Enterococcus faecalis is a ubiquitous intestinal symbiont and common early colonizer of the neonatal gut. Although colonization with E. faecalis has been previously associated with decreased pathology of necrotizing enterocolitis (NEC), these bacteria have been also implicated as opportunistic pathogens. Here we characterized 21 strains of E. faecalis, naturally occurring in 4-day-old rats, for potentially pathogenic properties and ability to colonize the neonatal gut. The strains differed in hemolysis, gelatin liquefaction, antibiotic resistance, biofilm formation, and ability to activate the pro-inflammatory transcription factor NF-κB in cultured enterocytes. Only 3 strains, BB70, 224, and BB24 appreciably colonized the neonatal intestine on day 4 after artificial introduction with the first feeding. The best colonizer, strain BB70, effectively displaced E. faecalis of maternal origin. Whereas BB70 and BB24 significantly increased NEC pathology, strain 224 significantly protected from NEC. Our results show that different strains of E. faecalis may be pathogenic or protective in experimental NEC.
Citation: Delaplain PT, Bell BA, Wang J, Isani M, Zhang E, Gayer CP, et al. (2019) Effects of artificially introduced Enterococcus faecalis strains in experimental necrotizing enterocolitis. PLoS ONE 14(11): e0216762. https://doi.org/10.1371/journal.pone.0216762
Editor: Nicholas J. Mantis, New York State Department of Health, UNITED STATES
Received: April 20, 2019; Accepted: October 20, 2019; Published: November 1, 2019
Copyright: © 2019 Delaplain et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work was supported by: H.R.F., grant AI 014032 from National Institute of Allergy and Infectious Diseases, https://www.niaid.nih.gov/. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Necrotizing enterocolitis (NEC) affects approximately 1 in 1000 live births and is one of the leading causes of mortality among preterm infants . Although the pathogenesis of this disease is not yet fully understood, it is broadly accepted that bacterial colonization of the immature intestine in combination with perinatal stresses such as formula feeding, hypoxia, and hypothermia play the leading role [2–4]. While no single pathogen is likely responsible for NEC, previous work has implicated clostridia , Cronobacter [6, 7], E. coli , and lactobacilli [9–11] as either NEC-promoting or protective colonizers of the neonatal intestine. Importantly, protective or pathogenic properties of these bacteria are strain-specific.
Probiotics, bacteria believed to be beneficial upon administration, have been extensively tried for prevention of NEC. Most of these trials are encouraging [12–14]. However, probiotics may cause adverse effects . Evidence-based recommendations for clinical use of probiotics in NEC have not yet been developed due to lack of standardization of bacterial species/strains, doses, and treatment regimens across different trials . A rational approach to probiotic therapy would be to identify commensals that effectively colonize the neonatal gut upon introduction and protect from NEC in animal models.
E. faecalis is a bacterial species of potential relevance to NEC. These bacteria constitute up to 1% of adult intestinal microbiome  and are readily transmitted from mothers to neonates in both humans [18, 19] and rodents [20, 21]. NEC patients tended to harbor lower percentages of E. faecalis in their microbiomes compared to healthy controls, but this tendency was not significant [22, 23]. Importantly, E. faecalis has been also implicated as pathogen . To gain insight into potential role of E. faecalis in the pathogenesis of experimental NEC, we isolated multiple strains of these bacteria from 4-day old rats and examined their ability to colonize the neonatal intestine and to alter NEC pathology. Only few strains colonized the intestine following artificial introduction with first feeding. One of these strains protected from NEC, whereas two others exacerbated NEC pathology.
Materials and methods
All animal experiments were approved by the CHLA Institutional Animal Care and Use Committee (IACUC). Timed-pregnant Sprague Dawley rats were obtained from either Envigo (Placentia, CA) or Charles River Laboratories (Hollister, CA). Envigo rats were only used as source of bacteria. Newborn rats were separated from dams at birth and were kept in an infant incubator (Ohio Medical Products, Madison, WI) at 30°C and 90% humidity. NEC was induced by formula feeding and hypoxia, according to our previously published protocol [20, 25, 26]. Neonatal rats are fed 200 μl of formula (15 g Similac 60/40, Ross Pediatrics Columbus, OH in 75 ml of Esbilac canine milk replacement, Pet-Ag Inc., Hampshire, IL) every 8 h for 4 d. Fresh formula is prepared daily, each new batch is tested for bacterial contamination by plating on blood agar and MRS, and care is taken with each feeding to prevent introduction of extraneous bacteria. Pups are subjected to hypoxia at the conclusion of each feeding (10 min in 95% N2 and 5% O2). On day 4, animals are euthanized by decapitation and terminal ileum is collected for NEC pathology score and plating of intestinal contents. Samples for pathology scoring are fixed in formalin, embedded in paraffin, sectioned and stained with hematoxylin-eosine. These are then scored by a pathologist blinded to treatment groups. NEC score is assigned based on the degree of observed injury to the intestinal epithelium based on a 5-point scale (0: no pathology; 1: epithelial sloughing and/or mild sub-mucosal edema; 2: damage to the tips of the villi; 3: damage to more than half of the villi; 4: complete obliteration of the epithelium). Samples collected for bacterial analysis are homogenized in PBS, serially diluted and plated onto diagnostic media within 2 h of collection. Adult animals are euthanized by CO2 asphyxiation. If animals were treated withE. faecalis bacteria, those were re-suspended in formula and given with the first feed. Cronobacter muytjensii 51329 was purchased from the American Type Culture Collection (Manassas, VA) and given to animals with the second feed.
Identification of bacteria
Independent isolates of E. faecalis were established from the intestinal contents of 4-day old rats, colony-purified and kept as frozen stock as described previously . To characterize populations of intestinal bacteria, freshly excised intestines were homogenized, serially diluted, and plated on blood agar (Sigma) for broad range of bacteria and MRS agar (Oxoid, Basinstoke, UK) for lactic bacteria. Plates were incubated for 4 d at 37°C in air (blood agar) or CO2 atmosphere (MRS agar). Emerging colonies were classified according to their appearance. Pure cultures for each colony type were purified by re-streaking and kept as frozen stocks. Bacterial species were identified by sequencing 16S rRNA gene PCR-amplified with 27F and 1492R primers at GeneWiz (Los Angeles, CA). Sequences were queried against NCBI non-redundant nucleotide (nt) database using the BLAST algorithm.
Bacterial culture and phenotypes
E. faecalis bacteria were grown at 37°C aerobically in Brain Heart Infusion (BHI), Tryptic Soy Broth (TSB), or Luria Broth (LB). For pouring plates, agar was added to 17 g/L. Selective media for isolating E. faecalis contained 0.4 g/L sodium azide. E. faecalis phenotypes were determined by replica plating onto diagnostic media including blood agar, gelatin liquefaction medium (5 g/L peptone, 3 g/L beef extract, 120 g/L gelatin), antibiotic agar (LB supplemented with 50 mg/L ampicillin, or 100 mg/ml kanamycin, or 30 mg/l rifaximin), β-galactosidase agar (LB supplemented with 30 mg/L X-gal and 2 mM IPTG), and sugar fermentation agar (LB supplemented with 0.2 mg/L Neutral Red and 1% appropriate sugar). Gelatin plates were incubated upright at room temperature. Bacterial culture density was determined by spectrophotometry at 600 nm. Correlation between OD600 and CFU/ml was determined by serial dilution and plating.
Restriction endonuclease analysis of bacterial DNA
Bacterial DNA was extracted from overnight culture by 5-min vortexing with 200 μm glass beads in TEN buffer (50 mM Tris pH 8.0, 100 mM NaCl, 10mM EDTA), overnight digestion at 50°C following addition of SDS and Proteinase K to 1% and 20 μg/ml, respectively, phenol/chloroform extraction, and ethanol precipitation. 5 μg DNA samples were digested with 10 u HindIII (New England Biolabs, Ipswich, MA) for 2 h at 37°C. Digestion products were resolved by electrophoresis through 0.8% agarose Tris-acetate gel. Gels were stained with ethidium bromide and photographed under UV light using GelDoc XR (Bio-Rad, Hercules, CA).
A modified crystal violet assay, as previously described [27–29], was used to quantify biofilm formation. Overnight cultures of E. faecalis were diluted 1:50 in fresh medium and inoculated into wells of a 96-well polystyrene plate. Following static 24 h incubation at 37°C, plate was rinsed 3 times with PBS and air dried. After 10 min fixing with 3:1 ethanol:acetic acid, biofilms were stained with 0.1% crystal violet for 15 min. Wells were then washed with water until effluent ran clear. Crystal violet was extracted with 10% acetic acid, samples transferred to a new 96-well plate and OD550 was measured on plate reader.
Binding of bacteria to enterocytes and activation of NF-κB
IEC-6 cells (rat intestinal epithelial cells) were obtained from ATCC and grown in DMEM+5% FBS as recommended by the supplier. Cells (passage 21–30) were used at 70–90% confluence. Bacteria grown overnight were diluted in DMEM and added to IEC-6 cells. After 30 min incubation at 37°C, cells were rinsed 3 times with ice-cold PBS, collected, serially diluted and plated on blood agar for bacterial quantification.
For activation of NF-κB, IEC-6 cells were treated with bacteria for 15 min, lysed on ice for 10 min with RIPA buffer (50 mM Tris-HCl pH 7.0, 100 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 1 mM PMSF). Lysates cleared by centrifugation at 10,000g for 10 min were mixed with 2x Laemmli buffer and boiled for 1 min. 20 μg protein samples were resolved on a 10% SDS-polyacrylamide gel. Transfer of protein onto nitrocellulose membrane, membrane blocking, incubation with primary antibody for IκBα (Cell Signaling, Danvers, MA) and secondary HRP-conjugated antibody were performed as recommended by antibody supplier. After extensive washing in PBS, membranes were impregnated with peroxide-luminol reagent and exposed to x-ray film.
Means for parametric data were compared using unpaired 2-sample t-test. Categorical and ordinal data were compared using χ2 test. All analyses were conducted in R v3.5.1 . Graphics were designed either in R or GraphPad Prism v8.
Diversity of Enterococcus faecalis in 4-day old rats
To examine potential relationship between E. faecalis and NEC, we sought to isolate enterococci naturally occurring in rats, identify different strains, and examine their properties in the rat model of NEC. Enterococci were isolated by blood agar plating of intestinal content from 4-day-old rats subjected to the NEC-inducing regimen of formula feeding and hypoxia. E. faecalis were identified by their resistance to azide, characteristic morphology upon Gram staining, and 16S rRNA gene sequencing. In our previous study, Enterococcus spp. was found in about 90% of intestinal samples. In the animals where enterococci were found, they constituted 17±3% of the bacterial populations . To characterize diversity of E. faecalis, we examined 147 independent isolates of these bacteria collected from 4-day-old offspring of Charles River and Envigo (formerly Harlan) rats during 2008–2013. All isolates were catalase-negative, glucose-, fructose-, and sucrose-fermenting, ampicillin- and tetracycline-sensitive. Characteristics that differed among isolates included colony morphology, hemolysis, gelatin liquefaction, β-galactosidase activity, resistance to kanamycin or rifaximin, and fermentation of sorbitol, mannitol, and arabinose (S1 Data File, Fig 1A). Each unique combination of these phenotypes defined a distinct strain. We thus identified 21 different strains of E. faecalis, each represented by between 1 and 59 isolates (S2 Data File, Fig 1B). All strains belonged to one of the two genomic groups, as revealed by patterns of HindIII DNA fragments (S2 Data File, Fig 1C). These results indicate considerable diversity of E. faecalis strains in laboratory rats.
Frequencies of different phenotypes (A) and different strains (B) within a group of E. faecalis isolates (n = 147) from 4-day old rats. (C) Patterns of genomic DNA HindIII fragments of indicated strains. Note dissimilarity of DNA patterns A (55–249, 49–171, and BB70) and B (265, BB24).
Potential pathogenic properties of E. faecalis strains identified in vitro
In order to narrow down the list of E. faecalis strains for in vivo studies, we examined strains’ potential pathogenic properties. Of the phenotypes described above, antibiotic resistance , hemolysis, and proteolysis  may contribute to pathogenicity. Another pathogenic phenotype of relevance to NEC could be the ability of bacteria to trigger mucosal inflammatory response. To characterize this phenotype, we treated IEC-6 cells (intestinal epithelioid cell line of rat origin) with each of the 21 strains of E. faecalis and examined activation of the pro-inflammatory transcription factor NF-κB by western blotting for the inhibitory subunit IκBα. Strains 25, 37, 49, and 82 caused degradation of IκBα (i.e. activation of NF-κB), whereas other strains caused partial degradation or no degradation (Fig 2).
IEC-6 cells were treated with 1010 cfu/ml each strain of E. faecalis and activation of the NF-κB pathway was determined by western blotting for IκBα. β-actin reprobes are included to demonstrate lane load. Representative blots of 3 independent experiments are shown.
Efficient binding to target cells may be one more phenotype associated with pathogenicity . To characterize binding of our E. faecalis strains to enterocytes, IEC-6 monolayers were incubated with bacteria, washed, and homogenized. Resulting homogenates were serially diluted and plated onto BHI-azide agar for E. faecalis counts. There were no significant differences in binding efficiency of different strains. On average, 11±1.4% of 108 cfu input, or 0.32±0.04 cfu per IEC-6 cell were bound for each of the 21 strains. Binding was weak–numbers of bound bacteria progressively decreased with additional washes (S1 Fig).
Biofilm formation, which may be associated with efficient colonization , is yet another potentially pathogenic phenotype. We measured biofilm formation in our E. faecalis strains by overnight culturing in polystyrene plates, washing off suspended bacteria, and biofilm staining with crystal violet (Fig 3A). Biofilm formation varied considerably among strains and was not associated with other phenotypes. Although BHI is a recommended culture medium for E. faecalis, it promoted the lowest average biofilm formation across strains compared to TSB or LB (Fig 3B). Thus, our strains of E. faecalis differed in inflammatory activation and biofilm formation properties, but not in enterocyte binding efficiency.
Maternal enterococci outcompete most artificially introduced strains of E. faecalis in colonization of newborn rats
To examine effects of different E. faecalis strains in experimental NEC, we introduced these bacteria to newborn rats on day 1 and scored NEC pathology on day 4 of the NEC-inducing regimen of formula feeding-hypoxia. Percentages of E. faecalis in intestinal microbiomes and strain composition of E. faecalis on day 4 were also determined. Two strains with contrasting combinations of potentially pathogenic phenotypes were chosen for initial experiments. Strain 8 is non-hemolytic, negative for gelatin liquefaction, NF-κB activation, and biofilm formation. Strain 82 is α-hemolytic, positive for gelatin liquefaction, NF-κB activation, and biofilm formation (S2 Data File). Newborn rats were given 108–1010 cfu of either strain 8 or strain 82 once, with first formula feed. Control animals were given equivalent amount of bacterial culture supernatant. After 4 d of formula feeding-hypoxia, animals were sacrificed, and intestinal content was plated on blood agar for total bacterial counts and BHI-azide for E. faecalis. E. faecalis strains (50–100 azide-resistant colonies per animal) were identified by replica plating onto diagnostic media. NEC was scored microscopically. Somewhat lower levels of disease (37.5%) in the control FFH group can be explained by the fact that Charles River rats used in these experiments are less susceptible to NEC than commonly used Harlan rats, attributable to microbiota differences between the two facilities . Interestingly, neither of the two strains was recovered from the inoculated animals; all enterococci isolated were thus of the maternal origin (S3 Data File). There were no significant differences in NEC scores between control group and animals inoculated with strains 8 or 82 (Table 1, S3 Data File). Thus, strains 8 or 82 failed to appreciably colonize neonatal rats upon artificial introduction. Inoculation with these strains did not have significant effect on NEC pathology at all doses tested (Table 1).
Identification of efficient colonizers among E. faecalis strains
One reason for the failure of strains 8 and 82 to effectively colonize the neonatal intestine may be adaptive disadvantage of bacteria grown to stationary phase in liquid BHI culture. Indeed, bacteria coming from mothers may successfully colonize the offspring because they are adapted to survival and growth in the organismal environment. In attempts to improve colonization, we tried different culture conditions including growth in medium optimal for biofilm formation (LB), pre-incubation in FBS, or starving bacteria in dilute TSB to induce dormant state. None of these treatments significantly promoted colonization (S3 Data File).
In another approach to improving colonization, we set out to determine whether some of our E. faecalis strains are inherently better colonizers than others. Newborn rats were given a combined 1010 cfu inoculum of all 21 strains mixed in equal proportions, and strain composition of the enterococci was determined on day 4. Strikingly, only 3 strains out of 21 turned out capable of at least some degree of colonization (S4 Data File). Strain BB70 was the best colonizer—it was found in all animals that received the mixed inoculum, and constituted, on average, 30±5.6% of enterococcal populations. 224 and BB24 colonized the neonates at significantly lower efficiencies (8±3.2% and 7±2.4%, p = 0.0017 and 0.0006). None of the input strains were recovered from control non-inoculated animals. We next evaluated efficiency of colonization with pure cultures of BB70, 224, and BB24. In all animals inoculated with 108 cfu or higher doses, E. faecalis populations were dominated by the input strain (Table 1). Inoculation with the mixture of the remaining 18 strains did not result in noticeable colonization with any one of them (S4 Data File), therefore failure of these strains to colonize was not due to competition with BB70, 224, or BB24. For comparison, the known well-colonizing NEC pathogen, C. muytjensii 51329 , demonstrated efficient colonization when introduced at the 106 cfu dose (Table 1). Our results indicate that most enterococcal strains failed to colonize newborn rat intestine upon introduction as pure culture. However, some strains could be quite efficient colonizers.
Effects of colonizing E. faecalis strains on NEC pathology
E. faecalis BB70 is negative for hemolysis, gelatin liquefaction, antibiotic resistance, and biofilm, therefore it was expected to be innocuous. However, animals inoculated with this strain had significantly higher NEC scores than control formula-hypoxia animals. 224, another presumably innocuous strain, significantly protected from NEC. Potentially pathogenic BB24 (positive for hemolysis and gelatin liquefaction) significantly exacerbated NEC (Table 1). Thus, the potentially virulent phenotypes that we identified in vitro do not necessarily predict pathogenicity in rat NEC.
We isolated and characterized 21 different strains of E. faecalis from neonatal rats. The strains differed in their colony appearance, hemolysis, gelatin liquefaction, antibiotic resistance, β-galactosidase, and fermentation of sorbitol, mannitol, and arabinose. The strains also differed in their ability to form biofilm and to activate the pro-inflammatory transcription factor NF-κB in cultured enterocytes. There were two genomic variants based on HindIII DNA fragment patterns. Only 3 out of 21 strains, 224, BB24, and BB70 appreciably colonized the GI tract of newborn rats upon artificial introduction with first feed. Whereas BB24 and BB70 exacerbated NEC pathology, strain 224 protected from NEC. These results provide an insight into the role of E. faecalis in the pathogenesis of experimental NEC.
The strain diversity that we observed was somewhat surprising. This diversity may indicate that laboratory rat populations harbor a multitude of E. faecalis strains with either equal adaptive fitness in the organismal environment, or specific adaptation to different ecological niches. The identification of new strains during the course of our inoculation experiments also suggests that strain composition at the suppliers’ facilities might have changed over the course of several years. Laboratory rats thus present an interesting model to examine significance of the previously described enterococcal diversity [33–35].
E. faecalis strains that we isolated originated from the specific pathogen-free environment, therefore none of them is a likely true pathogen. Nevertheless, some of the strains’ phenotypes, such as hemolysis, gelatin liquefaction, antibiotic resistance, biofilm formation, or activation of pro-inflammatory signaling could be associated with opportunistic pathogenicity in the context of NEC. We identified strains possessing multiple potentially pathogenic traits, such as 82, as well as strains with one or zero pathogenic traits, such as 8 or BB70. We hypothesized that the former will behave as pathogens and the latter as innocuous or protective symbionts in the rat model of NEC. However, the fact that presumably innocuous BB70 turned in fact pathogenic is contrary to this hypothesis. It is possible that BB70 possesses an unknown pathogenic trait not identified by our experiments. Importantly, our results demonstrate that pathogenic properties of E. faecalis in experimental NEC are strain-specific.
The failure of the majority of our strains to colonize the neonatal intestine upon early introduction was a surprising finding in view of the fact that all the strains were isolated from 4-day-old rats and thus had previous history of successful neonatal colonization. Artificial colonization did not improve significantly by inducing dormancy, culturing in media that promoted biofilm formation, or pre-incubation of bacteria with FBS. A plausible explanation for the poor colonization with bacterial cultures is that maternal enterococci, but not cultured bacteria, are adapted to the organismal environment and therefore have higher probability of survival upon transfer to the neonates. Strain BB70, and to lesser extent 224 and BB24, were exceptional: they successfully competed with maternally acquired E. faecalis strains. It is possible that in vivo survival of bacteria depends on host-induced genes, and such genes may be constitutively expressed in BB70, 224 and BB24. Our findings indicate that failure of cultured bacteria to establish intestinal colonization may be a serious limitation to probiotic therapy. Finding probiotic strains similar to BB70 in colonization ability may be a way of overcoming this limitation.
S1 Fig. Binding of E. faecalis strains to IEC-6 cells.
S1 Data File. Phenotypic characterization of 147 isolates of naturally-occurring E. faecalis.
S2 Data File. Characteristics of 21 unique strains of E. faecalis.
S3 Data File. Bacterial populations and NEC scores of 4-day-old rats following various treatments.
We thank Monica Williams for help in isolating E. faecalis strains; Alec Borsook and Alex Li for bacterial binding experiments. This study was supported by NIH grant AI 014032 to H.R.F.
- 1. Stoll BJ, Hansen NI, Bell EF, Walsh MC, Carlo WA, Shankaran S, et al. Trends in Care Practices, Morbidity, and Mortality of Extremely Preterm Neonates, 1993–2012. Jama. 2015;314(10):1039–51. Epub 2015/09/09. pmid:26348753; PubMed Central PMCID: PMC4787615.
- 2. Nino DF, Sodhi CP, Hackam DJ. Necrotizing enterocolitis: new insights into pathogenesis and mechanisms. Nat Rev Gastroenterol Hepatol. 2016;13(10):590–600. Epub 2016/08/19. pmid:27534694; PubMed Central PMCID: PMC5124124.
- 3. Grishin A, Bowling J, Bell B, Wang J, Ford HR. Roles of nitric oxide and intestinal microbiota in the pathogenesis of necrotizing enterocolitis. J Pediatr Surg. 2016;51(1):13–7. Epub 2015/11/19. pmid:26577908; PubMed Central PMCID: PMC4894644.
- 4. Grishin A, Papillon S, Bell B, Wang J, Ford HR. The role of the intestinal microbiota in the pathogenesis of necrotizing enterocolitis. Seminars in pediatric surgery. 2013;22(2):69–75. Epub 2013/04/25. pmid:23611609; PubMed Central PMCID: PMC3647029.
- 5. Schonherr-Hellec S, Klein GL, Delannoy J, Ferraris L, Roze JC, Butel MJ, et al. Clostridial Strain-Specific Characteristics Associated with Necrotizing Enterocolitis. Applied and environmental microbiology. 2018;84(7). Epub 2018/01/21. pmid:29352082; PubMed Central PMCID: PMC5861827.
- 6. Hunter CJ, Singamsetty VK, Chokshi NK, Boyle P, Camerini V, Grishin AV, et al. Enterobacter sakazakii enhances epithelial cell injury by inducing apoptosis in a rat model of necrotizing enterocolitis. The Journal of infectious diseases. 2008;198(4):586–93. Epub 2008/07/01. pmid:18588483; PubMed Central PMCID: PMC2497445.
- 7. Liu Q, Mittal R, Emami CN, Iversen C, Ford HR, Prasadarao NV. Human isolates of Cronobacter sakazakii bind efficiently to intestinal epithelial cells in vitro to induce monolayer permeability and apoptosis. The Journal of surgical research. 2012;176(2):437–47. Epub 2012/01/10. pmid:22221600; PubMed Central PMCID: PMC3323755.
- 8. Thomas DM, Bell B, Papillon S, Delaplain P, Lim J, Golden J, et al. Colonization with Escherichia coli EC 25 protects neonatal rats from necrotizing enterocolitis. PloS one. 2017;12(11):e0188211. Epub 2017/12/01. pmid:29190745; PubMed Central PMCID: PMC5708813.
- 9. Blackwood BP, Yuan CY, Wood DR, Nicolas JD, Grothaus JS, Hunter CJ. Probiotic Lactobacillus Species Strengthen Intestinal Barrier Function and Tight Junction Integrity in Experimental Necrotizing Enterocolitis. Journal of probiotics & health. 2017;5(1). Epub 2017/06/24. pmid:28638850; PubMed Central PMCID: PMC5475283.
- 10. Hoang TK, He B, Wang T, Tran DQ, Rhoads JM, Liu Y. Protective effect of Lactobacillus reuteri DSM 17938 against experimental necrotizing enterocolitis is mediated by Toll-like receptor 2. American journal of physiology Gastrointestinal and liver physiology. 2018;315(2):G231–g40. Epub 2018/04/13. pmid:29648878; PubMed Central PMCID: PMC6139641.
- 11. Olson JK, Navarro JB, Allen JM, McCulloh CJ, Mashburn-Warren L, Wang Y, et al. An enhanced Lactobacillus reuteri biofilm formulation that increases protection against experimental necrotizing enterocolitis. American journal of physiology Gastrointestinal and liver physiology. 2018;315(3):G408–G19. Epub 2018/06/01. pmid:29848024; PubMed Central PMCID: PMC6415713.
- 12. AlFaleh K, Anabrees J. Probiotics for prevention of necrotizing enterocolitis in preterm infants. Cochrane Database Syst Rev. 2014;(4):CD005496. Epub 2014/04/12. pmid:24723255.
- 13. Athalye-Jape G, Rao S, Patole S. Effects of probiotics on experimental necrotizing enterocolitis: a systematic review and meta-analysis. Pediatric research. 2018;83(1–1):16–22. Epub 2017/09/28. pmid:28949953.
- 14. Sawh SC, Deshpande S, Jansen S, Reynaert CJ, Jones PM. Prevention of necrotizing enterocolitis with probiotics: a systematic review and meta-analysis. PeerJ. 2016;4:e2429. Epub 2016/10/21. pmid:27761306; PubMed Central PMCID: PMC5068355.
- 15. Molinaro M, Aiazzi M, La Torre A, Cini E, Banfi R. [Lactobacillus Rhamnosus sepsis in a preterm infant associated with probiotic integrator use: a case report.]. Recenti Prog Med. 2016;107(9):485–6. Epub 2016/10/12. pmid:27727257.
- 16. Pell LG, Loutet MG, Roth DE, Sherman PM. Arguments against routine administration of probiotics for NEC prevention. Current opinion in pediatrics. 2019;31(2):195–201. Epub 2019/01/10. pmid:30624281.
- 17. Dubin K, Pamer EG. Enterococci and Their Interactions with the Intestinal Microbiome. Microbiol Spectr. 2014;5(6). Epub 2014/11/01. pmid:29125098; PubMed Central PMCID: PMC5691600.
- 18. Orrhage K, Nord CE. Factors controlling the bacterial colonization of the intestine in breastfed infants. Acta Paediatr Suppl. 1999;88(430):47–57. Epub 1999/11/24. pmid:10569223.
- 19. P OC, Giri R, Hoedt EC, McGuckin MA, Begun J, Morrison M. Enterococcus faecalis AHG0090 is a Genetically Tractable Bacterium and Produces a Secreted Peptidic Bioactive that Suppresses Nuclear Factor Kappa B Activation in Human Gut Epithelial Cells. Frontiers in immunology. 2018;9:790. Epub 2018/05/04. pmid:29720977; PubMed Central PMCID: PMC5915459.
- 20. Isani M, Bell BA, Delaplain PT, Bowling JD, Golden JM, Elizee M, et al. Lactobacillus murinus HF12 colonizes neonatal gut and protects rats from necrotizing enterocolitis. PloS one. 2018;13(6):e0196710. Epub 2018/06/23. pmid:29933378; PubMed Central PMCID: PMC6014650.
- 21. Barnes AMT, Dale JL, Chen Y, Manias DA, Greenwood Quaintance KE, Karau MK, et al. Enterococcus faecalis readily colonizes the entire gastrointestinal tract and forms biofilms in a germ-free mouse model. Virulence. 2017;8(3):282–96. Epub 2016/08/27. pmid:27562711; PubMed Central PMCID: PMC5411234.
- 22. Normann E, Fahlen A, Engstrand L, Lilja HE. Intestinal microbial profiles in extremely preterm infants with and without necrotizing enterocolitis. Acta paediatrica (Oslo, Norway: 1992). 2013;102(2):129–36. Epub 2012/10/23. pmid:23082780.
- 23. Stewart CJ, Marrs EC, Magorrian S, Nelson A, Lanyon C, Perry JD, et al. The preterm gut microbiota: changes associated with necrotizing enterocolitis and infection. Acta paediatrica (Oslo, Norway: 1992). 2012;101(11):1121–7. Epub 2012/08/01. pmid:22845166.
- 24. Tendolkar PM, Baghdayan AS, Shankar N. Pathogenic enterococci: new developments in the 21st century. Cellular and molecular life sciences: CMLS. 2003;60(12):2622–36. Epub 2003/12/20. pmid:14685687.
- 25. Papillon S, Castle SL, Gayer CP, Ford HR. Necrotizing enterocolitis: contemporary management and outcomes. Advances in pediatrics. 2013;60(1):263–79. Epub 2013/09/07. pmid:24007848.
- 26. Nadler EP, Dickinson E, Knisely A, Zhang XR, Boyle P, Beer-Stolz D, et al. Expression of inducible nitric oxide synthase and interleukin-12 in experimental necrotizing enterocolitis. The Journal of surgical research. 2000;92(1):71–7. Epub 2000/06/23. pmid:10864485.
- 27. O'Toole GA, Kolter R. Initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic analysis. Mol Microbiol. 1998;28(3):449–61. Epub 1998/06/19. pmid:9632250.
- 28. Toledo-Arana A, Valle J, Solano C, Arrizubieta MJ, Cucarella C, Lamata M, et al. The enterococcal surface protein, Esp, is involved in Enterococcus faecalis biofilm formation. Applied and environmental microbiology. 2001;67(10):4538–45. Epub 2001/09/26. pmid:11571153; PubMed Central PMCID: PMC93200.
- 29. Stepanovic S, Vukovic D, Dakic I, Savic B, Svabic-Vlahovic M. A modified microtiter-plate test for quantification of staphylococcal biofilm formation. J Microbiol Methods. 2000;40(2):175–9. Epub 2000/03/04. pmid:10699673.
- 30. R Development Core Team. R: A language and environment for statisical computing. Vienna, Austria: R Foundation for Statistical Computing; 2010.
- 31. Anderson AC, Andisha H, Hellwig E, Jonas D, Vach K, Al-Ahmad A. Antibiotic Resistance Genes and Antibiotic Susceptibility of Oral Enterococcus faecalis Isolates Compared to Isolates from Hospitalized Patients and Food. Advances in experimental medicine and biology. 2018;1057:47–62. Epub 2017/06/12. pmid:28601926.
- 32. Maharshak N, Huh EY, Paiboonrungruang C, Shanahan M, Thurlow L, Herzog J, et al. Enterococcus faecalis Gelatinase Mediates Intestinal Permeability via Protease-Activated Receptor 2. Infection and immunity. 2015;83(7):2762–70. Epub 2015/04/29. pmid:25916983; PubMed Central PMCID: PMC4468563.
- 33. Chen PW, Tseng SY, Huang MS. Antibiotic Susceptibility of Commensal Bacteria from Human Milk. Current microbiology. 2016;72(2):113–9. Epub 2015/10/24. pmid:26494365.
- 34. Landete JM, Peiroten A, Medina M, Arques JL, Rodriguez-Minguez E. Virulence and Antibiotic Resistance of Enterococci Isolated from Healthy Breastfed Infants. Microbial drug resistance (Larchmont, NY). 2018;24(1):63–9. Epub 2017/07/15. pmid:28708453.
- 35. He Q, Hou Q, Wang Y, Li J, Li W, Kwok LY, et al. Comparative genomic analysis of Enterococcus faecalis: insights into their environmental adaptations. BMC genomics. 2018;19(1):527. Epub 2018/07/13. pmid:29996769; PubMed Central PMCID: PMC6042284.