Seminars in Pediatric Surgery (2005) 14, 49-57

Disordered enterocyte signaling and intestinal barrier

dysfunction in the pathogenesis of necrotizing
enterocolitis
David J. Hackam, Jeffrey S. Upperman, Anatoly Grishin, Henri R. Ford

From the Division of Pediatric Surgery, Children’s Hospital of Pittsburgh, and the Departments of Surgery, Cell Biology
and Physiology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania.

KEYWORDS Necrotizing enterocolitis (NEC) is the leading cause of death from gastrointestinal disease in neonates,
Nitric oxide; and is characterized by the development of diffuse intestinal necrosis in the stressed, pre-term infant.
Intestinal restitution; Systemic stress causes a breakdown in the intestinal mucosal barrier, which leads to translocation of
Enterocyte migration; bacteria and endotoxin and the initiation of a signaling response within the enterocyte. This review
Endotoxin; summarizes recent evidence defining a clear role that defective enterocyte signaling plays in the
Cox-2; pathogenesis of NEC through the following mechanisms: 1) The localized production of nitric oxide by
RhoA; villus enterocytes results in an increase in enterocyte apoptosis and impaired proliferation; 2) The
Intestinal barrier translocation of endotoxin results in a PI3K-dependent activation of RhoA-GTPase within the entero-
function cyte leading to decreased enterocyte migration and impaired restitution; 3) Dysregulated sodium-proton
exchange within the enterocyte by endotoxin renders the enterocyte monolayer more susceptible to
damage in the face of the acidic microenvironment characteristic of systemic sepsis; and 4) Endotoxin
causes a p38-dependent release of the pro-inflammatory molecule COX-2 by the enterocyte, which
potentiates the systemic inflammatory response. An understanding of the mechanisms by which
disordered enterocyte signaling contributes to the pathogenesis of barrier failure and NEC—through
these and other mechanims—may lead to the identification of novel therapeutic approaches for this
devastating disease.
© 2005 Elsevier Inc. All rights reserved.

Necrotizing enterocolitis (NEC) is the leading cause of
death from gastrointestinal disease in neonates.1 Complica-
tions related to NEC represent the major indications for
surgical intervention in neonates, and NEC remains a lead-
ing cause of intestinal failure in the pediatric population.2
The high incience of NEC in premature infants3 and the lack
of effective treatment strategies suggest that novel therapeu-
tic approaches are required to improve the survival of in-
fants afflicted with this disease. Elucidation of the molecular
Address reprint requests and correspondence: Henri R. Ford, Di-
vision of Pediatric Surgery, 4A 486 Desoto Wing, Children’s Hospital of
Pittsburgh, 3705 Fifth Avenue, Pittsburgh, Pennsylvania 15217.
E-mail address: Henri.ford@chp.edu.
mechanisms contributing to the pathogenesis of NEC is
essential to derive new treatment modalities.
Studies from our laboratory and the laboratories of others
have indicated a clear role for defective enterocyte signaling
in the pathogenesis of NEC. Rather than serving as a mere
absorptive surface for nutrients, the enterocytes form a tight
epithelial barrier that restricts the passage of microbial
pathogens and regulates mucosal antigen sampling. The
dynamic interface between the enterocyte monolayer and
the intestinal microbial flora has profound implications for
the host immune system. The enterocytes are capable of
functioning as immune effector cells because they can sense
the presence of pathogenic organisms, and when stimulated

1055-8586/$ -see front matter © 2005 Elsevier Inc. All rights reserved.
doi:10.1053/j.sempedsurg.2004.10.025
50 Seminars in Pediatric Surgery, Vol 14, No 1, February 2005

mediators, and underscores the importance of understanding

Clinical observation Biologic relevance the signaling cascades, within the enterocyte and within
• Premature infant • Susceptible host other immune cells, that lead to NEC.
Each of these clinical observations and biological corre-
• Systemic stress • Gut barrier damage
lates reflects the importance of enterocyte signaling and the
• Pneumatosis • Bacterial translocation development of intestinal barrier failure in the pathogenesis
of this disease. We will now describe our general approach
• Systemic inflammation • Cytokine release to understanding how barrier failure leads to NEC, then
Figure 1 Clinical and biological correlates in the pathogenesis explore various signaling cascades that lead to barrier dys-
of NEC. function in NEC.

Hypothesis: Intestinal barrier failure results in

by such microbes, can mount an immune response resulting
in the release of pro-inflammatory cytokines and other me- the development of NEC
diators. This cascade must be tightly regulated, as pertur-
bations in enterocyte signaling can lead to disruption of the We and others have proposed the following unifying
epithelial barrier, bacterial translocation and activation of hypothesis regarding the pathogenesis of NEC (see Figure
the inflammatory cascade resulting in systemic sepsis and 2). In the setting of an episode of perinatal stress, such as
full-blown NEC. An understanding of the mechanisms gov- respiratory distress syndrome or systemic hypoxia, the pre-
erning enterocyte signaling in the neonate, and the effects of mature infant suffers a period of intestinal ischemia that
disordered enterocyte signaling on intestinal function, are
critical in the elucidation of the mechanisms of NEC. This
review will explore the mechanisms by which disordered
enterocyte signaling can lead to the early steps resulting in
intestinal inflammation, mucosal necrosis and systemic sep-
sis—the hallmark of NEC. We will begin by listing a variety
of clinical observations that implicate various putative fac-
tors in the development of NEC. Subsequently, we will
propose a model whereby disordered enterocyte signaling
can lead to a breakdown of the mucosal barrier, and suggest
possible therapeutic approaches to combat this devastating
disease.

From clinical observations to scientific

investigation

A variety of clinical features of NEC have shed light on

possible mechanisms contributing to this disease. These are
listed in Figure 1, and are summarized below. First, NEC,
by its very definition, occurs predominantly in premature
infants. The premature infant represents a vulnerable host in
whom disordered enterocyte signaling can be expected to
exert dramatic effects, not observed in older children and
adults. Second, NEC is known to develop after a systemic
stress, such as systemic hypoxia, which may result from a
nuchal cord, or from associated congenital heart disease.
Such a hypoxic insult can lead to direct damage to the
intestinal mucosa. Third, NEC is characterized by the find-
ing of pneumatosis within the wall of the involved intestine
in the vast majority of cases. Pneumatosis results from the
release of bacterial gases within the wall of the intestine and
illustrates both the importance of bacterial colonization, and
the potential consequences of bacterial-enterocyte interac-
tions in the pathogenesis of this disease. Finally, advanced
NEC is associated with profound systemic sepsis. The sys-
temic inflammatory response that characterizes NEC re- Figure 2 The role of intestinal barrier failure in the pathogenesis
flects the release of pro-inflammatory cytokines and other of NEC.
Hackam et al Pathogenesis of Necrotizing Enterocolitis
results in mucosal injury and disruption of the neonatal gut
barrier: a functional and anatomic region of the intestinal
lining that normally limits transmucosal passage of micro-
bial pathogens (see below). As a result, the damaged intes-
tinal mucosa can be readily breached by indigenous micro-
organisms that translocate across it. The translocated
bacteria then initiate an inflammatory cascade characterized
by the release of various mediators which, in turn, may be
responsible for the systemic manifestations of NEC. To
fully understand the pathogenesis of NEC, it is essential to
define the mechanisms responsible for maintaining the gut
barrier in the healthy neonate.
Components of the neonatal mucosal barrier
The mucosal barrier consists of a complex defense sys-
tem that is designed to prevent enteric bacteria and associ-
ated toxins from invading the body.4 Destruction of the
epithelial barrier can occur through a variety of mecha-
nisms, including hypoxia, bacterial or viral infections. Cen-
tral to the maintenance of barrier function are a variety of
repair mechanisms designed to counteract these destructive
insults. Disruption of these repair mechanisms can result in
uncontrolled transfer of pathogenic antigen or bacteria
across the intestinal mucosa. These antigenic stimuli, in
turn, initiate a systemic inflammatory response that leads to
the development of NEC.5 The effects of these stimuli on
the enterocyte and the mechanism by which bacterial– epi-
thelial interactions can lead to barrier dysfunction are de-
scribed below.
Mechanical factors contributing to barrier integrity
Various mechanical barriers restrict the amount of antigen
reaching the epithelial surface of the intestine. These include
intestinal peristalsis, the mucus coat, and secreted antimicro-
bial factors such as secretory IgA (sIgA) and intestinal trefoil
factor. Active peristalsis limits exposure time of antigens to the
intestinal mucosa and promotes expulsion of antigen–antibody
complexes formed in the mucus coat.6 The mucous layer,
which is composed of the glycoprotein, mucin, secreted by
goblet cells, limits adhesion of pathogenic antigens or mi-
crobes to the intestinal epithelium.7 Goblet cells also secrete
peptide-based reagents termed “trefoil” peptides,8 which main-
tain the integrity of the mucosal layer. Secretory IgA released
in the intestinal lumen acts as an antiseptic paint, binding
bacteria.9 The intestinal epithelium itself serves as a physical
barrier to antigenic and microbial transport. The microvillus
architecture along with its negatively charged surface restricts
the transmucosal transport of charged molecules larger than 25
nm.10 Tight junctions exist between adjacent enterocytes and
form a barrier to the diffusion of large molecules.11 Impair-
ment of these mechanical factors or disruption of underlying
repair mechanisms, as we shall see later, permits uncontrolled
transfer of antigen or microbes across the damaged epithelium,
and results in the development of NEC.
The above-mentioned barrier mechanisms are not fully de-
veloped at birth. Coordinated peristaltic activity develops late
51
in gestation and may not be fully developed until the eighth
month of gestation,12 which promotes stasis and increases the
likelihood of bacterial adhesion to the epithelium; the first and
necessary step in the process of transmucosal passage. The
mucous layer is relatively deficient in preterm infants due to
the immaturity of the intestinal goblet cells,13 which further
increases the propensity for adherence of pathogenic organ-
isms and their subsequent translocation. Synthesis of IgA does
not occur for several weeks postnatally which results in de-
creased clearance of pathogenic bacteria.14 Evidence suggests
that the lipid composition of the enterocyte plasma membrane
changes with development, such that the composition of phos-
pholipid and fatty acyl groups in the newborn membrane con-
tributes to its permeability to large antigens.15 Because of these
perturbations in the preterm intestinal barrier, the infant gut is
more permeable to macromolecular antigenic complexes com-
pared with adults. This phenomenon was shown experimen-
tally by Udall and coworkers who fed rabbits radiolabeled
albumin and measured plasma radioactivity at birth, 1, 2, 6
weeks and 1 year of age.16 They demonstrated a marked
increase in transport of radiolabeled albumin in the younger
age groups. Similar findings have been demonstrated in human
neonates,17 which underscore their susceptibility to transloca-
tion of pathogenic antigens and the development of NEC.
Bacterial factors
In the neonate, indigenous bacteria colonize the mucosal
surfaces and the intestinal lumen in a typical pattern called
succession, which is divided into several phases.18,19 The first
phase lasts from birth to 2 weeks. During this phase, Gram-
positive, nonspore-forming anaerobes appear and include pre-
dominantly Bifidobacteria in breastfed infants, and Lactoba-
cilli in formula-fed infants. The second phase occurs from the
end of the first phase to the consumption of solid food. During
this phase, Bacteroides progressively increase in number until
the flora resembles that of adults. The indigenous intestinal
microflora, especially the anaerobes, serve a protective role by
carpeting the intestinal epithelium, thereby preventing adhe-
sion of pathogenic bacteria. Because there is a relative paucity
of anaerobes during phase I of succession, pathogenic bacteria
may be more likely to colonize the neonatal gut, and thus
contribute to the pathogenesis of NEC, which typically devel-
ops during the first 2 weeks of life.20
Cellular processes regulating mucosal healing
The mucosal barrier stands as a dynamic interface between
luminal toxins and the immune cells in the subepithelial tissue.
A central feature of this barrier is its ability to undergo self-
repair, which is necessary to maintain its integrity and protect
the host from the dysregulated transfer of antigens. One of the
cardinal pathologic features of NEC is the presence of defects
in the intestinal mucosal monolayer. These lesions result in a
loss of barrier integrity, which leads to the translocation of
bacteria and activation of the host immune system. In response
to mucosal injury, a repair process is immediately initiated.
52
Repair of the intestinal mucosa after injury occurs in two
phases: epithelial restitution, which involves the migration of
enterocytes from healthy areas to the sites of injury; and pro-
liferation, in which new enterocytes arise from stem cells in the
crypts. The response to injury occurs in two temporally dis-
parate phases:8,21 epithelial restitution occurs within minutes to
hours and provides a “temporary” seal to the damaged muco-
sa;22 while proliferation occurs over days, and replaces the lost
cells.23,24 Disruption of both restitution and proliferation could
theoretically augment the duration and severity of NEC by
impairing the ability of the mucosa to heal.25
The role of nitric oxide in the pathogenesis
of NEC
Biochemical properties of nitric oxide
Nitric oxide (NO) is a short-lived free radical that reacts
with a variety of biologically active substances.26 Such
reactions result in both local and systemic effects that mod-
ulate the inflammatory response in a variety of tissues.27
The vast and diverse responses attributable to NO justified
its recent naming as the molecule of the year.28 The syn-
thesis of NO is regulated by NO synthase (NOS), which
catalyzes the oxidation of the amino acid L-arginine to
release citrulline and NO. Although diverse molecular tar-
gets of NO have been identified, the fastest and most bio-
logically relevant reaction of NO is with superoxide to
produce the potent oxidant peroxynitrite.29 Peroxynitrite is a
key intermediate that is generated at inflammatory sites in
vivo, and is responsible for mediating tissue injury, in part,
through lipid peroxidation.30
Isoforms of nitric oxide synthase
Three isoforms of NOS exist: neuronal (nNOS) and
endothelial (eNOS), which are expressed constitutively; and
the inducible isoform, iNOS.31 The constitutive and induc-
ible isoforms share approximately 50% sequence homol-
ogy.32 The constitutive isoforms, which generate low con-
centrations of NO, are regulated by intracellular calcium
fluxes via calmodulin. By contrast, iNOS is induced by a
variety of cytokines, growth factors and inflammatory stim-
uli, which lead to release of high levels of NO. All three
NOS isoforms are expressed in the gastrointestinal tract.
The constitutive forms of NOS are expressed by enteric
nerves within the myenteric plexus, gastric epithelial cells
and villus enterocytes.33 By contrast, iNOS expression and
activity within the intestinal epithelium is normally low,
although it may be increased up to 15-fold after stimulation
with lipopolysaccharide.33 The induction of iNOS expres-
sion by inflammatory mediators suggests that NO may par-
ticipate in the pathogenesis of NEC.
Seminars in Pediatric Surgery, Vol 14, No 1, February 2005
Figure 3 The role of nitric oxide in the pathogenesis of necro-
tizing enterocolitis.
A role for NO in the pathogenesis of NEC
According to our overriding hypothesis, pro-inflamma-
tory mediators act on the enterocyte monolayer and contrib-
ute to epithelial injury, leading to further barrier injury and
the development of NEC. An understanding of the molec-
ular components of this “cytokine– enterocyte axis” is crit-
ical in elucidating the pathogenesis of NEC. Data from our
laboratory suggest an important role for NO in damaging
the enterocyte monolayer leading to the development of
NEC. In support of this hypothesis, Ford and coworkers
compared fifteen infants undergoing intestinal resection for
NEC with six infants of similar age undergoing intestinal
resection for a variety of other disorders and found that
iNOS mRNA was detected significantly more often in pa-
tients with NEC compared with controls.34 In situ hybrid-
ization and immunohistochemistry showed that the entero-
cytes were the predominant source of iNOS activity in the
intestine of the NEC patients.34
Further studies have been performed to establish a link
between NO release and barrier failure in the pathogenesis
of NEC. To do so, we have utilized a neonatal rat model of
intestinal inflammation resembling human NEC which in-
volves the administration of a rodent formula by gavage and
thrice daily hypoxic treatments.35 By the fourth day of
treatment, the infant rats develop patchy intestinal (epithe-
lial) necrosis, edema, neutrophil infiltration and necrosis,
which resemble the histopathologic findings in human
NEC.35 Importantly, rats with experimental NEC demon-
strate significant upregulation of iNOS mRNA and protein,
consistent with the findings in the human disease.35,36
Mechanisms of barrier disruption by NO:
Enhanced injury and inhibited repair
What are the mechanisms by which NO could participate
in the pathogenesis of NEC? Studies from our laboratory
have indicated that NO can not only cause direct epithelial
injury, but it can also disrupt the ability of the mucosa to
repair itself. These findings are summarized in Figure 3.
Ford and coworkers showed that extensive apoptosis was
Hackam et al Pathogenesis of Necrotizing Enterocolitis
seen in the enterocytes in the apical villi of infants with
NEC; this observation correlated with the degree of nitro-
tyrosine immunostaining—an in vivo marker of NO release
and reactivity in tissues.34 This finding suggested the pos-
sibility that NO could lead to enterocyte apoptosis. In sup-
port of this hypothesis, peroxynitrite was found to induce
apoptosis in the rat intestinal crypt enterocyte cell line
(IEC-6) in both a time-dependent and dose-dependent man-
ner.37 In explaining the mechanisms involved, we found
that the pro-apoptotic factor, pro-caspase 3, is cleaved and
activated in IEC-6 cells after peroxynitrite exposure.38 In-
hibition of NF-␬B enhanced peroxynitrite-induced entero-
cyte apoptosis, suggesting that NF-␬B may upregulate a
protective factor.37 Moreover, NO has been found to inhibit
mitochondrial transmembrane potential, which may lead to
a reduction in cellular energy stores, and ultimately to cell
death.39 In support of this theory, the glutathione antioxi-
dant system, which serves to preserve energy stores in the
face of oxidative stress, was found to play an important role
in cell survival in an animal model of NEC.40 Together,
these data indicate that NO may contribute to intestinal
injury in NEC by enhancing enterocyte apoptosis. This
concept is supported by our recent findings that strategies
that reduce the release of NO were found to ameliorate the
extent of NEC and reduce enterocyte apoptosis (Cetin and
coworkers, manuscript in preparation).
In addition to the cytopathic effects of NO on the en-
terocyte, we have also shown that NO may impair the
capacity of the mucosa to heal. For instance, peroxynitrite
was found to cause a significant decrease in enterocyte
proliferation, likely by disrupting the Src signaling path-
way.41 We have also shown that enterocyte migration is
inhibited by NO, which leads to a profound reduction in
epithelial restitution (Cetin and coworkers, manuscript in
preparation). Together, these findings indicate that NO in-
hibits the two major pathways required for mucosal healing,
suggesting a causative role in the development of NEC.
Based on these findings, we propose that NO induces
accelerated enterocyte apoptosis at the intestinal villus tip.
Because enterocyte migration and proliferation are impaired
by NO, we propose that NO causes sustained epithelial
damage as evidenced by a persistent area of denuded mu-
cosa. At these sites, bacteria can therefore traverse the
epithelium and initiate or amplify a systemic inflammatory
response that ultimately results in the development of full
blown NEC (Figure 3).
The role of endotoxin in the pathogenesis of
necrotizing enterocolitis
Evidence for the role of endotoxin in the
pathogenesis of intestinal inflammation
The earliest event that occurs after disruption of the
intestinal epithelial barrier is the translocation of bacteria
53
Figure 4 Mechanisms by which endotoxin contributes to barrier
dysfunction in the pathogenesis of necrotizing enterocolitis.
and bacterial by-products across the basement membrane.
This phenomenon results in the activation of the host im-
mune system and the initiation of an inflammatory response.
One of the first pro-inflammatory molecules to cross the
intestinal barrier is endotoxin (lipopolysaccharide, LPS),
which is a principal component of the outer cell wall of
Gram-negative bacteria. Mounting evidence implicates LPS
in the pathogenesis of diseases of intestinal inflammation,
including NEC. Systemic administration of LPS induces gut
inflammation in animals, and circulating LPS is increased in
patients with NEC.42 Once bacterial translocation has oc-
curred, high concentrations of LPS could interact with the
basolateral surface of intestinal epithelial cells directly, and
adversely affect enterocyte function. Indeed, intestinal per-
meability is impaired in cultured epithelia after treatment
with LPS43,44 and bacterial products can impair intestinal
function in vivo, as evidenced by reduced peristalsis, en-
hanced translocation of bacteria and reduced gut immune
function.45 Together, these data support a possible role for
endotoxin in the pathogenesis of NEC. The potential mech-
anisms whereby endotoxin could impair barrier function in
the development of NEC are summarized in Figure 4, and
explained in further detail below.
Recognition of lipopolysaccharide by enterocytes
Recent studies have shed light on how cells recognize
LPS. Mutations in the Drosophila family of Toll receptors
resulted in fatal infection, which led to the identification of
a novel class of homologous proteins in mammalian cells
(Toll-like receptors (TLRs).46,47 TLR4 was found to be the
receptor for LPS through positional cloning of the Lps gene
in LPS resistant mice.48,49 Recognition of LPS by TLR4
requires several accessory molecules, including a serum
protein, LPS-binding protein (LBP), which transfers LPS
monomers to CD14, a GPI-linked component of the LPS
receptor complex and other accessory proteins including
54
MD-1.50 For enterocytes to respond directly to LPS, the
presence of the LPS receptor on the surface of these cells is
a prerequisite. TLR4 was recently found to be expressed by
an enterocyte cell line.51,52 We have shown that enterocytes
can bind LPS, and can internalize fluorescein-labeled endo-
toxin through TLR4.53 Recently, Hornef and coworkers
showed that LPS signaling within enterocytes occurs within
Golgi, and that internalized endotoxin colocalizes with
TLR4 within the cell.51,52 In view of the capacity of entero-
cytes to recognize and signal in response to endotoxin, we
sought to define the mechanisms by which endotoxin sig-
naling could contribute to the pathogenesis of NEC.
Endotoxin inhibits enterocyte migration
As described above, one of the earliest events that occur
in response to mucosal injury is epithelial restitution, in
which healthy enterocytes migrate to sites of injury to
bridge the mucosal defect. Translocation of bacteria and
endotoxin would be expected to occur shortly after a mu-
cosal defect is formed. In view of the importance of epithe-
lial restitution in maintaining barrier integrity, we sought to
investigate whether enterocyte migration was impaired in
NEC, and whether endotoxin had any effect on enterocyte
migration. In an animal model of NEC, we recently found
that intestinal restitution was significantly impaired com-
pared with control animals.53 Therefore, we sought to de-
termine the mechanisms governing enterocyte migration
under basal conditions and after an endotoxin challenge.
Strikingly, endotoxin treatment significantly inhibited epi-
thelial restitution, as measured by impaired IEC-6 cell mi-
gration across a scraped wound.53 To determine the mech-
anisms by which LPS could impair enterocyte migration,
we examined the effect of LPS on the activity of the small
molecular weight protein, Rho-A-GTPase, which we had
determined to play a major role in the formation of stress
fibers in enterocytes.53 LPS was found to increase RhoA
activity in a PI3K-dependent manner, leading to an increase
in phosphorylation of focal adhesion kinase and an en-
hanced number of focal adhesions.53 Importantly, endotoxin
caused a progressive, RhoA-dependent increase in cell-
matrix tension/contractility which correlated with the ob-
served impairment in enterocyte migration. These findings
led us to postulate that endotoxin inhibits enterocyte migra-
tion through a RhoA-dependent increase in focal adhesions
and enhanced cell adhesiveness, which may participate in
the impaired epithelial restitution observed in experimental
NEC.
Endotoxin differentially modulates the
basolateral and apical sodium/proton exchangers
(NHE) in enterocytes
Normal enterocyte function requires tight control of in-
tracellular pH (pHi), and perturbations in pHi cause disrup-
tion in cellular activity.54-57 NEC is characterized by sys-
Seminars in Pediatric Surgery, Vol 14, No 1, February 2005
temic hypoperfusion which leads to metabolic acidosis and
the passive diffusion of protons into the enterocyte.58 This
process results in a trend toward cytoplasmic acidification.
To counter this process, cellular pH regulatory mechanisms
function to transport the protons out of the cell and maintain
pHi homeostasis.59 Impairments in these pH regulatory
mechanisms by endotoxin during conditions of systemic
hypoperfusion could therefore lead to enterocyte acidifica-
tion, barrier disruption and further barrier dysfunction. In
view of the importance of pHi regulation for the mainte-
nance of an intact gut barrier, we sought to define the
mechanisms governing pHi regulation in enterocytes during
conditions of endotoxin stimulation.
The major pHi regulatory mechanism in mammalian
cells is the sodium-proton exchanger, or NHE.60 All seven
NHE isoforms have similar topology, with 12 transmem-
brane domains and a cytoplasmic tail that may modify
exchanger activity through protein modification.61 Although
different isoforms have different tissue expression, NHE1 is
expressed ubiquitously, and NHE1 to 3 are expressed in
epithelia.61 NHE maintains cytoplasmic pH through the
electroneutral exchange of intracellular protons for extra-
cellular sodium. Importantly, in the acidic microenviron-
ment that characterizes NEC, impairments in NHE activity
by endotoxin at the basolateral surface of the enterocyte
after translocation has occurred would lead to cellular acid-
ification and a disruption of the intestinal barrier.62
We recently demonstrated that LPS caused a dose-de-
pendent reduction in basolateral NHE1 activity in entero-
cytes, but had no effect on apical NHE3 activity.63 This
effect could not be explained by reduced expression or
impaired plasma membrane localization of NHE isoforms.
Strikingly, LPS-mediated NHE1 impairment caused marked
cytoplasmic acidification under conditions of extracellular
acidosis, whereas functional NHE1 maintained cytoplasmic
pH homeostasis in control cells. These data provide further
evidence that LPS could modulate enterocyte function after
LPS translocation and suggests an additional mechanism
leading to barrier disruption in NEC (Figure 4).63
Endotoxin initiates an inflammatory signaling
cascade within the enterocyte leading to the
release of cyclooxygenase 2
From the foregoing studies, it is apparent that endotoxin
initiates a signaling cascade within the enterocyte, the net
effect of which is a marked disruption in the integrity of the
gut barrier. And although NEC is characterized by the influx
of inflammatory cells into the site of mucosal injury, we
postulate that endotoxin and other pro-inflammatory medi-
ators can have adverse effects on the enterocyte monolayer
before the influx of neutrophils and monocytes has oc-
curred. By extension, therefore, the enterocyte monolayer
may be endowed with immunological properties itself, and
the release of cytokines from the enterocytes in response to
endotoxin stimulation may contribute to the pathogenesis of
Hackam et al Pathogenesis of Necrotizing Enterocolitis
NEC. In support of this concept, treatment of enterocytes
with endotoxin leads to the release of nitric oxide and
interferon gamma.64 Cyclooxygenase-2 (COX-2) is in-
volved in the biosynthesis of prostaglandins, potent pro-
inflammatory molecules that are released at high levels at
sites of inflammation.65 Because NEC is associated with
high levels of prostaglandins, we hypothesized that COX-2
could be released by enterocytes in response to endotoxin
stimulation. In support of this hypothesis, COX-2 expres-
sion was significantly increased in IEC-6 enterocytes in
response to endotoxin treatment.66 We further hypothesized
that LPS initiates a signaling cascade in the enterocyte,
leading to COX-2 expression. Accordingly, LPS treatment
led to a rapid and transient activation of the 3 major mito-
gen-activated protein kinases (MAPKs): extracellular-regu-
lated kinase (ERK), c-Jun N-terminal kinase (JNK), and
p38. SB203580, a specific inhibitor of p38, but not U0126
(ERK inhibitor) or SP600125 (JNK inhibitor), blocked the
endotoxin-induced accumulation of COX-2 protein.66 This
response was also blocked by expression of dominant-neg-
ative p38 but not by the dominant-negative ERK con-
struct.66 Together, these results indicate that the p38 signal-
ing cascade is activated in enterocytes in response to
endotoxin, and provide a rationale for investigating the role
of COX-2 antagonists in the treatment of NEC (Figure 4).
Novel therapeutic approaches for NEC
Current treatment of NEC involves antibiotics, gut rest, total
parenteral nutrition and surgical resection or peritoneal
drainage as indicated. Each of these modalities represents a
stop-gap measure to prevent the septic state from over-
whelming the patient. However, none of the currently avail-
able treatment modalities addresses the fundamental biolog-
ical processes that underlie the development of NEC.
Perhaps this explains why the overall mortality of patients
with NEC has changed very little over the past 25 years. As
described above, the observation that impaired intestinal
barrier function is the engine that drives the pathogenesis of
NEC now raises the specter for the development of specific
novel therapeutic modalities to combat this disease. For
instance, the ability to neutralize the high levels of intestinal
NO could conceivably minimize the degree of enterocyte
apoptosis and restore epithelial restitution and proliferation,
thereby restoring barrier integrity. Likewise, agents that
specifically target RhoA-GTPase in the gut, or restore ba-
solateral sodium–proton exchange, could ameliorate the ef-
fects of endotoxin on barrier function. The fact that endo-
toxin-treated enterocytes express high levels of COX-2
raises the intriguing possibility that agents that modify
COX-2 activity or expression may reduce the extent or
severity of NEC in patients at risk of developing this dis-
ease. Although such therapeutic agents remain far from
clinical trials, the approaches we have examined will hope-
fully provide a sound rationale for the development of such
55
drugs, and therefore herald the development of a novel class
of therapeutics for this devastating disease.
Summary
Necrotizing enterocolitis is the leading cause of death from
gastrointestinal disease in the newborn period. Studies per-
formed by ourselves and others have led to the conclusion
that intestinal barrier failure plays a critical role in the
development of this disease. After an initial stressful insult,
such as hypoxic injury, bacterial translocation occurs, which
leads to the activation of the host immune system and the
release of pro-inflammatory molecules. One such molecule
is nitric oxide, which damages the intestinal barrier through
accelerated enterocyte apoptosis, and by initiating a signal-
ing cascade within the enterocyte leading to impaired heal-
ing. One of the first molecules to translocate from the apical
to the basolateral surface is endotoxin. After exposure to
high levels of LPS, enterocyte migration is impaired, cyto-
plasmic pH regulation is thwarted, and a p38 MAPK cas-
cade is initiated leading to the release of COX-2 by the
enterocytes. As a result there is further disruption of the
intestinal barrier, ongoing bacterial translocation, immuno-
cyte activation, and systemic sepsis. Novel therapies that are
designed to neutralize these pro-inflammatory processes
may provide the possibility for an entirely new approach in
the management of this most complex, and often frustrating,
disorder.
Acknowledgments
This work is supported by National Institutes of Health
grants K08 GM65583-01 to DJH, R01-AI-49473 and AI-
14032 to HRF, and KO8-GM00696 01 to JSU. The authors
gratefully acknowledge Drs. Selma Cetin, Faisal Qureshi,
Cynthia Leaphart, Natasha Kelly, Ruben Zamora and X-R
Zhang as well as Jun Li, Kerri Friend, Patricia Boyle,
Catarina Wong and J. Wang whose dedication, effort, ex-
pertise and passion for understanding NEC are reflected in
the experiments described in this review.
References
1. Henry MC, Moss RL: Current issues in the management of necrotizing
enterocolitis. Semin Perinatol 28:221-233, 2004
2. Bakonyi NA, Takegawa B, Ortolan E, et al: Demographic of short gut
syndrome: Increasing demand is not followed by referral of potential
candidates for small bowel transplantation. Transplant Proc 36:259-
260, 2004
3. Network TCN: Variations in incidence of necrotizing enterocolitis in
Canadian neonatal intensive care units. J Pediatr Gastroenterol Nutr
39:366-372, 2004
4. Israel E, Walker W: Host defense development in gut and related
disorders. Pediatr Clin North Am 35:1-15, 1988
56
5. Malin M, Isolauri E, Pikkarainen P, et al: Enhanced absorption of
macromolecules. A secondary factor in Crohn’s disease. Dig Dis Sci
41:1423-1428, 1996
6. Snyder JD, Walker WA: Structure and function of intestinal mucin:
Developmental aspects. Int Arch Allergy Appl Immunol 82:351-356,
1987
7. Allen A, Cunliffe WJ, Pearson JP, et al: Studies on gastrointestinal
mucus. Scand J Gastroenterol 93:101, 1984
8. Feil W, Lacy E, Wong Y, et al: Rapid epithelial restitution of human
and rabbit colonic mucosa. Gastroenterology 97:685-701, 1989
9. van der Waaij LA, Mesander G, Limburg PC, et al: Direct flow
cytometry of anaerobic bacteria in human feces. Cytometry 16:270-
279, 1994
10. Shibata Y, Arima T, Yamamoto T: Regular structures on the microvil-
lar surface membrane of ileal epithelial cells in suckling rat intestine.
J Ultrastruct Res 85:70-81, 1983
11. Frizzell RA, Schultz SG: Ionic conductances of extracellular shunt
pathway in rabbit ileum. Influence of shunt on transmural sodium
transport and electrical potential differences. J Gen Physiol 59:318-
346, 1972
12. Takita S: Automaticity of the alimentary tract: Observations on the
fetal alimentary tract. Nippon Heikatsukin Gakkai Zasshi 6:79-86,
1970
13. Ryley HC, Rennie D, Bradley DM: The composition of a mucus
glycoprotein from meconium of cystic fibrosis, healthy pre-term and
full-term neonates. Clin Chim Acta 135:49-56, 1983
14. Ogra S, Weintraub D, Ogra P: Immunologic aspects of human co-
lostrum and milk. III. Fate and absorption of cellular and soluble
components in the gastrointestinal tract of the newborn. J Immunol
119:245-248, 1977
15. Pang KY, Bresson JL, Walker WA: Development of the gastrointes-
tinal mucosal barrier. Evidence for structural differences in microvillus
membranes from newborn and adult rabbits. Biochim Biophys Acta
727:201-208, 1983
16. Udall JN, Pang K, Fritze L, et al: Development of gastrointestinal
mucosal barrier. I. The effect of age on intestinal permeability to
macromolecules. Pediatr Res 15:241-244, 1981
17. Axelsson I, Jakobsson I, Lindberg T, et al: Macromolecular absorption
in preterm and term infants. Acta Paediatr Scand 78:532-537, 1989
18. Cooperstock MS, Zedd AJ: Intestinal flora of infants. Human Intestinal
Microflora in Health and Disease 1983:79-100
19. Hentges DJ: The anaerobic microflora of the human body. Clin Infect
Dis 16:S175-S180, 1993 (suppl 4)
20. Freter R, Brickner H, Botney M, et al: Mechanisms that control
bacterial populations in continuous-flow culture models of mouse large
intestinal flora. Infect Immun 39:676-685, 1983
21. Dieckgraefe B, Stenson W, Alpers D: Gastrointestinal epithelial re-
sponse to injury. Curr Opin Gastroenterol 12:109-114, 1996
22. Waller D, Thomas N, Self T: Epithelial restitution in the large intestine
of the rat following insult with bile salts. Virchows Arch A Pathol
Anat Histopathol 414:77-81, 1988
23. Rutten M, Ito S: Morphology and electrophysiology of guinea pig
gastric mucosal repair in vitro. Am J Physiol 244G:171-182, 1983
24. Ciacci C, Lind S, Podolsky D: Transforming growth factor ␤ regula-
tion of migration in wounded rat intestinal epithelial monolayers.
Gastroenterology 105:93-101, 1993
25. Podolsky D: Healing the epithelium: Solving the problem from two
sides. J Gastroenterol 32:122-126, 1997
26. Kim PK, Zuckerbraun BS, Otterbein LE, et al: ‘Til cell death do us
part: Nitric oxide and mechanisms of hepatotoxicity. Biol Chem 385:
11-15, 2004
27. Schwentker A, Billiar TR: Nitric oxide and wound repair. Surg Clin
North Am 83:521-530, 2003
28. Koshland DE Jr: Molecule of the year. Science 262:1953, 1993
29. Beckamn JS, Koppenol WH: Nitric oxide, superoxide, and peroxyni-
trite: The good, the bad and the ugly. Am J Physiol 271:C1424-C1428,
1996
Seminars in Pediatric Surgery, Vol 14, No 1, February 2005
30. Kong S-K, Yim MB, Stadtman ER, et al: Peroxynitrite disables the
tyrosine phosphorylation regulatory mechanism: Lymphocyte-specific
tyrosine kinase fails to phosphorylate nitrated cdc2(6-20)NH2 peptide.
Proc Natl Acad Sci U S A 93:3377-3382, 1996
31. Alderton WK, Cooper CE, Knowles RG: Nitric oxide synthases: Struc-
ture, function and inhibition. Biochem J 357:593-615, 2001
32. Knowles RG, Moncada S: Nitric oxide synthases in mammals. Bio-
chem J 298:249-258, 1994
33. Tepperman BL, Brown J, Whittle BJ: Nitric oxide synthase induction
and intestinal cell viability in rats. Am J Physiol 265:G214-G218, 1993
34. Ford HR, Watkins S, Reblock K, et al: The role of inflammatory
cytokines and nitric oxide in the pathogenesis of necrotizing entero-
colitis. J Pediatr Surg 32:275-282, 1997
35. Nadler EP, Dickinson E, Knisely A, et al: Expression of inducible
nitric oxide synthase and interleukin-12 in experimental necrotizing
enterocolitis. J Surg Res 92:71-77, 2000
36. Nadler EP, Upperman JS, Ford HR: Controversies in the management
of necrotizing enterocolitis. Surg Infect (Larchmt) 2:113-119; discus-
sion 119-120, 2001
37. Potoka DA, Upperman JS, Nadler EP, et al: NF-kappaB inhibition
enhances peroxynitrite-induced enterocyte apoptosis. J Surg Res 106:
7-14, 2002
38. Upperman JS, Nadle EP, Hoffman RA, et al: Peroxynitrite induces
enterocyte apoptosis in vitro by modulating mitochondrial membrane
potential and PARS cleavage. Surg Forum L:81-82, 1999
39. Richter C, Schweizer M, Cossarizza A, et al: Control of apoptosis by
the cellular ATP level. FEBS Lett 378:107-110, 1996
40. Kelly N, Friend K, Boyle P, et al: The role of the glutathione antiox-
idant system in gut barrier failure in a rodent model of experimental
necrotizing enterocolitis. Surgery 136:557-566, 2004
41. Potoka DA, Upperman JS, Zhang XR, et al: Peroxynitrite inhibits
enterocyte proliferation and modulates Src kinase activity in vitro.
Am J Physiol Gastrointest Liver Physiol 285:G861-G869, 2003
42. Noerr B: Current controversies in the understanding of necrotizing
enterocolitis. Adv Neonatal Care 3:107-120, 2003
43. Atabai K, Ishigaki M, Geiser T, et al: Keratinocyte growth factor can
enhance alveolar epithelial repair by nonmitogenic mechanisms. Am J
Physiol Lung Cell Mol Physiol 283:L163-L169, 2002
44. Halpern MD, Holubec H, Dominguez JA, et al: Hepatic inflammatory
mediators contribute to intestinal damage in necrotizing enterocolitis.
Am J Physiol Gastrointest Liver Physiol 284:G695-G702, 2003
45. Nieuwenhuijzen G, Deitch E, Goris RJ: The relationship between
gut-derived bacteria and the development of the multiple organ dys-
function syndrome. J Anat 189:537-548, 1996
46. Akira S, Takeda K, Kaisho T: Toll-like receptors: Critical proteins
linking innate and acquired immunity. Nat Immunol 2:675-680, 2001
47. Lemaitre B, Nicolas E, Michaut L, et al: The dorsoventral regulatory
gene cassette spätzle/Toll/cactus controls the potent antifungal re-
sponse in Drosophila adults. Cell 86:973-983, 1996
48. Poltorak A, He X, Smirnova I, et al: Defective LPS signaling in
C3H/HeJ and C57BL/10ScCr mice: Mutations in Tlr4 gene. Science
282:2082-2088, 1998
49. Qureshi ST, Lariviere L, Leveque G, et al: Endotoxin-tolerant mice
have mutations in Toll-like receptor 4 (Tlr4). J Exp Med 189:615-625,
1999
50. Backhed F, Hornef M: Toll-like receptor 4-mediated signaling by
epithelial surfaces: Necessity or threat? Microbes Infect 5:951-959,
2003
51. Hornef MW, Frisan T, Vandewalle A, et al: Toll-like receptor 4 resides
in the Golgi apparatus and colocalizes with internalized lipopolysac-
charide in intestinal epithelial cells. J Exp Med 195:559-570, 2002
52. Hornef MW, Normark BH, Vandewalle A, et al: Intracellular recog-
nition of lipopolysaccharide by Toll-like receptor 4 in intestinal epi-
thelial cells. J Exp Med. 198:1225-1235, 2003
53. Cetin S, Ford HR, Sysko LR, et al: Endotoxin inhibits intestinal
epithelial restitution through activation of Rho-GTPase and increased
focal adhesions. J Biol Chem 279:24592-24600, 2004
Hackam et al Pathogenesis of Necrotizing Enterocolitis
54. Musch MW, Bookstein C, Rocha F, et al: Region-specific adaptation of
apical Na/H exchangers after extensive proximal small bowel resection.
Am J Physiol Gastrointest Liver Physiol 283:G975-G985, 2002
55. Nemeth ZH, Deitch EA, Szabo C, et al: Na⫹/H⫹ exchanger blockade
inhibits enterocyte inflammatory response and protects against colitis.
Am J Physiol Gastrointest Liver Physiol 283:G122-G132, 2002
56. Dube B, Luke H, Aumailley M, et al: Hyaluronan reduces migration
and proliferation in CHO cells. Biochim Biophys Acta 1538:283-289,
2001
57. Falcone J, Richard A, Shin CE, et al: Differential Expression of Ileal
Na⫹/H⫹ Exchanger Isoforms after Enterectomy*1. J Surg Res 86:
192-197, 1999
58. Carcillo JA, Fields AI: Clinical practice parameters for hemodynamic
support of pediatric and neonatal patients in septic shock. Crit Care
Med 30:1365-1378, 2002
59. Hackam DJ, Grinstein S, Rotstein O: Intracellular pH regulation in
leukocytes: Mechanisms and functional significance. Shock 5:17-21,
1996
57
60. Orlowski J, Grinstein S: Na⫹/H⫹ exchangers of mammalian cells.
J Biol Chem 272:22373-22376, 1997
61. Hayashi H, Szaszi K, Grinstein S: Multiple modes of regulation of
Na⫹/H⫹ exchangers. Ann N Y Acad Sci 976:248-258, 2002
62. Sjaastad MD, Wenzl E, Machen TE: pHi dependence of Na-H ex-
change and H delivery in IEC-6 cells. Am J Physiol 262:C164-C170,
1992
63. Cetin S, Dunklebarger J, Li J, et al: Endotoxin differentially modulates
the basolateral and apical sodium/proton exchangers (NHE) in entero-
cytes. Surgery 136:375-383, 2004
64. Forsythe RM, Xu DZ, Lu Q, et al: Lipopolysaccharide-induced en-
terocyte-derived nitric oxide induces intestinal monolayer permeabil-
ity in an autocrine fashion. Shock 17:180-184, 2002
65. Wallace JL: Prostaglandin biology in inflammatory bowel disease.
Gastroenterol Clin North Am 30:971-980, 2001
66. Grishin A, Wang J, Hackam D, et al: p38 MAP kinase mediates
endotoxin-induced expression of cyclooxygenase-2 in enterocytes.
Surgery 136:329-335, 2004