The ISME Journal (2009) 3, 944–954 & 2009 International Society for Microbial Ecology All rights reserved 1751-7362/09 $32.00 www.nature.com/ismej ORIGINAL ARTICLE 16S rRNA gene-based analysis of fecal microbiota from preterm infants with and without necrotizing enterocolitis Yunwei Wang1,6, Jeanette D Hoenig2,6, Kathryn J Malin2, Sanaa Qamar2, Elaine O Petrof 3, Jun Sun4, Dionysios A Antonopoulos5, Eugene B Chang1 and Erika C Claud1,2 1Department of Medicine, University of Chicago, Chicago, IL, USA; 2Department of Pediatrics, University of Chicago, Chicago, IL, USA; 3Department of Medicine, Queen’s University, Kingston, Ontario, Canada; 4Department of Medicine, University of Rochester, Rochester, NY, USA and 5Biosciences Division, Institute for Genomics and Systems Biology, Argonne National Laboratory, Argonne, IL, USA Neonatal necrotizing enterocolitis (NEC) is an inflammatory intestinal disorder affecting preterm infants. Intestinal bacteria have an important function; however no causative pathogen has been identified. The purpose of this study was to determine if there are differences in microbial patterns that may be critical to the development of this disease. Fecal samples from 20 preterm infants, 10 with NEC and 10 matched controls (including 4 twin pairs) were obtained from patients in a single site level III neonatal intensive care unit. Bacterial DNA from individual fecal samples was PCR- amplified and subjected to terminal restriction fragment length polymorphism analysis and library sequencing of the 16S rRNA gene to characterize diversity and structure of the enteric microbiota. The distribution of samples from NEC patients distinctly clustered separately from controls. Intestinal bacterial colonization in all preterm infants was notable for low diversity. Patients with NEC had even less diversity, an increase in abundance of Gammaproteobacteria, a decrease in other bacteria species, and had received a higher mean number of previous days of antibiotics. Our results suggest that NEC is associated with severe lack of microbiota diversity that may accentuate the impact of single dominant microorganisms favored by empiric and widespread use of antibiotics. The ISME Journal (2009) 3, 944–954; doi:10.1038/ismej.2009.37; published online 16 April 2009 Subject Category: microbe–microbe and microbe–host interactions Keywords: necrotizing enterocolitis; clone library; operational taxonomical units; Gamma- proteobacteria Introduction inappropriate pro-inflammatory response and bacterial colonization (Claud and Walker, 2001). Necrotizing enterocolitis (NEC) is an acquired Because the risk factors for NEC are issues common gastrointestinal disease that primarily affects pre- to all premature infants, it is impossible to predict mature babies. It is the most common gastrointest- which infants will develop this devastating disease. inal emergency in the newborn and may lead to As opposed to the adult intestinal microbiota that death in severely affected infants. The incidence is comprise more than 1013 microorganisms, the new- 1–3 per 1000 live births, occurring in 10% of infants born gut is sterile at birth. Although the full-term born under 1500 g and representing 2–5% of neonate becomes rapidly colonized with a diverse neonatal intensive care unit (NICU) admissions flora, preterm infants have a delayed colonization, (Kosloske, 1994). Mortality has been reported to fewer bacterial species present and are more range from 9% to 28% (Stoll, 1994). Although the susceptible to colonization with pathogenic bacteria exact cause is unknown, the critical elements are (Goldmann and JLeclair Macone, 1978; Gewolb thought to be prematurity, enteral feeding, an et al., 1999; Schwiertz et al., 2003). Bacteria are believed to be important in the Correspondence: EC Claud, Department of Pediatrics, University pathogenesis of NEC; however no specific pathogen of Chicago, 5841 S Maryland Ave, MC6060, Chicago, IL 60637, has been identified. Previous studies, however, USA. have been limited by the inability of conventional E-mail: [email protected] microbiological cultivation techniques to thor- 6Both are Co-first authors. Received 13 January 2009; revised 6 March 2009; accepted 15 oughly characterize the human gastrointestinal March 2009; published online 16 April 2009 microbiota. It has been reported that 80% of the Microbial patterns in necrotizing enterocolitis Y Wang et al 945 human colonic microbiota are not detected by Table 1 Clinical characteristics of the hospitalized preterm conventional culture methods (Eckburg et al., infants in this study 2005). Taxonomical classification and noncultiva- Characteristics NEC infants Control infants tion molecular profiling of microbiomes are now (n ¼ 10) (n ¼ 10) possible by sequencing the highly conserved 16S small subunit bacterial ribosomal RNA gene (rRNA), Male/Female 6/4 7/3 allowing identification of previously undetectable Gestational age (weeks) 25–32 26–32 microbes (Relman, 1999). This approach therefore Sample collection age 5–49 4–49 provides a more complete picture of the composi- (days) Delivery: C-section/ 8/2 9/1 tion of human intestinal microbiota. For this study, vaginal we hypothesized that there are differences in Feeding: breast milk/ 4/6 6/4 molecular microbial profiles of preterm infants formula without and with NEC, which may be critical in the etiopathogenesis of disease. Abbreviation: NEC, necrotizing enterocolitis. Materials and methods DNA extraction Patient characteristics and sample collection For DNA extraction, 50 mg frozen fecal sample was Subjects were recruited from a single level III NICU dissolved in 1 ml extraction buffer (50 mM Tris (pH at the University of Chicago. All patients underwent 7.4), 100 mM EDTA (pH 8.0), 400 mM NaCl, 0.5% routine NICU care as determined by the managing SDS) containing 20 ml proteinase K (20 mg mlÀ1). A service. A total of 21 fecal samples from 20 patients slurry (500 ml) of 0.1-mm diameter zirconia/silica were obtained including 10 infants with NEC and beads (BioSpec Products, Bartlesville, OK, USA) 10 infants without NEC as control. The study set was added into the extraction tubes and a Mini- included four sets of twins. In these twin pairs, one Beadbeater-8K Cell Disrupter (BioSpec Products) twin developed NEC (N1-N4) and the other was was used to lyse the microbial cells. After overnight spared (C1-C4) thus serving as an ideal control. incubation at 55 1C, standard DNA extraction with Controls (C5-C10) were matched to NEC (N5-N10) by phenol/chloroform/isoamyl alcohol, and precipita- gestational age and day of life at NEC diagnosis. tion with ethanol were performed. Isolated DNA was Control sample 9 was collected from an infant who dissolved in TE buffer and stored at À801C. later went on to develop NEC; however no follow-up sample after NEC diagnosis was available for this infant due to parent refusal. Only patients with PCR and terminal restriction fragment length definite or advanced NEC, corresponding to Bell polymorphism analysis stages II and III were included in the NEC cohort 16S rRNA gene sequences were amplified from DNA (Bell et al., 1978). For patient N10 with NEC, an samples using broad-range primers 8F (50-AGAG additional sample was analyzed (CN10), which had TTTGATCCTGGCTCAG-30) labeled with 60-carboxy- been prospectively obtained 3 days before disease fluorescein (6-FAM) and 1492R (50-GGTTACCTTGT diagnosis. TACGACTT-30) for the conserved 16S bacterial Included subjects were born between 25 and 32 domain. PCR reactions were performed for 30 cycles weeks gestation and the patient age when the fecal using Takara high-fidelity Ex Taq (Takara Mirus Bio, sample was obtained ranged from 4 to 49 days. Madison, WI, USA) with an annealing temperature Thirteen male and seven female infants were of 58 1C. Replicate PCR products were pooled and included. Seventeen patients were delivered by verified by electrophoresis and purified by precipi- cesarean section and three vaginally. Ten of the tation. FAM-labeled PCR products were then infants were exclusively formula-fed and ten re- digested by restriction enzyme MspI (New England ceived breast milk. Study patient characteristics Biolabs Inc., Ipswich, MA, USA), mixed with were not significantly different between NEC and GeneScan-500 Size Standard (Applied Biosystems, control groups (Table 1). Carlsbad, CA, USA) and sequenced by capillary The study was approved by institutional review electrophoresis. board for human studies of the University of Chicago and informed consent was obtained from patient parents. Fecal samples for NEC patients were Clone library and sequencing collected as soon as possible after the diagnosis was Unlabeled PCR products were purified using the made, with an average collection time of o1 day QIAquick gel extraction kit (Qiagen, Valencia, CA, after the diagnosis of NEC. Nurses collected the fecal USA) and cloned into pCR-2.1-TOPO vectors using sample directly from the diaper into the collection the TOPO-TA cloning kit (Invitrogen, Carlsbad, CA, tube using the wooden end of a sterile cotton swab. USA) for sequencing as per manufacturer’s instruc- The sample was immediately frozen. All samples tions. From each library, 288 colonies were picked were stored at À80 1C until processed. randomly and processed for sequencing. The 50 end The ISME Journal Microbial patterns in necrotizing enterocolitis Y Wang et al 946 of the plasmid inserts was sequenced unidirection- gene sequences were imported into the ARB soft- ally using 8F forward primers