Lactobacillus paracasei A survives gastrointestinal passage and affects the fecal microbiota of healthy infants

Research in Microbiology 157 (2006) 857–866 www.elsevier.com/locate/resmic

Lactobacillus paracasei A survives gastrointestinal passage and affects the fecal microbiota of healthy infants Marta Marzotto a , Claudio Maffeis b , Thomas Paternoster a , Rossano Ferrario c , Lucia Rizzotti a , Maristella Pellegrino b , Franco Dellaglio a , Sandra Torriani a,∗ a Department of Science and Technology, University of Verona, Strada Le Grazie, 15, 37134 Verona, Italy b Department of Mother & Child, Biology-Genetics, Pediatrics Unit, University of Verona, Piazzale L. A. Scuro, 10, 37134 Verona, Italy c PLADA Industriale S.r.l., via Donizzetti, Monguzzo, Como, Italy

Received 22 December 2005; accepted 28 June 2006 Available online 2 August 2006

Abstract This study focuses on the potentiality of a putative probiotic strain, Lactobacillus paracasei A, to survive gastrointestinal (GI) passage and modulate the resident microbiota of healthy infants. In a placebo-controlled study, 26 children aged 12–24 months received 100 g/day of either fermented milk containing strain A or pasteurized yogurt for four weeks. Fecal samples were analyzed before starting the administration, after 1, 3 and 4 weeks of consumption and after washout. The fate of strain A was followed by means of a newly developed PCR targeting a strain-specific genomic marker. The composition and dynamics of fecal microbial communities during the study were analyzed by culturing on selective media and by the PCR-denaturing gradient gel electrophoresis (DGGE) technique using universal and group-specific (Lactobacillus and Bifidobacterium) primers. The variation in enzymatic activities in infant feces during probiotic consumption was also analyzed. Strain A survived in fecal samples in most (92%) of the infants examined after 1 week of consumption, and temporarily dominated the intestinal Lactobacillus community. The administration of L. paracasei A led to a significant increment in the Lactobacillus population, while a moderate effect upon the main bacterial groups in the GI ecosystem was observed. Strain A also affected the diversity of the Lactobacillus and Bifidobacterium populations. The fecal bacterial structure of 1–2-year-old infants seems to combine neonate and adult-like features. The microbiota of these subjects promptly responded to probiotic consumption, later restoring the endogenous equilibrium. © 2006 Elsevier Masson SAS. All rights reserved. Keywords: Lactobacillus; Probiotic strains; Fermented milk; Infants; Gastrointestinal content; PCR-DGGE

1. Introduction Yogurt and fermented milk enriched with probiotic bacteria are commonly consumed baby foods by 6 months of age. Probiotic bacteria may provide protective effects to the host through modulation of the immune response and endogenous microbiota of the gut; these actions appear strictly correlated, especially in infants [15,23]. The balancing action of probiotics upon the intestinal microbiota involves an augmentation in bacterial components, especially lactobacilli and bifidobacteria, that may be beneficial to the host, and a reduction in poten* Corresponding author.

E-mail address: [email protected] (S. Torriani). 0923-2508/$ – see front matter © 2006 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.resmic.2006.06.007

tially harmful microorganisms such as coliforms and clostridia, thus reducing the risk of diarrhea, inflammatory and allergic diseases [11]. A number of studies have shown that the microbial composition of the newborn intestinal ecosystem quickly evolves during the first 12 months of life, but remains unstable for a longer period, until it resembles that of adults by about 24 months [9,18]. Hence it can be supposed that microbiota of infants aged 12–24 months may be more easily influenced by intrinsic and environmental factors than that of older children or adults. In particular, infants of this age usually attend day care centers and are potentially more frequently exposed to the risk of bacterial infections. Remarkably, only a limited number of investigations have addressed the influence of probiotic consumption on the micro-

858

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

biota composition of healthy infants of this age [12,31]. Some studies have applied molecular techniques, such as polymerase chain reaction (PCR) coupled with denaturing gradient gel electrophoresis (DGGE) for typing the composition of fecal and intestinal microbiota, but only a few subjects were analyzed [14,26]. Many intervention and ecological studies have focused on the microbial intestinal composition of allergic infants [4,16] or on subjects affected by gastrointestinal (GI) pathologies, e.g. rotavirus diarrhea [10,27]. Several studies documented the protective effect of probiotic consumption on infants attending day care centers [13,24,28]. However, such studies were aimed at analyzing the incidence and severity of diarrhea or other diseases, and not at investigating the interaction between probiotics and the intestinal microbiota. Moreover, the viability of probiotic strains in the infant gut and their temporary colonization capacities were not specifically investigated. The properties and human health effects of different bacterial strains must be assessed in a case-by-case manner, since each may have unique properties, e.g. differing areas of adherence (site-specific), specific immunological effects, etc., and interactions with the host or with intestinal microbiota may be distinct from each other. Bertazzoni Minelli et al. [2] demonstrated that Lactobacillus paracasei A shows high resistance to GI stresses in vitro and good adhesion capacity to the intestinal cell line Caco2. Moreover, oral administration to rat models of a fermented milk product containing strain A induced stimulation of the intestinal Lactobacillus community, an increment in the activity of glycolytic enzymes in feces and an enhancement of some blood biochemical parameters correlated with hepatic functions. The aim of the present report was to further evaluate the probiotic potential of L. paracasei A in humans. In a randomized placebo-controlled study, healthy infants aged 12–24 months from daycare centers received a fermented milk product containing strain A in a 4-week intervention period. Interest was focused on two aspects: (i) the potentiality of strain A to survive GI passage; and (ii) its ability to modulate the host resident fecal microbiota. Since a specific and reliable identification method was required to prove the survival capacity of ingested L. paracasei A, an assay based on PCR was specifically set up to detect and identify such a strain. A combination of traditional bacteriological and biochemical methods and advanced molecular approaches was applied to perform a more comprehensive analysis of the composition, dynamics and activity of the resident fecal microbiota of healthy infants. Particular attention was paid to monitoring diversity and changes in Lactobacillus and Bifidobacterium communities which may have been due to an interaction with the ingested strain.

lus acidophilus DSM 20079T , Lactobacillus amylovorus DSM 20531T , Lactobacillus delbrueckii subsp. bulgaricus ATCC 11842T , Lactobacillus brevis LMG 7944T , L. paracasei NCDO 151T , Lactobacillus casei ATCC 334, L. casei ATCC 393T , Lactobacillus crispatus LMG 9479T , Lactobacillus curvatus LMG 9198T , Lactobacillus fermentum LMG 6902T , Lactobacillus fuchuensis DSM 14340T , Lactobacillus gasseri LMG 9203T , Lactobacillus helveticus ATCC 15009T , Lactobacillus jensenii LMG 64145T , Lactobacillus johnsonii LMG 9436T , Lactobacillus plantarum ATCC 14917T , Lactobacillus reuteri LMG 9213T , Lactobacillus rhamnosus LMG 8400T , Lactobacillus ruminis LMG 10756T , Lactobacillus sakei subsp. sakei LMG 9468T and Lactobacillus salivarius LMG 9477T . An additional 39 L. paracasei strains of different origin belonged to the culture collection of the Department of Science and Technology, University of Verona. All Lactobacillus strains were grown in MRS broth (Oxoid, Milan, Italy) at 37 ◦ C for 30 h in anaerobiosis obtained with the Anaerocult A system (Merck, Darmstadt, Germany). For Bifidobacterium, the following culture collection strains were used: Bifidobacterium breve LMG 11042T , Bifidobacterium longum biotype infantis LMG 11046T , Bifidobacterium longum biotype longum LMG 10547T , Bifidobacterium bifidum LMG 11041T , Bifidobacterium animalis subsp. lactis DSM 10140T , Bifidobacterium adolescentis DSM 20083T , Bifidobacterium animalis subsp. animalis LMG 10508T , Bifidobacterium pseudocatenulatum LMG 10505T , Bifidobacterium catenulatum LMG 11043T , Bifidobacterium angulatum LMG 11039T and Bifidobacterium dentium LMG 11045T . These strains were grown anaerobically in MRS broth with 0.05% L-cysteine hydrochloride at 37 ◦ C for 48 h. 2.2. Test and control products

2. Materials and methods

The fermented milk products used in this study were manufactured from partially skimmed milk inoculated with a protosymbiotic culture of L. delbrueckii subsp. bulgaricus and Streptococcus thermophilus at the Plada Industriale S.r.l. dairy plant. The probiotic product was prepared blending this classic yogurt to standardized milk fermented with strain A at a final ratio of 1:10, w/w. The standardized milk contained 93.5% (w/w) whole concentrated milk, 1.5% (w/w) milk proteins and 3% (w/w) sucrose, and fermentation by L. paracasei A occurred for 8 h at 37 ◦ C. Pasteurized classic yogurt was used as a control. The tested fermented milk product contained L. delbrueckii subsp. bulgaricus, S. thermophilus and L. paracasei A at a concentration of approximately 7.7 log10 CFU g−1 , 8.6 log10 CFU g−1 and 8.2 log10 CFU g−1 , respectively, while the control yogurt did not contain viable cells.

2.1. Bacterial strains and growth conditions

2.3. Subjects and study design

L. paracasei A was previously described by Bertazzoni Minelli et al. [2]. Reference Lactobacillus strains obtained from international culture collections were as follows: Lactobacil-

Twenty-six healthy infants aged 12–24 months from four daycare centers were included in this study. They were enrolled in accordance with criteria of good health and absence of

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

medical treatments. A feedback form daily assessing the consumption of fermented milk products was completed for each child; the questionnaire also included observations and other problems encountered during the study period. Infants were randomly allocated to two groups: L. paracasei (or probiotic) group A (n = 13) and control (or placebo) group P (n = 13) which received a daily dose of 100 g of either the probiotic product or the pasteurized yogurt. The subjects of the same day nursery were assigned to the same experimental group. The study was carried out in a blind fashion. The subjects received the products for four weeks and this period was followed by a week of washout. The study was approved by the Ethical Committee of the University Hospital of Verona (Italy) and had the informed consent of the infants’s parents.

859

randomly chosen from both groups by means of the API ZYM System (Biomérieux SA), as described by Zoppi et al. [41]. Enzymatic activities were assessed by comparative visual analysis, scoring on a colorimetric scale from 1 to 5 (most strongly colored). 2.5. DNA extraction from pure cultures and fecal samples Total genomic DNA was purified from 1 ml pure cultures of isolates and reference strains as reported by Marmur [19]. Total bacterial DNA from fecal samples was extracted as described by Dal Bello et al. [7]. Nucleic acids were obtained from 500 µl of fecal homogenate and resuspended in 50 µl of TE buffer (10 mM Tris, 1 mM EDTA, pH 8.0). DNA concentration was determined by a fluorometric assay (Versa Fluor Fluorometer, Biorad, Italy).

2.4. Microbiological and chemical analysis of fecal samples 2.6. Identification of Lactobacillus isolates Fecal samples were collected in sterile tubes, kept under anaerobiosis before starting administration (T0), at the end of the first (T1), third (T3) and fourth week (T4) of administration and 1 week after the supplementation period (T5). Processing of samples was carried out within 2 h after collection as follows: once diluted (100 g l−1 ) in pre-reduced physiological solution (0.9%, w/v, NaCl, 0.05% L-cysteine hydrochloride), each sample was homogenized with a hand-held glass homogenizer and filtered through a porous membrane (∅ = 0.5 mm). Each homogenate was divided into two aliquots: one was immediately analyzed, whilst the second was maintained at −20 ◦ C for molecular analysis. Serial dilutions and plating procedures were performed in an anaerobic box [2]. Counts of the main fecal bacterial groups were determined using the following selective agar media provided by Oxoid (Hampshire, UK): RCM and Rogosa (bifidobacteria), Slanetz-Bartley (enterococci), Schaedler plus 10 mg l−1 kanamycin and 100 mg l−1 vancomycin (Bacteroides), RCM (clostridia), MacConkey (enterobacteria), Schaedler (total anaerobes) and BHI (total aerobes). Incubation conditions were according to Oxoid instructions. Identification of colonies grown on selective media was confirmed by morphological observation and phenotypic characterization of representative isolates, randomly selected from plates using the API tests (Biomérieux SA, Merci l’Etoile, France) (API 20 E for enterobacteria and Gram-negative bacteria; rapid ID 32 A for total anaerobes; rapid ID 32 STREP for enterococci and streptococci) and following the manufacturer’s instructions. Counts of fecal lactobacilli were determined on MRS–Van (MRS plus 25 mg l−1 vancomycin HCl) agar plates after 30 h incubation at 37 ◦ C in anaerobic environment. Microbial counts were expressed as log10 CFU g−1 wet feces. Student’s t test was used for statistical analysis of microbial counts, comparing both the probiotic-treated group versus the control group and the times of sampling versus the times 0. The statistical significance was fixed at P < 0.01 or P < 0.05. Measurements of enzymatic activities were carried out at each time of sampling on the fecal suspension of 8 subjects

Representative colonies of lactobacilli were picked from MRS–Van agar plates and cultured in MRS broth for 48 h at 37 ◦ C. Identification of isolates was achieved by means of PCRbased assays. RAPD-PCR analysis was performed with the primer D8635 according to Akopyanz et al. [1]. The electrophoretic patterns were analyzed using the Pearson product moment correlation coefficient and the unweighted pair group with mathematical average clustering algorithm (UPGMA) through the GelCompar software package (version 4.0, Applied Maths, Kortijk, Belgium). In order to substantiate RAPD-PCR results, species-specific PCR protocols previously published were used for identification of the following bacterial species: L. paracasei/casei, L. gasseri, L. johnsonii, L. acidophilus [37], L. plantarum [35], L. reuteri, L. salivarius, L. fermentum [5], L. rhamnosus [33], L. curvatus, L. sakei [3], L. delbrueckii subsp. bulgaricus [34]. 2.7. Development of a PCR assay specific for L. paracasei A To identify molecular markers specific for L. paracasei A, several random primers were singly tested in RAPD-PCR assays. Primer RP [39] proved to be suitable for this aim. A specific RAPD-PCR-derived fragment of 250 bp was sequenced after cloning in the pGEM-T vector (Promega, Madison, WI). The obtained nucleotide sequence (Accession number AM162273) was searched for coding regions and nucleotide and protein sequence homologies using the BLAST and FASTA programs. Primers LcA-Fw (5 -AGCACCCACGTTCAAGAA-3 ) and LcA-Rv (5 -CACACGGGGACACCAAGTT-3 ) were designed by using Oligo 3.4 software. The PCR reaction mixture (20 µl) contained reaction buffer 1× (Polymed, Firenze, Italy), 1.5 mM MgCl2 , 200 µM dNTPs, 0.4 µM each primer (Sigma Aldrich, Milano, Italy) and 1 U Taq DNA polymerase (Polymed). Either 30 ng DNA extracted from pure isolates or 150 ng DNA from fecal samples were used as target. The amplification program consisted of an initial

860

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

denaturation of 5 min at 94 ◦ C; 30 cycles of 90 s at 94 ◦ C, 25 s at 63 ◦ C and 1 min at 72 ◦ C and a final extension of 7 min at 72 ◦ C. The detection limits of this PCR assay were assessed using as templates DNA extracted from fecal suspensions prepared by mixing the fecal samples of a subject who had never received strain A with serial dilutions of L. paracasei A cells (3 to 8 log10 CFU g−1 feces). 2.8. PCR-DGGE analysis Fecal DNAs extracted from infants in probiotic group A and control group P were randomly mixed in 4 pools (A1, A2, A3 and A4) and 3 pools (P1, P2 and P3), respectively. Pooled DNAs were used as templates in PCR-DGGE reactions. DNA amplification from total intestinal microbiota was obtained with the universal primers HDA1-GC and HDA2 [37]. DNA from fecal lactobacilli and related genera (32 ng) was amplified with primers SDBacto11 and SGLab677 [14] at an annealing temperature of 56 ◦ C in a final volume of 30 µl. Fecal bifidobacteria DNA was amplified with the genus-specific primers Bif164 and Bif662 [17]. The diluted amplification products of lactobacilli and bifidobacteria (1:10) were further amplified by a nested PCR reaction using primers HDA1-GC and HDA2. Resolution of the PCR amplicons was performed by DGGE using the Bio-Rad DCode Universal Mutation Detection System (Bio-Rad Laboratories Inc., Hercules, CA) [8]. The denaturing gradients of urea and formamide used for discrimination of amplicons from total microbiota, lactobacilli and bifidobacteria were 30–50%, 30–60% and 40–50%, respectively. A 100% denaturing solution contained 40% (v/v) formamide and 7.0 M urea. DGGE ladders for the identification of Lactobacillus- and Bifidobacterium-like populations were prepared by mixing equal amounts of amplicons obtained from selected reference species using primers HDA1 and HDA2-GC. Selected unknown DGGE bands were excised from the acrylamide gels and re-amplified with primers HDA1 and HDA2. After purification, sequencing was carried out at the Bio Molecular Research Center (BMR), University of Padova (Italy), with primer HDA1 [8]. Sequence identities were determined by the BLAST program in the GENBANK database. DGGE electrophoretic patterns were analyzed by TotalLab software (Phoretix, Newcastle, UK), using the “1D image analysis” tool. The data field analyzed was the peak height showing the maximum value of the profile of the band not including the background. Similarities among the DGGE profiles were determined with GelCompar software package version 4.0. Similarity indices were calculated with the Pearson product moment correlation coefficient and Ward’s algorithm was used to construct the corresponding dendrogram. The analysis was performed three times.

3. Results 3.1. Subjects and compliance Thirteen infants in probiotic group A and 13 infants in control group P were included in the analysis since they had completed the study without intercurrent illness. Five infants at T3 and 9 infants at T4 included in the L. paracasei group A ingested the probiotic yogurt only 2–3 times per week. Compliance was verified by isolating strain A from at least one fecal sample of infants in the test group, as explained below. The infants receiving milk fermented with L. paracasei A showed no adverse effects during the experimental period, like those receiving the placebo. 3.2. Specific PCR for L. paracasei A The primary goal of the present work was the development of a PCR-based method for achieving conclusive identification and detection of L. paracasei A. To select a target sequence specific for such strain, several RAPD-PCR analyses with different random primers were conducted on a wide collection of L. paracasei strains of different origin (vegetables, artisanal cheeses, commercial probiotic fermented milks), including strain A. Visual comparison of RAPD-PCR profiles showed the presence of a 250 bp band discriminative for strain A in the fingerprint produced with primer RP (data not shown). This band was cloned and sequenced to verify its identity; no significant homologies at the nucleotide or protein level were found with genes from Lactobacillus spp. or other bacteria. The highest identity value was retrieved at the amino acid level with a putative DNA helicase from Plasmodium falciparium (identities 26%). Specific primers LcA-Fw and LcA-Rv were selected from the terminal regions of the 250 bp fragment sequence; L. paracasei A provided an amplicon of the expected size (235 bp), while no amplification products were obtained when DNA from 20 different Lactobacillus species and 39 L. paracasei strains were tested. 3.3. Specific identification/detection of L. paracasei A in fecal samples To detect and quantify viable cells of L. paracasei A in fecal samples, amplifications with primers LcA-Fw and LcA-Rv were applied to Lactobacillus isolates from MRS–Van plates. The resulting mean counts of viable L. paracasei A during treatment are reported in Fig. 1. We observed that 92% of subjects retained viable probiotic cells after 1 week of consumption (T1). At this sampling time, strain A counts ranged between values of 4.3 and 8.2 log10 CFU g−1 of feces, suggesting a host-specific effect on colonization. The percentage of positive samples decreased during treatment up to 16% after 1 week of washout (T5). No positive samples were found in the two groups before the start of administration. The strain-specific PCR was also used on DNA directly extracted from fecal samples. Preliminarily, detection limits of the

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

PCR assay were assessed adding different amounts of strain A cells to fecal suspensions, as described in Materials and methods. On agarose gel the 235 bp amplification product was detected up to a concentration of 5 log10 CFU g−1 wet feces of strain A (data not shown). All DNA samples from infants carrying viable strain A gave positive PCR signals with few exceptions (Fig. 1): 4 sam-

861

ples did not produce the expected band due to the presence of L. paracasei A counts lower than the detection limit of the reaction. Four samples in which strain A was not isolated, however, produced amplification signals, suggesting that the probiotic cells in feces were dead or had a non-cultivable status. No L. paracasei-A-specific amplicons were obtained from infant feces belonging to placebo group P. 3.4. Microbiological and chemical analyses of fecal samples

Fig. 1. Selective plate counts of L. paracasei A isolated from infant fecal samples during the probiotic supplementation trial. Filled circles represent mean values. The two series of percentages indicate the number of infants in whom L. paracasei A was detected out of the total number of infants in the probiotic group A (n = 13); bacteria were identified by strain-specific PCR carried out on DNA extracted from isolates grown on MRS–Van plates (italic font) and from fecal samples (bold font).

The composition of microbial populations in baby fecal samples during the trial, as determined by selective plating, is summarized in Fig. 2. The main effect of L. paracasei A consumption on infant fecal bacteria consisted of an increment in lactobacillus number. A significant increase (P < 0.01) in the mean counts of lactobacilli over that measured when consuming pasteurized yogurt was observed after 1 week of probiotic consumption (T1); the same trend was maintained after 3 weeks (T3), while the number came close to the initial one at the sampling time (T4) which followed, and after the washout (T5). Lower and more variable lactobacilli counts were generally observed in the placebo group. Mean bifidobacteria counts in the probiotic group A did not show significant increment during L. paracasei A consumption over that measured in the starting sample (T0), but the observed values were higher than those of the placebo group in samples T1, T4 and T5 (P < 0.05). A significant decrease (P < 0.05) in the clostridia count was recorded in infant feces after 4 weeks of treatment (T4) with

Fig. 2. Microbial composition of fecal samples determined by selective plating before (T0), after 1, 3 and 4 weeks (T1, T3, T4) of treatment and subsequent to washout (T5). Infants received fermented milk containing L. paracasei A (probiotic group A) or pasteurized yogurt (placebo group P). Values are expressed as mean ± SD, n = 13. Details of administration are given in Section 2.3. a, b Means and SD are significantly different, with P < 0.01 or P < 0.05, respectively; Student’s t -test T0 versus Tn, where Tn means the time of sampling (T1-T5). c Means and SD are significantly different (P < 0.05); Student’s t -test probiotic group A versus placebo.

862

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

Table 1 Lactobacillus species identified by RAPD-PCR or by Lactobacillus-group specific PCR-DGGE at different sampling times in feces from infants who received fermented milk containing L. paracasei A Species identified

Detection at sampling time (identification method) T0

T1

T3

T4

T5

L. amylovorus L. brevis L. casei/paracasei L. curvatus L. d. bulgaricus L. fermentum L. fuchuensis L. gasseri L. helveticus L. johnsonii L. plantarum L. reuteri L. rhamnosus L. ruminis L. sakei L. salivarius Lc. mesenteroides S. thermophilus

− +(r) +(r/d) − +(d) +(r/d) +(r) +(r) − +(r) +(r) − +(r) +(r) − +(r) +(d) +(d)

+(r) − +(r/d) − − +(d) − − − +(r) +(r) +(r) +(r) +(r) +(d) − +(d) +(d)

+(r) − +(r/d) − +(d) +(r) − − +(r) +(r) +(r) − +(r) − +(d) +(r) +(d) +(d)

− − +(r/d) +(r) +(r) +(d) − +(r) +(r) − − +(r) +(r) − +(d) − +(d) +(d)

− − +(r/d) +(r) +(d) +(d) +(r) − − +(r) − − +(r) − +(d) − +(d) +(d)

Amplicon IDa (ac GenBank, % identity)

L21, L25, L30, LP9 (AF375931, 99–100%) L40 (AY050171, 99%) LP7 (AF375932, 97%)

L28 (AY442936, 94%); LP3 (AY442936, 99%) L18 (AF375902, 100%) L20, L27, LP1 (AY188354, 100%)

+ or −, detected or not detected, respectively. r, RAPD fingerprints were performed on template DNA of Lactobacillus spp. isolated on MRS–Van from infant feces in probiotic group A. d, PCR-DGGE carried out with primers SD-Bact011/SG-Lab677 and nested HDA1-GC/HDA2 on template DNA from pooled fecal samples of infants in probiotic group A. a Values between brackets indicate the ID label/s of DGGE bands reported in Fig. 3a, the accession numbers of the closest relative species as determined by comparative sequence analysis and the identity % obtained.

strain A. Consumption of L. paracasei A did not affect the counts of enterococci, total anaerobes or Bacteroides. The analysis of fecal enzymatic activities showed that L. paracasei A administration did not induce a significant increase in the α- and β-galactosidase and α-glucosidase reactions compared to placebo (data not shown). β-Glucosidase and β-Glucuronidase activities were set on negative values during the treatment period in both studied groups. The other enzymatic activities tested were absent or unchanged during the experimentation. Fecal pH values were not affected by consumption of probiotic fermented milk, ranging from 5.94 ± 0.6 to 6.77 ± 0.64 in probiotic group A and from 6.39 ± 0.63 to 6.75 ± 0.44 in placebo group P during the trial course. 3.5. Examination of fecal samples by PCR-DGGE No significant changes in the DGGE profiles of the dominant bacteria in the total microbial community (primer couple HDA1-GC/HDA2) were observed during or after L. paracasei A consumption (T1, T3, T4 and T5), when compared to starting samples (T0) or to patterns of the placebo group (data not shown). In particular, the use of pooled samples was aimed at minimizing individual changes and revealing possible similar variations in the composition of bacterial species. Sequence analysis showed that the dominant genera of infant feces were bifidobacteria, enterococci, clostridia and ruminococci. Clostridia-related bands did not change during experimentation, confirming the results of the culturing methods. Notably, we identified S. thermophilus as a dominant species of the

4 pools of probiotic group A and in 2 placebo P pools. No Bacteroides-like bands were identified in the DGGE profiles of the two groups. The effects of L. paracasei A consumption upon the fecal population of lactobacilli and related genera (Streptococcus, Leuconostoc and Lactococcus) were analyzed by lactobacillilike group-specific PCR-DGGE. Fig. 3a represents the DGGE profiles obtained from the samples. The species identified by sequence analysis are summarized in Table 1. A few pools harbored dominant endogenous L. paracasei-like species, as shown by the presence of specific bands at T0 and in placebo groups. The analysis of patterns from pools in probiotic group A showed a clear-cut increment in L. paracasei band after 1 week (T1) or 3 weeks (T3) of strain A intake. However, the L. paracasei returned to the initial concentration after the end of consumption (T5), in accordance with plate counts (Fig. 2). Other species identified in infant microbiota were L. sakei, L. delbrueckii subsp. bulgaricus and L. fermentum, which generally incremented during probiotic intake. Moreover, we detected Lactococcus lactis, Leuconostoc mesenteroides and S. thermophilus as dominant species in infant fecal samples. The latter species was identified in all tested samples and the type of treatment influenced its evolution throughout the experiment: in A pools the band intensity increased, but was reduced after the washout (T5) (bands L20 and L27, Fig. 3a). Conversely, in samples from subjects who received pasteurized yogurt, the intensity of bands corresponding to S. thermophilus changed at random. Finally, we recognized bands of non-cultivatable bacteria.

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

863

Fig. 3. DGGE analysis of PCR products from pooled fecal samples of infants after L. paracasei A (A1, A2, A3, A4) or placebo (P1, P2, P3) consumption using (a) specific primers for the Lactobacillus group (SDBacto11/SGLab677, nested with HDA1-GC/HDA2) and (b) the Bifidobacterium-specific primer couple (Bif164/Bif662, nested with HDA1-GC/HDA2). Fecal samples were taken before the start of administration (lanes T0), after 1, 3 and 4 weeks (lanes T1, T3, T4, respectively) and after 1 week of washing out (lanes T5). Lane A: L. paracasei A. Lane M: lactobacilli identification ladder (L1, L. fermentum; L2, L. sakei; L3, L. delbruekii subsp. bulgaricus; L4, S. thermophilus; L5, L. paracasei group). Lanes M1 and M2: bifidobacteria identification ladders (B1, B. longum bv. longum; B2, B. longum bv. infantis; B3, B. bifidum; B4, B. animalis subsp. lactis/B. adolescentis; B5, B. animalis subsp. animalis; B6, B. breve; B7, B. catenulatum/pseudocatenulatum; B8, B. angulatum; B9, B. dentium). Numbered arrows refer to bands sequenced, and correspond to amplicon IDs in Table 1, except for: U, uncultured bacterium; LP2, Lactococcus lactis (AY348313, 100%); LP12, Enterococcus sp. (AY184239, 94%); B10, B. thermoacidophilum (AY166524, 98%).

Similarities among DGGE profiles obtained from lactobacilli-like community at different sampling times were calculated through construction of a dendrogram based on Pearson correlation coefficients (data not shown). The profiles shared high correlation values (90%), confirming the absence of significant alterations in Lactobacillus-like profiles during the experiment and the presence of many stable bands in almost all samples. The largest differences were observed between the patterns of samples belonging to group A, collected before (T0) and during probiotic treatment (T1, T3, T4). The latter, however, were not clustered in the dendrogram, indicating the absence of a correlation between the time of probiotic consumption in the different subjects and the type of DGGE pattern. The bifidobacterial population in feces of infants consuming L. paracasei A or placebo was also selectively examined by means of genus-specific PCR-DGGE. Nested-PCR retrieved clear and discriminative profiles which allowed clear identification of environmental bands by comparison with the ladders (Fig. 3b). Bifidobacterium infantis and B. bifidum were detected in almost all samples of the two groups. We frequently

identified the species B. longum and B. catenulatum as well. The presence of Bifidobacterium thermoacidophilum was detected in one placebo pool (P2). In the placebo samples we did not observe significant changes correlated with consumption of pasteurized yogurt. Conversely, the profiles of samples from L. paracasei group A showed more variations. However, the evolution of some bifidobacteria species differed in each pool. For instance, the intensity of the band corresponding to B. longum increased during probiotic consumption in pools A1 and A3, but fluctuated in pool A2 and decreased in pool A4. Divergent dynamics were also noted for the species B. catenulatum and B. adolescentis. 3.6. Identification by RAPD-PCR of lactobacilli from L. paracasei A-treated infants Using RAPD-PCR, we characterized more than 150 Lactobacillus isolates at different sampling times from feces of infants who received L. paracasei A. Identification was obtained by construction of a dendrogram based on correlation

864

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

coefficients attributed to RAPD fingerprints of fecal isolates and representative type strains (data not shown). The recognized species are reported in Table 1. This analysis revealed the complex (16 species) and varied composition of the Lactobacillus community, which contained typical intestinal species (e.g. L. ruminis, L. reuteri, L. rhamnosus and L. gasseri), as well as food-associated species (e.g. L. paracasei, L. plantarum, L. fermentum). Variability was observed among the species isolated throughout the trial, though L. casei/paracasei was found to be the prevalent species (43% of the isolates). Notably, the autochthonous species were generally maintained throughout the experiment and recurred after probiotic intake had been stopped (L. rhamnosus, L. johnsonii). Yogurt species L. delbrueckii subsp. bulgaricus was isolated only sporadically. As shown in Table 1, a higher number of species were identified with the culture-dependent method than with the PCRDGGE approach. This result may be due to the detection limit of the PCR-DGGE technique. Moreover, different species can be hidden in DGGE analysis as comigrating in the same position of the gel, such as L. sakei, L. fuchuensis and L. curvatus. 4. Discussion The ability of probiotic strains to survive transit and to colonize the GI tract is considered an important biomarker for providing potential health benefits. Thus, the assessment of this trait constitutes a significant element in characterization of strains for human use. Modern molecular tools enable reliable identification and detection of strain-specific markers which are useful for tracing the presence of probiotic bacteria in the microbiota. However, data were compiled in only a restricted number of studies in which ingested probiotic strains were marked up in the GI tract by specific PCR [33], real-time PCR [36] or by recognition using monoclonal antibodies [40]. Other authors have used fingerprinting methods to identify, at the strain level, probiotics isolated from feces [32]. In the present study, the fate of orally administered L. paracasei A in infants was investigated by means of a strain-specific detection test based on PCR amplification of a selected marker derived from RAPD-PCR fingerprinting assay. As previously reported in the literature [25,33], RAPD-PCR is proving to be a successful technique for recognizing unique sequences for a selected strain because of its ability to detect differences among bacteria at the genomic level. Particular care was used in verifying, in the optimized PCR assay, the absence of crossreactions towards endogenous and other food-derived L. paracasei strains which may be present in the infant intestine. Moreover, the selected primer pair allowed the reliable detection of L. paracasei A in complex microbial communities with or without previous cultivation. In particular, we investigated the behavior of L. paracasei A and the response of infant microbiota at different times during probiotic intake, providing interesting and precise information about the dynamics of the resident microbiota following the introduction of an exogenous strain; this is particularly relevant in young subjects in whom the GI microbial community is not yet in stable equilibrium, which is typical of healthy adults.

L. paracasei A was shown to survive in fecal samples in most of the infants examined (92%), particularly after 1 week of oral consumption of fermented milk, demonstrating the ability of the analyzed strain to pass the GI barrier and to temporarily dominate the intestinal Lactobacillus community of different subjects, with diverse intestinal environments. These data also indicate that the microbial community of these infants might favor the colonizing ability of L. paracasei A, perhaps because equilibrium among the species was not completely stabilized. However, colonization of the probiotic strain was reduced during the following weeks of consumption (T3-T4) and after 1 week of washout, as observed by plate counts and DGGE profiles (L. paracasei-like bands). This was partly due to a reduced probiotic intake in some infants and to a natural rebalancing effect of the endogenous microbial population, which seemed to react to the presence of exogenous L. paracasei A. Similar trends have been observed by Guerin-Danan et al. [12]. In particular, strain A-containing milk was well tolerated by all infants and no GI discomfort correlated with probiotic intake was reported during the trial. Although not yet understood in their complex mechanisms, such dynamics could be considered a natural response of healthy microbiota in which the relationships among the species of the community are defined and well functioning. The evolution of the dominant bacterial groups in the intestinal microbiota was monitored by microbiological analysis and by PCR-DGGE at different times during probiotic consumption. In general, no drastic changes in bacterial composition were observed with the two methods, as previously reported in numerous studies performed on healthy subjects [6,21,29,30,32]. The increment in fecal lactobacilli was the main effect due to strain A consumption in the studied infants, as revealed by microbiological analysis. For this reason, the present study focused on this group of bacteria using different analytical approaches to investigate its composition throughout the trial. By means of group-specific DGGE we identified the dominant species of microbiota, and further data were retrieved by genetic characterization (RAPD-PCR) of a wide number of isolates. This culture-independent method has been considered to give an accurate reflection of Lactobacillus diversity [14,22, 38]; even so, in this study, the use of a culture-based analysis led to obtainment of a more accurate characterization of the active bacterial community compared to DGGE. The technique, however, detected other relevant species that were not isolated by culturing methods, such as Leuconostoc mesenteroides and S. thermophilus, which play a potential role in the GI tract. The phase of Lactobacillus colonization represents an interesting element in the formation and establishment of intestinal microbiota in humans. At present, the sources of lactobacilli and the impact of fermented foods upon microbial modulation in infants have not been completely elucidated. In the present study, it was demonstrated that infants aged 12–24 months harbored a complex and active Lactobacillus population, including species that are considered truly autochthonous and foodassociated. Further studies aimed at demonstrating the origins of so-called food-related species in microbiota of infants will

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

be of interest to provide new insights into the dynamics of intestinal colonization and to clarify the true impact of fermented foods in the shaping of microbiota. The analysis of the behavior of milk-derived strain A is a remarkable result in this context. With regard to the general microbial population in infant feces, plate count results and DGGE profiles were characterized by the presence of bifidobacteria as dominant bacteria, suggesting that this situation, considered typical of neonates, is also maintained in older infants [9]. In addition to detecting typical bifidobacteria of baby gut (B. infantis, B. longum and B. bifidum), analysis of the main Bifidobacterium species by genus-specific DGGE detected the species B. catenulatum as being dominant, though the latter has generally been found to be associated with the adult rather than the infant intestine [20]. The dynamics of the bifidobacteria community was indirectly influenced by L. paracasei A, but the effect on many species appeared strain-specific, and was not similar in different subjects that harbor typical bacterial populations. Ruminococci and clostridia were found to be relevant components of infant microbiota, proving their importance in intestinal microecology. However, their role should be further investigated. In conclusion, the new probiotic strain L. paracasei A was tested in healthy infants and revealed colonization capacity in fecal microbiota by means of a strain-specific detection test. The effect of L. paracasei A on the main bacterial groups in the GI ecosystem was moderate, while it led to a significant increment in the Lactobacillus population. The effect of the presence of the probiotic strain in the infants examined was maximum after the first week of consumption; the situation tended toward normalization in the last period of the experiment or after the washout. Characterization of fecal microbiota by means of traditional microbiological and culture-independent methods revealed that the bacterial structure of infants aged 12–14 months is in a transitional state, combining neonate- and adult-like features. The microbiota of these subjects promptly responded to probiotic consumption, later restoring the endogenous equilibrium. Acknowledgements We are grateful to S. Fasoli for her skillful collaboration. References [1] N. Akopyanz, N.O. Bukanov, T.U. Westbloom, S. Kresovich, D.E. Berg, DNA diversity among clinical isolates of Helicobacter pylori detected by PCR-based RAPD fingerprinting, Nucleic Acids Res. 20 (1992) 5137– 5142. [2] E. Bertazzoni Minelli, A. Benini, M. Marzotto, A. Sbarbati, O. Ruzzenente, R. Ferrario, H. Hendriks, F. Dellaglio, Assessment of novel probiotic Lactobacillus paracasei strains for the production of functional dairy foods, Int. Dairy J. 14 (2004) 723–736. [3] F. Berthier, S.D. Ehrlich, Rapid species identification within two groups of closely related lactobacilli using PCR primers that target the 16S/23S rRNA spacer region, FEMS Microbiol. Lett. 161 (1998) 97–106. [4] B. Bjorksten, E. Sepp, K. Julge, T. Voor, M. Mikelsaar, Allergy development and the intestinal microflora during the first year of life, J. Allergy Clin. Immunol. 108 (2001) 516–520.

865

[5] P. Chagnaud, K. Machinis, L.A. Coutte, A. Marecat, A. Mercenier, Rapid PCR-based procedure to identify lactic acid bacteria: Application to six common Lactobacillus species, J. Microbiol. Methods 44 (2001) 139–148. [6] R. Crittenden, M. Saarela, J. Mättö, A.C. Ouwehand, S. Salminen, L. Pelto, E.E. Vaughan, W.M. De Vos, A. Von Wright, R. Fondén, T. Mattila-Sandholm, Lactobacillus paracasei subsp. paracasei F19: Survival, ecology and safety in the human intestinal tract—A survey of feeding studies within the PROBDEMO project, Microb. Ecol. Health Dis. 14 (2002) S22–S26. [7] F. Dal Bello, J. Walter, W.P. Hammes, C. Hertel, Increased complexity of the species composition of lactic acid bacteria in human faeces revealed by alternative incubation condition, Microb. Ecol. 45 (2003) 455–463. [8] S. Fasoli, M. Marzotto, L. Rizzotti, F. Rossi, F. Dellaglio, S. Torriani, Bacterial composition of commercial probiotic products as evaluated by PCR-DGGE analysis, Int. J. Food Microbiol. 82 (2003) 59–70. [9] C.F. Favier, E.E. Vaughan, W.M. de Vos, A.D. Akkermans, Molecular monitoring of succession of bacterial communities in human neonates, Appl. Environ. Microbiol. 6 (2002) 219–226. [10] S. Guandalini, L. Pensabene, M.A. Zikri, J.A. Dias, L.G. Casali, H. Hoekstra, S. Kolacek, K. Massar, D. Micetic-Turk, A. Papadopoulou, J.S. de Sousa, B. Sandhu, H. Szajewska, Z. Weizman, Lactobacillus GG administered in oral rehydration solution to children with acute diarrhea: A multicenter European trial, J. Pediatr. Gastroenterol. Nutr. 30 (2000) 54–60. [11] F. Guarner, J.R. Malagelada, Gut flora in health and disease, Lancet 361 (2003) 512–519. [12] C. Guerin-Danan, C. Chabanet, C. Pedone, F. Popot, P. Vaissade, C. Bouley, O. Szylit, C. Andrieux, Milk fermented with yogurt cultures and Lactobacillus casei compared with yogurt and gelled milk: Influence on intestinal microflora in healthy infants, Am. J. Clin. Nutr. 67 (1998) 111–117. [13] K. Hatakka, E. Savilahti, A. Ponka, J.H. Meurman, T. Poussa, L. Nase, M. Saxelin, R. Korpela, Effect of long term consumption of probiotic milk on infections in children attending day care centres: Double blind, randomized trial, Br. Med. J. 322 (2001) 1327–1331. [14] H.G.J. Heilig, H. Zoetendal, E.E. Vaughan, P. Marteau, A.D.L. Akkermans, W.M. de Vos, Molecular diversity of Lactobacillus spp. and other lactic acid bacteria in the human intestine as determined by specific amplification of 16S ribosomal DNA, Appl. Environ. Microbiol. 68 (2002) 114–123. [15] E. Isolauri, S. Salminen, A.C. Ouwehand, Microbial–gut interactions in health and disease: Probiotics, Best. Pract. Res. Clin. Gastroenterol. 18 (2004) 299–313. [16] M. Kalliomaki, P. Kirjavainen, E. Eerola, P. Kero, S. Salminen, E. Isolauri, Distinct patterns of neonatal gut microflora in infants in whom atopy was and was not developing, J. Allergy Clin. Immunol. 107 (2001) 129–134. [17] R.G. Kok, A. de Waal, F. Schut, G.W. Welling, G. Weenk, K.J. Hellingwerf, Specific detection and analysis of a probiotic Bifidobacterium strain in infant feces, Appl. Environ. Microbiol. 62 (1996) 3668–3672. [18] R.J. Mackie, A. Sghir, H.R. Gaskin, Developmental microbial ecology of the neonatal gastrointestinal tract, Am. J. Clin. Nutr. 69 (1999) S1035– S1045. [19] J. Marmur, A procedure for the isolation of deoxyribonucleic acid from microorganisms, J. Mol. Biol. 3 (1961) 208–218. [20] T. Matsuki, K. Watanabe, R. Tanaka, M. Fukuda, H. Oyaizu, Distribution of bifidobacterial species in human intestinal microflora examined with 16S rRNA-gene-targeted species-specific primers, Appl. Environ. Microbiol. 65 (1999) 4506–4512. [21] A. Montesi, R. Garcia-Albiach, M.J. Pozuelo, C. Pintado, I. Goni, R. Rotger, Molecular and microbiological analysis of caecal microbiota in rats fed with diets supplemented either with prebiotics or probiotics, Int. J. Food Microbiol. 98 (2005) 281–289. [22] D.S. Nielsen, P.L. Moller, V. Rosenfeldt, A. Paerregaard, K.F. Michaelsen, M. Jakobsen, Case study of the distribution of mucosa-associated Bifidobacterium species, Lactobacillus species, and other lactic acid bacteria in the human colon, Appl. Environ. Microbiol. 69 (2003) 7545–7548. [23] A.C. Ouwehand, S. Salminen, E. Isolauri, Probiotics: An overview of beneficial effects, Antonie van Leeuwenhoek 82 (2002) 279–289. [24] C.A. Pedone, A.O. Bernabeu, E.R. Postaire, C.F. Bouley, P. Reinert, The effect of supplementation with milk fermented by Lactobacillus paraca-

866

[25] [26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

M. Marzotto et al. / Research in Microbiology 157 (2006) 857–866

sei (strain DN-114 001) on acute diarrhoea in children attending day care centres, Int. J. Clin. Pract. 53 (1999) 179–184. F. Quere, A. Deschamps, M.C. Urdaci, DNA probe and PCR-specific reaction for Lactobacillus plantarum, J. Appl. Microbiol. 82 (1997) 783–790. T. Requena, J. Burton, T. Matsuki, K. Munro, M.A. Simon, R. Tanaka, K. Watanabe, G.W. Tannock, Identification, detection, and enumeration of human Bifidobacterium species by PCR targeting the transaldolase gene, Appl. Environ. Microbiol. 68 (2002) 2420–2427. J.M. Saavedra, N.A. Bauman, I. Oung, J.A. Perman, R.H. Yolken, Feeding of Bifidobacterium bifidum and Streptococcus thermophilus to infants in hospital for prevention of diarrhoea and shedding of rotavirus, Lancet 344 (1994) 1046–1049. J.M. Saavedra, A. Abi-Hanna, N. Moore, R.H. Yolken, Long-term consumption of infant formulas containing live probiotic bacteria: Tolerance and safety, Am. J. Clin. Nutr. 79 (2004) 261–267. J.M. Simpson, V.J. McCracken, H.R. Gaskins, R.I. Mackie, Denaturing gradient gel electrophoresis analysis of 16S ribosomal DNA amplicons to monitor changes in fecal bacterial populations of weaning pigs after introduction of Lactobacillus reuteri strain MM53, Appl. Environ. Microbiol. 66 (2000) 4705–47814. S. Spanhaak, R. Havenaar, G. Schaafsma, The effect of consumption of milk fermented by Lactobacillus paracasei strain Shirota on the intestinal microflora and immune parameters in humans, Eu. J. Clin. Nutr. 52 (1998) 899–907. A. Sullivan, R. Bennet, M. Viitanen, A.C. Palmgren, C.E. Nord, Influence of Lactobacillus F19 on intestinal microflora in children and elderly persons and impact on Helicobacter pylori infections, Microb. Ecol. Health Dis. 14 (2002) S17–S21. G.W. Tannock, K. Munro, H.J.M. Harmsen, G.W. Welling, J. Smart, P.K. Gopal, Analysis of the fecal microflora of human subjects consuming a probiotic containing Lactobacillus rhamnosus DR20, Appl. Environ. Microbiol. 66 (2000) 2578–2588. A. Tilsala-Timisjarvi, T. Alatossava, Strain-specific identification of probiotic Lactobacillus rhamnosus with randomly amplified polymorphic

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

DNA-derived PCR primers, Appl. Environ. Microbiol. 64 (1998) 4816– 4819. S. Torriani, G. Zapparoli, F. Dellaglio, Use of PCR-based methods for rapid differentiation of Lactobacillus delbrueckii subsp. bulgaricus and L. delbrueckii subsp. lactis, Appl. Environ. Microbiol. 65 (1999) 4351– 4356. S. Torriani, G.E. Felis, F. Dellaglio, Differentiation of Lactobacillus plantarum, L. pentosus, and L. paraplantarum by recA gene sequence analysis and multiplex PCR assay with recA gene-derived primers, Appl. Environ. Microbiol. 67 (2001) 3450–3454. B. Vitali, M. Candela, D. Matteuzzi, P. Brigidi, Quantitative detection of probiotic Bifidobacterium strains in bacterial mixtures by using real-time PCR, Syst. Appl. Microbiol. 26 (2003) 269–276. J. Walter, G.W. Tannock, A. Tilsala-Timisjarvi, S. Rodtong, D.M. Loach, K. Munro, T. Alatossava, Detection and identification of gastrointestinal Lactobacillus species by using denaturing gradient gel electrophoresis and species-specific PCR primers, Appl. Environ. Microbiol. 66 (2000) 297– 303. J. Walter, C. Hertel, G.W. Tannock, C.M. Lis, K. Munro, W.P. Hammes, Detection of Lactobacillus, Pediococcus, Leuconostoc, and Weissella species in human feces by using group-specific PCR primers and denaturing gradient gel electrophoresis, Appl. Environ. Microbiol. 67 (2001) 2578–2585. L.J.H. Ward, M.J. Timmins, Differentiation of Lactobacillus paracasei, Lactobacillus paraparacasei and Lactobacillus rhamnosus by polymerase chain reaction, Lett. Appl. Microbiol. 29 (1999) 90–92. N. Yuki, K. Watanabe, A. Mike, Y. Tagami, R. Tanaka, M. Ohwaki, M. Morotomi, Survival of a probiotic, Lactobacillus casei strain Shirota, in the gastrointestinal tract: Selective isolation from faeces and identification using monoclonal antibodies, Int. J. Food Microbiol. 48 (1999) 51–57. G. Zoppi, M. Cinquetti, A. Benini, E. Bonamini, E. Bertazzoni Minelli, Modulation of the intestinal ecosystem by probiotics and lactulose in children during treatment with ceftriaxone, Curr. Ther. Res. 62 (2001) 418– 435.