Hypothesis: inappropriate colonization of the

premature intestine can cause neonatal

necrotizing enterocolitis
ERIKA C. CLAUD AND W. ALLAN WALKER1
Developmental Gastroenterology Laboratory, Combined Program in Pediatric Gastroenterology and
Nutrition, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA

ABSTRACT Neonatal necrotizing enterocolitis (NEC) intestinal colonization. Before birth, the infant gut is
is a major cause of morbidity in preterm infants. We sterile. Thereafter, it acquires bacteria from the envi-
hypothesize that the intestinal injury in this disease is a ronment. The intestine is initially colonized with a
consequence of synergy among three of the major risk complex flora that reflects maternal vaginal and large
factors for NEC: prematurity, enteral feeding, and intestinal flora (3, 4). The preterm infant is next
bacterial colonization. Together these factors result in exposed to bacteria in the NICU, with colonization
an exaggerated inflammatory response, leading to isch- frequently affected by the use of broad-spectrum anti-
emic bowel necrosis. Human milk may decrease the biotics (5–7). Although a range of aerobic and anaer-
incidence of NEC by decreasing pathogenic bacterial obic flora colonizes normal infants by 10 days of age,
colonization, promoting growth of nonpathogenic infants in the NICU undergo a delayed colonization
flora, promoting maturation of the intestinal barrier, with a limited number of bacterial species, which tend
and ameliorating the proinflammatory response.— to be virulent (2, 8).
Claud, E. C., Walker, W. A. Hypothesis: inappropriate Feeding is another variable in the acquisition of
colonization of the premature intestine can cause neo- intestinal flora. In breast-fed infants, bifidobacterium is
natal necrotizing enterocolitis. FASEB J. 15, 1398 –1403 a primary organism, with lactobacillus and streptococ-
(2001) cus as minor components. In formula-fed infants, sim-
ilar amounts of bacteroides and bifidobacterium are
Key Words: necrotizing enterocolitis 䡠 prematurity 䡠 intestinal found with minor components of the more pathogenic
inflammation species staphylococcus, Escherichia coli, and clostridia
(3). Since binding of pathogenic organisms can be
influenced by the underlying microbial ecology
NEONATAL NECROTIZING ENTEROCOLITIS through competition for binding sites or nutrients,
production of inhibiting agents, alteration in pH, and
Necrotizing enterocolitis (NEC) is the most com- synthesis of growth factors, promotion of the growth of
mon gastrointestinal emergency in the neonatal inten- competitive nonpathogenic strains of bacteria may pro-
sive care unit (NICU). It is characterized by gastro- tect the infant (8 –10).
intestinal dysfunction progressing to pneumatosis intes-
tinalis, pneumoperitoneum, systemic shock, and rapid
death in severe cases (1). Despite its significant effect
RESPONSE OF THE INTESTINE TO BACTERIAL
on preterm morbidity, the exact etiology and pathogen-
COLONIZATION
esis of this disease have not been clearly delineated.
The most common risk factors cited are prematurity,
enteral feeding, and bacterial colonization (2). Here Microbial pathogens have complex interactions with
we intend to provide evidence to support our hypoth- their host cells, resulting in a variety of responses that
esis that the intestinal injury in NEC may be the result can lead to injury. Signal transduction leading to
of synergy of these three risk factors, in which feeding nuclear factor ␬B (NF-␬B) activation is a key element in
results in colonization of the uniquely susceptible pre- the intestinal epithelial cell innate immune response
mature intestine with pathogenic bacteria, resulting in (11). In its inactive state, NF-␬B is bound to the
an exaggerated inflammatory response. inhibitory protein I␬B. Once I␬B is phosphorylated, it
can be degraded by ubiquitination, freeing NF-␬B to
translocate to the nucleus and stimulate transcription
COLONIZATION OF THE NEWBORN of inflammatory mediators (12). In cultured intestinal
INTESTINE
1
Correspondence: Developmental Gastroenterology Labora-
Environment, variation in pH, intestinal peristalsis, tory, Massachusetts General Hospital, 149 13th St., Charlestown,
bacterial opposition, and the type of feeding all affect MA 02129, USA. E-mail: walker@helix.mgh.harvard.edu

1398 0892-6638/01/0015-1398 © FASEB

epithelial cells, it has been shown that Salmonella typhi- factors such as lactoferrin and lysozyme. Together,
murium activates signal transduction pathways leading these features may partially explain this difference in
to activation of NF-␬B (11, 13, 14). Other pathogens, protection of the newborn infant.
including E. coli, Yersinia, Shigella, and Listeria, trigger
signal transduction pathways that degrade I␬B, freeing
NF-␬B to translocate to the nucleus and up-regulate THE ROLE OF BACTERIA AND
expression of proinflammatory mediators. In contrast, INFLAMMATION IN NEC
not only can colonization with commensal organisms
competitively reduce colonization by these pathological Bacteria have long been suspected of having a role in
organisms, it may also have anti-inflammatory proper- NEC; however, only 30% of NEC patients have positive
ties. A recent study by Madara et al. has shown that blood cultures (6). Epidemics of NEC have occurred, but
nonpathogenic Salmonella prevents degradation of I␬B, NEC usually occurs sporadically and has not been associ-
limiting further transcription of inflammatory media- ated with a particular pathogen (9). Rather than a direct
tors (12). infection, this disease process may be a result of secondary
inflammatory responses to the organisms (15).
Histopathologic sections of intestine from infants
IMMATURE GASTROINTESTINAL IMMUNE with NEC reveal a wide variation. Most show evidence
SYSTEM OF THE PREMATURE INFANT of necrosis and inflammation in addition to bacterial
overgrowth (1, 20, 18). The cycle of intestinal injury
Ninety percent of infants with NEC are born preterm, begins with neutrophil migration, which can mechani-
and it appears that an immature gastrointestinal im- cally disrupt tight junctions between intestinal epithe-
mune defense system places a preterm infant at higher lial cells, increasing transepithelial permeability (21).
risk for NEC. The premature infant has lower gastric Neutrophil activation can lead to inflammatory me-
acid production and lower levels of protective mucus diator release resulting in vasoconstriction, ischemia/
(5). These infants have also been noted to have lower reperfusion injury, oxygen free radical release, and
proteolytic enzyme activity, leading to incomplete further disruption of tight junctions and the intestinal
breakdown of toxins (5). Decreased motility, lower barrier (22).
levels of B and T lymphocytes, and lower levels of Ultimately there is exacerbation of an inflammatory
secretory IgA also increase bacterial adherence to the cascade with increased tissue and serum levels of in-
intestinal mucosa and susceptibility to infection (1, 5, flammatory mediators such as tumor necrosis factor
15). Coupled with increased intestinal mucosal perme- (TNF), interleukin 1 (IL-1), IL-6, IL-8, and platelet-
ability, this potentially leads to the delivery of whole activating factor (PAF) (23, 24). It has been shown that
bacteria and endotoxin to the bloodstream (1). PAF levels are elevated in infants with NEC (23). In rats,
exogenous PAF given intravenously results in ischemic
bowel necrosis (25). Furthermore, studies using a rat
FEEDING AND INTESTINAL IMMATURITY model of NEC showed that PAF receptor blockade
reduced the incidence of NEC (26 –29). Many stimuli
Ninety-five percent of affected infants have been enter- that have been associated with NEC can increase PAF
ally fed, and NEC has never been identified in utero production including hypoxia, enteral feeding (30),
(16). The fetal gut is exposed to amniotic fluid contain- acidosis, TNF, and endotoxin. Once activated, PAF
ing hormones and peptides that may have a role in initiates production of other inflammatory mediators
intestinal maturation and preparation for postnatal (31). This process is exacerbated in preterm infants by
feeding. The preterm infant may not have completed the fact that acetylhydrolase, the enzyme responsible
this maturation process when initially fed. Preterm for PAF degradation, is present in decreased amounts
infants are unable to fully digest carbohydrates and in infants. Acetylhydrolase has also specifically been
proteins, leading to the production of organic acids, shown to be decreased in infants with NEC (Fig. 1;
which may be harmful to the developing intestine (17). 32, 33).
Undigested casein, the protein in formula, can func-
tion as a chemoattractant for neutrophils (18).
Several studies have shown that formula-fed infants INCREASED SUSCEPTIBILITY OF THE
have a higher incidence of NEC than breast-fed infants IMMATURE ENTEROCYTE TO BACTERIAL
(19). A prospective multicenter study of 926 preterm COLONIZATION
infants found a 6- to 10-fold increase in the incidence of
NEC in formula-fed infants compared with breast milk- Our laboratory has studied the interaction between the
fed infants (16). Breast milk contains factors that enterocyte and bacteria and shown that many aspects
enhance intestinal maturation. In addition, human are developmentally regulated. Cell surface glycoconju-
milk provides passive immunity factors such as poly- gates serve as adhesion sites for a variety of microbes.
meric IgA (pIgA) and macrophages that may bind Creation of specific carbohydrate receptors through
antigens, bacteria, or endotoxin and decrease mucosal glycosylation is a developmentally regulated process
translocation. It also contains nonspecific antimicrobial that is important in intestinal colonization and defense.

GUT COLONIZATION AND NEC 1399

 trophils from intravascular to interstitial sites, which
can further exacerbate the inflammatory process. This
excessive proinflammatory response may explain why
primarily preterm infants develop NEC (36).

INCOMPLETE METABOLISM OF FOOD

SUBSTRATE AND INTESTINAL INJURY

The carbohydrate in milk is lactose, which can be

fermented to short-chain fatty acids, hydrogen, lactate,
and CO2 when incompletely metabolized (37). As
mentioned, preterm infants often cannot completely
metabolize carbohydrates and proteins. Formula-fed
infants produce butyrate in the small intestine and
breast-fed infants produce lactic acid. High concentra-
Figure 1. Plasma PAF levels (open bars), plasma TNF levels
(horizontally hatched bars), and plasma acetylhydrolase ac- tions of short-chain fatty acids may have a toxic effect
tivity (vertically hatched bars) in NEC patients and control on the small intestine. They have been shown to lower
subjects. Plasma PAF and TNF levels were higher in NEC the pH of the small intestine, causing mucosal break-
patients than in control patients. Plasma acetylhydrolase down and increasing susceptibility to bacterial translo-
activity was lower in NEC patients than in age-matched cation (18). Furthermore, we have shown that priming
control subjects. Reproduced with permission from Caplan with the short-chain fatty acid butyrate can increase
and Hsueh (33)
IL-8 secretion from enterocytes in response to LPS or
IL1-␤ (38).
Our laboratory has found that newborn intestinal mi-
crovillous preparations have increased sialic acid and
N-acetylglucosamine residues whereas adult prepara- ROLE OF BREAST MILK IN PREVENTION
tions have increased mannose, glucose, and fucose. OF NEC
When glycosyltransferase activity was examined in mice,
Lucas and Cole showed in their prospective study a
a correlating increase in ␣ 2,6-sialyltransferase activity
decline in the incidence of NEC with advancing gesta-
was noted in the newborn period, declining at the time
of weaning. In contrast, ␣ 1,2/1,3 fucosyltransferase
activity was at low levels in the newborn period and
increased at weaning (34, 35). These observations have
also recently been demonstrated in our laboratory in
the human fetus (D. Dai, unpublished data). These
differences may be responsible for increased patho-
genic colonization in preterm infants. We have shown
that pathogenic organisms adhere and translocate
across the intestine to a greater extent in immature vs.
mature animals.

EXAGGERATED RESPONSE OF THE IMMATURE

ENTEROCYTE TO GRAM-NEGATIVE
ORGANISMS

As previously discussed, formula-fed infants are more

likely to be colonized with gram-negative organisms
such as E. coli and Klebsiella than are breast-fed infants
(3). Our laboratory has shown that the enterocyte
response to gram-negative organisms is exaggerated in
fetal cells. IL-8 secretion and mRNA in response to LPS
and IL-1␤ were significantly increased in H4 cells
(human fetal intestinal cell line) and fetal organ cul-
tures as compared with Caco2 (adult intestinal cell
line) or older infant organ culture preparations (Fig. Figure 2. Depiction of IL-8 secretion (A) and IL-8 mRNA
2). Immunohistochemical staining showed that the IL-8 induction (B) in fetal and infant organ culture preparations
was secreted by the enterocyte, not by lymphoid cells. in response to LPS (50 ␮g/ml) or media alone as control.
IL-8 is a chemokine that stimulates migration of neu- Reproduced with permission from Nanthakumar et al. (36)

1400 Vol. 15 June 2001 The FASEB Journal CLAUD AND WALKER
tional age in breast milk-fed but not in formula-fed methasone before PAF injection to induce ischemic
infants. This suggests that maturity of the intestinal bowel necrosis increased acetylhydrolase levels and
mucosal barrier improved by advancing gestational age prevented intestinal injury (51). Other clinical and
and breast milk is important (16). Furthermore, data animal studies have indicated that postnatal cortisone
support the notion that by modulating intestinal colo- may also induce maturation; however, recent analyses
nization and supporting inherent host gastrointestinal raising questions about adverse neurological outcome
defense, breast milk may decrease the incidence of in the face of postnatal steroids give one pause (46, 52).
NEC. Supplementation with growth factors such as EGF or
Breast milk interferes with adherence of pathogenic TGF may also aid in gut maturation.
bacteria by providing pIgA and oligosaccharides. Poly- Probiotics are another area of intense research. A
meric IgA is produced by the immune system of the probiotic is a live anaerobic bacterial food supplement
mother specifically against enteric pathogens to which that benefits the recipient by improving the intestinal
she has been exposed. Oligosaccharides function as flora balance. Probiotics have been shown to be bene-
receptor analog decoys, binding bacteria before they ficial in other conditions, such as C. difficile pseudo-
can adhere to glycoconjugates on the microvillous membranous colitis and rotavirus diarrhea, and have
membrane (39). To promote colonization with non- been shown to restore normal flora after antibiotic
pathogenic bacteria, breast milk contains growth fac- administration (53). Other studies have shown that
tors for bifidobacterium (40). orally administered nonenteropathogenic bacteria can
Breast milk also provides hormones, growth factors, colonize the newborn gut and reduce colonization with
and nucleotides that facilitate the maturation of the pathogenic strains (54, 55). Animal studies have sug-
intestinal mucosal barrier. This may decrease bacterial gested that the administration of bifidobacterium de-
translocation across the intestine. Cortisol, growth hor- creases the incidence of NEC through modulation of
mone, insulin-like growth factor, erythropoietin, and the inflammatory cascade (56). One clinical study has
thyroid hormone have all been identified in breast shown a decreased incidence of NEC with oral admin-
milk. Epidermal growth factor (EGF) is found in high- istration of lactobacillus and bifidobacterium to human
est quantity in colostrum from mothers of preterm infants (57). Although probiotics appear safe in infants,
infants and is a growth stimulator of epithelial cells. their efficacy in preventing NEC in infants still requires
Transforming growth factor (TGF) is found in breast further investigation.
milk and has an immunomodulatory role in addition to Exogenous nucleotides as a formula supplement
inducing differentiation of intestinal epithelial cells have been suggested. Nucleotides are important metab-
(41). Breast milk also provides glutamine and nucleo- olites that serve as a ‘conditional essential nutrient’ in
tides that affect cell growth (42). times of stress. They have a role in controlling cell
To ameliorate the proinflammatory response, breast turnover and a role in immunity (58). There is an
milk contains an IL-1 receptor antagonist, TGF-␤, and increased requirement for nucleotides in rapidly pro-
IL-10. Breast milk also contains acetylhydrolase, which liferating cells such as intestinal epithelial cells and they
metabolizes PAF to inactive lyso-PAF. It has been shown are present in higher concentrations in breast milk
that acetylhydrolase activity is higher in preterm than than in infant formulas (42). They are taken up by
term human milk (43). Furthermore, breast milk may intestinal cell lines and organ culture preparations and
be beneficial by providing antioxidant factors such as have a role in intestinal epithelial cell proliferation and
vitamin E, beta carotene, and ascorbic acid (2, 44). differentiation (59, 60).

OTHER APPROACHES SUMMARY

For infants for whom breast milk is not available, we Necrotizing enterocolitis is a devastating condition with
speculate that it may be prudent to mimic properties of high morbidity and mortality. It specifically affects
breast milk. Several studies have shown that prenatal preterm infants and appears to be the result of an
steroids reduce morbidity from NEC (45, 46). In ani- inflammatory response to pathogenic organisms. The
mal studies, pups from pregnant rats injected with balance of pro- and anti-inflammatory influences ap-
dexamethasone had more mature and ordered mi- pears to be the fundamental issue. We have provided
crovillous membranes, increased jejunal sucrase, in- evidence to support our hypothesis that intestinal in-
creased salivary amylase, increased gastric acid secre- jury in NEC may be the consequence of synergy of
tion, increased activity of gastric pepsinogen, increased prematurity, enteral feeding, and bacterial coloniza-
pancreatic amylase, and decreased bacterial transloca- tion. Feeding may result in colonization of the prema-
tion (47, 48). Animal studies have also shown that ture intestine with pathogenic bacteria, resulting in an
prenatal steroids alter bacterial colonization, resulting exacerbated inflammatory response. Preterm infants
in decreased bacterial counts and fewer gram-negative are uniquely susceptible because of an immature im-
organisms, possibly by maturation of glycosyltransferase mune system unable to sufficiently protect against
activity (49, 50). A study by Furakawa et al. showed that pathogenic organisms. In addition, immature glycosyl-
in a rat model, pretreatment of the animal with dexa- ation patterns may actually promote colonization by

GUT COLONIZATION AND NEC 1401

pathogenic bacteria, and the immune response of the 16. Lucas, A., and Cole, T. J. (1990) Breast milk and neonatal
necrotising enterocolitis. Lancet 336, 1519 –1523
immature enterocyte may itself increase injury by ex-
17. Clark, D. A., and Miller, M. J. (1990) Intraluminal pathogenesis
cessive cytokine production in response to gram-nega- of necrotizing enterocolitis. J. Pediatr. 117, S64 –S67
tive organisms. We suggest that breast milk feeding can 18. Carbonaro, C. A., Clark, D. A., and Elseviers, D. (1988) A
reduce colonization by pathogenic organisms and in- bacterial pathogenicity determinant associated with necrotizing
enterocolitis. Microb. Pathog. 5, 427– 436
duce colonization by commensal organisms. This may 19. Beeby, P. J., and Jeffery, H. (1992) Risk factors for necrotising
modulate inflammatory reactions, decreasing intestinal enterocolitis: the influence of gestational age. Arch. Dis. Child.
injury. 67, 432– 435
20. Ballance, W. A., Dahms, B. B., Shenker, N., and Kliegman, R. M.
(1990) Pathology of neonatal necrotizing enterocolitis: a ten-
Supported by grants from the National Institutes of Health year experience. J. Pediatr. 117, S6 –S13
(RO1-HD31852, R37-HD12437, and PO1-DK33506). 21. Madara, J. L. (1990) Warner-Lambert/Parke-Davis Award lec-
ture. Pathobiology of the intestinal epithelial barrier. Am. J.
Pathol. 137, 1273–1281
22. Kinugasa, T., Sakaguchi, T., Gu, S., and Reinecker, H. C. (2000)
REFERENCES Claudins regulate the intestinal barrier in response to immune
mediators. Gastroenterology 118, 1001–1011
1. Faix, R. G., and Adams, J. T. (1994) Neonatal necrotizing 23. Caplan, M. S., Sun, X. M., Hseuh, W., and Hageman, J. R.
enterocolitis: current concepts and controversies. Adv. Pediatr. (1990) Role of platelet activating factor and tumor necrosis
Infect. Dis. 9, 1–36 factor-alpha in neonatal necrotizing enterocolitis. J. Pediatr. 116,
2. Kosloske, A. M. (1994) Epidemiology of necrotizing enterocoli- 960 –964
tis. Acta Paediatr. Suppl. 396, 2–7 24. Harris, M. C., Costarino, A. T., Jr., Sullivan, J. S. Dulkerian, S.,
3. Harmsen, H. J., Wildeboer-Veloo, A. C., Raangs, G. C., Wagen- McCawley, L., Corcoran, L., Butler, S., and Kilpatrick, L. (1994)
dorp, A. A., Klijn, N., Bindels, J. G., and Welling, G. W. (2000) Cytokine elevations in critically ill infants with sepsis and
Analysis of intestinal flora development in breast-fed and for- necrotizing enterocolitis. J. Pediatr. 124, 105–111
mula-fed infants by using molecular identification and detection 25. Gonzalez-Crussi, F., and Hsueh, W. (1983) Experimental model
methods. J. Pediatr. Gastroenterol. Nutr. 30, 61– 67 of ischemic bowel necrosis. The role of platelet-activating factor
4. Gronlund, M. M., Lehtonen, O. P., Eerola, E., and Kero, P. and endotoxin. Am. J. Pathol. 112, 127–135
(1999) Fecal microflora in healthy infants born by different 26. Caplan, M. S., Kelly, A., and Hsueh, W. (1992) Endotoxin and
methods of delivery: permanent changes in intestinal flora after hypoxia-induced intestinal necrosis in rats: the role of platelet
cesarean delivery. J. Pediatr. Gastroenterol. Nutr. 28, 19 –25 activating factor. Pediatr. Res. 31, 428 – 434
5. Udall, J. N., Jr. (1990) Gastrointestinal host defense and necro- 27. Caplan, M. S., Hedlund, E., Adler, L., Lickerman, M., and
tizing enterocolitis. J. Pediatr. 117, S33–S43 Hsueh, W. (1997) The platelet-activating factor receptor antag-
6. Hoy, C., Millar, M. R., MacKay, P., Godwin, P. G., Langdale, V., onist WEB 2170 prevents neonatal necrotizing enterocolitis in
and Levene, M. I. (1990) Quantitative changes in faecal micro- rats. J. Pediatr. Gastroenterol. Nutr. 24, 296 –301
flora preceding necrotising enterocolitis in premature neo- 28. Hsueh, W., Gonzalez-Crussi, F., and Arroyave, J. L. (1987)
nates. Arch. Dis. Child. 65, 1057–1059 Platelet-activating factor: an endogenous mediator for bowel
7. Murray, R. D. (1990) Effects of bacterial fermentation end necrosis in endotoxemia. FASEB J. 1, 403– 405
products on intestinal function: implications for intestinal dys- 29. Wallace, J. L., Steel, G., Whittle, B. J., Lagente, V., and Vargaftig,
function. J. Pediatr. 117, S59 –S63 B. (1987) Evidence for platelet-activating factor as a mediator of
8. Orrhage, K., and Nord, C. E. (1999) Factors controlling the endotoxin-induced gastrointestinal damage in the rat. Effects of
bacterial colonization of the intestine in breast-fed infants. Acta three platelet-activating factor antagonists. Gastroenterology 93,
Paediatr. Suppl. 88, 47–57 765–773
9. Panigrahi, P., Gupta, S., Gewolb, I. H., and Morris, J. G., Jr. 30. MacKendrick, W., Hill, N., Hsueh, W., and Caplan, M. (1993)
(1994) Occurrence of necrotizing enterocolitis may be depen- Increase in plasma platelet-activating factor levels in enterally
dent on patterns of bacterial adherence and intestinal coloni- fed preterm infants. Biol. Neonate 64, 89 –95
zation: studies in Caco-2 tissue culture and weanling rabbit 31. Caplan, M. S., and MacKendrick, W. (1994) Inflammatory
models. Pediatr. Res. 36, 115–121 mediators and intestinal injury. Clin. Perinatol. 21, 235–246
10. Lodinova-Zadnikova, R., Slavikova, M., Tlaskalova-Hogenova, 32. Caplan, M., Hsueh, W., Kelly, A., and Donovan, M. (1990)
H., Adlerberth, I., Hanson, W. A., Wold, A., Carlsson, B., Serum PAF acetylhydrolase increases during neonatal matura-
Svanborg, C., and Mellander, L. (1991) The antibody response tion. Prostaglandins 39, 705–714
in breast-fed and non-breast-fed infants after artificial coloniza- 33. Caplan, M. S., and Hsueh, W. (1990) Necrotizing enterocolitis:
tion of the intestine with Escherichia coli O83. Pediatr. Res. 29, role of platelet activating factor, endotoxin, and tumor necrosis
396 –399 factor. J. Pediatr. 117, S47–S51
11. Elewaut, D., DiDonato, J. A., Kim, J. M., Truong, F., Eckmann, 34. Chu, S. H., and Walker, W. A. (1986) Developmental changes in
L., and Kagnoff, M. F. (1999) NF-kappa B is a central regulator the activities of sialyl- and fucosyltransferases in rat small intes-
of the intestinal epithelial cell innate immune response induced tine. Biochim. Biophys. Acta 883, 496 –500
by infection with enteroinvasive bacteria. J. Immunol. 163, 1457– 35. Dai, D., Nanthkumar, N. N., Newburg, D. S., and Walker, W. A.
1466 (2000) Role of oligosaccharides and glycoconjugates in intesti-
12. Neish, A. S., Gewirtz, A. T., Zeng, H., Young, A. N., Hobert, nal host defense. J. Pediatr. Gastroenterol. Nutr. 30, S23–S33
M. E., Karmali, V., Rao, A. S., and Madara, J. L. (2000) 36. Nanthakumar, N. N., Fusunyan, R. D., Sanderson, I., and
Prokaryotic regulation of epithelial responses by inhibition of Walker, W. A. (2000) Inflammation in the developing human
IkappaB-alpha Science 289, 1560 –1563 intestine: a possible pathophysiologic contribution to necrotiz-
13. Hobbie, S., Chen, L. M., Davis, R. J., and Galan, J. E. (1997) ing enterocolitis. Proc. Natl. Acad. Sci. USA 97, 6043– 6048
Involvement of mitogen-activated protein kinase pathways in 37. Kien, C. L. (1990) Colonic fermentation of carbohydrate in the
the nuclear responses and cytokine production induced by premature infant: possible relevance to necrotizing enterocoli-
Salmonella typhimurium in cultured intestinal epithelial cells. tis. J. Pediatr. 117, S52–S58
J. Immunol. 159, 5550 –5559 38. Fusunyan, R. D., Quinn, J. J., Ohno, Y., MacDermott, R. P., and
14. Gewirtz, A. T., Rao, A. S., Simon, P. O., Jr., Merlin, D., Carnes, Sanderson, I. R. (1998) Butyrate enhances interleukin (IL)-8
D., Madara, J. L., and Neish, A. S. (2000) Salmonella typhi- secretion by intestinal epithelial cells in response to IL-1beta
murium induces epithelial IL-8 expression via Ca2⫹-mediated and lipopolysaccharide. Pediatr. Res. 43, 84 –90
activation of the NF-kappaB pathway. J. Clin. Invest. 105, 79 –92 39. Newburg, D. S. (2000) Oligosaccharides in human milk and
15. Caplan, M. S., and MacKendrick, W. (1993) Necrotizing entero- bacterial colonization. J. Pediatr. Gastroenterol. Nutr. 30, S8 –S17
colitis: a review of pathogenetic mechanisms and implications 40. Petschow, B. W., Talbott, R. D., and Batema, R. P. (1999) Ability
for prevention. Pediatr. Pathol. 13, 357–369 of lactoferrin to promote the growth of Bifidobacterium spp. in

1402 Vol. 15 June 2001 The FASEB Journal CLAUD AND WALKER
 vitro is independent of receptor binding capacity and iron 51. Furukawa, M., Lee, E. L., and Johnston, J. M. (1993) Platelet-
saturation level. J. Med. Microbiol. 48, 541–549 activating factor-induced ischemic bowel necrosis: the effect of
41. Bernt, K. M., and Walker, W. A. (1999) Human milk as a carrier platelet-activating factor acetylhydrolase. Pediatr. Res. 34, 237–
of biochemical messages. Acta Paediatr. Suppl. 88, 27– 41 241
42. Barness, L. A. (1994) Dietary sources of nucleotides—from 52. Yeh, T. F., Lin, Y. J., Huang, C. C., Chen, Y. J., Lin, C. H., Lin,
breast milk to weaning. J. Nutr. 124, 128S–130S H. C., Hsieh, W. S., and Lien. Y. J. (1998) Early dexamethasone
43. Moya, F. R., Eguchi, H., Zhao, B., Furukawa, M., Sfeir, J., Osorio, therapy in preterm infants: a follow-up study. Pediatrics 101, E7
M., Ogawa, Y., and Johnston, J. M. (1994) Platelet-activating 53. Bennet, R., Nord, C. E., and Zetterstrom, R. (1992) Transient
factor acetylhydrolase in term and preterm human milk: a colonization of the gut of newborn infants by orally adminis-
preliminary report. J. Pediatr. Gastroenterol. Nutr. 19, 236 –239 tered bifidobacteria and lactobacilli. Acta Paediatr. 81, 784 –787
44. Aynsley-Green, A., Lucas, A., Lawson, G. R., and Bloom, S. R. 54. Lodinova-Zadnikova, R., and Sonnenborn, U. (1997) Effect of
(1990) Gut hormones and regulatory peptides in relation to preventive administration of a nonpathogenic Escherichia coli
enteral feeding, gastroenteritis, and necrotizing enterocolitis in strain on the colonization of the intestine with microbial
infancy. J. Pediatr. 117, S24 –S32 pathogens in newborn infants. Biol. Neonate 71, 224 –232
45. Crowley, P., Chalmers, I., and Keirse, M. J. (1990) The effects of 55. Rastegar Lari, A., Gold, F., Borderon, J. C., Laugier, J., and
corticosteroid administration before preterm delivery: an over- Lafont, J. P. (1990) Implantation and in vivo antagonistic effects
view of the evidence from controlled trials. Br. J. Obstet. Gynaecol. of antibiotic-susceptible Escherichia coli strains administered to
97, 11–25 premature newborns. Biol. Neonate 58, 73–78
46. Halac, E., Halac, J., Begue, E. F., Casanas, J. M., Indiveri, D. R., 56. Caplan, M. S., and Jilling, T. (2000) Neonatal necrotizing
Petit, J. F., Figueroa, M. J., Olmas, J. M., Rodriguez, L. A., enterocolitis: possible role of probiotic supplementation. J. Pe-
Obregon, R. J., et al. (1990) Prenatal and postnatal corticoste- diatr. Gastroenterol. Nutr. 30, S18 –S22
roid therapy to prevent neonatal necrotizing enterocolitis: a 57. Hoyos, A. B. (1999) Reduced incidence of necrotizing entero-
controlled trial. J. Pediatr. 117, 132–138 colitis associated with enteral administration of Lactobacillus
47. Israel, E. J., Schiffrin, E. J., Carter, E. A., Freiberg, E., and acidophilus and Bifidobacterium infantis to neonates in an
Walker, W. A. (1990) Prevention of necrotizing enterocolitis in intensive care unit. Int. J. Infect. Dis. 3, 197–202
the rat with prenatal cortisone. Gastroenterogy 99, 1333–1338 58. Walker, W. A. (1996) Exogenous nucleotides and gastrointesti-
48. Neu, J., Ozaki, C. K., and Angelides, K. J. (1986) Glucocorticoid- nal immunity. Transplant Proc. 28, 2438 –2441
mediated alteration of fluidity of brush border membrane in rat 59. Sanderson, I. R., and He, Y. (1994) Nucleotide uptake and
small intestine. Pediatr. Res. 20, 79 – 82 metabolism by intestinal epithelial cells. J. Nutr. 124, 131S–137S
49. Schiffrin, E. J., Carter, E. A., Walker, W. A., Frieberg, E., 60. Tanaka, M., Lee, K., Martinez-Augustin, O., He, Y., Sanderson,
Benjamin, J., and Israel, E. J. (1993) Influence of prenatal I. R., and Walker, W. A. (1996) Exogenous nucleotides alter the
corticosteroids on bacterial colonization in the newborn rat. proliferation, differentiation and apoptosis of human small
J. Pediatr. Gastroenterol. Nutr. 17, 271–275 intestinal epithelium. J. Nutr. 126, 424 – 433
50. Ozaki, C. K., Chu, S. H., and Walker, W. A. (1989) Developmen-
tal changes in galactosyltransferase activity in the rat small Received for publication January 19, 2001.
intestine. Biochim. Biophys. Acta 991, 243–247 Accepted for publication February 15, 2001.

GUT COLONIZATION AND NEC 1403