ORIGINAL ARTICLE: GASTROENTEROLOGY

Bacterial Community Structure Associated With Elective Cesarean Section Versus Vaginal Delivery in Chinese Newborns Dong Liu, Jialin Yu, Luquan Li, Qing Ai, Jinxing Feng, Chao Song, and Hongdong Li

See ‘‘Delivery Mode Shaped the Gut Microbiome in Chinese Newborns’’ by Yang and Hu on page 149.

ABSTRACT Objectives: Increasing attention is being paid to the potential for cesarean birth to influence the taxa of the bacteria that compose the infant intestinal microbiota. The present study characterized the diversity of the intestinal microbiota in newborn infants delivered vaginally (VD) or by cesarean section (CD). Methods: A cross-sectional study was performed using fecal specimens collected on days 2 and 4 of postnatal life from 25 VD infants and 16 CD infants. Profiles of the fecal microbiota were analyzed using polymerase chain reaction (PCR)–denaturing gradient gel electrophoresis in combination with 16S ribosomal RNA (rRNA) gene sequencing of the clones corresponding to the degenerating gradient gel electrophoresis (DGGE) bands. Results: On days 2 and 4 of postnatal life, VD and CD infants did not differ in the richness and evenness of the fecal bacterial community; however, the taxa of the fecal microbiota were significantly different between the 2 groups. In VD infants, Escherichia coli, Bacteroides sp, and Bifidobacterium longum were the dominant microbes. In CD infants, Staphylococcus sp, Clostridium sp, Enterobacter sp, and Streptococcus sp were more common. Conclusions: These results demonstrate that delivery method has a profound influence on the structure of the intestinal microbiota in Chinese newborn infants. This is in accordance with data reported in other regions. Key Words: birth mode, denaturing gradient gel electrophoresis, fecal microbiota, newborn infant, sequencing

(JPGN 2015;60: 240–246)

Received December 17, 2013; accepted October 10, 2014. From the Department of Neonatology, the Ministry of Education Key Laboratory of Child Development and Disorders, the Key Laboratory of Pediatrics in Chongqing, and Chongqing International Science, and the Technology Cooperation Center for Child Development and Disorders, Children’s Hospital, Chongqing Medical University, Chongqing, People’s Republic of China. Address correspondence and reprint requests to Jialin Yu, Department of Neonatology, Children’s Hospital, Chongqing Medical University, Chongqing 400014, People’s Republic of China (e-mail: yujialin486@ hotmail.com). The study was supported by the National Natural Science Foundation of China (No. 81370744). The authors report no conflicts of interest. Copyright # 2015 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition DOI: 10.1097/MPG.0000000000000606

T

he rising rate of cesarean sections during the last decade in many countries, including China, has raised concerns about the short- and long-term consequences of this mode of delivery on human health (1). Cesarean section was demonstrated to correlate with the development of allergic diseases such as asthma and allergic rhinitis (2). The occurrence of these conditions is believed to involve aberrant development of the immune system (3). The gut microbiota is known to contribute in key ways to the maturation of the human immune system, and could play an important role in the development of immune-associated conditions (4). Thus, the development of ‘‘normal’’ intestinal microbiota early in life could be a major component of future health outcomes (3). To date, the relation between the infant intestinal microbiota and mode of delivery has been characterized using bacteria cultivation–dependent technology and molecular methods (5–11). Early molecular approaches were generally based on amplifying specific bacteria species (pathogen specific). Current molecular methods are able to identify fastidious or nonculturable bacteria that cannot be detected using traditional culture methods. Although previous work has provided valuable insights into patterns of infant microbiome development, considerable discrepancies remain between studies. These differences may result from variations in the infants enrolled, the sampling time, or the methods used. For example, based on culturing methods, Gronlund et al found that vaginally delivered (VD) infants were more often colonized with Bacteroides fragilis than cesarean-delivered (CD) infants (5). In other studies using the same methods, Escherichia coli was observed to be more prevalent in VD infants (6,7); when molecular approaches were used, the dominant bacteria in VD infants were Bifidobacterium species (8–10). More recently, sequencing the 16S ribosomal RNA (rRNA) gene (or 16S ribosomal DNA [rDNA]), which exists in eubacteria, has allowed investigators to detect the diverse variable region of this gene sequence in different bacteria using an unbiased approach. In a study using a clone library of 16S rDNA, the dominant bacteria found in VD infants were Acinetobacter sp, Bifidobacterium sp, and Staphylococcus sp, whereas Citrobacter sp, E coli, and Clostridium difficile were more common in CD infants (12). Using this technique, Dominguez-Bello et al investigated the microbiota of the meconium of newborn infants with pyrosequencing technology. They showed a higher presence of Lactobacillus, Prevotella, and Sneathia spp in the meconium of VD neonates, and Staphylococcus, Corynebacterium, and Propionibacterium spp in the meconium of CD neonates (13). Another study based on pyrosequencing indicated that Escherichia, Shigella, and Bacteroides spp were underrepresented in CD infants (14). The diversity of the intestinal microbiota has been reported to be influenced with the mode of delivery (12–14). ‘‘Diversity’’ provides details about the richness and evenness of the bacterial microbiota, and its composition. An attempt to describe the diversity of the microbiota was made using pyrosequencing in newborns

240 JPGN
Volume 60, Number 2, February 2015 Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.

JPGN




Volume 60, Number 2, February 2015

in South America (13). Although the present study is a pivotal study, it is likely that diversity analyses will vary depending on the hospital setting; therefore, region-specific investigations are warranted. Degenerating gradient gel electrophoresis (DGGE), which is used to detect variations in genes, may be used to analyze the diversity of the fecal microbiota by detecting the variation in 16S rRNA genes from different bacteria. This technique is widely used to study microbial ecology and evaluate the diversity of a bacterial community. In combination, 16S rDNA polymerase chain reaction (PCR), DGGE, and sequencing may provide details of a bacterial community, including the taxa of the bacteria. Thus, in the present study, we used 16S rDNA PCR-DGGE and sequencing to describe the influence of the mode of delivery on the diversity of the intestinal microbiota during early colonization in full-term Chinese infants.

METHODS Patients The study protocol was approved by the institutional review board of the Children’s Hospital, Chongqing Medical University. Informed written consent was obtained from the parents or guardians of the infants enrolled. A total of 41 fecal specimens were collected from full-term infants from Chongqing Maternal and Child Care Service Centre, including 25 VD infants and 16 CD infants. All of the cesarean sections were elective for personal reasons, and the mothers were prophylactically administered flucloxacillin before the operation. VD and CD infants were matched for factors known to affect the intestinal microbiota, including gestational age and feeding type. Information about the gestational age, body length, head circumference, and birth weight of these infants was collected. Infants who were born premature, had active infections, received antibiotics, or had anatomical or physiological anomalies of the intestine were excluded from the study.

Sample Collection

Bacteria Associated With Delivery Method in Chinese Newborns

DGGE Analysis DNA was electrophoresed on polyacrylamide gels with a denaturing gradient ranging from 35% to 65% in 1 Tris-acetic acid-EDTA buffer at 608C for 16 hours at 85 Vusing DCode Universal Mutation System (Bio-Rad, Hercules, CA) according to the manufacturer’s instructions. On day 2, 5 of the 14 VD infants were used as a lane marker to provide a reliable between-gel comparison of VD and CD infants. These samples were loaded on both gels to adjust for migration variation. After electrophoresis, the gels were incubated in 1 Tris-acetic acid-EDTA containing SYBR Green I (Biotech, Shanghai, China) for 30 minutes and scanned using a Benchtop 3UV Transilluminator (UVP, Inc, Upland, CA). Visible bands on polyacrylamide gels were excised with sterile blades, placed in 30 mL of sterile water, and incubated overnight at 48C. Each sample was analyzed at least twice using PCR-DGGE.

TA Cloning To identify the bacterial species corresponding to the DGGE bands, the DNA recovered from the excised DGGE band was repurified and TA cloning was conducted. Extracted DNA from the DGGE bands was used for PCR amplification following the procedures described above. PCR products were purified by electrophoresis with 2% agar gel. The bands on the gel were excised and DNA was recovered using the Gel DNA Extraction Kit (Takara) according to the manufacturer’s instructions. A total of 2 mL of the DNA dissolved in 30 mL of Solution I was used to conduct TA cloning using the pMD 18-T Vector Kit (Takara) according to the manufacturer’s instructions. Plasmids containing PCR amplicons were transferred to DH5a competent cells (E coli) (Tiangen, China) and cultured on ampicillin-resistant lysogeny broth (LB) media overnight at 378C.

Confirmation of the Cloned DNA by Colony PCR

Genomic DNA was extracted from stool samples using QIAamp Stool Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. In the heating step, the lysis temperature was increased to 958C for 5 minutes to enhance cell wall breakdown of Gram-positive bacteria. A total of 100 mL of whole genomic DNA (including bacteria genomic DNA) dissolved in Tris-EDTA buffer was collected and stored at 208C.

Clones were picked out from the ampicillin-resistant LB media and blended in 10 mL sterile water. Colony PCR was performed to confirm the cloned DNA using the following primers: M13-RV (50 -CAGGAAACAGCTATGAC-30 ) and M13-M3 (50 -GTAAAACGACGGCCAGT-30 ). The PCR solution was prepared with 1 mL template (the monoclone blended in sterile water), 1 mL of each primer (10 mM), and 12.5 mL Premix Taq (Takara), and adjusted with sterile water to a final volume of 25 mL. The PCR solution was incubated at 948C for 7 minutes. Subsequently, 30 cycles of PCR were performed consisting of denaturation: 30 seconds at 948C; annealing: 30 seconds at 608C; and extension: 30 seconds at 728C. After the final cycle, the solution was incubated for 5 minutes at 728C. Agarose gel electrophoresis was carried out to confirm the length of the amplicons.

PCR Amplification

Sequencing

The variable V3 region of the 16S rRNA gene was amplified with the following primer sets: 357f (50 -CCTACGGGAGGCAGCAG-30 ) and 518r (50 -ATTACCGCGGCTGCTGG-30 ). A 40-bp GC clamp was attached to the 50 end of the 357f primer to prevent complete separation of PCR amplicons during DGGE analysis. The PCR mixture was prepared with 25 mL of Premix Taq (Takara, Kyoto, Japan), 1 mL of each primer (10 mM), and 5 mL of DNA extract, and adjusted with sterile water to a final volume of 50 mL. PCR amplifications were performed using a PCR amplifier (Eppendorf, Hamburg, Germany) and touchdown PCR, as previously described (15).

After confirmation that the V3 region of the bacterial 16S rDNA had been successfully inserted, each clone was incubated in 1 mL of liquid ampicillin-resistant LB media overnight at 38C and sent to Sangon Biotech for sequencing. The sequencing was performed with M13þ (47) primers on an ABI 3730xl DNA Analyzer using BigDye Terminator version 3.1 (ABI, Abilene, TX). The sequences were analyzed and a Basic Local Alignment Search Tool (BLAST) on National Center for Biotechnology Information was performed (http://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Web& PAGE_TYPE=BlastHome) for sequence alignments and percentage identity.

Infant fecal samples were collected in sterile collection tubes immediately after defecation on days 2 and 4 of postnatal life, and stored at 808C until processed.

DNA Extraction

www.jpgn.org

241

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.

Yu et al

JPGN

Analysis of Microbial Profiles DGGE profiles were analyzed using Quantity One software (Bio-Rad); a 2% tolerance was selected when generating gel comparisons. Diversity indices were calculated using Bio-Dap software (http://nhsbig.inhs.uiuc.edu/wes/populations.html). Shannon index (H0 ) and Simpson index (D) were used to assess the richness and dominance of the bacterial taxa demonstrated by the DGGE profiles. The Shannon evenness index (E) was used to evaluate the equitability of the microbiota. The Dice coefficient was calculated by Quantity One to assess pairwise community similarities. Measurements of these diversity indices have been previously described (16,17). Dendrograms based on the Dice coefficients were constructed using the unweighted pair group method with arithmetic averages (UPGMA) using Quantity One software (Bio-Rad).

Statistical Analysis The results of clinical characteristics and diversity indices were analyzed using the independent t test. The prevalence of bacteria in VD and CD groups was compared using Fisher exact test. Normality of the distribution was investigated using the Kolmogorov-Smirnov test before applying parametric tests. Analyses were performed using SPSS version 19 (IBM SPSS Statistics, Armonk, NY); P < 0.05 was considered statistically significant.

RESULTS Clinical Characteristics A total of 41 fecal samples were collected: for full-term VD infants (n ¼ 20), samples were collected on day 2 (n ¼ 14) and day 4 (n ¼ 6) of postnatal life; for full-term CD infants (n ¼ 21), samples were collected on day 2 (n ¼ 11) and day 4 (n ¼ 10) of postnatal life. Differences in the duration of hospital stay (2 days for VD infants; 3–4 days for CD infants) made it difficult to continuously collect fecal samples from VD infants from day 2 to 4. Given that the surrounding environment could affect the development of the intestinal microbiota, infants who were discharged from the hospital before obtaining fecal samples were not enrolled in the present study. For this reason, the fecal samples collected reflect a crosssectional sampling of different groups of newborns. Thus, the results of the present study do not contain any analysis of longitudinal changes in the intestinal microbiota over time. The clinical characteristics of the infants enrolled in the present study are shown in Table 1. There were no statistically significant differences between VD and CD infants in sex, gestational age, body length, head circumstances, birth weight, and




Volume 60, Number 2, February 2015

feeding method. All of the infants enrolled in the present study received mixed feeding (ie, a combination of breast-feeding and formula feeding) during the first 4 days of postnatal life.

DGGE Fingerprints The DGGE banding patterns of fecal samples collected from VD infants (Figs. 1A and 2A, lanes 1–5) and CD infants (Fig. 2A, lanes 6–16) are shown on day 2. The independent DGGE profiles collected on day 2 from 14 VD infants were compared on a single DGGE gel (Fig. 1A). The mean number of bands in the DGGE profiles of samples collected on day 2 was 14  5 bands in VD infants and 11  3 bands in CD infants. The difference in band number was not statistically significant (P ¼ 0.529). Subsequently, the DGGE banding patterns of VD and CD infants from day 4 were compared (Fig. 3A). The mean number of bands in the DGGE profiles collected on day 4 was 11  1 in VD infants and 15  2 in CD infants. The difference in band number was not statistically significant (P ¼ 0.366) (Table 2).

Diversity Analysis of the Intestinal Microbiota The diversity indices of the DGGE profiles of the fecal samples collected from the 2 groups of infants on days 2 and 4 are shown in Table 2. The difference in the mean values of these diversity indices was compared between the groups using the independent t test. No significant difference in the diversity indices was observed, which indicated that bacterial diversity was not disrupted by the mode of delivery.

Diversity Analysis and Phylogenetic Analysis of the Intestinal Microbiota The similarity of bacterial taxa among individual infants and dissimilarity between groups are shown in the form of dendrograms, which were constructed using the UPGMA method based on Dice coefficients (Figs. 1B, 2B, and 3B). In VD infants, on day 2, the relatedness of the Dice coefficients ranged from 0% to 68.6% with an average value of 28.0% (Fig. 1B). In a pooled group of 5 VD and 11 CD infants, DGGE profiles on day 2 were similar (Fig. 2B). The Dice coefficients of the CD infants ranged from 0% to 61.9% with an average value of 28.6% (Fig. 2B). On day 4, the DGGE fingerprint was similar between VD and CD infants, with Dice coefficients ranging from 21.2% to 63.7% in VD infants and 27.6% to 83.5% in CD infants (Fig. 3B). Infants are located in 1 cluster of the dendrogram because of high Dice coefficients; this indicated similar band patterns between them. Two main clusters were identified in Figure 1B: v2, v3, v6,



TABLE 1. Clinical characteristics of infants enrolled

Day 2 Characteristics n (boys/girls) Gestational age, day Body length, cm Head circumstances, cm Birth weight, g Mixed feeding, n

Day 4

VD

CD

VD

CD

14 (8/6) 278.0  7.9 50.3  1.0 34.3  0.6 3301.0  329.7 14

11 (7/4) 272.8  6.8 50.1  2.5 34.5  0.9 3364.5  501.3 11

6 (3/4) 274.7  6.7 50.2  1.3 34.3  1.3 3191.7  226.9 6

10 (6/4) 273.4  8.4 49.0  1.8 34.2  1.5 3192.6  401.7 10

CD ¼ cesarean-delivered infants; SD ¼ standard deviation; VD ¼ vaginally delivered infants.  Values are mean  SD or n.

242

www.jpgn.org

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.

JPGN




Volume 60, Number 2, February 2015

Bacteria Associated With Delivery Method in Chinese Newborns A

A v1

v2 v3

v4 v5

v9 v10 v11 v12 v13 c1 c2 c3 c4 c5 c6 c7 c8 c9 c10 c11

v6 v7 v8 v9 v10v11 v12 v13 v14

1 11

2 5

4

6 7 3 8 9 10

B 0.21

0.40

0.60

0.80

1.00 c2

B

0.18

0.40

0.60

0.80

1.00

c7 c3

v14 v5

c6 v13 c10

v1

c5

v4

v10

v8 v7 v2

v11 v9 c1 v12

v12

c4

v6

c11

v11 v9 v10 v3

FIGURE 1. A, DGGE analysis of fecal samples collected on day 2 from vaginally delivered (VD) infants. The alphanumeric infant identifier is shown across the top of the lanes. ‘‘v’’ refers to infants enrolled in the VD group. B, Dendrogram constructed using the UPGMA clustering method based on Dice coefficients. The scale at the top represents similarity (1.00 ¼ 100% similar).

v9, v11, and v12 were located in 1 cluster and v1, v4, v7, v8, and v13 were located in another. The low similarity of DGGE profiles from v14 and v5 compared with those from other infants made it difficult to locate them in either cluster. In Figure 2B, the VD infants v12 and v13 are coclustered with infants c1 to c11 from the CD group; however, 3 of the 5 VD infants (v9–v11) were clustered into 1 group that was distinct from the CD infants. The similarity value between the VD and CD infants was only 21%. Although a marked difference in DGGE band patterns was observed between the 2 groups of infants, the band patterns within the CD group were also greatly variable on day 2. www.jpgn.org

c9

v13

c8

FIGURE 2. A, DGGE analysis of fecal samples collected on day 2 from vaginally delivered (VD) and cesarean-delivered (CD) infants. The alphanumeric infant identifier is shown across the top of the lanes. ‘‘v’’ refers to infants in the VD group and ‘‘c’’ refers to infants in the CD group. The bands marked with Arabic numerals to their left side were excised and sequenced. The sequencing results are shown in Table 3. B, Dendrogram constructed using the UPGMA clustering method based on Dice coefficients. The scale at the top represents similarity (1.00 ¼ 100% similar). The serial numbers of the CD group are highlighted to clearly distinguish them from the VD group.

Marked differences were observed between VD and CD infants on day 4, with 4 of the 6 VD infants clustered into 1 group that was distinct from the CD infants. The similarity value between the VD and CD infants was only 29%, which was consistent with the results of day 2. Taken together, these results indicate that DGGE fingerprint patterns are greatly dissimilar between the infants delivered via VD or CD; however, there was a high degree (60%) of intragroup similarity within the VD and CD groups. Therefore, the composition of microbiota in VD infants more closely resembled that in other VD infants instead of that in CD infants, who were more likely to be colonized with a different type of bacterial taxon.

243

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.

Yu et al

A

JPGN v1’ v2’ v3’ v4’ v5’ v6’ c1’ c2’ c3’ c4’ c5’ c6’ c7’ c8’ c9’ c10’

3 4 5 2 1 13 6

7 8 9 10 11 12

B

0.29

0.40

0.50

0.60

0.70

0.80

1.00 v3’




Volume 60, Number 2, February 2015

Sequencing Results Sequence identities derived from DGGE bands were compared between the VD and CD infants on day 2 (Table 3) and day 4 (Table 4). The band size was 234 bp, and an average of 8 clones from each band were sequenced. On day 2, sequence analysis from the 16S rDNA of the DGGE bands suggested a significantly higher prevalence of E coli (13/14 vs 4/11) and Bacteroides sp (including B fragilis and Parabacteroides distasonis) (13/14 vs 5/11) in fecal samples of VD infants as compared with in fecal samples of those born via cesarean section (P < 0.05). In contrast, Staphylococcus sp (8/11 vs 0/14, including S epidermidis), Clostridium sp (5/11 vs 0/ 14), and Enterobacter sp (5/11 vs 0/14) were more common in CD infants (P < 0.01). Streptococcus sp (4/11) was observed in the stool specimens of both VD and CD infants on day 2, and its prevalence was similar between the 2 groups. According to the sequencing results obtained from a separate group of infants on day 4, VD infants were more often colonized with Bifidobacterium sp (5/6 vs 0/10) and Bacteroides sp (4/6 vs 1/ 10), whereas Clostridium sp (7/10 vs 0/6) and Streptococcus sp (7/ 10 vs 0/6) were more prevalent in CD infants (P < 0.05). Lactobacillus sp, Enterococcus sp, Veillonella sp, and Neisseria mucosa strain were found at similar frequencies in both groups of infants. It is noteworthy that E coli, which was common on day 2, especially in VD infants, was not found on day 4.

v6’

DISCUSSION

v2’ v1’ v5’ v4’ c2’ c1’ c8’ c6’ c9’ c7’ c5’ c3’ c10’ c4’

FIGURE 3. A, DGGE analysis of fecal samples collected on day 4. The infants enrolled at day 4 are not the same as those enrolled at day 2. The alphanumeric infant identifier is shown across the top of the lanes. ‘‘v’’ refers to infants in the vaginal delivery (VD) group and ‘‘c’’ refers to infants in the cesarean delivery (CD) group. The bands marked with Arabic numerals to their left side were excised and sequenced. The sequencing results are shown in Table 3. B, Dendrogram constructed using the UPGMA clustering method based on Dice coefficients. The scale at the top represents similarity (1.00 ¼ 100% similar). The serial numbers of the CD group are highlighted to clearly distinguish them from the VD group.

The results of the present study demonstrate that in Chinese infants, the mode of delivery has a profound impact on the intestinal microbiota. Although there was a similar level of microbial diversity in VD and CD infants, the composition of the microbiota was changed by the mode of delivery at days 2 and 4 of postnatal life. DGGE fingerprints of the newborn infants were characterized by several bands, demonstrating some diversity of their fecal samples, which is different from the results of a study by Biasucci et al (10). The primers used for PCR amplification may be the main cause of this discrepancy. The present study used primers corresponding to the V3 region of the 16S rRNA gene. Such primers have been reported to produce more informative DGGE profiles than primers corresponding to the V6–V8 region, which were used by Biasucci et al (10). The amplicon of theV3 region is shorter than that of the V6–V8 region; however, we chose to use primers corresponding to the V3 region, given the satisfactory sequencing results from the amplicons of the V3 region and the deficiency of the V6–V8 region in the DGGE analysis. Information describing the impact of mode of delivery on the diversity of the microbiota in newborn infants has been reported elsewhere using different methodology (13,14). In the present study, diversity indices such as the Shannon index (H0 ), Shannon evenness (E), Simpson index (D), and Dice coefficient were applied

TABLE 2. Diversity indices of DGGE profiles from VD and CD infants on days 2 and 4



Day 2 Index Shannon index (H0 ) Shannon evenness (E) Simpson index (D) Dice index (%) Diversity (bands)

Day 4

VD

CD

P

VD

CD

P

1.94  0.29 0.95  0.25 0.15  0.05 28.0  15.5 14  5

2.30  0.25 0.97  0.15 0.11  0.28 28.6  15.2 11  3

2.544 1.691 2.559 0.231 0.529

2.32  0.12 0.98  0.08 0.10  0.01 45.5  11.4 11  1

2.64  0.14 0.99  0.05 0.07  0.10 50.2  13.5 15  2

4.682 0.096 4.833 1.195 0.366

CD ¼ cesarean-delivered infants; SD ¼ standard deviation; VD ¼ vaginally delivered infants.  Values are mean  SD.

244

www.jpgn.org

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.

JPGN




Volume 60, Number 2, February 2015

Bacteria Associated With Delivery Method in Chinese Newborns



TABLE 3. Sequencing results of DGGE bands Bandy

NCBI BLAST result

Identity, %

D2–1 D2–2 D2–3 D2–4 D2–5 D2–6 D2–7 D2–8 D2–9 D2–10 D2–11 D4–1 D4–2 D4–3 D4–4 D4–5 D4–6 D4–7 D4–8 D4–9 D4–10 D4–11 D4–12 D4-13

Bacteroides fragilis Bacteroides sp Escherichia coli Parabacteroides distasonis Staphylococcus sp Uncultured Streptococcus sp Enterobacter sp Enterobacter sp Uncultured Clostridium sp Uncultured Clostridium sp Staphylococcus sp Uncultured Lactobacillus sp Uncultured Enterococcus sp Bifidobacterium longum Uncultured Bacteroides sp Uncultured bacterium clone Uncultured bacterium clone Veillonella sp Veillonella sp Veillonella sp Streptococcus sp Clostridium sp Uncultured Clostridium sp Neisseria mucosa strain

97 98 100 100 100 100 100 100 99 99 99 100 99 100 100 100 100 100 99 99 100 100 100 99

BLAST ¼ Basic Local Alignment Search Tool; NCBI ¼ National Center for Biotechnology Information.  Sequencing results from DGGE fingerprints of fecal specimens on days 2 and 4 are given. y D2 ¼ day 2; D4 ¼ day 4. The number indicates the band excised from the gel and is shown in Figs. 2A and 3A, respectively.

to evaluate the infant microbiota from the perspective of richness, equitability, and similarity. In contrast to previous reports, no significant differences were observed between VD and CD groups (13,14); however, the 16S rDNA PCR-DGGE method used in the TABLE 4. Dominant intestinal flora in stool samples obtained on days 2 and 4 from VD and CD infants Prevalence of positive samples Day 2

Organisms Bacteroides sp Escherichia coli Staphylococcus sp Streptococcus sp Enterobacter sp Clostridium sp Lactobacillus sp Enterococcus sp Bifidobacterium longum Veillonella sp Neisseria mucosa strain

VD (n ¼ 14)

Day 4

CD (n ¼ 11)

VD (n ¼ 6)

4 5

4 0 0 0 0 0 4 2  5 4 3



11  13 0 1 0 0 0 0 0 0 0



8

4  5  5 0 0 0 0 0



CD (n ¼ 10) 1 0 0  7 0  7 4 6 0 10 9

Some bacteria were classified in the genus, for example, Parabacteroides distasonis and Bacteroides fragilis were classified in Bacteroides sp. CD ¼ cesarean-delivered infants; VD ¼ vaginally delivered infants.    P < 0.05; P < 0.01; P < 0.001.

www.jpgn.org

present study can only provide a profile of the dominant microbiota (approximately >1% relative abundance) (18), which may limit the generalizability of these data. Future studies confirming the diversity of the microbiota in these infants with more accurate technology, such as pyrosequencing, are warranted. During vaginal delivery, the maternal microbiota, including E coli, Bacteroides sp, and Bifidobacterium sp, may be the main source of the infant microbiota because defecation during vaginal delivery is common. Facultative anaerobes are the first to colonize the infant intestine and create an anaerobic environment for subsequent colonization by strict anaerobes (19). The present study and others (6,7) have described facultative anaerobes such as E coli (13/14) and anaerobic bacteria including Bacteroides sp in fecal samples of VD infants on day 2 of postnatal life, and the presence of Bifidobacterium sp on day 4 (5). Interestingly, Bifidobacterium sp and E coli were not detected in the same fecal samples. Bifidobacterium sp can inhibit the growth of E coli by producing a large amount of acetic acid (20). Thus, expanding Bifidobacterium sp may have caused the numbers of E coli bacterium to fall below the limit of detection of the 16S rDNA PCR-DGGE method by day 4. Unexpectedly, Lactobacillus sp, which can be found in the vagina of healthy women and transmitted in maternal feces (21), was not found in fecal samples from VD infants on day 2, and there was no significant difference between VD and CD infants on day 4. If the hypothesis that vaginal delivery provides the initial inoculum for bacteria colonization in the intestine of VD infants is supported, the presence of Lactobacillus sp should have been significantly more common in VD infants in the present study (19). Additional work is needed to elucidate how Lactobacillus sp colonizes the infant intestine and to identify the primary source of the inoculum. There were marked differences in the bacterial composition of the microbiota between the VD and CD groups. The pattern was emerging on day 2 and apparent on day 4 of postnatal life. In the present study and others, facultative anaerobes such as Staphylococcus sp and Enterobacter sp were more prevalent in the intestinal microbiota of CD infants (7,13). Sterile techniques are used during cesarean sections; therefore, it is likely that bacteria spread in neonatal care units seeds the CD infant intestinal microbiota (22). Contact with parents and caregivers may be the primary source of bacteria such as Veillonella sp and a N mucosa strain of oropharynx inhabitants, which were found to be similar in both VD and CD infants on day 4 of postnatal life (23,24). Streptococcus sp was significantly more common in CD than in VD infants (25). Similarly, S mutans, which plays a major role in the etiology of dental caries, was acquired earlier in CD infants than in those born via vaginal delivery (26); however, little is known about the influence of the mode of delivery on the presence of other species of Streptococcus sp in infants. In CD infants, the colonization of Streptococcus sp appears to be correlated with the absence of Staphylococcus sp and Enterobacter sp, which is similar to the relation between E coli and Bifidobacterium sp in VD infants. It is unclear whether this results from a competitive inhibition between these bacteria (27), or is just a reflection of interindividual variability. Consistent with previous reports using culture methods and specific PCR amplification, Clostridium sp was found to be dominant in CD infants on both days 2 and 4 (5,8). Further investigation is needed to confirm and extend our observations about the composition of the infant microbiota following VD or CD birth, and to identify any temporal changes in the infant microbiota. Although the results obtained in the present study are largely in agreement with previous work, considerable controversy exists about how the mode of delivery affects the composition of the infant microbiome. A study conducted by Pandey et al demonstrated a

245

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.

Yu et al

JPGN

higher prevalence of Citrobacter sp, E coli, and C difficile in CD infants, and Acinetobacter sp, Bifidobacterium sp, and Staphylococcus sp in VD infants. They also reported that the diversity of the microbiome was more complex in CD infants (12). The differences between the present study and others are likely because of several factors. First, the results represent different infants and sampling times. It is possible to describe internally clear and consistent messages from previous studies, but it remains important to replicate these findings in diverse populations to account for regional differences (7). The present study was a small-scale, cross-sectional study of limited duration. Therefore, these findings should be interpreted as a preliminary investigation of the intestinal microbiota composition in Chinese infants. Regional differences in the hospital setting and management of deliveries (CD: emergency and elective; length of stay: on average 3 days for VD and 4 days for CD infants in the present study) may account for many of the discrepancies between data in the present study and previous reports. Second, the study techniques may account for differences in results. For example, the 16S rDNA clone library used by Pandey et al may permit only an initial survey of diversity and identification of bacterial taxa (12). The diversity may not be exhaustively sampled because of insufficient clone sequencing from the 16S rDNA clone library. For example, environmental samples such as soil may require >40,000 clones to document 50% of the microbial richness (28). Finally, unlike most previous work, which studied infants fed solely breast milk, all of the infants in our study were receiving a mix of breast milk and formula, when breast milk was insufficient (mixed feeding). In addition, because there are no contraindications for feeding in full-term infants, all of the neonates in the present study were fed on demand. In conclusion, the present study demonstrated a close relation between mode of delivery and intestinal microbiota in Chinese newborn infants. DGGE profiles of bacteria in Chinese newborns show that there are differences in community structure depending on the mode of delivery. Mode of delivery had the greatest impact on the structure of the intestinal microbiota, instead of on the diversity during the first 4 days of the infant’s life. Because the importance of elective CD is increasing in China, future long-term multicenter studies evaluating samples obtained from a larger population using more advanced technology, such as pyrosequencing, are necessary to more fully describe the impact of the mode of delivery on the intestinal microbiota of Chinese infants. Acknowledgment: We thank Ying Dong for critical comments on the article.

REFERENCES 1. Sufang G, Padmadas SS, Fengmin Z, et al. Delivery settings and caesarean section rates in China. Bull World Health Organ 2007; 85:755–62. 2. Renz-Polster H, David MR, Buist AS, et al. Caesarean section delivery and the risk of allergic disorders in childhood. Clin Exp Allergy 2005;35:1466–72. 3. Neu J, Rushing J. Cesarean versus vaginal delivery: long-term infant outcomes and the hygiene hypothesis. Clin Perinatol 2011;38:321–31. 4. Guarner F, Malagelada JR. Gut flora in health and disease. Lancet 2003;361:512–9. 5. 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. 6. Hallstrom M, Eerola E, Vuento R, et al. Effects of mode of delivery and necrotising enterocolitis on the intestinal microflora in preterm infants. Eur J Clin Microbiol Infect Dis 2004;23:463–70.

246




Volume 60, Number 2, February 2015

7. Adlerberth I, Lindberg E, Aberg N, et al. Reduced enterobacterial and increased staphylococcal colonization of the infantile bowel: an effect of hygienic lifestyle? Pediatr Res 2006;59:96–101. 8. Penders J, Thijs C, Vink C, et al. Factors influencing the composition of the intestinal microbiota in early infancy. Pediatrics 2006;118:511–21. 9. Mikami K, Takahashi H, Kimura M, et al. Influence of maternal bifidobacteria on the establishment of bifidobacteria colonizing the gut in infants. Pediatr Res 2009;65:669–74. 10. Biasucci G, Rubini M, Riboni S, et al. Mode of delivery affects the bacterial community in the newborn gut. Early Hum Dev 2010;86 (suppl 1):13–5. 11. Cilieborg MS, Boye M, Sangild PT. Bacterial colonization and gut development in preterm neonates. Early Hum Dev 2012;88 (suppl 1): S41–9. 12. Pandey PK, Verma P, Kumar H, et al. Comparative analysis of fecal microflora of healthy full-term Indian infants born with different methods of delivery (vaginal vs cesarean): Acinetobacter sp. prevalence in vaginally born infants. J Biosci 2012;37:989–98. 13. Dominguez-Bello MG, Costello EK, Contreras M, et al. Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns. Proc Natl Acad Sci U S A 2010;107:11971–5. 14. Azad MB, Konya T, Maughan H, et al. Gut microbiota of healthy Canadian infants: profiles by mode of delivery and infant diet at 4 months. CMAJ 2013;185:385–94. 15. Yu Z, Morrison M. Comparisons of different hypervariable regions of rrs genes for use in fingerprinting of microbial communities by PCR-denaturing gradient gel electrophoresis. Appl Environ Microbiol 2004;70:4800–6. 16. Dice LR. Measures of the amount of ecologic association between species. Ecology 1945;26:297–302. 17. Hill TC, Walsh KA, Harris JA, et al. Using ecological diversity measures with bacterial communities. FEMS Microbiol Ecol 2003;43:1–11. 18. Muyzer G, de Waal EC, Uitterlinden AG. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 1993;59:695–700. 19. Thompson-Chagoyan OC, Maldonado J, Gil A. Colonization and impact of disease and other factors on intestinal microbiota. Dig Dis Sci 2007;52:2069–77. 20. Bezirtzoglou E, Romond C. Occurrence of Bifidobacterium in the feces of newborns delivered by cesarean section. Biol Neonate 1990;58:247– 51. 21. Tannock GW, Fuller R, Smith SL, et al. Plasmid profiling of members of the family Enterobacteriaceae, lactobacilli, and bifidobacteria to study the transmission of bacteria from mother to infant. J Clin Microbiol 1990;28:1225–8. 22. Fryklund B, Tullus K, Berglund B, et al. Importance of the environment and the faecal flora of infants, nursing staff and parents as sources of gram-negative bacteria colonizing newborns in three neonatal wards. Infection 1992;20:253–7. 23. Knapp JS, Hook EW 3rd. Prevalence and persistence of Neisseria cinerea and other Neisseria spp. in adults. J Clin Microbiol 1988; 26:896–900. 24. Belda-Ferre P, Alcaraz LD, Cabrera-Rubio R, et al. The oral metagenome in health and disease. ISME J 2012;6:46–56. 25. Bezirtzoglou E. The intestinal microflora during the first weeks of life. Anaerobe 1997;3:173–7. 26. Li Y, Caufield PW, Dasanayake AP, et al. Mode of delivery and other maternal factors influence the acquisition of Streptococcus mutans in infants. J Dent Res 2005;84:806–11. 27. Uehara Y, Kikuchi K, Nakamura T, et al. H(2)O(2) produced by viridans group streptococci may contribute to inhibition of methicillin-resistant Staphylococcus aureus colonization of oral cavities in newborns. Clin Infect Dis 2001;32:1408–13. 28. Rastogi G, Sani RK. Molecular Techniques to Assess Microbial Community Structure, Function, and Dynamics in the Environment. Berlin: Springer; 2011.

www.jpgn.org

Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.