Advanced
Development of a Novel Long-Range 16S rRNA Universal Primer Set for Metagenomic Analysis of Gastrointestinal Microbiota in Newborn Infants
Development of a Novel Long-Range 16S rRNA Universal Primer Set for Metagenomic Analysis of Gastrointestinal Microbiota in Newborn Infants
Journal of Microbiology and Biotechnology. 2014. Jun, 24(6): 812-822
Copyright © 2014, The Korean Society For Microbiology And Biotechnology
  • Received : March 12, 2014
  • Accepted : April 08, 2014
  • Published : June 28, 2014
Download
PDF
e-PUB
PubReader
PPT
Export by style
Article
Author
Metrics
Cited by
TagCloud
About the Authors
Hye-Jin Ku
Ju-Hoon Lee
juhlee@khu.ac.kr

Abstract
Metagenomic analysis of the human intestinal microbiota has extended our understanding of the role of these bacteria in improving human intestinal health; however, a number of reports have shown that current total fecal DNA extraction methods and 16S rRNA universal primer sets could affect the species coverage and resolution of these analyses. Here, we improved the extraction method for total DNA from human fecal samples by optimization of the lysis buffer, boiling time (10 min), and bead-beating time (0 min). In addition, we developed a new longrange 16S rRNA universal PCR primer set targeting the V6 to V9 regions with a 580 bp DNA product length. This new 16S rRNA primer set was evaluated by comparison with two previously developed 16S rRNA universal primer sets and showed high species coverage and resolution. The optimized total fecal DNA extraction method and newly designed long-range 16S rRNA universal primer set will be useful for the highly accurate metagenomic analysis of adult and infant intestinal microbiota with minimization of any bias.
Keywords
Introduction
The gastrointestinal tract (GIT) of newborn infants is rapidly colonized by members of a variety of bacterial genera after delivery [4 , 27] . It has been generally reported that the intestinal microbiota performs important functions for the human host, including protection of infant intestinal health by immune stimulation [18] and prevention of infantile diarrhea [15] . The composition of the initial intestinal microbiota in infants may be simple; the major first colonizers belong to the phylum Firmicutes , and then the phyla Actinobacteria and Bacteroidetes become dominant a few weeks after delivery [20] . However, the timing and procedure of the formation of an adult-like intestinal microbiota in infants are not yet understood. It is thought that introduction of solid foods to infants may be involved in the transition of the infant intestinal microbiota to an adult-like microbiota consisting of four major phyla, Firmicutes , Bacteroidetes , Actinobacteria , and Proteobacteria [22 , 26] .
After initial development of the intestinal microbiota in infants, the main factors determining the composition of the intestinal microbiota may be the type of feeding (breastfed or formula-fed) and delivery mode (vaginal delivery or Caesarean section [C-section]) [23] . Previous studies of infant microbiotas showed that the phylum Actinobacteria (including Bifidobacterium ) is generally dominant in breastfed infants whereas the phylum Proteobacteria (including E. coli ) is dominant in formula-fed infants [10 , 33] , suggesting that human breast milk may influence the composition of the intestinal microbiota in infants. In addition, comparative analysis of fecal pH values showed that the fecal pH of breast-fed infants (<5.2) is lower than that of formula-fed infants (>7.0), probably as a result of the production of acetic and lactic acids by bifidobacteria [5] . As previously mentioned, the infant delivery mode is also an important factor determining the composition of the intestinal microbiota in infants. Interestingly, although the composition of intestinal microbiota in vaginally delivered infants is very similar to that of the mother’s vagina, the composition in infants delivered via C-section is similar to that of the mother’s skin [7] . Intensive investigations of the initial acquisition and subsequent colonization of bacteria in the infant GIT may provide significant information concerning postnatal intestinal maturation [23] . The culturing method that has traditionally been used to analyze the composition of the intestinal microbiota suggests that there are approximately 400 bacterial species in the human intestine [17] . However, this method is laborious and time-consuming and can detect only intestinal bacteria that can be cultured under the conditions used. To overcome this limitation, a nonculturing method based on the 16S rRNA gene sequencing technique has also been widely used to determine the composition of the human intestinal microbiota. This nonculturing method indicated that the human intestinal microbiota is composed of approximately 500 different bacterial species and that 75% of all detected intestinal bacteria are nonculturable [8 , 9] . Since the development of next-generation sequencing (NGS), a metagenomic approach using NGS of bacterial 16S rRNAs from human fecal samples was suggested and has provided an enormous amount of information about the microbiota composition in human gut [13 , 28] . Recently, metagenomic analysis using fecal samples from 124 European adults showed that the human intestinal microbiota consists of at least 1,150 different bacterial species and that 160 core bacterial species are shared by all individuals [28] .
Before the NGS era, 16S RNA universal primer sets targeted large regions (>1 kb) of bacterial 16S rRNA gene sequences, including the primer sets Bact-8F-Bact-1391R [9] , Bact-8F- Bact-1510R [13] , and Bact-8F-T7-1391R [26] . At that time, specific bacterial 16S rRNA regions were amplified, cloned, randomly picked, and sequenced via a shotgun sequencing approach. After the introduction of NGS, newly developed 16S rRNA universal primer sets targeted more narrow regions (<300 bp) of bacterial 16S rRNA sequences, including 8F/338R targeting the V1-V2 region [20 , 31] , 784F/1061R targeting the V5-V6 region [2] , 338F/533R targeting the V3 region, and 967F/1046R targeting the V6 region [19] , because of NGS limits (<300 bp for 454 FLX Pyrosequencer; Roche, Mannheim, Germany). However, previous studies reported a bias in the metagenomic composition of the human intestinal microbiota, probably because of inefficient bacterial DNA isolation/purification and biased coverage of 16S rRNA universal primer sets that resulted in extremely low detection rates of the phylum Actinobacteria in most cases [20 , 31] . In contrast, another study reported that the 784F/1061R 16S rRNA universal primer set preferentially detected the phylum Actinobacteria, but detected a very low composition of the phylum Bacteroidetes [2] . These findings suggest that the universal primer set is a key factor determining the detection coverage of intestinal microbiota [26] . Furthermore, although NGS technology has advanced rapidly and is now capable of sequencing up to 1 kb of DNA [14 , 25] , the 16S rRNA PCR products for NGS are still too small (generally <300 bp). Therefore, a longrange bacterial universal primer set amplifying >500 bp of 16S rRNA region should be developed to increase the resolution and accuracy of metagenomic analysis of the human intestinal microbiota.
In this study, to improve the efficiency of fecal DNA isolation, we optimized the fecal DNA isolation and purification method by modification of the general method used for isolation of fecal DNA. Furthermore, a novel longrange 16S rRNA universal primer set targeting the V6-V9 regions was developed and tested with fecal samples from infants to improve the species coverage and the resolution of metagenomic composition analysis of infant intestinal microbiotas. In addition to enhancing the metagenomic analysis of human intestinal microbiotas, this new universal primer set was also specialized for the metagenomic analysis of infant intestinal microbiotas. This new fecal DNA isolation method and the novel long-range 16S rRNA universal primer set will contribute to further studies and understanding of the composition and role of human intestinal microbiotas, especially the infant intestinal microbiota.
Materials and Methods
- Fecal Samples, Bacterial Strains, and Growth Conditions
The fecal samples used in this study were obtained from a 23-year-old healthy female and a 3-day-old male infant who was born by natural delivery in Bundang CHA Medical Center, Seongnam, South Korea. The fecal samples were stored at -80℃ for further DNA extraction. The intestinal bacterial strains used for construction of the simulated intestinal microbiota and their growth conditions are listed in Table 1 . Anaerobic conditions were generated using the BBL Anaerobic System (BD, Cockeysville, MD, USA). E. coli DH5 (Invitrogen, Carlsbad, CA, USA) was used as a competent cell for standard heat-shock transformation [29] and was grown with shaking at 37℃ in Luria-Bertani (LB) medium. The pGEM-T Easy Vector System (Promega, Madison, WI, USA) and 50 μg/ml ampicillin sulfate (Sigma, St. Louis, MO, USA; final concentration) were used for cloning and selection, respectively.
Bacterial strains used forin vitrosimulation of the human intestinal microbiota.
PPT Slide
Lager Image
acMRS, de Man-Rogosa-Sharpe medium (Difco, Detroit, MI, USA) supplemented with 0.05% L-cysteine·HCl; M17G, M17 medium supplemented with 0.5% D-glucose; LB, Luria-Bertani medium; RCM, Reinforced Clostridial Medium; NB, Nutrient medium. bATCC, American Type Culture Collection; KCTC, Korean Collection for Type Culture; DCC, Dairy Culture Collection, Food Omics Laboratory of Kyung Hee University.
- Fecal DNA Extraction and Optimization
For general extraction of total DNA from fecal samples, the QIAamp DNA Stool mini kit (Qiagen, Valencia, CA, USA) was used according to the manufacturer’s instructions. To optimize the DNA extraction from human fecal samples, we modified the manufacturer’s recommended conditions of boiling/bead-beating and the composition of lysis buffer. To compare the efficiency of bacterial DNA isolation between the commercial kit and the optimized method, two different total fecal DNA samples were prepared from the same fecal sample using the commercial kit and the optimized method, respectively. Then, each total fecal DNA sample was serially diluted up to 10 -9 with molecular biology grade water. PCR amplification was conducted with the new 16S rRNA universal primer set and serially diluted total DNA samples in the range of 10 0 to 10 -9 .
- Primer Design and Coverage
To develop a new long-range 16S rRNA universal primer set, the complete 16S rRNA sequences of 26 intestinal bacterial strains covering four major intestinal phyla, Firmicutes , Bacteroidetes , Proteobacteria , and Actinobacteria , were obtained from the GenBank database [3] ( Table 2 ). Multiple alignments were conducted using the ClustalX2 program [21] . The selected primer set and the previously reported universal primer sets (8F/338R and 784F/1061R) were synthesized and purified by Macrogen, Seoul, South Korea. To verify the primer coverage of three different 16S rRNA universal primers (8F/338R, 784F/1061R, and 926F/1505R developed in this study) in silico , 1,578 complete 16S rRNA sequences were obtained from the complete bacterial genome sequences in the RDP database [6] , and perfect matches with those 16S rRNA universal primers were counted and compared.
List of 26 major intestinal bacteria for 16S rRNA sequences.
PPT Slide
Lager Image
List of 26 major intestinal bacteria for 16S rRNA sequences.
- Bacterial DNA Extraction and In Vitro Simulation of Human Intestinal Microbiota
For extraction of bacterial DNA from pure culture, 1 ml of each bacterial cell culture was pelleted at 4ºC for 5 min at 16,000 × g and the pellet was washed twice and resuspended in 200 μl of molecular biology grade water. The bacterial cells were disrupted using a Mini-Beadbeator-16 (BioSpec, Bartlesville, OK, USA) for 30 sec with 0.15 mm sterile zirconia beads (Next Advanced, Averill Park, NY, USA), and D NA i n the supernatant was quantified using a NanoDrop 1000 spectrophotometer (Thermo Scientific, Waltham, MA, USA) and used as a template for PCR. For simulation of the human intestinal microbiota, we selected 21 major intestinal bacteria from four major phyla, Firmicutes, Bacteroidetes, Proteobacteria, and Actinobacteria. An equal amount of each bacterial DNA (100 ng) was mixed and 50 ng of the resulting mixture was used as a DNA template in PCR.
- 16S rRNA PCR, Cloning, and Capillary Sequencing
PCR was performed using a C1000 Thermal Cycler (Bio-Rad, Hercules, CA, USA). The reaction mixtures (final volume, 25 μl) consisted of 1 μl of DNA template, 0.6 μM of each primer set ( Table 3 ), 0.2 mM of dNTP mix (Promega), and 2.5 U of Taq DNA polymerase (New England Biolabs, Ipswich, MA, USA). The PCR conditions for the 926F/1505R primer set were as follows: 95℃ for 5 min; 30 cycles of 30 sec at 95℃, 30 sec at a specific annealing temperature for each primer set ( Table 3 ), 30 sec at 72℃; and 5 min at 72℃. PCR conditions for 8F/338R and 784F/1061R were as previously reported [2 , 31] . The amplified 16S rRNA PCR products produced using each primer set were purified with an AxyPrep DNA Gel Extraction Kit (Axygen, Tewksbury, MA, USA) and cloned into the pGEM-T Easy Vector for transformation into competent E. coli DH5α. Fifty colonies for simulation of the human intestinal microbiota and 100 colonies for detection and identification of infant fecal DNA were randomly picked from the selective LB agar medium, respectively. The plasmid DNA was isolated and purified from the selected transformants using a Plasmid Mini Prep Kit (Zymo Research, Irvine, CA, USA). Sequencing of the insert DNAs was conducted using sequencing primers, T7 forward primer or SP6 reverse primer, provided by Macrogen.
Universal primers targeting the 16S rRNA sequence.
PPT Slide
Lager Image
aR, A/G; H, A/T/C; Y, C/T.
- Bacterial Identification
Sequences of amplified 16S rRNA PCR products were compared with 16S rRNA sequences in the NCBI-BLAST database for bacterial identification [1] .
Results and Discussion
- Optimization of Fecal DNA Extraction
For accurate analysis of the intestinal microbiota without any bias, it is very important to obtain sufficient purified total genomic DNA from fecal samples [24 , 34] . To date, several commercial kits for extraction of bacterial DNA from fecal samples are available and their performance has been compared [24] . In this study, the QIAamp DNA Stool mini kit was found to be the most effective for total fecal DNA extraction and was therefore used for optimization of total fecal DNA extraction from a fecal sample of a 23-yearold healthy woman. Because bacterial cell lysis is a key process for optimal total fecal DNA extraction, we optimized the lysis buffer, boiling time, and bead-beating time. A previous report suggested that lysis buffer consisting of 500 mM NaCl, 50 mM Tris-HCl (pH 8.0), 50 mM EDTA (pH 8.0), and 4% sodium dodecyl sulfate (SDS) is more effective for total fecal DNA extraction than the ASL buffer provided in the kit [34] . The boiling time and bead-beating time were also optimized. After resuspension of 0.25 g of fecal sample with 1 ml of lysis buffer, the solution was incubated in boiling water for different incubation times (0, 5, and 10 min at 100℃) and treated with a bead-beater for different times (0, 0.25, 0.5, 1, 3, and 5 min). Agarose gel electrophoresis revealed that 10 min was the most effective boiling time for total fecal DNA extraction without beadbeating because the total DNA extracted from the fecal samples under these condition was less sheared ( Fig. 1 A). Agarose gel electrophoresis of total DNA after boiling for 10 min and different times of bead-beating showed that there was no difference in shearing of total fecal DNA among various bead-beating times, suggesting that beadbeating is not important for total fecal DNA extraction ( Fig. 1 B). Based on these results, the optimum conditions for pretreatment of fecal samples were 10 min incubation in boiling water and no bead-beating. Furthermore, the yield of total fecal DNA using the QIAamp Stool mini kit after this optimized pretreatment of the fecal sample showed approximately 8-fold higher yield than that without pretreatment of the sample (~120 μg vs. ~15 μg), substantiating these pretreatment conditions ( Fig. 2 ). Therefore, the efficiency of total fecal DNA extraction using the QIAamp Stool mini kit could be improved with this pretreatment, and these optimal conditions were used for subsequent total fecal DNA extractions in this study.
PPT Slide
Lager Image
Optimization of pretreatment for fecal DNA extraction from a fecal sample of a young woman. (A) Optimization of boiling time as a pretreatment condition. M, 1 kb DNA ladder (Bioneer, Daejeon, Korea); 1, no boiling; 2, 5 min; 3, 10 min. (B) Optimization of bead-beating time. M, 1 kb DNA ladder (Bioneer); 1, no bead-beating; 2, 0.25 min; 3, 0.5 min; 4, 1 min; 5, 3 min; 6, 5 min.
PPT Slide
Lager Image
Agarose gel electrophoresis of total genomic DNA extracted from the fecal samples using the manufacturer’s standard protocol of Qiagen QIAamp Stool mini kit (Lane 1) and the optimized protocol (Lane 2). M, 1 kb DNA ladder (Bioneer).
To verify the efficiency of bacterial DNA isolation between the commercial kit and the optimized method, the ranges and limits of PCR amplification with the serially diluted total fecal DNA samples were compared. Therefore, we isolated and purified two total fecal DNA samples from the same fecal sample of the 23-year-old healthy woman using the commercial kit and the optimized total DNA isolation method, respectively. To verify the amount of bacterial DNA in each total DNA sample, each DNA sample was serially diluted and PCR amplification was conducted using this novel 16S rRNA universal primer set. Interestingly, whereas specific PCR products (~580 bp) were observed up to 10 -4 dilution in the DNA sample prepared by the commercial kit, they were observed up to 10 -8 dilution in the DNA sample prepared by the optimized total DNA isolation method, substantiating that the optimized total DNA isolation method can isolate and purify a higher amount of bacterial total DNA from the same fecal sample ( Fig. 3 ). Therefore, the optimized method for isolation and purification of total DNA from fecal samples may be better than the commercial kit.
PPT Slide
Lager Image
PCR evaluation of extracted bacterial DNA amount from a fecal sample by the optimized total DNA extraction method (A) and the commercial kit (B). The size of PCR product is 580 bp. The 926F/1505R primer set and serially diluted fecal DNA samples (100 to 10-9) were used as a PCR primer set and PCR templates for amplification. M, 1 kb DNA ladder (Bioneer).
- Development of Long-Range 16S rRNA Universal Primer Set 926F/1505R
To elucidate and analyze the composition of the human intestinal microbiota, several 16S rRNA universal primer sets have been developed, such as 8F/338R targeting the V1-V2 region [20 , 31] , 338F/533R targeting the V3 region [19] , and 784F/1061R targeting the V5-V6 region [2] . Although these universal primer sets could detect a wide range of intestinal bacteria, they also showed some bias or preference for detection of specific intestinal bacteria. In addition, short PCR products and the resultant short DNA sequences may reduce the detection resolution of highly varied intestinal bacteria. For example, although the 8F/338R primer set detected various intestinal bacteria by comparison of 16S rRNA sequences at the V1-V2 region, bifidobacteria were hardly detected using this primer set, probably due to a biased preference for bacterial detection or inefficient total fecal DNA extraction [20] . In contrast, the 784F/1061R primer set detected many kinds of Actinobacteria , but hardly detected Bacteroidetes [2] , suggesting that a novel long-range 16S rRNA universal primer set should be developed to overcome this bias and low detection resolution of intestinal bacteria detection.
PPT Slide
Lager Image
Graphic diagram of conserved and variable regions in the 16S rRNA sequence of E. coli ATCC 8739. C1-C10 and V1-V9 indicate conserved and variable regions, respectively. A to J indicate the candidate regions for primer design. The exact binding locations of 926F and 1505R in the F and J regions, respectively, are indicated.
To develop a novel long-range 16S rRNA universal primer set with unbiased detection efficiency for all human intestinal bacteria, 16S rRNA sequences of 26 major human intestinal bacteria were obtained from the GenBank database ( Table 2 ). Multiple alignment of these 16S rRNA sequences revealed 9 variable (V1 to V9) and 10 conserved (C1 to C10) regions ( Fig. 4 ). Among the conserved regions, 10 candidate regions for primer design were chosen (A to J). Six candidate regions, A, B, D, G, H, and I, have relatively low homology among the selected bacterial 16S rRNA sequences, and their conserved regions are too narrow to design the 16S rRNA universal primer (data not shown). Therefore, four candidate regions, C, E, F, and J, were selected and the best combinations for primer design were considered. To select the best combination of two candidate regions for 16S rRNA universal primer design, we considered several factors such as the length of the PCR product, the coverage of variable regions, and one-step sequencing capability by the GS-FLX+ approach. When the PCR product and coverage of variable regions are longer, the species coverage and resolution will be better. However, present GS-FLX+ technology only guarantees up to 600-700 bp with DNA sequence fidelity [30] . Therefore, the C-F and E-J combinations (~700 and ~800 bp, respectively) were eliminated from the candidate collections because one-step NGS cannot be guaranteed. In addition, the C-E combination yields an ~500 bp PCR product and covers only two variable regions, V3 and V4. Therefore, the F-J combination is the best selection for 16S rRNA universal primer design, providing an ~600 bp PCR product for GS-FLX+ one-step sequencing and covering four different variable regions, V6 to V9, for better species coverage and resolution. The new 16S rRNA universal primers were designed to this selected conserved region and designated the 926F/1505R primer set ( Fig. 4 ).
To confirm the high primer coverage of 926F/1505R developed in this study comparing with two other previously developed 16S rRNA universal primer sets (8F/338R and 784F/1061R), primer sequences were compared with complete 16S rRNA gene sequences from bacterial genomes and perfect matches were counted ( Table 4 ). Among 1,578 rRNA gene sequences, 926F primer and 1505R primer covered 1,154 (73.1%) and 1,231 (78.0%) sequences, respectively, which was higher than the other two primer sets, suggesting that the 926F/1505R primer set has a higher primer coverage than the other two primer sets. This novel primer set was also tested and evaluated with total DNA from individual intestinal bacteria species, a mixture of 21 intestinal bacteria, and a fecal sample from a 1-week-old infant.
Primer coverage of three different 16S rRNA universal primer sets (926F/1505R, 8F/338R, and 784F/1061R).
PPT Slide
Lager Image
Primer coverage of three different 16S rRNA universal primer sets (926F/1505R, 8F/338R, and 784F/1061R).
- Comparative Analysis of Detection Coverage of the 926F/1505R Primer Set and Other 16S rRNA Universal Primer Sets Using Selected Individual Intestinal Bacteria
The human intestinal microbiota consists of four major phyla, Firmicutes , Bacteroidetes , Proteobacteria , and Actinobacteria . Therefore, 11 species from Firmicutes , 2 species from Bacteroidetes, 4 species from Proteobacteria , and 4 species from Actinobacteria were chosen to represent the major human intestinal bacteria ( Table 1 ). These 21 human intestinal bacterial species were purchased from the American Type Culture Collection (ATCC) and Korean Collection for Type Culture (KCTC) to test the detection coverage of the 926F/1505R primer set using selected individual intestinal bacterium. In addition, two previously developed 16S rRNA universal primer sets (8F/338R and 784F/1061R) were chosen for comparison with the novel primer set (926F/1505R) ( Table 3 ). PCRs with three different universal primer sets showed that all selected primer sets detected all 21 intestinal bacteria, suggesting that the species coverage and resolution of these three universal primer sets are acceptable for detection of major human intestinal bacteria. To further evaluate their species coverage and resolution, PCR detection with each selected universal primer set was verified with the simulated human intestinal microbiota and a real fecal sample.
- Evaluation and Optimization of a 926F/1505R Primer Set Using Total DNA from a Simulated Intestinal Microbiota and a Fecal Sample from a 1-Week-Old Infant
PCRs were performed with three different 16S rRNA universal primer sets and total DNA extracted from a simulated human intestinal microbiota. After cloning the PCR products into an E. coli cloning vector, 50 colonies were randomly selected for each universal primer set. The 8F/338R and 784F/1061R universal primer sets detected 7 different species (4 from Firmicutes , 3 from Proteobacteria ) and 11 different species (5 from Firmicutes , 1 from Bacteroidetes , 4 from Proteobacteria , and 1 from Actinobacteria ), respectively, among 21 intestinal bacteria. The newly designed 926F/1505R universal primer set detected 12 species (7 from Firmicutes, 1 from Bacteroidetes , and 4 from Proteobacteria ) from the simulated intestinal microbiota ( Table 5 ). These results indicate that the species coverage and resolution of the 784F/1061R and 926F/1505R primer sets are better than those of the 8F/338R primer set.
Detection of intestinal bacteria from thein vitrosimulated intestinal microbiota.
PPT Slide
Lager Image
Detection of intestinal bacteria from the in vitro simulated intestinal microbiota.
Detection of intestinal bacteria from infant total fecal DNA.
PPT Slide
Lager Image
Detection of intestinal bacteria from infant total fecal DNA.
To further evaluate these primer sets, DNA was extracted from a fecal sample of a 1-week-old infant and used for PCR. One hundred colonies were randomly selected and insert DNAs were sequenced for bacterial identification. For the 8F/338R universal primer set, the sequencing result indicated 84% Streptococcus salivarius subsp. thermophilus , 8% Clostridiales sp., 4% Anaerostipes sp., and 4% Bifidobacterium longum subsp. infantis ( Table 6 ). The 784F/1061R universal primer set detected 56% Bacteroidetes sp., 28% of Bifidobacterium longum subsp. infantis , 4% Streptococcus salivarius subsp. thermophilus , 4 % E. coli , 4% Kocuria sp., and 4% uncultured bacteria. The sequencing result using the 926F/1505R universal primer set indicated 38% Streptococcus salivarius subsp. thermophilus , 36% Bifidobacterium longum subsp. infantis , 9% Clostridiales sp., 3% Bacteroides vulgatus , 1% E. coli , and 13% uncultured bacteria.
Similar to the previous result with the simulated intestinal microbiota, the 8F/338R primer set detected the fewest kinds of intestinal bacteria and could not detect any uncultured bacteria, suggesting that the species coverage and resolution of this universal primer set is relatively poor. Previously, Koenig et al . [20] reported that this primer set has a very low detection efficiency for the Actinobacteria phylum, including Bifidobacterium , consistent with the result of only 4% detection of Bifidobacterium ( Table 6 ). In contrast, a previous report using 784F/1061R [2] showed that this primer set can detect high numbers of Bifidobacterium . Consistent with this report, random PCR with this primer set revealed that it can detect high numbers of Bacteroidetes and Actinobacteria , suggesting a possible detection preference for these two phyla. Therefore, these primer sets may have biased detection coverage and may not be useful for broad-range detection of various intestinal bacteria. However, the newly designed 926F/1505R primer set revealed broad-range detection coverage of the infant intestinal microbiota and detected the highest number of uncultured bacteria, suggesting that this primer set may be better than the other two primer sets, without any biased detection. Interestingly, the 926F/1505R primer set detected the highest number of Streptococcus thermophilus and Bifidobacterium in the infant fecal sample, indicating that these two bacteria constitute over 70% of the infant intestinal microbiota. A recent report using metagenomic analysis of infant intestinal microbiotas showed that Streptococcus thermophilus and Bifidobacterium are the major dominant bacteria [32] , supporting our data. On the basis of comparative coverage and resolution analysis of three 16S rRNA universal primer sets, the newly designed 926F/1505R primer set is highly recommended for metagenomic compositional analysis of infantile intestinal microbiotas, with broad-range detection coverage and high resolution.
To date, many 16S rRNA universal primer sets have been developed and optimized for metagenomic analysis of human adult and infant intestinal microbiotas [13 , 20 , 28] . Although metagenomic analysis using these primer sets has been widely performed to determine the composition of intestinal microbiota, the results have often been biased towards a specific phylum or genus [2 , 20 , 26] . It has been suggested that this bias might be due to non-optimized preparation of total fecal DNA and/or biased bacterial detection by specific 16S rRNA universal primer sets. In this study, total fecal DNA extraction was optimized to minimize the loss of fecal DNA during preparation, by modification of the pretreatment conditions such as bacterial cell lysis buffer, boiling time, and bead-beating time. In addition, a novel 16S rRNA universal primer set (926F/1505R) was designed and compared with two other widely used universal primer sets (8F/338R and 784F/1061R) to evaluate its broad-range detection coverage and resolution. Our data suggest that this newly optimized total fecal DNA extraction method and new primer set may be very valuable for increasing the accuracy and fidelity in the metagenomic compositional analysis of human and infant intestinal microbiotas with minimization of any bias.
Acknowledgements
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (No. NRF-2012R1A1A1009859)
References
Altschul SF , Gish W , Miller W , Myers EW , Lipman DJ 1990 Basic local alignment search tool. J. Mol. Biol. 215 403 - 410    DOI : 10.1016/S0022-2836(05)80360-2
Andersson AF , Lindberg M , Jakobsson H , Backhed F , Nyren P , Engstrand L 2008 Comparative analysis of human gut microbiota by barcoded pyrosequencing. PLoS One 3 e2836 -    DOI : 10.1371/journal.pone.0002836
Benson DA , Cavanaugh M , Clark K , Karsch-Mizrachi I , Lipman DJ , Ostell J , Sayers EW 2013 GenBank. Nucleic Acids Res. 41 D36 - D42    DOI : 10.1093/nar/gks1195
Bezirtzoglou E 1997 The intestinal microflora during the first weeks of life. Anaerobe 3 173 - 177    DOI : 10.1006/anae.1997.0102
Bullen C , Tearle P , Willis A 1976 Bifidobacteria in the intestinal tract of infants: an in-vivo study. J. Med. Microbiol. 9 325 - 333    DOI : 10.1099/00222615-9-3-325
Cole JR , Wang Q , Fish JA , Chai B , McGarrell DM , Sun Y 2014 Ribosomal Database Project: data and tools for high throughput rRNA analysis. Nucleic Acids Res. 42 D633 - D642    DOI : 10.1093/nar/gkt1244
Dominguez-Bello MG , Costello EK , Contreras M , Magris M , Hidalgo G , Fierer N , Knight R 2010 Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns. Proc. Natl. Acad. Sci. USA 107 11971 - 11975    DOI : 10.1073/pnas.1002601107
Duncan S , Louis P , Flint H 2007 Cultivable bacterial diversity from the human colon. Lett. Appl. Microbiol. 44 343 - 350    DOI : 10.1111/j.1472-765X.2007.02129.x
Eckburg PB , Bik EM , Bernstein CN , Purdom E , Dethlefsen L , Sargent M 2005 Diversity of the human intestinal microbial flora. Science 308 1635 - 1638    DOI : 10.1126/science.1110591
Fan W , Huo G , Li X , Yang L , Duan C , Wang T , Chen J 2013 Diversity of the intestinal microbiota in different patterns of feeding infants by Illumina high-throughput sequencing. World J. Microbiol. Biotechnol. 29 2365 - 2372    DOI : 10.1007/s11274-013-1404-3
Fleming DW , Cochi SL , MacDonald KL , Brondum J , Hayes PS , Plikaytis BD 1985 Pasteurized milk as a vehicle of infection in an outbreak of listeriosis. N. Engl. J. Med. 312 404 - 407    DOI : 10.1056/NEJM198502143120704
Gasson MJ 1983 Plasmid complements of Streptococcus lactis NCDO 712 and other lactic streptococci after protoplastinduced curing. J. Bacteriol. 154 1 - 9
Gill SR , Pop M , DeBoy RT , Eckburg PB , Turnbaugh PJ , Samuel BS 2006 Metagenomic analysis of the human distal gut microbiome. Science 312 1355 - 1359    DOI : 10.1126/science.1124234
Grumbt B , Eck S , Hinrichsen T , Hirv K 2013 Diagnostic applications of next generation sequencing in immunogenetics and molecular oncology. Transfus. Med. Hemother. 40 196 - 206    DOI : 10.1159/000351267
Guarner F , Malagelada JR 2003 Gut flora in health and disease. Lancet 361 512 - 519    DOI : 10.1016/S0140-6736(03)12489-0
Hoiseth SK , Stocker B 1981 Aromatic-dependent Salmonella Typhimurium are non-virulent and effective as live vaccines. Nature 291 238 - 239    DOI : 10.1038/291238a0
Holzapfel WH , Haberer P , Snel J , Schillinger U , Huis in’t Veld JH 1998 Overview of gut flora and probiotics. Int. J. Food Microbiol. 41 85 - 101    DOI : 10.1016/S0168-1605(98)00044-0
Hooper LV , Macpherson AJ 2010 Immune adaptations that maintain homeostasis with the intestinal microbiota. Nat. Rev. Immunol. 10 159 - 169    DOI : 10.1038/nri2710
Huse SM , Dethlefsen L , Huber JA , Welch DM , Relman DA , Sogin ML 2008 Exploring microbial diversity and taxonomy using SSU rRNA hypervariable tag sequencing. PLoS Genetics 4 e1000255 -    DOI : 10.1371/journal.pgen.1000255
Koenig JE , Spor A , Scalfone N , Fricker AD , Stombaugh J , Knight R 2011 Succession of microbial consortia in the developing infant gut microbiome. Proc. Natl. Acad. Sci. USA 108 4578 - 4585    DOI : 10.1073/pnas.1000081107
Larkin MA , Blackshields G , Brown NP , Chenna R , McGettigan PA , McWilliam H 2007 Clustal W and Clustal X version 2.0. Bioinformatics 23 2947 - 2948    DOI : 10.1093/bioinformatics/btm404
Lee J-H , O’Sullivan DJ 2010 Genomic insights into bifidobacteria. Microbiol. Mol. Biol. Rev. 74 378 - 416    DOI : 10.1128/MMBR.00004-10
Matamoros S , Gras-Leguen C , Le Vacon F , Potel G , de La Cochetiere M-F 2013 Development of intestinal microbiota in infants and its impact on health. Trends Microbiol. 21 167 - 173    DOI : 10.1016/j.tim.2012.12.001
McOrist AL , Jackson M , Bird AR 2002 A comparison of five methods for extraction of bacterial DNA from human faecal samples. J. Microbiol. Methods 50 131 - 139    DOI : 10.1016/S0167-7012(02)00018-0
Onmus-Leone F , Hang J , Clifford RJ , Yang Y , Riley MC , Kuschner RA 2013 Enhanced de novo assembly of high throughput pyrosequencing data using whole genome mapping. PloS One 8 e61762 -    DOI : 10.1371/journal.pone.0061762
Palmer C , Bik EM , DiGiulio DB , Relman DA , Brown PO 2007 Development of the human infant intestinal microbiota. PLoS Biol. 5 e177 -    DOI : 10.1371/journal.pbio.0050177
Penders J , Thijs C , Vink C , Stelma FF , Snijders B , Kummeling I 2006 Factors influencing the composition of the intestinal microbiota in early infancy. Pediatrics 118 511 - 521    DOI : 10.1542/peds.2005-2824
Qin J , Li R , Raes J , Arumugam M , Burgdorf KS , Manichanh C 2010 A human gut microbial gene catalogue established by metagenomic sequencing. Nature 464 59 - 65    DOI : 10.1038/nature08821
Sambrook J , Fritsch EF , Maniatis T 1989 Molecular Cloning. Cold Spring Harbor Laboratory Press New York
Shokralla S , Spall JL , Gibson JF , Hajibabaei M 2012 Next-generation sequencing technologies for environmental DNA research. Mol. Ecol. 21 1794 - 1805    DOI : 10.1111/j.1365-294X.2012.05538.x
Turnbaugh PJ , Hamady M , Yatsunenko T , Cantarel BL , Duncan A , Ley RE 2009 A core gut microbiome in obese and lean twins. Nature 457 480 - 484    DOI : 10.1038/nature07540
Turroni F , Peano C , Pass DA , Foroni E , Severgnini M , Claesson MJ 2012 Diversity of bifidobacteria within the infant gut microbiota. PLoS One 7 e36957 -    DOI : 10.1371/journal.pone.0036957
Yoshioka H , Iseki K , Fujita K 1983 Development and differences of intestinal flora in the neonatal period in breast-fed and bottle-fed infants. Pediatrics 72 317 - 321
Yu Z , Morrison M 2004 Improved extraction of PCR-quality community DNA from digesta and fecal samples. Biotechniques 36 808 - 813