40 volunteers were recruited to the study; however, 3 dropped out because of antibiotic treatment and one because of acute diarrhoea. 2.2. Study design The study was a randomized, parallel-group, open-label intervention with no placebo group since the different product forms (capsules, yoghurt, and cheese) were compared with each other.

However, analysis of the study samples was blinded. Volunteers were randomized to 1 of 3 groups using computer-conducted, random permuted blocks with three persons' blocks. The 9-week study included a 4-week run-in period, 2-week intervention, and 3-week follow-up period. At the beginning of the run-in phase, the participants received a list of all probiotic products available on the market and were forbidden to consume any of them. Otherwise, they were instructed to continue their normal diet and lifestyle. During the intervention period, the volunteers consumed the probiotic study products. Saliva and faecal samples were collected at the end of the run-in period (day 0 of the intervention), on days 7 and 14 during the intervention, and on days 1, 3, 5, 7, 10, 14, 17, and 21 during the follow-up period. Venous blood puncture samples were drawn at the end of the run-in and intervention periods for safety control. During these research visits, the study diaries of the volunteers were checked (these reported the consumption of the product, medications, possible consumption of forbidden probiotics, and abnormal symptoms) and they received their study intervention products. 2.3. Probiotic products The probiotic multispecies combination (LGG® Extra, Valio Ltd, Helsinki, Finland) consisted of four bacterial strains; Lactobacillus rhamnosus GG (ATCC 53103), L. rhamnosus LC705 (DSM 7061), Propionibacterium freudenreichii subsp. shermanii JS (DSM 7067), and Bifidobacterium animalis subsp. lactis Bb12 (DSM 15954). All products were taken once a day during breakfast. The bacteria were administered in 3 product forms: hard cellulose capsules, low-fat, low-lactose yoghurt, and low-fat semi-hard cheese. The volunteers took either two capsules (total dose 1.9 × 1010 cfu), 200 g yoghurt (total dose 3.0 × 1010 cfu), or 35 g cheese (total dose 5.0 × 109 cfu) daily. The doses of the individual strains and product nutritional information are presented in Table 1. The cfu of L. rhamnosus GG and LC705 in the products were analyzed on modified MRS agar (De Man et al., 1960) where glucose was replaced with lactose and 50 μg/ml vancomycin was added. The plates were incubated at 37 ± 1 °C under anaerobic conditions for 3 days. The colonies of the GG strain are light and transparent (lactose negative) on this agar, and LC705 forms white colonies (lactose positive). The cfu of P. freudenreichii subsp. shermanii JS were analyzed on buffered propionibacteria agar plates (5.0 g tryptone, 10.0 g yeast extract, 10.0 g β-glycerophosphate [C3H7Na206P∙5H2O] and 16.8 ml sodium lactate [50%], per liter [pH 7.2]) by incubation at 30 °C under anaerobic conditions for 5–7 days. Typical colonies of JS strain are yellowish brown. B. animalis subsp. lactis Bb12 counts were analyzed on raffinose agar plates (Hartemink Table 1 The nutritional value of yoghurt and cheese, the daily doses of the individual strains (Lactobacillus rhamnosus GG and LC705, Propionibacterium freudenreichii subsp. shermanii JS, and Bifidobacterium animalis subsp. lactis Bb12), and the total probiotic daily dose in 2 capsules, 200 g yoghurt, or 35 g cheese. Nutritional values/100 g Energy, kJ/kcal Protein, g Carbohydrates, g Lactose, g Fat, g

Capsule n.d. n.d. n.d. n.d. n.d.

b

Yoghurta

Cheese

290/69 3.35 12.9 0.80 0.44

1100/260 31 0 0 15

4.7 × 109 3.3 × 109 7.5 × 109 1.4 × 1010 3.0 × 1010

2.8 × 109 4.2 × 108 1.7 × 109 4.2 × 107–1.2 × 106 5.0 × 109

Daily doses of the probiotic strains, cfuc L. rhamnosus GG L. rhamnosus LC705 P. freudenreichii subsp. shermanii JS B. animalis subsp. lactis Bb12 Total

5.2 × 109 7.4 × 109 4.2 × 109 1.8 × 109 1.9 × 1010

a Yoghurt also contained a yoghurt-starter culture with Streptococcus thermophilus and Lactobacillus delbrückii subsp. bulgaricus. b Not determined. c Colony forming units.

M. Saxelin et al. / International Journal of Food Microbiology 144 (2010) 293–300

et al., 1996) by incubation at 37 °C under anaerobic conditions for 3 days. Colonies are typically yellowish-green surrounded by yellow precipitate haloes. The viability of L. rhamnosus GG and LC705, and P. freudenreichii subsp. shermanii JS strains were practically unchanged during the intervention in all products (data not shown). However, the viability of B. animalis subsp. lactis Bb12 in cheese decreased from the original 4.2 × 107 cfu to 1.6 × 106 cfu/daily dose after one week, possibly because the sliced cheese was packed under a protective gas and after opening the package, bifidobacteria were challenged with oxygen. To compensate for the decrease in bifidobacteria, new cheese packages were opened by the participants every 4 days. 2.4. Blood samples The general health of the volunteers and the safety of the intervention were confirmed by blood analyses at the ends of the run-in and intervention periods. Blood samples were drawn after fasting for 12 h and analyzed for hemoglobin, hematocrit, aspartateamino-transferase (s-ASAT), alanine-amino-transferase (s-ALAT), Creactive protein (CRP), serum total cholesterol (s-Chol), serum highdensity lipoprotein (s-HDL), serum low-density lipoprotein (s-LDL), and serum triglyceride (s-Tg). All blood samples were analyzed by a certified clinical laboratory, United Laboratories Ltd., Helsinki, Finland.

295

significance in baseline characteristics between the product groups was evaluated by analysis of variance or the Fisher–Freeman–Halton test. The blood analyses were tested using paired samples t-test within the groups. The comparison of gastrointestinal symptoms between the intervention and the follow-up period was made with the McNemar test (number of the participants with any symptoms) and the Fisher–Pitman permutation test (the duration of symptoms). The comparison between groups in quantities of the probiotic strains at the end of intervention was made using the Mann–Whitney or Kruskal–Wallis test followed by Dwass–Steel–Chritchlow–Finger-test for pair wise comparison. The change in the quantities of the probiotic strains within groups was tested with Wilcoxon signed rank test for related samples. The observed survival of the strains was illustrated using the life table techniques. The survival curves between the product groups were tested by Wilcoxon– Gehan test. 2.9. Ethics All the participants provided written informed consent and were told that they could withdraw from the study at any time. The study followed good clinical practices and the Helsinki Declaration. The Human Ethics Committee of the Joint Authority for the Hospital District of Helsinki and Uusimaa (HUS) approved the study protocol. 3. Results

2.5. Saliva samples 3.1. Characteristics of participants The participants in the yoghurt and cheese groups collected unstimulated saliva samples in the morning of the specified days, before breakfast and brushing their teeth. Use of disinfecting mouth wash and tooth pastes with trichlosane was forbidden. The samples were frozen immediately at −20 °C and transferred to −70 °C within 2 days. No saliva samples were collected from the group consuming capsules because the capsules were ingested without chewing; thus, the probiotics were not released into the oral environment. 2.6. Faecal samples Spot faecal samples were collected on the previous night or in the morning of the specified days (11 samples/participant), and transported to the study freezers or frozen in the home freezer (−20 °C) wherefrom they were transferred to −70 °C within 2 days. 2.7. Analyses of the probiotic strains in saliva and faecal samples Strain-specific, real-time quantitative PCR (qPCR) assays were used to quantify the two Lactobacillus and the Propionibacterium strains. (Ahlroos and Tynkkynen, 2009; Karjalainen et al., 2010). For B. animalis subsp. lactis Bb12 strain, a subspecies-specific primer pair (Ventura et al., 2001) was used in qPCR as described by Myllyluoma et al. (2007). DNA was extracted from faecal samples as described previously (Ahlroos and Tynkkynen, 2009). Standard curves consisted of a series of 10-fold dilutions of target strain genomic DNA, between 0.1 pg and 1000 ng, being equivalent to ca. 38–40 and 3.8 × 108 to 4.0 × 108 target genomes, respectively. The saliva samples (0.1 ml) were first diluted with sterile water then the cells were pelleted by centrifugation. Before starting the intervention, the adequacy of the detection methods for saliva samples and the storage stability of the DNA in frozen saliva samples were tested in a pilot study. 2.8. Statistical analyses The data are expressed as means with standard deviation (SD) or as medians with interquartile ranges (IQR). The most important outcomes are presented with 95% confidence intervals (CI). The statistical

The comparison of the baseline characteristics of the 36 volunteers who completed the study is presented in Table 2. The groups did not differ with respect to sex ratio, age, body mass index, or chronic medications. However, there was a difference (P b 0.05) in their consumption of probiotic products prior to the study. None of the participants smoked, and alcohol was consumed weekly or less frequently. Disinfecting mouth washes were not used by any participant. Diagnosed chronic diseases were reported by 16 participants and included allergy (n = 3), hypertension (n = 4), and elevated serum cholesterol (n = 2). 3.2. Safety, compliance, and gastrointestinal symptoms All blood sample tests for haematology, inflammation, lipid metabolism, and liver functions showed the participants had a good health status, and the intervention did not influence their health (Table 3). The compliance based on diary reports was excellent; only 1 participant forgot to take the probiotic capsules 1 time during the intervention phase. All participants refrained from consuming probiotic products other than the intervention probiotic product during the entire study period. Participants reported any abdominal symptoms (pain, diarrhoea, constipation, and flatulence) in a diary. Table 2 Baseline characteristics of the study volunteers (n = 12 in each group). Product group

M/F Mean age, (SD) y Mean body mass index (SD) kg/m2

Capsule

Yoghurt

Cheese

4/8 43 (12) 24.8 (3.1)

4/8 34 (12) 23.4 (3.6)

3/9 42 (14) 26.0 (4.3)

2 (17) 5 (42) 2 (25) 2 (17)

1 (8) 7 (58) 2 (17) 2 (17)

Consumption of probiotic products, prior to the studya Not at all, n (%) Monthly, n (%) Weekly, n (%) Daily, n (%)

6 (50) 0 (0) 1 (8) 5 (42)

a Significant difference between the groups (P b 0.05) based on Fisher–Freeman– Halton test.

296

M. Saxelin et al. / International Journal of Food Microbiology 144 (2010) 293–300

Table 3 The results of the blood analyses at the beginning and the end of the intervention period. The 4 strains in a probiotic combination (L. rhamnosus GG and LC705, P. freudenreichii subsp. shermanii JS, and B. animalis subsp. lactis Bb12) were administered in 3 product forms; as capsules, yoghurt, and cheese for 2 weeks. Product groups

Capsules (n = 12) Beginning End Yoghurt (n = 12) Beginning End Cheese (n = 12) Beginning End

s-Chol, mmol/L

s-HDL, mmol/L

s-LDL, mmol/L

s-Tg, mmol/L

s-ALAT, U/L

s-ASAT, U/L

s-CRP, mg/L

B-Hb, g/L

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

5.1 (0.7) 4.9 (0.8)

2.1 (0.5) 2.1 (0.5)

2.5 (0.6) 2.4 (0.6)

1.1 (0.5) 1.0 (0.4)

20.8 (8.9) 21.6 (7.8)

25.7 (7.4) 28.0 (12.8)

1.0 (1.5) 1.1 (1.2)

143 (10.4) 144 (11.0)

5.0 (0.9) 4.9 (1.0)

2.0 (0.5) 1.9 (0.4)

2.5 (0.8) 2.5 (0.8)

1.0 (0.5) 1.0 (0.6)

19.8 (7.4) 19.2 (8.4)

25.3 (4.9) 23.9 (5.8)

1.4 (1.9) 1.3 (1.8)

143 (11.7) 142 (11.3)

5.1 (0.8) 5.1 (0.7)

1.9 (0.6) 1.9 (0.8)

2.6 (1.0) 2.7 (0.8)

1.2 (0.7) 1.3 (0.6)

34.9 (27.8) 39.8 (34.7)

31.2 (1.0) 30.9 (10.7)

1.7 (2.7) 2.5 (3.3)

140 (10.9) 138 (8.9)

Reference values: s-Chol, b5; s-HDL, N1; s-LDL, b3; s-Tg, b2; s-ALAT, female 10–45, male 10–70; s-ASAT, female, 15–35, male 15–45; s-CRP, b 10.0; B-Hb, female 117–155, male 134–167.

3.3. Detection limits of the strains Strain-specific detection limits for the L. rhamnosus GG, P. freudenreichii subsp. shermanii JS, and B. animalis subsp. lactis Bb12 were 2.5 × 102 and 1.0 × 104 genome copies/g for saliva and the wet weight of stool samples, respectively, and 2.5 × 103 and 1.0 × 105 genome copies/g for L. rhamnosus LC705 in saliva and stool samples, respectively. 3.4. Recovery of the strains in saliva samples In the yoghurt and cheese groups, the recovery of the strains in saliva samples was analyzed. At the end of the run-in period, 4 of 24 participants (17%) carried L. rhamnosus GG in their saliva, but during the intervention period GG was recovered in 88% of the volunteers at least once (median 4.22. [IQR: 2.10, 4.96], log10 genome copies/g saliva at the end of the intervention). There was no difference between the yoghurt and cheese groups with regard to the quantities of GG measured at the end of the intervention period. Other strains were only occasionally recovered. One participant in the yoghurt group harboured P. freudenreichii subsp. shermanii JS and B. animalis subsp. lactis Bb12, and 1 participant in the cheese group carried the Bb12 strain; L. rhamnosus LC705 was never recovered in saliva. At the beginning of the follow-up period, L. rhamnosus GG was frequently recovered in saliva, and was present in about half of the samples on day 3 (Fig. 1). There was no difference between the yoghurt and cheese groups concerning the follow-up salivary recovery of GG. Three participants had GG in saliva from the end of the run-in period to the end of the follow-up period. Eight persons did not carry GG at any time during the follow-up period. Also, B. animalis subsp. lactis Bb12 was recovered but only on the first day of follow-up, and from the same 2 participants who carried it during the intervention. 3.5. Recovery of the strains in faecal samples during the intervention period and the matrix effect

matrix effect since no difference was detected in the faecal GG quantities between the product groups at the end of the intervention. Before the intervention, L. rhamnosus LC705 was present in 17% of the participants (6/36). The strain was recovered during the intervention in faecal samples of all but 1 participant. The concentration increased during the intervention period to a median 8.67 log10 genome copies/g (IQR = 8.48, 9.05) at the end of the intervention (Fig. 2B). There was no matrix effect since there was no difference between the product groups in the quantities of faecal LC705 at the end of the intervention. P. freudenreichii subsp. shermanii was recovered in faecal samples of 3 participants at the end of run-in period (8%). During the intervention it was recovered in faecal samples from all participants (Fig. 2C), but in 1 person only on day 14. There was a matrix effect since a difference was detected in the quantities of JS strain between the product groups (P b 0.01) at the end of the intervention. The highest faecal JS quantity was obtained from participants in the yoghurt group, and the lowest from participants in the cheese group (median 8.01 log10 [IQR = 7.89, 8.25] vs. 7.24 log10 [IQR = 7.01, 7.43] genome copies/g, respectively; P b 0.01) at the end of the intervention. B. animalis subsp. lactis Bb12 was recovered in 7 (19%) participants at the end of the run-in period and from all the participants during the intervention (Fig. 2D). The product matrix influenced the faecal Bb12 quantity at the end of the intervention period (P b 0.001). The quantity 100 Yoghurt

90

Cheese 80

Survival of GG in saliva, %

No difference was observed between the intervention and follow-up periods in the number of participants with symptoms, or in the duration of symptoms.

70 60 50 40 30 20 10

Before the intervention, L. rhamnosus GG was already present in 36% of the volunteers (14/36), and during the intervention the strain was recovered from faecal samples of all participants in high quantities. The concentration of the strain increased in participants from all the product groups during the intervention period (P b 0.01 for each group), to a median of 8.08 log10 genome copies/g (IQR = 7.78, 8.39) at the end of the intervention period (Fig. 2A). There was no

0 0 1

3

5

7

10

14

17

21

Follow-up, days Fig. 1. Survival of L. rhamnosus GG (%) in saliva samples during the follow-up period after yoghurt and cheese consumption (n = 12 in both groups). The figure shows the percentage share of persons having GG in saliva from all (n = 12) the participants.

M. Saxelin et al. / International Journal of Food Microbiology 144 (2010) 293–300

B

10,0 9,5

Quantity of GG in fecal samples, log10 genome copies/g

10,0

Capsule Yoghurt Cheese

9,0

Quantity of LC705 in fecal samples, log10 genome copies/g

A

8,5 8,0 7,5 7,0 6,5 6,0 5,5 5,0 4,5

Capsule Yoghurt Cheese

9,5 9,0 8,5 8,0 7,5 7,0 6,5 6,0 5,5

5,0 < det.limit

4,0 < det.limit 0

7

14

0

Intervention, days

7

14

Intervention, days

C

D 9,0

11,0

Capsule Yoghurt Cheese

Capsule Yoghurt Cheese

10,5

Quantity of Bb12 in fecal samples, log10 genome copies/g

8,5

Quantity of PJS in fecal samples, log10 genome copies/g

297

8,0 7,5 7,0 6,5 6,0 5,5 5,0 4,5

10,0 9,5 9,0 8,5 8,0 7,5 7,0 6,5 6,0 5,5 5,0 4,5

4,0 < det.limit

4,0 < det.limit 0

7

14

0

Intervention, days

7

14

Intervention, days

Fig. 2. The quantities (log10 genome copies/g) of L. rhamnosus GG (A), L. rhamnosus LC705 (B), P. freudenreichii subsp. shermanii JS (C), and B. animalis subsp. lactis Bb12 (D) in faecal samples taken before (day 0) and during the 2-week intervention period (days 7 and 14) in the 3 product forms; capsule, yoghurt and cheese (n = 12 in each group). Boxes show interquartile range with the median (square); whiskers are the 10th and 90th percentiles; dots represent outliers.

was significantly higher (P b 0.001) in the yoghurt group (median 9.89 log10 genome copies/g [IQR = 9.65, 10.14]) than in the capsule group (median 7.79 log10 genome copies/g [IQR = 7.30, 8.13]), or in the cheese group (median 7.26 log10 genome copies/g [IQR = 7.15, 7.68]). 3.6. Faecal recovery of the strains after ceasing the intervention and the matrix effect After successful recovery during the intervention period the excretion of the strains was followed for up to 21 days after stopping their consumption. There was a difference in the excretion times (P b 0.001) detected during the follow-up period. The longest excretion time was detected for L. rhamnosus GG (median 17 days; 95% CI, 10 to 21 days), followed by B.animalis subsp. lactis Bb12 (median 7 days; 95% CI, 5 to14 days), P. freudenreichii subsp. shermanii JS (median 7 days; 95% CI, 5 to 7 days) and L. rhamnosus LC705 (median 5 days; 95% CI, 3 to 5 days). At the end of the 3-week follow-up period, 28% of participants (10/36) still carried the GG strain but only 3 participants carried Bb12, 1 person carried JS, and none carried the LC705 strain.

The product matrix had no effect on the persistence of L. rhamnosus GG and LC705 (Fig. 3A and B). However, there was a difference in the survival for P. freudenreichii subsp. shermanii JS (P b 0.05, Fig. 3C), depending on the product consumed. The median excretion time was 7 days (95% CI, 5 to 10 days) in the capsule and yoghurt groups, but 5 days (95% CI, 3 to 7 days) in the cheese group. Also, the survival for B. animalis subsp. lactis Bb12 differed between the product forms (P b 0.05, Fig. 3D). The median excretion time was 7 days (95% CI, 5 to 7 days) in the capsule group, 17 days (95% CI, 10 to 17 days) in the yoghurt group, and 3 days (95% CI, 1 to 14 days) in the cheese group. 4. Discussion The study population of the present trial represented generally healthy adults based on participant interviews conducted prior to the study and the results of a blood analyses. There were differences in the consumption of probiotic products prior to the study but these differences did not have an effect on the results because of the long run-in period. The probiotic interventions did not influence the

298

M. Saxelin et al. / International Journal of Food Microbiology 144 (2010) 293–300

A

B 100

Yoghurt Cheese

80 70 60 50 40 30 20

Capsule Yoghurt

90

Survival of LC705 in fecal samples, %

90

Survival of GG in fecal samples, %

100

Capsule

Cheese 80 70 60 50 40 30 20 10

10

0

0 0 1

3

5

7

10

14

17

21

0 1

3

5

Follow-up, days

10

14

17

21

Follow-up, days

C

D 100

100

Capsule Yoghurt Cheese

80 70 60 50 40 30 20

Capsule Yoghurt

90

Survival of Bb12 in fecal samples, %

90

Survival of PJS in fecal samples, %

7

Cheese 80 70 60 50 40 30 20 10

10

0

0 0 1

3

5

7

10

14

17

21

Follow-up, days

0 1

3

5

7

10

14

17

21

Follow-up, days

Fig. 3. Survival of L. rhamnosus GG (A), L. rhamnosus LC705 (B), P. freudenreichii subsp. shermanii JS (C), and B. animalis subsp. lactis Bb12 (D) strains in faecal samples during the follow-up period. The strains were administered in 3 product forms; capsule, yoghurt, and cheese (n = 12 in each group). The figure shows the percentage share of persons having GG in faeces from all (n = 12) the participants. The administration matrix influenced the duration of the excretion times of PJS and Bb12 (P b 0.05 for both).

clinical chemistry values of the serum samples, which supports the safety of the intervention. The first contact probiotics consumed in food have with the human mucosa takes place in the oral cavity. A few studies have shown that probiotics can reduce the risk of dental decay and relieve gingivitis (Krasse et al., 2006; Meurman and Stamatova, 2007). These mechanisms of the effects are not known, but the adhesion capacity and persistence on the oral mucosa may be important. In the present study, the presence of all 4 administered strains was analyzed in saliva samples. L. rhamnosus GG was the only strain that was frequently recovered. Interestingly, it was shown earlier to reduce the risk of dental caries (Näse et al., 2001). Also, both L. rhamnosus GG and B. animalis subsp. lactis Bb12 were previously reported to adhere to the parotid-salivacoated hydroxyapatite and to reduce the adhesion of Streptococcus mutans. The suggested mechanism was based on the binding and degradation of salivary agglutinin gp340, which is the receptor for S.

mutans adhesion (Haukioja et al., 2008). Together, these results indicate that GG may have a greater potential to affect the oral environment compared with the other 3 strains assessed in the study. Several aspects of probiotics can enhance gastrointestinal persistence. For Lactobacillus crispatus, an aggregating phenotype is important to maintain colonization capacity since a non-aggregative mutant was not recovered in faecal samples or in colon biopsy specimens (Cesena et al., 2001; Voltan et al., 2007). For L. rhamnosus GG a gene cluster, spaCBA, was shown to determine the adhesion capacity to human intestinal mucus. L. rhamnosus LC705 missed the gene cluster, and is poorly adhesive (Kankainen et al., 2009; von Ossowski et al., 2010). During the intervention period of the present study, the faecal recovery frequency of every strain was nearly or equal to 100% in all the product groups. This is in agreement with earlier studies where the same strains were recovered in all stool samples during the administration of about 1010 cfu/day in capsules

M. Saxelin et al. / International Journal of Food Microbiology 144 (2010) 293–300

or about 1011 cfu/day in a dairy-based drink (Kajander et al., 2005; Kekkonen et al., 2008; Myllyluoma et al., 2005). In previous studies, the frequency of faecal recovery of the B. animalis subsp. lactis Bb12 strain was less than that found in the present study, with 87% and 79% recovery in volunteers who consumed a daily dose of 1011 cfu in capsules or yoghurts, respectively (Larsen et al., 2006; Mättö et al., 2006), but in only half of those consuming 1010 cfu in capsules (Larsen et al., 2006). Recovery frequency was 100% in our study. The discrepancy between the previous and the present results can be explained by the analytical differences. Generally, strain-specific real-time qPCR methods give 1–2 log higher genome copy numbers/g faeces compared with culturing methods (Ahlroos and Tynkkynen, 2009; Dommels et al., 2009; Fujimoto et al., 2008). Although the qPCR-based methods may also count dormant and dead bacteria, it is recognized that inactivated bacteria may also have healthpromoting effects (Gill and Rutherfurd, 2001; Kaila et al., 1995; Lopez et al., 2008). The quantity of faecal probiotics may be relevant for their health-promoting effects, e.g. the quantity of faecal B. animalis subsp. lactis Bb12 was inversely correlated with the production of interferon-γ (Christensen et al., 2006). The effect of food matrix on the intestinal survival of probiotic bacteria is insufficiently studied. Rochet et al. (2008) did not see any difference in the faecal level of B. animalis strain, when 6 × 1010– 2 × 1011 cfu were administered in fermented milk or in freeze-dried form, but the food matrix improved significantly the survival of L. plantarum MF1298 (Klingberg and Budde, 2006) and L. rhamnosus GG (Saxelin et al., 1991; Saxelin et al., 1993), when lower doses (6 × 109 cfu and 1–2 × 109 cfu, respectively) were used. Fresh dairy products are the most common product forms of probiotic delivery, but ripened cheeses have also been successfully tested as a carrier matrix (Heller, 2001; Saxelin, 2008). In the present study, the product matrix, i.e. capsules, yoghurt, or cheese, did not influence the intestinal survival of the lactobacilli strains; both L. rhamnosus GG and LC705 concentrations were as high (N108 genome copies/g faeces) at the end of the intervention period. Also, the P. freudenreichii subsp. shermanii JS survived well in the gastrointestinal environment. Surprisingly, the faecal quantity of JS was lower when consumed in cheese, the natural habitat of the strain, although even this lower level of faecal propionibacteria (N107 genome copies/g) may be considered rather high. Also, there was a significant difference between the product forms concerning faecal B. animalis subsp. lactis Bb12. The highest quantities were recovered in the samples from the yoghurt group, followed by the capsule group, and finally the cheese group (N7 × 109, N6 × 107, and N1 × 107 genome copies/g faeces, respectively). Bifidobacteria retain their viability better in fermented milk, which partially protects against oxygen stress but also supports the growth of fastidious bifidobacteria by supplying a growth substrate and by helping them adapt to acidic conditions. Milk also buffers gastric acidity (Petschow and Talbott, 1990). In the present study, the original daily dose of Bb12 strain in cheese was remarkably lower compared with capsules and yoghurt, and exposure to oxygen reduced viability. Thus, the lower counts of Bb12 in stool samples from cheese consumers can be explained by the lower daily dose consumed compared with the other product forms. The faecal recovery time of the 4 probiotic strains varied considerably during the follow-up period. The longest excretion time was for L. rhamnosus GG (median 17 days) followed by B. animalis subsp. lactis Bb12 and P. freudenreichii subsp. shermanii JS (median 7 days, each). L. rhamnosus LC705 disappeared sooner (median 5 days). When the results are compared with the in vitro mucus adhesion of these strains, it seemed that the more adherent the strain the longer the excretion time. L. rhamnosus GG is strongly adherent to human intestinal mucus in vitro followed by Bb12, but JS and LC705 are only weakly adherent (Collado et al., 2006; Ouwehand et al., 2000). Furthermore, L. rhamnosus GG, but not LC705, was recently shown to have pili structures protruding from the cell wall

299

that determine the adhesion capacity on human intestinal mucus (Kankainen et al., 2009). Although the analytical detection limit of LC705 strain in samples was 1 log10 higher and may have influenced the survival results of the strain, the poorer adhesive capacity to the intestinal lining is a more relevant explanation for the short excretion time of this strain. Interestingly, B. animalis subsp. lactis Bb12 was excreted for a longer time when it was consumed in yoghurt compared with the other two product forms. An earlier in vitro study showed that a yoghurt-starter strain of Lactobacillus delbrückii ssp. bulgaricus in addition to L. rhamnosus GG enhanced the adhesion of B. animalis subsp. lactis Bb12 to the intestinal mucus (Ouwehand et al., 2000) and a similar phenomenon may have influenced the results of this study. It is also known that propionibacteria are bifidogenic, and the similar excretion times for strains JS and Bb12 in faeces may indicate that they support each other's growth and adhesion. Although the daily dose of Bb12 was a little higher in yoghurt than in other product forms, it was not supposed to prolong the excretion time. In practice, the excretion time of probiotics is not very relevant because the products are generally consumed daily or at least several times a week. However, differing strengths of colonization of probiotic strains was demonstrated, and the results indicate that the transient colonization of L. rhamnosus GG can last several weeks. It is worth of reminding, however, that the matrix effect in this study was demonstrated with a daily dose of 1010 cfu, and it is not valid for considerably lower doses. The results may differ with low doses that hardly yield counts above the detection limit of a strain in faecal samples, but any exact data for this is not available. To summarize, all 3 product forms (capsules, yoghurt and cheese) were good vehicles for the administration of the 4 strains (L. rhamnosus GG and LC705, P. freudenreichii subsp. shermanii JS and B. animalis subssp. lactis Bb12) in a probiotic combination. Consumption of probiotics in a yoghurt matrix yielded high quantities of all the strains in faecal samples, and capsules provided a comparable means of administration. This important finding supports the idea that the results of clinical studies performed using probiotics in capsule/ powder form can be extrapolated to fermented milk products, provided the probiotic remains viable in such products and the dose is high enough. When using ripened cheese as a vehicle, protecting anaerobic strains against atmospheric oxygen is vital to maintain their viability. The time of faecal excretion varied considerably depending on the strain, and in some cases on the product matrix.

Acknowledgments We wish to thank the study participants for their important contributions. Ms. Lea Tuulio is acknowledged for the preparation and microbial analysis of yoghurt; Jarna Tanskanen, PhD, and Ms Pirkko Piispanen for the microbial analyses of cheese; and Anna-Maija Ahonen, M.Sc., for providing and analyzing the capsules. Valio Ltd is acknowledged for the financial support of the study.

References Ahlroos, T., Tynkkynen, S., 2009. Quantitative strain-specific detection of Lactobacillus rhamnosus GG in human faecal samples by real-time PCR. Journal of Applied Microbiology 106, 506–514. Alander, M., Satokari, R., Korpela, R., Saxelin, M., Vilpponen-Salmela, T., Mattila-Sandholm, T., von Wright, A., 1999. Persistence of colonization of human colonic mucosa by a probiotic strain, Lactobacillus rhamnosus GG, after oral consumption. Applied and Environmental Microbiology 65, 351–354. Cesena, C., Morelli, L., Alander, M., Siljander, T., Tuomola, E., Salminen, S., MattilaSandholm, T., Vilpponen-Salmela, T., von Wright, A., 2001. Lactobacillus crispatus and its nonaggregating mutant in human colonization trials. Journal of Dairy Science 84, 1001–1010. Christensen, H.R., Larsen, C.N., Kaestel, P., Rosholm, L.B., Sternberg, C., Michaelsen, K.F., Frokiaer, H., 2006. Immunomodulating potential of supplementation with probiotics: a dose-response study in healthy young adults. FEMS Immunology and Medical Microbiology 47, 380–390.

300

M. Saxelin et al. / International Journal of Food Microbiology 144 (2010) 293–300

Collado, M.C., Jalonen, L., Meriluoto, J., Salminen, S., 2006. Protection mechanism of probiotic combination against human pathogens: in vitro adhesion to human intestinal mucus. Asia Pacific Journal of Clinical Nutrition 15, 570–575. De Man, J.C., Rogosa, M., Sharpe, M.E., 1960. A medium for the cultivation of lactobacilli. Journal of Applied Bacteriology 23, 130–135. Dommels, Y.E., Kemperman, R.A., Zebregs, Y.E., Draaisma, R.B., Jol, A., Wolvers, D.A., Vaughan, E.E., Albers, R., 2009. Survival of Lactobacillus reuteri DSM 17938 and Lactobacillus rhamnosus GG in the human gastrointestinal tract with daily consumption of a low-fat probiotic spread. Applied and Environmental Microbiology 75, 6198–6204. Fujimoto, J., Matsuki, T., Sasamoto, M., Tomii, Y., Watanabe, K., 2008. Identification and quantification of Lactobacillus casei strain Shirota in human feces with strainspecific primers derived from randomly amplified polymorphic DNA. International Journal of Food Microbiology 126, 210–215. Ghadimi, D., Folster-Holst, R., de Vrese, M., Winkler, P., Heller, K.J., Schrezenmeir, J., 2008. Effects of probiotic bacteria and their genomic DNA on TH1/TH2-cytokine production by peripheral blood mononuclear cells (PBMCs) of healthy and allergic subjects. Immunobiology 213, 677–692. Gill, H.S., Rutherfurd, K.J., 2001. Viability and dose-response studies on the effects of the immunoenhancing lactic acid bacterium Lactobacillus rhamnosus in mice. British Journal of Nutrition 86, 285–289. Haarman, M., Knol, J., 2006. Quantitative real-time PCR analysis of fecal Lactobacillus species in infants receiving a prebiotic infant formula. Applied and Environmental Microbiology 72, 2359–2365. Hartemink, R., Kok, B.J., Weenk, G.H., Rombouts, F.M., 1996. Raffinose-Bifidobacterium (RB) agar, a new selective medium for bifidobacteria. Journal of Microbiological Methods 27, 33–43. Hatakka, K., Savilahti, E., Pönkä, A., Meurman, J.H., Poussa, T., Näse, L., Saxelin, M., Korpela, R., 2001. Effect of long term consumption of probiotic milk on infections in children attending day care centres: double blind, randomised trial. British Medical Journal 322 (7298), 1327. Haukioja, A., Loimaranta, V., Tenovuo, J., 2008. Probiotic bacteria affect the composition of salivary pellicle and streptococcal adhesion in vitro. Oral Microbiology and Immunology 23, 336–343. Heller, K.J., 2001. Probiotic bacteria in fermented foods: product characteristics and starter organisms. American Journal of Clinical Nutrition 73 (2 Suppl), 374S–379S. Kaila, M., Isolauri, E., Saxelin, M., Arvilommi, H., Vesikari, T., 1995. Viable versus inactivated Lactobacillus strain GG in acute rotavirus diarrhoea. Archives of Diseases in Childhood 72, 51–53. Kajander, K., Hatakka, K., Poussa, T., Färkkilä, M., Korpela, R., 2005. A probiotic mixture alleviates symptoms in irritable bowel syndrome patients: a controlled 6-month intervention. Alimentary and Pharmacological Therapy 22, 387–394. Kajander, K., Myllyluoma, E., Rajilic-Stojanovic, M., Kyrönpalo, S., Rasmussen, M., Järvenpää, S., Zoetendal, E.G., de Vos, W.M., Vapaatalo, H., Korpela, R., 2008. Clinical trial: multispecies probiotic supplementation alleviates the symptoms of irritable bowel syndrome and stabilizes intestinal microbiota. Alimentary and Pharmacological Therapy 27, 48–57. Kankainen, M., Paulin, L., Tynkkynen, S., von Ossowski, I., Reunanen, J., Partanen, P., Satokari, R., Vesterlund, S., Hendrickx, A.P., Lebeer, S., De Keersmaecker, S.C., Vanderleyden, J., Hämäläinen, T., Laukkanen, S., Salovuori, N., Ritari, J., Alatalo, E., Korpela, R., Mattila-Sandholm, T., Lassig, A., Hatakka, K., Kinnunen, K.T., Karjalainen, H., Saxelin, M., Laakso, K., Surakka, A., Palva, A., Salusjärvi, T., Auvinen, P., de Vos, W.M., 2009. Comparative genomic analysis of Lactobacillus rhamnosus GG reveals pili containing a human-mucus binding protein. Proceedings of National Academy of Sciences of the United States of America 106, 17193–17198. Karjalainen, H., Ahlroos, T., Myllyluoma, E., Tynkkynen, S., 2010. Real-time PCR assays for strain-specific quantification of probiotic strains in human faecal samples. Unpublished. Kekkonen, R.A., Lummela, N., Karjalainen, H., Latvala, S., Tynkkynen, S., Järvenpää, S., Kautiainen, H., Julkunen, I., Vapaatalo, H., Korpela, R., 2008. Probiotic intervention has strain-specific anti-inflammatory effects in healthy adults. World Journal of Gastroenterology 14, 2029–2036. Klingberg, T.D., Budde, B.B., 2006. The survival and persistence in the human gastrointestinal tract of five potential probiotic lactobacilli consumed as freezedried cultures or as probiotic sausage. International Journal of Food Microbiology 109, 157–159. Krasse, P., Carlsson, B., Dahl, C., Paulsson, A., Nilsson, A., Sinkiewicz, G., 2006. Decreased gum bleeding and reduced gingivitis by the probiotic Lactobacillus reuteri. Swedish Dental Journal 30, 55–60. Lahtinen, S.J., Gueimonde, M., Ouwehand, A.C., Reinikainen, J.P., Salminen, S.J., 2006. Comparison of four methods to enumerate probiotic bifidobacteria in a fermented food product. Food Microbiology 23, 571–577. Larsen, C.N., Nielsen, S., Kaestel, P., Brockmann, E., Bennedsen, M., Christensen, H.R., Eskesen, D.C., Jacobsen, B.L., Michaelsen, K.F., 2006. Dose-response study of probiotic bacteria Bifidobacterium animalis subsp lactis BB-12 and Lactobacillus paracasei subsp paracasei CRL-341 in healthy young adults. European Journal of Clinical Nutrition 60, 1284–1293. Lievin-Le Moal, V., Sarrazin-Davila, L.E., Servin, A.L., 2007. An experimental study and a randomized, double-blind, placebo-controlled clinical trial to evaluate the antisecretory activity of Lactobacillus acidophilus strain LB against nonrotavirus diarrhea. Pediatrics 120, e795–803. Lievin-Le Moal, V., Servin, A.L., 2006. The front line of enteric host defense against unwelcome intrusion of harmful microorganisms: mucins, antimicrobial peptides, and microbiota. Clinical Microbiology Reviews 19, 315–337.

Lopez, M., Li, N., Kataria, J., Russell, M., Neu, J., 2008. Live and ultraviolet-inactivated Lactobacillus rhamnosus GG decrease flagellin-induced interleukin-8 production in Caco-2 cells. Journal of Nutrition 138, 2264–2268. Maruo, T., Sakamoto, M., Toda, T., Benno, Y., 2006. Monitoring the cell number of Lactococcus lactis subsp. cremoris FC in human feces by real-time PCR with strainspecific primers designed using the RAPD technique. International Journal of Food Microbiology 110, 69–76. Matsuki, T., Watanabe, K., Fujimoto, J., Kado, Y., Takada, T., Matsumoto, K., Tanaka, R., 2004. Quantitative PCR with 16S rRNA-gene-targeted species-specific primers for analysis of human intestinal bifidobacteria. Applied and Environmental Microbiology 70, 167–173. Meurman, J.H., Stamatova, I., 2007. Probiotics: contributions to oral health. Oral Diseases 13, 443–451. Myllyluoma, E., Kajander, K., Mikkola, H., Kyrönpalo, S., Rasmussen, M., Kankuri, E., Sipponen, P., Vapaatalo, H., Korpela, R., 2007. Probiotic intervention decreases serum gastrin-17 in Helicobacter pylori infection. Digestive and Liver Diseases 39, 516–523. Myllyluoma, E., Veijola, L., Ahlroos, T., Tynkkynen, S., Kankuri, E., Vapaatalo, H., Rautelin, H., Korpela, R., 2005. Probiotic supplementation improves tolerance to Helicobacter pylori eradication therapy — a placebo-controlled, double-blind randomized pilot study. Alimentary and Pharmacological Therapy 21, 1263–1272. Mättö, J., Fonden, R., Tolvanen, T., von Wright, A., Vilpponen-Salmela, T., Satokari, R., Saarela, M., 2006. Intestinal survival and persistence of probiotic Lactobacillus and Bifidobacterium strains administered in triple-strain yogurt. International Dairy Journal 16, 1174–1180. Näse, L., Hatakka, K., Savilahti, E., Saxelin, M., Pönkä, A., Poussa, T., Korpela, R., Meurman, J.H., 2001. Effect of long-term consumption of a probiotic bacterium, Lactobacillus rhamnosus GG, in milk on dental caries and caries risk in children. Caries Research 35, 412–420. Ouwehand, A.C., Isolauri, E., Kirjavainen, P.V., Tölkkö, S., Salminen, S.J., 2000. The mucus binding of Bifidobacterium lactis Bb12 is enhanced in the presence of Lactobacillus GG and Lact. delbrueckii subsp. bulgaricus. Letters in Applied Microbiology 30, 10–13. Petschow, B.W., Talbott, R.D., 1990. Growth promotion of Bifidobacterium species by whey and casein fractions from human and bovine milk. Journal of Clinical Microbiology 28, 287–292. Requena, T., Burton, J., Matsuki, T., Munro, K., Simon, M.A., Tanaka, R., Watanabe, K., Tannock, G.W., 2002. Identification, detection and enumeration of human Bifidobacterium species by PCR targeting the transaldolase gene. Applied and Environmental Microbiology 68, 2420–2427. Rinttilä, T., Kassinen, A., Malinen, E., Krogius, L., Palva, A., 2004. Development of an extensive set of 16S rDNA-targeted primers for quantification of pathogenic and indigenous bacteria in faecal samples by real-time PCR. Journal of Applied Microbiology 97, 1166–1177. Rochet, V., Rigottier-Gois, L., Ledaire, A., Andrieux, C., Sutren, M., Rabot, S., Mogenet, A., Bresson, J.-L., Cools, S., Picard, C., Goupil-Feuillerat, N., Doré, J., 2008. Survival of Bifidobacterium animalis DN-173 010 in the faecal microbiota after administration in lyophilised form or in fermented product — a randomised study in healthy adults. Journal of Molecular Microbiology and Biotechnology 14, 128–136. Saxelin, M., 2008. Probiotic formulations and applications, the current probiotics market, and changes in the marketplace: a European perspective. Clinical Infectious Diseases 46 (Suppl 2), S76–79. Saxelin, M., Ahokas, M., Salminen, S., 1993. Dose response on the faecal colonization of Lactobacillus strain GG administered in two different formulations. Microbial Ecology in Health and Disease 6, 119–122. Saxelin, M., Elo, S., Salminen, S., Vapaatalo, H., 1991. Dose response colonization of faeces after oral administration of Lactobacillus casei strain GG. Microbial Ecology in Health and Disease 4, 209–214. Saxelin, M., Tynkkynen, S., Salusjärvi, T., Kajander, K., Myllyluoma, E., Mattila-Sandholm, T., Korpela, R., in press. Developing a multispecies probiotic combination. International Journal of Probiotics and Prebiotics. Saxelin, M., Pessi, T., Salminen, S., 1995. Fecal recovery following oral administration of Lactobacillus strain GG (ATCC 53103) in gelatine capsules to healthy volunteers. International Journal of Food Microbiology 25, 199–203. Sharp, M.D., McMahon, D.J., Broadbent, J.R., 2008. Comparative evaluation of yogurt and low-fat cheddar cheese as delivery media for probiotic Lactobacillus casei. Journal of Food Science 73, M375–377. Ventura, M., Reniero, R., Zink, R., 2001. Specific identification and targeted characterization of Bifidobacterium lactis from different environmental isolates by a combined multiplex-PCR approach. Applied and Environmental Microbiology 67, 2760–2765. Voltan, S., Castagliuolo, I., Elli, M., Longo, S., Brun, P., D'Inca, R., Porzionato, A., Macchi, V., Palu, G., Sturniolo, G.C., Morelli, L., Martines, D., 2007. Aggregating phenotype in Lactobacillus crispatus determines intestinal colonization and TLR2 and TLR4 modulation in murine colonic mucosa. Clinical and Vaccine Immunology 14, 1138–1148. von Ossowski, I., Reunanen, J., Satokari, R., Vesterlund, S., Kankainen, M., Huhtinen, H., Tynkkynen, S., Salminen, S., de Vos, W.M., Palva, A., 2010. Mucosal adhesion properties of the probiotic Lactobacillus rhamnosus GG SpaCBA and SpaFED pilin subunits. Applied and Environmental Microbiology 76, 2049–2057. Yli-Knuuttila, H., Snäll, J., Kari, K., Meurman, J.H., 2006. Colonization of Lactobacillus rhamnosus GG in the oral cavity. Oral Microbiology and Immunology 21, 129–131.