Secondary Logo

Journal Logo

Intestinal Microbiota of 6-week-old Infants Across Europe: Geographic Influence Beyond Delivery Mode, Breast-feeding, and Antibiotics

Fallani, Matteo*; Young, David; Scott, Jane; Norin, Elisabeth§; Amarri, Sergio||; Adam, Rüdiger; Aguilera, Marga#; Khanna, Sheila; Gil, Angel#; Edwards, Christine A; Doré, Joël*and Other Members of the INFABIO Team

Journal of Pediatric Gastroenterology and Nutrition: July 2010 - Volume 51 - Issue 1 - p 77–84
doi: 10.1097/MPG.0b013e3181d1b11e
Original Articles: Hepatology and Nutrition
Free

Objectives: There are many differences in diet and lifestyle across Europe that may influence the development of the infant gut microbiota. This work aimed to assess the impact of geographic area, mode of delivery, feeding method, and antibiotic treatment on the fecal microbiota of infants from 5 European countries with different lifestyle characteristics: Sweden, Scotland, Germany, Italy, and Spain.

Patients and Methods: Fecal samples from 606 infants (age 6 weeks) recruited within the European project INFABIO were analyzed by fluorescent in situ hybridization combined with flow cytometry using a panel of 10 rRNA targeted group- and species-specific oligonucleotide probes. Information on factors potentially affecting gut microbiota composition was collected with questionnaires and associations were evaluated with multivariate analyses.

Results: The Bifidobacterium genus was predominant (40% average proportion of total detectable bacteria), followed by Bacteroides (11.4%) and enterobacteria (7.5%). Northern European countries were associated with higher proportions of bifidobacteria in infant feces, whereas a more diverse microbiota with more bacteroides characterized southern countries. Bifidobacteria dominated the microbiota of breast-fed infants, whereas formula-fed babies had significantly higher proportions of Bacteroides and members of the Clostridium coccoides and Lactobacillus groups. Newborns delivered by cesarean section or from mothers treated with antibiotics perinatally had lower proportions of Bacteroides and members of the Atopobium cluster.

Conclusions: Delivery mode and feeding method influenced the fecal microbiota of European infants at 6 weeks, as expected, but the effect of country of birth was more pronounced, with dominant bifidobacteria in northern countries and greater early diversification in southern European countries.

*Institut National de la Recherche Agronomique, Unité d'Écologie et de Physiologie du Système Digestif, Jouy en Josas, France

Department of Statistics and Modelling Science, University of Strathclyde, Glasgow, UK

Department of Human Nutrition, University of Glasgow, Yorkhill Hospitals, UK

§Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden

||Department of Paediatrics, University of Modena and Reggio Emilia, Modena, Italy

Department of General Paediatrics, University Children's Hospital, Düsseldorf, Germany

#Department of Biochemistry and Molecular Biology, University of Granada, Granada, Spain.

Received 13 August, 2008

Accepted 13 March, 2009

Address correspondence and reprint requests to Christine A. Edwards, Human Nutrition Section, University of Glasgow, Yorkhill Hospitals, Glasgow G3 8SJ, UK (e-mail: c.a.edwards@clinmed.gla.ac.uk).

This study was carried out with financial support from the Commission of the European Communities, specifically the RTD program “Quality of Life and Management of Living Resources,” QLK1 2002 02606, “INFABIO,” coordinated by Professor Christine Edwards (University of Glasgow, Glasgow, UK). It does not necessarily reflect its views and in no way anticipates the commission's future policy in this area. Other members of the INFABIO project involved Benoit Thezé (Institut National de la Recherche Agronomique, Unité d'Écologie et de Physiologie du Système Digestif, Jouy en Josas, France), Khalid M. Khan (Department of Human Nutrition, University of Glasgow, Yorkhill Hospitals, UK), Agneta Uusijarvi (Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden), Francesca Benatti (Department of Paediatrics, University of Modena and Reggio Emilia, Modena, Italy), Rosa Maria De Mola (Department of Paediatrics, University of Modena and Reggio Emilia, Modena, Italy), Horst Schroten (Department of General Paediatrics, University Children's Hospital, Düsseldorf, Germany), José Maldonado (Department of Biochemistry and Molecular Biology, University of Granada, Granada, Spain), Oscar Thompson (Department of Biochemistry and Molecular Biology, University of Granada, Granada, Spain), José Maria Vieites (Department of Biochemistry and Molecular Biology, University of Granada, Granada, Spain), and Marco Ferrer (Department of Biochemistry and Molecular Biology, University of Granada, Granada, Spain).

The authors report no conflicts of interest.

The human intestinal microbiota is a complex ecosystem, consisting of several hundred different bacterial species. This microbiota plays an important role in human health and nutrition by producing nutrients, preventing colonization of the gut by potential pathogenic microorganisms (1), and preserving the health of the host through interactions with the developing immune system (2). The microbiota in early life has been linked to allergy risk (3) and infection. A more recent study (4) suggested that the composition of the infant gut microbiota may be related to later risk of obesity.

Major changes in the intestinal microbial composition occur in early life. Sterile in utero, the gastrointestinal tract of the newborn infant is rapidly colonized at birth by a myriad of maternal vaginal and fecal bacteria and other sources from its environment (5). The first few weeks after birth correspond to critical stages of gut colonization. Bacterial colonization of the gastrointestinal tract is influenced by numerous factors including diet, environment, antibiotic treatment, mucosal maturation, and age.

Characterization of the microbiota represents a first step in understanding the contribution of the microbial community inhabiting the gastrointestinal tract to human physiology (6). Most previous studies, which generated our current understanding of gut microbiology and ecology of infants, relied almost exclusively on the use of culturing methods. However, these methods are not applicable to large sets of samples and are limited by their inability to detect noncultivable bacteria, representing 60% to 70% of all bacteria (7,8). New techniques based on bacterial RNA and DNA have been developed for investigating, identifying, and quantifying the intestinal microbiota. Molecular tools have mainly targeted ribosomal RNA and, more specifically, 16S rRNA (9). Fluorescent in situ hybridization (FISH) combined with flow cytometry (FC) is a particularly high-throughput method based on 16S rRNA probe hybridization, reliable for characterizing the composition of fecal microbiota in epidemiological studies (10–12).

In the past few years, several studies based on culture-independent techniques have investigated the molecular composition of the infant intestinal microbiota. These studies based on FISH (2,13), denaturing gel gradient electrophoresis (14), quantitative real-time polymerase chain reaction (15–17), or terminal-restriction fragment length polymorphism (18) showed a high variability in the first year of life of infants, depending on their mode of delivery, age, time of weaning, and feeding method, with particularly high numbers of bifidobacteria and Bacteroides and presence of lactobacilli/enterococci and coliforms. However, most of these studies have so far been restricted to small cohorts of infants and all with infants recruited within a single geographic region or country.

Given the increasing number of studies comparing infants from different countries to investigate the role of bacteria in the development of conditions such as allergy, it is important to understand the impact of country of birth on the development of the microbiota in infants without allergy. Large-scale molecular studies have compared the gut microbiota compositions of adults (19) and elderly adults (20) across Europe. The present work was undertaken to determine, using FISH-FC, the fecal microbiota composition of 606 young infants from 5 European countries with different lifestyle characteristics and to correlate this composition to country of origin, mode of delivery, feeding method, and perinatal antibiotic treatment.

Back to Top | Article Outline

PATIENTS AND METHODS

Sample and Data Collection

Fecal samples were available from 606 infants participating in the Diet and Environment longitudinal study of the European project INFABIO (www.gla.ac.uk/infabio). Infants were recruited in 5 different European centers: 158 from Glasgow (UK), 125 from Reggio Emilia (Italy), 116 from Stockholm (Sweden), 109 from Granada (Spain), and 98 from Düsseldorf (Germany). Mothers of newborn infants filled in a first questionnaire recording events concerning pregnancy and delivery. Once the baby was 6 weeks old, mothers filled in a second questionnaire concerning feeding method, health history of the baby at approximately 6 weeks, antibiotic treatment of infants or breast-feeding mother, and so forth.

Back to Top | Article Outline

Fecal Samples and Cell Fixation

Each fecal specimen, collected from the 606 infants at approximately 6 weeks of age, was placed in a sterile plastic box and maintained under anaerobic conditions at 4°C using an anaerocultA (Merck, Nogent sur Marne, France) for a maximum of 4 hours before processing for cell fixation as previously described (21). There was much discussion in the planning of the project of how to keep the conditions of collection and processing of the samples as standardized as possible but allowing for local conditions and practices at each center. Thus, soon after fecal samples were passed, they were placed into the anaerobic jar and maintained under anaerobic conditions and processed for cell fixation within 4 hours of passage. In some cases, research staff collected the sample in the diaper from the infant's home as soon as parents contacted the laboratory. In others, parents delivered the samples under anaerobic conditions quickly to the laboratory, or the infant provided the sample at the clinic, and this was placed in an anaerobic jar and sent to the laboratory and processed in the same way and time frame as all of the other samples. Sample fixation kits were provided to each collection center. Feces were homogenized by mechanical kneading for 3 minutes and aliquots of 1 g (wet weight) were added to 9 mL of anaerobic phosphate-buffered saline (PBS). The suspension was mixed to complete homogeneity in a 50-mL stoppered sterile glass jar and aliquots of 0.2 mL of the suspension were added to 0.6 mL of 4% paraformaldehyde (PFA) in PBS. After 1 night at 4°C, suspensions fixed in PFA were stored at −70°C and shipped on dry ice approximately every 3 months for analysis at a single location (INRA, Jouy en Josas, France).

Back to Top | Article Outline

FISH Method

As described previously (10,11), 400 μL of fixed suspension was mixed with 600 μL of PBS. Before hybridization, cells were always pelleted at 8000 g for 3 minutes in a microcentrifuge tube and resuspended in a volume of 1 mL. After washing in Tris-EDTA buffer (100 mmol/L Tris-HCl pH 8.0, 50 mmol/L EDTA pH 8.0), pellets were suspended in Tris-EDTA buffer containing 1 mg/mL of lysozyme (Serva, Heidelberg, Germany) and incubated for 10 minutes at room temperature. Cells were then washed in PBS to remove lysozyme and equilibrated in the hybridization solution (900 mmol/L NaCl, 20 mmol/L Tris-HCl pH 8.0, 0.01% sodium dodecyl sulfate pH 7.2, 30% formamide). A 50-μL aliquot of this suspension was used for FISH with control and group-specific probes. Hybridization was performed in a 96-well microtiter plate overnight at 35°C in the hybridization solution containing 4 ng/μL (final concentration) of the appropriate labeled probes. Following hybridization, 150 μL of hybridization solution was added in each well and cells were pelleted at 4000 g for 15 minutes. Washing was performed to remove nonspecific binding of the probe by incubating the bacterial cells at 37°C for 20 minutes in the washing solution (64 mmol/L NaCl, 20 mmol/L Tris-HCl pH 8.0, 5 mmol/L EDTA pH 8.0, 0.01% sodium dodecyl sulfate pH 7.2). Cells were finally pelleted at 4000 g for 15 minutes and resuspended in PBS. An aliquot of 100 μL was added to 0.4 mL of FACS FLOW for FC detection.

Back to Top | Article Outline

Data Acquisition by Flow Cytometry

Data acquisition was performed with a FACS Calibur flow cytometer (Becton Dickinson, Erembodegem-Aalst, Belgium) equipped with an air-cooled argon ion laser providing 15 mW at 488 nm combined with a 635-nm red-diode laser, as described previously (10,11). All of the parameters were collected as logarithmic signals. The 488-nm laser was used to measure the forward angle light scatter (FSC, in the 488-nm band pass filter), the side angle light scatter (SSC, in the 488-nm band pass) and the green fluorescence intensity conferred by fluorescein isothiocyanate–labeled probes (FL1, in the 530-nm band pass filter). The red-diode laser was used to detect the red fluorescence conferred by Cy5-labeled probes (FL4, in a 660-nm band pass filter). The acquisition threshold was set in the side scatter channel. The rate of events in the flow was generally below 3000 events per second. A total of 100,000 events were stored in list mode files for each sample. Subsequent analyses were conducted using the CellQuest Software (Becton Dickinson). Analyses were performed as described previously (22) by creating density plots and delineating regions to encompass double-labeled bacteria.

Back to Top | Article Outline

Oligonucletide Probes

The EUB 338 probe, conserved within the domain bacteria (23), was used as a positive control of hybridization. Conversely, the NON 338 probe (24) was used as a negative control. These oligonucleotide probes were covalently linked at their 5′ end either with FITC or with Cy5 (Thermo Electron, Ulm, Germany). A panel of 10 group- and species-specific probes covalently linked with Cy5 at their 5′ end was used to assess the microbiota composition (Table 1) (25–31). Enumeration of the different bacterial groups or species was performed by FISH-FC by combining in the same tube 1 specific probe labeled with Cy5 together with the EUB 338 FITC probe.

TABLE 1

TABLE 1

Back to Top | Article Outline

Statistical Analysis

Data are expressed as average and standard deviation of the proportions of cells that hybridized with each of the 10 oligonuclotide probes relative to the total bacteria. A value of zero was used to calculate the means whenever a microbial group was undetected or detected below the threshold of sensitivity of 0.4% (25). A general linear model was used to investigate outcome variables of interest (proportions of bacterial groups detected) across other main effects (country effect, delivery method, antibiotic treatment on infant and on mother during pregnancy) while correcting for feeding method. We did not correct for socioeconomic status because this has not been suggested to influence the microbiota at 6 weeks beyond its impact on breast-feeding rates in some countries. The same model was used to compare the outcome variables of interest across the 3 different feeding methods (fully breast-fed, formula fed, mixed fed) while correcting for the country effect. Comparisons were made using the Bonferroni correction factor to compensate for multiple testing. The geographical distribution observed in the composition of dominant fecal microbiota was assessed using principal component analyses (PCAs) by comparing the principal components across country of birth using analysis of variance. All analyses were performed using Minitab (version 14, Minitab Ltd, Coventry, UK) with a significance level of 5%.

Back to Top | Article Outline

RESULTS

Infant Feeding Practice

The average proportion of fully breast-fed infants of the total of 606 infants considered was 51.5%, whereas the proportion of formula-fed infants was 30.1% and the proportion of mixed-fed infants was 18.4% of the total infants. During this study, some infant formula manufacturers began including prebiotics in infant formula. It was not the intention of this study to investigate the impact of prebiotics on the fecal microbiota. The infant formula fed to the infants in this study varied between infants and centers and it was not possible to identify for certain which infants had received prebiotics.

Back to Top | Article Outline

Assessment of the Microbiota Composition of Infant Fecal Samples With FISH Combined With Flow Cytometry

Nearly all of the samples received (98%) could be analyzed. There was insufficient fecal material in the other 2%. When the data of all 606 infants were considered together, the predominant group by far detected was Bifidobacterium, with 40% of all detectable bacteria (±30.6%), followed by Bacteroides (11.4% ± 17.6%) and Enterobacteria (7.5% ± 15.9%). The Clostridium coccoides group, the main predominant group of the adult gut microbiota, presented proportions of only 5.5% (±11.5%), whereas Clostridium perfringens + Clostridium difficile species represented 3% of the total (±8.4%). Atopobium cluster, Streptococcus group, and Lactobacillus group represented 2.1% ± 6.2%, 1.6% ± 3.5%, and 1.2% ± 4.0% of the total, respectively, whereas the Clostridium leptum group, also a majority in the adult microbiota, presented average proportions at the limit of detection of the method (0.4% ± 2.3%). When the proportions of the bacterial cells detected were added together, a mean of 72.7% ± 24.5% was obtained with the panel of 10 nonoverlapping phylogenetic probes.

Back to Top | Article Outline

Impact of Country of Birth

The proportions of the different bacterial groups detected in the different countries are given in Table 2. Looking at the different bacterial groups, the center of origin was found to have a high impact, particularly for bifidobacteria, Bacteroides, and enterobacteria. A lesser impact was observed for members of the C coccoides, Lactobacillus, and Streptococcus groups, whereas no difference between countries was found for the less represented C leptum group, the Atopobium cluster, and the C difficile and C perfringens species. Samples from infants born in Granada, where breast-feeding rate was 43.1% at the time the samples were collected, presented significantly greater proportions of Bacteroides and enterobacteria, and significantly lower proportions of bifidobacteria, compared with all other countries (Table 2). Granada samples also presented lower proportions of C coccoides members compared with Glasgow (P < 0.001) and Düsseldorf (P = 0.002) and higher proportions of members of the Lactobacillus group compared with Glasgow (P < 0.001), Stockholm (P = 0.003), and Reggio Emilia (P = 0.007). On the contrary, Stockholm (breast-feeding rate 75.9%) presented significantly higher proportions of bifidobacteria compared with all of the other centers, Glasgow (breast-feeding rate 48.7%) showed higher proportions of members of the C coccoides group compared with all of the centers except Düsseldorf (breast-feeding rate 45%), whereas Reggio Emilia (breast-feeding rate 62.4%) had higher proportions of enterobacteria compared with Glasgow (P = 0.029) and Stockholm (P = 0.012) and higher proportions of members of the Streptococcus group compared with Glasgow (P = 0.047) and Düsseldorf (P = 0.010). No significant difference was found between centers for the sum of the proportions of the different bacterial groups analyzed. The geographic trend in microbiota composition was further supported by PCA (Fig. 1A). Infants' microbiota were distributed along the axes of the first and second principal components (accounting for 22% and 14% of variability, respectively) as a function of the country of birth with an apparent north–south distribution. The overall sum of detected bacterial groups best explained the principal component 1, and proportions of bifidobacteria and bacteroides were inversely associated along principal component 2 in such a way that northern countries were characterized by a preweaned infant fecal microbiota highly dominated by bifidobacteria, whereas southern countries showed more Bacteroides and the highest early diversification of infant microbiota. The possibility of a confounding effect of feeding method on the impact of country of birth was excluded by a PCA showing that infant microbiota in Stockholm (Fig. 1B) and Granada (Fig. 1C) did not cluster as a function of feeding method.

Table 2

Table 2

FIGURE 1

FIGURE 1

Back to Top | Article Outline

Impact of the Feeding Method

Comparisons of the different bacterial groups detected with FISH-FC across the three feeding methods (fully breast-feeding, formula feeding, and mixed feeding) were determined using a linear model while correcting for the country effect. The adjusted mean values for each feeding type are shown in Figure 2. Breast-fed infants presented significantly greater proportions of bifidobacteria (44.8% vs 29.9%, P < 0.001) and significantly lower proportions of Bacteroides (8.8% vs 15.9%, P < 0.001), C coccoides (3.7% vs 6.9%, P = 0.014), and Lactobacillus groups (0.9% vs 1.9%, P = 0.046) compared with formula-fed babies. Breast-fed infants also presented significantly lower proportions of Bacteroides compared with mixed-fed babies (8.8% vs 13.8%, P = 0.034), whereas proportions of bifidobacteria were still significantly higher in mixed-fed infants than in formula-fed (40.9% vs 29.9%, P = 0.007) infants. Finally, formula-fed babies also presented lower proportions of C perfringens species compared with breast-fed infants (4.6% vs 24.3%, P = 0.006), while no significant difference was observed for C difficile species alone and C perfringens and C difficile detected together.

FIGURE 2

FIGURE 2

Back to Top | Article Outline

Impact of the Delivery Method

The mode of delivery had an impact on some bacterial groups. The adjusted mean values of the different bacterial groups detected for each delivery method are shown in Figure 3. Compared with cesarean section, vaginal delivery (67% of the total) was associated with higher average proportions of Bacteroides (16.1% vs 6.9%, P < 0.001) and members of the Atopobium cluster (2.9% vs 0.8%, P < 0.001) and lower proportions of members of the C coccoides group (4.5% vs 8.2%, P < 0.001) and the Streptococcus group (1.4% vs 1.9%, P = 0.048). Vaginally delivered infants also presented a greater proportion for the sum of detected groups compared with the other babies (75.4 vs 67.6, P < 0.001). There was no effect of the mode of delivery on the relative proportions of bifidobacteria.

FIGURE 3

FIGURE 3

Back to Top | Article Outline

Effect of Antibiotic Treatments

Newborns who received antibiotics (only 7% of the 606 children investigated) presented significantly higher proportions of enterobacteria (16.6%) compared with those without treatment (6.8%) (P < 0.001). On the contrary, when mothers received antibiotic treatment perinatally and/or during breast-feeding, infants presented significantly lower average proportions of Bacteroides (11.4% vs 15.0%, P = 0.029) and members of the Atopobium cluster (1.5% vs 2.6%, P = 0.044), as well as for the total sum of detected groups (69.6% vs 76.1%, P = 0.005), compared with those whose mother received no treatment during pregnancy. There was a wide range of antibiotics received by individual infants and mothers and differences between countries. Many mothers could not identify which antibiotic was used. It was, therefore, not possible to carry out any detailed analysis.

Back to Top | Article Outline

DISCUSSION

The aim of this study was to assess the gut microbiota composition of the young European infant by analyzing 606 fecal samples obtained from babies at approximately 6 weeks from 5 countries with different lifestyle characteristics. The present study also investigated the impact of some important variables such as geographic origin, feeding method, mode of delivery, and antibiotic treatment on the early development of the intestinal microbiota of children. There are many differences in diet and lifestyle characteristics across Europe. For example, in Scotland, breast-feeding rates are lower (32) and many infants are weaned before 3 months (33). In contrast, in Scandinavian countries, breast-feeding rates are high and infants are weaned later (34,35). One of the main ways that diet and environment influence the infant is through their effects on the gut microbiota and its metabolism (36), which may have important effects on the health of the infant and later on the longer-term health of the child and adult.

In this study, we confirmed previously published work (13–17,37–39) that bifidobacteria are the predominant group detected in the feces of preweaned infants, followed by Bacteroides and enterobacteria. More recent studies based on molecular approaches also confirmed these findings (13–17,40–42). Dore et al (40) detected rRNA proportions of 30% to 40% for bifidobacteria in 2-month-old babies by dot-blot hybridization, whereas Martin et al (41) found a proportion of 21.7% of bifidobacteria and 34.6% of enterobacteria in preweaning fecal samples. Harmsen et al (13) found that infants aged 20 days harbored a microbiota in which bifidobacteria was predominant (30%–80%), particularly in breast-fed infants, but Bacteroides were also important (20%–60%), particularly in formula-fed babies, while Escherichia coli represented around 5% to 10% of the total.

A strong impact of the geographic origin was observed in the present study. This effect was particularly observed for Bacteroides, bifidobacteria, and enterobacteria. Our observations suggest a possible “geographic gradient” in the composition of the gut microbiota in Europe, where the extremes, north (Glasgow and Stockholm) and south (Granada and Reggio Emilia), would present the highest number of differences. It should be noted, however, that the centers from which the infants were recruited in the present study may not be representative of the country in which they were based because only 1 center was used in each case.

The north–south gradient was characterized by higher proportions of bifidobacteria, Atopobium, C perfringens + C difficile, and sum of total detectable bacteria in north European countries, and by higher proportions of Bacteroides, enterobacteria, and lactobacilli in south European countries, whereas C coccoides, C leptum, and streptococci remained unaffected by the country of birth at the age considered. To our knowledge, this is the first cross-sectional study comparing the impact of country of origin on the development of the gut microbiota of babies born in different European countries. A few previous studies have compared the microbiota composition for 2 countries and the majority considered infants born in developing countries as well (42,43). Sepp et al (44) reported high counts of lactobacilli and eubacteria in Estonian infants and increased numbers of clostridia in Swedish babies, with bifidobacteria and anaerobic cocci equally prevailing in both groups, which they related to risk of allergy. It is well established that in Western industrialized countries, routine hygienic procedures aimed at reducing the spread of bacteria in maternity and neonatal wards have strongly influenced the colonization pattern of newborn infants, whereas infants born in developing countries are exposed to a heavier bacterial load from birth and this condition influences the colonization pattern of the gut. The low colonization rate of enterobacteria in infants born in Swedish hospitals, reported by Lundequist et al (38), is probably related to these practices. Meanwhile, 2 recent studies investigated cross-sectional differences (19,20) in microbiota in young and older adults across Europe. In young adults (19), no significant differences with respect to geographic origin were found, but the research included only volunteers from central to northern European countries (Denmark, United Kingdom, the Netherlands, France, Germany). In the other study (20), which compared young adults and elderly adult volunteers from Sweden, France, Germany, and Italy, a significant difference was observed for the Bifidobacterium group, with the population in Italy having 2- to 3-fold higher proportions of bifidobacteria than that in any other country. However, the factors that determine the complex and relatively stable microbiota of adult populations, in which diet and other lifestyle factors may vary considerably, may be different from those that influence the early colonization of the infant gut at 6 weeks of age.

The geographic origin was far more important for the composition of the preweaned infant microbiota than any other parameter. Yet, in accordance with previous culture-based studies (37,38,45–47) and more recent molecular studies (13,16,41), we observed an important impact of the feeding method on the early development of the infant gut microbiota. Breast milk, even in mixed feeding, clearly favored bifidobacteria, whereas in its absence, a more diversified microbiota was established, with higher proportions of Bacteroides and members of the C coccoides and Lactobacillus groups. Comparing several individual studies, Tannock (47) found that numbers of clostridia were always lower in breast-fed babies and that clostridia was the only group predictive of formula feeding. Our observations agreed with this concept, focusing on C coccoides as a potential indicator group.

The mode of delivery also significantly influenced the microbiota composition of the newborns' intestine. Upon vaginal delivery, the infant is predominantly exposed to vagina and fecal bacteria of maternal origin. Conversely, infants born by cesarean delivery have an initial exposure to environmental bacteria from equipment, air, other infants, and nursing staff. Vaginally delivered babies presented higher proportions of Bacteroides and members of the Atopobium, as well as added proportions of detectable bacteria, and lower proportions of members of the C coccoides and the Streptococcus groups compared with those born by cesarean section. Interestingly, the latter showed the same trend as infants born to mothers who received antibiotics during late pregnancy and/or while breast-feeding. A previous study (48) reported a considerable delay in the establishment of a stable microbiota in infants born by cesarean section, characterized by a low incidence of Bacteroides spp and a low isolation rate of other bacteria. Gronlund et al (49) also reported a delay in fecal colonization and a low number of Bacteroides fragilis in cesarean-delivered infants. In a recent study, Penders et al (16), investigating the fecal microbiota of 1032 Dutch infants by quantitative real-time polymerase chain reaction, observed that infants born by cesarean section had lower numbers of Bacteroides and bifidobacteria and were more often colonized with C difficile than were vaginally born infants. Hence, antibiotic treatment and cesarean delivery may promote the same suboptimal development of the microbiota in early infancy.

Concerning detection of C difficile and C perfringens species, an average of 3.0% ± 8.4% was found with the 2 probes coupled together, whereas separate detection of the 2 species on 46 samples, chosen among those with high combined counts, showed 0.7% and 12.4% average proportions, respectively. C perfringens was formerly detected within 2 days of life (39) and differences concerning these species were observed, using selective culture media, as a function of the feeding method (17,37,39), mode of delivery (50), and geographic origin (45). In the present study, no differences were observed for C difficile and C perfringens species detected together for any of the variables investigated. Only breast-fed babies were found to have significantly higher proportions of C perfringens species compared with formula-fed infants (P = 0.006, 24.3% vs 4.6%), but for a subset of the samples investigated.

In conclusion, in this large-scale study, we highlighted the impact of geographic origin, feeding method, delivery mode, and antibiotic treatment on the composition of the fecal microbiota of European infants at 6 weeks of age. Above all, the colonic microbiota of the young healthy infant appeared different across Europe. The potential of a south-to-north gradient among the European countries investigated needs to be researched further to establish the gradient using more centers and to determine the factors responsible.

Back to Top | Article Outline

REFERENCES

1. Guarner F, Malagelada JR. Gut flora in health and disease. Lancet 2003; 361:512–519.
2. Kirjavainen PV, Arvola T, Salminen SJ, et al. Aberrant composition of gut microbiota of allergic infants: a target of bifidobacterial therapy at weaning? Gut 2002; 51:51–55.
3. Penders J, Thijs C, van de Brandt PA, et al. Gut microflora composition and development of atopic manifestations in infancy: the KOALA birth cohort study. Gut 2007; 56:661–667.
4. Kalliomaki M, Collado MC, Salminen S, et al. Early differences in fecal microbiota composition in children may predict overweight. Am J Clin Nutr 2008; 87:534–538.
5. Hooper LV. Bacterial contributions to mammalian gut development. Trends Microbiol 2004; 12:129–134.
6. Hooper LV, Midtvedt T, Gordon JI. How host–microbial interactions shape the nutrient environment of the mammalian intestine. Annu Rev Nutr 2002; 22:283–307.
7. Harmsen HJM, Gibson GR, Elfferich P, et al. Comparison of viable cell counts and fluorescence in situ hybridization using specific rRNA-based probes for the quantification of human fecal bacteria. FEMS Microbiol Lett 2000; 183:125–129.
8. Suau A, Bonnet R, Sutren M, et al. Direct analysis of genes encoding 16S rRNA from complex communities reveals many novel molecular species within the human gut. Appl Environ Microbiol 1999; 65:4799–4807.
9. Amann RI, Ludwig W, Schleifer KH. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev 1995; 59:143–169.
10. Rigottier-Gois L, Le Bourhis AG, Gramet G, et al. Fluorescent hybridisation combined with flow cytometry and hybridisation of total RNA to analyse the composition of microbial communities in human faeces using 16S rRNA probes. FEMS Microbiol Ecol 2003; 43:237–245.
11. Rigottier-Gois L, Rochet V, Garrec N, et al. Enumeration of Bacteroides species in human faeces by fluorescent in situ hybridisation combined with flow cytometry using 16S rRNA probes. Syst Appl Microbiol 2003; 26:110–118.
12. Zoetendal EG, Ben-Amor K, Harmsen HJM, et al. Quantification of uncultured Ruminococcus obeum-like bacteria in human fecal samples by fluorescent in situ hybridization and flow cytometry using 16S rRNA-targeted probes. Appl Environ Microbiol 2002; 68:4225–4232.
13. Harmsen HJM, Wildeboer-Veloo ACM, Raangs GC, et al. Analysis of intestinal flora development in breast-fed and formula-fed infants by using molecular identification and detection methods. J Pediatr Gastroenterol Nutr 2000; 30:61–67.
14. Favier CF, Vaughan EE, De Vos WM, et al. Molecular monitoring of succession of bacterial communities in human neonates. Appl Environ Microbiol 2002; 68:219–226.
15. Hopkins MJ, Macfarlane GT, Furrie E, et al. Characterisation of intestinal bacteria in infant stools using real-time PCR and northern hybridisation analyses. FEMS Microbiol Ecol 2005; 54:77–85.
16. Penders J, Thijs C, Vink C, et al. Factors influencing the composition of the intestinal microbiota in early infancy. Pediatrics 2006; 118:511–521.
17. Penders J, Vink C, Driessen C, et al. Quantification of Bifidobacterium spp., Escherichia coli and Clostridium difficile in faecal samples of breast-fed and formula-fed infants by real-time PCR. FEMS Microbiol Lett 2005; 243:141–147.
18. Sakata S, Tonooka T, Ishizeki S, et al. Culture-independent analysis of fecal microbiota in infants, with special reference to Bifidobacterium species. FEMS Microbiol Lett 2005; 243:417–423.
19. Lay C, Rigottier-Gois L, Holmstrom K, et al. Colonic microbiota signatures across five northern European countries. Appl Environ Microbiol 2005; 71:4153–4155.
20. Mueller S, Saunier K, Hanisch C, et al. Differences in fecal microbiota in different European study populations in relation to age, gender, and country: a cross-sectional study. Appl Environ Microbiol 2006; 72:1027–1033.
21. Rochet V, Rigottier-Gois L, Beguet F, et al. Composition of human intestinal flora analysed by fluorescent in situ hybridisation using group-specific 16S rRNA-targeted oligonucleotide probes. Genet Select Evol 2001; 33:S339–S352.
22. Lay C, Sutren M, Rochet V, et al. Design and validation of 16S rRNA probes to enumerate members of the Clostridium leptum subgroup in human faecal microbiota. Environ Microbiol 2005; 7:933–946.
23. Amann RI, Krumholz L, Stahl DA. Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J Bacteriol 1990; 172:762–770.
24. Wallner G, Amann R, Beisker W. Optimizing fluorescent in situ hybridization with ribosomal-RNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry 1993; 14:136–143.
25. Fallani M, Rigottier-Gois L, Aguilera M, et al. Clostridium difficile and Clostridium perfringens species detected in infant faecal microbiota using 16S rRNA targeted probes. J Microbiol Methods 2006; 67:150–161.
26. Langendijk PS, Schut F, Jansen GJ, et al. Quantitative fluorescence in situ hybridization of Bifidobacterium spp with genis specific 16S ribosomal-RNA targeted probes and its application in fecal samples. Appl Environ Microbiol 1995; 61:3069–3075.
27. Manz W, Amman R, Ludwig W, et al. Application of a suite of 16S rRNA-specific probes designed to investigate bacteria of the phylum cytophaga-flavobacter-bacteroides in the natural environment. Microbiology UK 1996; 142:1097–1106.
28. Sghir A, Gramet G, Suau A, et al. Quantification of bacterial groups within human fecal flora by oligonucleotide probe hybridization. Appl Environ Microbiol 2000; 66:2263–2266.
29. Franks AH, Harmsen HJM, Raangs GC, et al. Variations in bacterial populations in human feces measured by fluorescent in situ hybridization with group specific 16S rRNA-targeted oligonucleotide probes. Appl Environ Microbiol 1998; 64:3336–3345.
30. Harmsen HJM, Elfferich P, Schut F, et al. A 16S rRNA-targeted probe for detection of Lactobacilli and Enterococci in faecal samples by fluorescent in situ hybridization. Microb Ecol Health Dis 1999; 11:3–12.
31. Harmsen HJM, Wildeboer-Veloo ACM, Grijpstra J, et al. Development of 16S rRNA-based probes doe the Coriobacterium group and the Atopobium cluster and their application for enumeration of Coriobaceteriaceae in human feces from volunteers of different age groups. Appl Environ Microbiol 2000; 66:4523–4527.
32. Tappin DM, Mackenzie JM, Brown AJ, et al. Comparison of breastfeeding rates in Scotland in 1990–1 and 197–8. BMJ 2001; 322:1335–1336.
33. Alder EM, Williams FLR, Anderson AS, et al. What influences the timing of introduction of solid foods to infants. Br J Nutr 2004; 92:527–531.
34. Mikkelsen A, Rinneljungquist L, Borres MP, et al. Do parents follow breast feeding and weaning recommendations given by pediatric nurses? A study with emphasis on introduction of cows milk protein in allergy risk families. J Pediatr Health Care 2007; 21:238–244.
35. Grjibovski AM, Ehrenblad B, Yngve A. Infant feeding in Sweden: sociodemographic determinants and associations with adiposity in childhood and adolescence. Int Breastfeed J 2008; 3:23.
36. Alm JS, Swartz J, Bjorksten B, et al. An anthroposophic lifestyle and intestinal microflora in infancy. Pediatr Allergy Immunol 2002; 13:402–411.
37. Benno Y, Sawada K, Mitsuoka T. The intestinal microflora of infants: composition of fecal flora in breast-fed and bottle-fed infants. Microbiol Immunol 1984; 28:975–986.
38. Lundequist B, Nord CE, Winberg J. The composition of the faecal microflora in breastfed and bottle fed infants from birth to eight weeks. Acta Paediatr Scand 1985; 74:45–51.
39. Stark PL, Lee A. The microbial ecology of the large bowel of breast-fed and formula-fed infants during the first year of life. J Med Microbiol 1982; 15:189–203.
40. Dore J, Sghir A, Hannequart-Gramet G, et al. Design and evaluation of a 16S rRNA-targeted oligonucleotide probe for specific detection and quantitation of human faecal Bacteroides populations. Syst Appl Microbiol 1998; 21:65–71.
41. Martin F, Savage SAH, Parrett AM, et al. Investigation of bacterial colonization of the colon in breast-fed infants using novel techniques. Proc Nutr Soc 2000; 59:64A.
42. Adlerberth I, Carlsson B, de Man P, et al. Intestinal colonization with Enterobacteriaceae in Pakistani and Swedish hospital-delivered infants. Acta Paediatr Scand 1991; 80:602–610.
43. Bennet R, Eriksson M, Tafari N, et al. Intestinal bacteria of newborn Ethiopian infants in relation to antibiotic treatment and colonisation by potentially pathogenic gram-negative bacteria. Scand J Infect Dis 1991; 23:63–69.
44. Sepp E, Julge K, Vasar M, et al. Intestinal microflora of Estonian and Swedish infants. Acta Paediatr 1997; 86:956–961.
45. Adlerberth I, Hanson LA, Wold AE. The ontogeny of the intestinal flora. In: Sanderson IR, Walker WA, editors. Development of the Gastrointestinal Tract. Hamilton, Canada: B.C. Decker; 1999.
46. Conway P. Development of intestinal microbiota. In: Mackie RI, White BA, Isaacson RE, editors. Gastrointestinal Microbiology. New York: Chapman & Hall; 1997. pp. 3–38.
47. Tannock GW. The acquisition of the normal microflora of the gastrointestinal tract. In: Gibson SAW, editor. Human health: the contribution of microorganisms. London: Springer-Verlag; 1994. p. 1–16.
48. Orrhage K, Nord CE. Factors controlling the bacterial colonization of the intestine in breastfed infants. Acta Paediatr 1999; 88:47–57.
49. Gronlund MM, Lehtonen OP, Eerola E, et al. Fecal microflora in healthy infants born by different methods of delivery: permanent changes in intestinal flora after cesarean delivery. J Pediatr Gastroenterol Nutr 1999; 28:19–25.
Keywords:

16S rRNA probes; flow cytometry; fluorescent in situ hybridization; geography; infant fecal microbiota; infant feeding

Copyright 2010 by ESPGHAN and NASPGHAN