Clinical Science (1997) 93,97- 108 (Printed in Great Britain)
97
Editorial Review
Epithelial migration in the colon: filling in the gaps
Andrew J. WILSON and Peter R GIBSON
University'of Melbourne, Department of Medicine, The Royal Melbourne Hospital, Victoria 3050, Australia
1. The efficient repair of gastrointestinal mucosal
injuries is essential in the preservation of the epithelial barrier to luminal antigens. Accumulated
evidence suggests that epithelial migration plays a
major part in this repair by rapidly resealing
defects induced by both physiological and pathological insults, a process termed restitution.
2. This migration has been modelled in various
ways, most commonly in mechanically wounded
monolayers of cell lines or cells in primary culture,
and in wounded human or animal tissue. Evidence
from these models indicates that migration is a
highly complex process, which is likely to involve
the tightly controlled spatial and temporal interaction of multiple factors: (i) extracellular molecules such as soluble factors (e.g. growth factors,
trefoil peptides, cytokines) and matrix components
(e.g. collagen, laminin, fibronectin); (ii) signalling
molecules activated by the interaction of these
factors with cell surface receptors (e.g. protein
kinases,
phospholipases,
low-molecular-weight
GTPases); (iii) factors which regulate adhesion to
other cells (e.g. E-cadherin) and to matrix components (e.g. integrins, hyaluronic acid receptors); (iv)
factors which regulate detachment from the extracellular matrix (e.g. urokinase-type plasminogen
activator, matrix metalloproteinases); and (iv) molecules which regulate cytoskeletal function (e.g. Rac),
which allows the formation of specialized cellular
processes termed lamellipodia.
3. The identification of physiologically relevant
factors that stimulate epithelial cell migration, and
a better understanding of their mechanism of
action, may be beneficial in the development of
novel therapeutic approaches in diseases such as
inflammatory bowel disease through the pharmacological or dietary enhancement of this migration.
I . INTRODUCTION
The epithelium of the gastrointestinal tract is a
major interface between the external environment
and the internal milieu. A variety of physiological
and pathological factors are known to disrupt this
epithelial barrier. A possible consequence is the
exposure of luminal macromolecules with proinflammatory properties to the mucosal immune system.
Efficient repair of epithelial breeches induced by
these factors may, therefore, be important for the
prevention, or resolution, of inflammation. When
the injury is confined to the mucosa, the first
response is the rapid induction of epithelial migration. While there have been recent reviews concerning this topic, they have been relatively brief,
concentrating on more specific points of interest
[l-31. A review encompassing all the recent
developments in understanding this vital response
has not been forthcoming. This review aims to correct that situation. It will identify areas of continuing
controversy, but will also present an overview of the
work done in models of cellular wounding that have
assisted in increasing understanding of the factors
that control cell migration. Finally, the role that cell
migration plays in the maintenance of homeostasis
and implications for pathological states, such as
inflammatory bowel disease, will be addressed.
2. MUCOSAL REPAIR IN THE COLON
The repair of mucosal injuries anywhere throughout the gastrointestinal tract is a biphasic process. In
order to restore mucosal barrier function, the rapid
onset of epithelial migration, termed restitution, is
followed by a regenerative phase. The time course
of this repair depends greatly on the severity and
tissue depth of the injury. Injuries confined to the
epithelium are resealed within only a few hours,
a time course that obviates cell proliferation
[4]. Direct evidence for this comes from a microautoradiographical study which demonstrated that
labelled cells take 20 h to migrate from the base of
the crypt to the surface epithelium IS]. Full histological resolution of deeper mucosal damage takes
days to weeks and is dependent upon increased proliferation and differentiation to restore normal
Key words: colon, epithelial migration, extracellular matrix, growth facton, restitution, trefoil peptides.
Abbreviations: EGF, epidermal growth factor, aFGF, acidic fibroblast growth factor; bFGF, basic FGF; HGF, hepatocyte growth factor; FAK, focal adhesion kinase; HA, hyaluronic
acid; IGF-I, insulin-like growth factor-I; IL, interleukin; KGF, keratinocyte growth factor; SCFAs, short chain fatty acids: SP, spasmolytic polypeptide: TGF, transforming growth
factor; u-PA, urokinase; u-PAR, urokinase receptor.
Correspondence: Professor Peter Gibson.
98
A. J. Wilson and P. R. Gibson
architecture [6]. Injuries that penetrate the submucosa are healed by a similar mechanism, except that
it may take a period of months because of the need
to develop tensile strength by extensive connective
tissue reorganization.
2. I Restitution
In 1985, Silen and Ito [7] defined restitution in
terms of the ability of the gastric mucosa to rapidly
repair minor epithelial injury, and suggested that it
may be an overlooked component in the maintenance of gastrointestinal health. Although certain
investigators had noted this phenomenon on previous occasions [8, 91, the majority of experimental
studies of wound healing before this time concentrated on the healing of surgical incisions. Because
of the harsh acidic luminal environment, its exposure to various drugs and foodstuffs, and the clinical
problems of gastric ulceration and erosions, the gastric mucosa has received the most attention with
respect to restitution. The colonic mucosa is also
exposed to a variety of noxious stimuli (section 2.2),
leading to studies which have identified restitution
as a relevant repair mechanism in the rat [lo] and
human [4] colon. The events involved in restitution,
as well as regional and species variations, have been
well characterized by light and electron microscopy
[ll]. Within minutes after physical or chemical
injury, cells adjacent to the denuded region flatten
and extend specialized cell processes (lamellipodia)
across the exposed stroma. By virtue of the processes described in section 4.2, they migrate across
the wound surface linked to the cells behind them,
as illustrated in Figure 1. Therefore, cells migrate
from all directions as a linked sheet to resurface
these epithelial defects. The migrating cells are protected by a necrotic layer consisting of dead cells,
plasma exudate, mucus, fibrin and possibly trefoil
peptides. The presence of this layer is essential,
since its removal by forceps or mucolytic agents
impairs restitution [12]. Another factor that appears
to assist the process is the reduction of the area
required for resurfacing, mediated by contraction of
subepithelial myofibroblasts [13] and possibly by
tractional forces generated by the epithelial cells
themselves [14].
The basal lamina is a connective tissue framework
that resides directly underneath epithelial cells and
acts as a scaffold for migrating cells. A commonly
used method to differentiate between superficial and
deep mucosal injury is the disruption of the basal
lamina. If the basal lamina is intact, restitution can
resurface the entire epithelium within 2-5 h [9,15].
Therefore, continuous contact between the migrating cells and basal lamina is required for the most
rapid type of repair. This is illustrated in an elegant
study in gastric biopsy samples taken from people
who had ingested ethanol [16]. Forty-five minutes
was evident in regions
after exposure, restitution
with an intact basal lamina, but not at sites of focal
haemorrhage, congestion or ischaemia. However,
the study was terminated after 4 h and it is possible
that restitution may have been observed over these
regions at later time points. Evidence from a
number of studies suggests that when the basal lamina is damaged, migration is delayed, but still occurs
before cell proliferation would be expected to take
place (9-12 h) [8, 171. This suggests that restitution
is a mechanism relevant for the repair of both
superficial and deeper mucosal erosions. The migration of flattened epithelial cells occurs in response
to duodenal and gastric wounding in rat models of
mucosal ulceration, suggesting that restitution is also
important for the initial resealing of mucosal ulcers
[612.2 Restitution and barrier maintenance
The colonic epithelium acts as a barrier to the
passive, uncontrolled, permeation of luminal macromolecules with toxic, antigenic, immunoadjuvant
and chemotactic properties. The components of this
barrier have been described elsewhere (reviewed in
[MI). Epithelial denudation is likely to result in the
exposure of these macromolecules to the mucosal
immune repertoire, which almost invariably leads to
mucosal inflammation [19]. Restitution, therefore,
surfactants, ammonia,sulphldes, phenols, trauma
*
.c
4
4
4
Lamina propria
Basal lamina
-
13
Bacterial toxins, chemotaxlns,
antlgens, lmmunoadjuvants
t
Fig. 1. After wounding of the mucosa by harmful luminal molecules
and by trauma, for example, epithelial defects are repaired by restitution. This process is characterized by the directed sheet migration of flattened cells, as demonstrated by the cell labelled with an asterisk, and helps
to restore the normal barrier function of the colonic epithelium.
Epahelial migration in the colon
can be viewed as a functional component of barrier
maintenance, a concept which is illustrated in Fig. 1.
This is significant because there is likely to be
ongoing superficial damage to the colonic mucosa
under normal physiological conditions. Direct
evidence for this phenomenon is limited, although
the resealing of membrane disruption is known to
be a common event in the intact gut [20]. Examples
of potentially injurious stimuli include exposure to
luminal factors such as phenols, surfactants,
ammonia and sulphides, and the physical trauma
associated with the passage of luminal contents
along the colon. Another possibility is that, as part
of normal cell turnover, epithelial cells may be exfoliated in sheets [21]. This is supported by DNA fragmentation studies in human and rat intestinal
mucosa, which demonstrate a cluster arrangement
for breaks in DNA [22]. The epithelium may also be
wounded by pathological stimuli, such as drugs or
radiation, and by the products of immune and
inflammatory processes.
The major role that restitution plays in the reestablishment of barrier function is to allow the reformation of tight junctions between cells. Tight
junctions regulate the paracellular permeation of
macromolecules, which is the major path of passive
solute flow. Restoration of tight-junction function
after injury has been assessed experimentally by
measuring the recovery of electrical parameters such
as tissue conductance and potential difference, by
permeation of macromolecules across the epithelium and by the transepithelial passage of bacteria.
A close correlation between the recovery of morphological integrity and that of electrical parameters
has been observed in some studies [e.g. 151, but not
in others [e.g. 41. The disparity of results may reflect
the fact that tissue is often mounted in Ussing
chambers and that mucosal viablity slowly deteriorates over time. There has been, in general, a closer
correlation when restitution is rapid. In deeper
injuries, where restitution is delayed, the loss of
potential difference is due to damage penetrating
deeper than the epithelium [23], and the reappearance of the tight-junction protein, ZO-1, is delayed
compared with histological recovery [24]. It is also
possible that morphological examination may not
reveal small defects in the epithelium that are manifest using functional measurements of epithelial
integrity.
3. MODELS OF RESTITUTION
In order to identify factors that control restitution, systems that model its characteristic migration
have been developed. In vivo studies have yielded
limited information because of the difficulty in
developing specialized operational techniques and
the confounding effects of inflammation, which is
induced by the influx of proinflammatory luminal
molecules. Relatively more simple assays are avail-
99
able. These include ex vivo Ussing chamber models
and in vitro models using tissue explants or gastrointestinal cells in primary culture, as well as cell
lines derived from normal or cancerous tissue.
3. I Ex vivo models
In isolated mucosa mounted in Ussing chambers,
restitution can be elicited by wounding with various
irritants, such as surfactants, hydrochloric acid and
ethanol. A factor of major significance is the reproducibility of the wounding produced. Extensive preliminary work to define appropriate reproducible
conditions for the tissue being studied is necessary.
In this regard, a recently described model using cyanoacrylate adhesive to denude surface epithelium
has potential advantages in both its apparent simplicity and reproducibility [25]. Even though it was
used in vivo in marsupialized rat colonic loops, it is
also likely to be successful in Ussing chamber
studies. This method is similar in principle to a
method described 9 years earlier, where a gelatincoated slide was used to denude rectal epithelium in
vivo [26].
An additional issue relates to the declining
viability of tissue ex vivo and the problems associated with it. Ideally, restitution should be completed
before the tissue begins to lose viability. However,
because the migration is so rapid in these systems,
the lag time associated with the onset of action of
both accelerants and inhibitors may result in observations contrary to current knowledge. For example,
the protein synthesis inhibitor, cycloheximide, does
not affect the progress of restitution over 4 h [27,
281, but does when the process is delayed [28].
There are several lines of evidence which suggest
that synthetic activity is intimately involved in the
initiation and propagation of migration (sections 4.1
and 5.1). Although an increasing body of work is
accruing using these systems, it cannot match the
veritable explosion in information derived from in
vitro models of cell migration.
3.2 In vitro models
Models of single-cell locomotion, such as chemotaxis and invasion assays, are not representative of
the sheet migration characteristic of restitution. This
has led to the development of wounding models utilizing mechanically wounded confluent monolayers
of gastrointestinal epithelial cells. These offer the
advantage of permitting the direct examination of
the action of modulating agents on migration. Consequently, they have identified a wide array of
factors that stimulate or inhibit migration and also
provided some clues in unlocking the mechanism by
which cells migrate. The role that cell proliferation
plays in the repair of these wounds is central to their
applicability as models of epithelial migration. How-
I00
A. J. Wilson and P. R. Gibson
ever, multiple lines of evidence suggest that proliferation plays only a minor role. Basal and stimulated
wound repair is unaffected by pretreatment with
DNA synthesis inhibitors such as mitomycin C
[29-311, and begins within a time frame inconsistent
with increased proliferation. Furthermore, immunohistochemical staining demonstrates only mildly
increased proliferative activity 24 h after wounding
[32]. This suggests that migration is the dominant
force mediating wound repair in the first 24 h. However, at later time points, it is likely that proliferation will play a larger role.
A variety of migration assays have been described
over the past 5 years, involving the mechanical denudation of confluent monolayers with a rotating silicon [32] or teflon [29] tip and vacuum suction [33]
to create circular wounds, and a razor blade to
create linear wounds [30]. Another system utilizes
the removal of a teflon ‘fence’ to elicit outwards
radial migration from a central circle of cells without direct injury to the monolayer [31]. A major
limitation of this model is the long period of culture
(6 days) before migration is assayed. The migration
evident in circular wounds, from all orientations
towards the centre, is likely to be a better model of
the events seen in the repair of mucosal erosions.
Another advantage of these circular wounding
models lies in the quantification of migration. The
diameter, or area, of the cell-free region can be
measured without marking the original wound edge.
In the razor-blade wounding assay, migration is
quantified as the number of, or area covered by,
cells migrating past an initial scratch. Scratching of
the substratum inhibits the movement of cells with
low basal migratory rates (A. Wilson, unpublished
work), suggesting that this method may not be suitable for all cell types. Furthermore, it has been
reported that such scratching actually removes
matrix components, as assessed by immunolocalization studies [34].
Major limitations of this work are the cells used
to assay migration and the substratum over which
they migrate. Cell lines can never be considered
‘normal’ and often originate from cancerous tissue.
Most of those used are poorly differentiated (e.g.
IEC-6 cells), suggesting that they cannot undergo
the same processes as polarized epithelial cells
during restitution (section 2.1). Exceptions are two
well-studied colon cancer cell lines, Caco-2 and T84,
which are able to spontaneously differentiate to a
well-differentiated phenotype and also express tight
junctions. However, the signalling pathways in neoplastic cells may be different to those in normal
cells, suggesting that results in cancer cell lines need
to be interpreted with caution. To overcome these
problems, epithelial cells in primary culture have
been used, but this has only been successfully
applied to the stomach [32, 351. Nevertheless, these
cells are not truly representative of the situation in
vivo because of the loss of their normal spatial
arrangements, while the isolation process itself may
alter their synthetic and secretory profile. Epithelial
cells in vitro have been grown on purified matrix
components found in basement membranes, such as
laminin, fibronectin and type IV collagen, and the
connective tissue matrix, matrigel. Even though the
latter is closest in character to an artificial basement
membrane, the migration of cells over any tissueculture substratum remains unrepresentative of the
physiological situation. The use of resected colonic
mucosal explants has also been described [36],
which, of the cells described in this section, most
closely model intestinal epithelium. This model
allows the three-dimensional aspects of the interactions of normal colonic epithelial cells with their
own matrix to be retained in an in vitro setting.
4. THE PROCESS OF CELL MIGRATION
4. I The initiation of migration
The processes involved in the initiation of migration remain poorly understood, although a number
of factors are likely to play a role (Figure 2). First,
after mechanical wounding of cell monolayers in
vitro and rat gastric mucosa in vivo, cells remaining
at the edge of the wound sustain and repair plasma
membrane disruptions, and then migrate into the
denuded area [20]. While not explicitly stated by the
authors, it is possible that disruption of the plasma
membrane may be a stimulus for migration. Possible
mechanisms could include the release of soluble
factors into the extracellular environment, which
may act as a primary wound signal to initiate migration [37], or the activation of signalling pathways
through cytoskeletal disruption. Evidence supporting
the latter is limited, although one study demonstrated that disruption of the cytoskeleton induces
expression of a protein commonly involved in migration, the urokinase receptor (u-PAR) [38]. Secondly,
another consequence of epithelial injury may be the
alteration to the cells’ environment. This could
involve changes in the organization of the surrounding extracellular matrix or loss of adjacent epithelial
cells, which may release cells from contact inhibition. Such changes are detected by the major regulators of cell-cell and cell-matrix contact, E-cadherin
and the integrins respectively (section 5.2). A
detailed analysis of the signalling events involved in
the initiation of migration is beyond the scope of
this review, although the involvement of key mediators will be briefly discussed in section 5.3. A common feature of the response to wounding of a
number of cell types is the early upregulation of
so-called ‘immediate response genes’, such as c-fos
and cjun (reviewed in [39]). The products of these
genes are transcription factors, which regulate the
expression of a number of genes encoding for proteins involved in migration. Thus, it appears likely
that the cellular response to wounding involves multiple rounds of both transcription and translation,
which is supported by evidence in v i m . The inhibi-
101
Epithelial migration in the colon
soluble factors, receptors, matrix components
4
J
f
Signalling molecules
eg protein kinases, phospholipases,
GTPases, 8-catenin
lntegrin
1
Cytoskeletal
disruption
+- ; -+
Release of soluble
factors
Physical forces
Fig. 2. Factors thought to be important in the initiation of cell migration after injury to the epithelium
tion of RNA and protein synthesis suppresses the
migration of a number of cell lines [29, 401.
4.2 The propagation of migration
In the past few years, substantial progress has
been made in understanding the events involved in
cell migration and the molecules that regulate them.
These recent advances are summarized in Fig. 3 and
described in detail below. It is known that migration
results from the tightly controlled interaction of
multiple factors that encompass changes in the
thermodynamic and mechanical profile of the cell
(reviewed in [41]). The movement of cells across a
denuded substratum requires the interaction of
soluble factors and their receptors, the signalling
elicited by this interaction and the subsequent biochemical changes which alter actin polymerization,
allowing lamellipodial extrusion. Atomic force
microscopy of the lamellipodia of migrating cells has
demonstrated that this is dependent upon the calcium-dependent turnover of plasma membrane components at the leading edge of migration [42]. This
turnover is thought to assist in locomotion by providing a fresh source of receptors for matrix components and motogenic factors, which are recycled
from the opposite pole of the cell from which the
lamellipodium extrudes [43]. The importance of the
cytoskeleton has been demonstrated by studies
which show that the microfilament inhibitor, cytochalasin [27, 36, 401, and the microtubule inhibitor,
colchicine [36], inhibit migration. Adhesive interactions between the cell and its substratum occur at
focal adhesions. The traction generated at these
sites acts as a fulcrum about which the cell body
moves forward by cytoskeletal contraction. A novel
mechanism, involving the binding of integrins to
rearward moving cytoskeletal components, has
recently been described for this phenomenon [44].
Whole-cell translocation is achieved when this is
coupled with the release of adhesions at the rear of
the cell, mediated by proteolytic enzymes such as
urokinase (u-PA). The molecular mechanisms
involved in the generation of the forces required for
cell movement have been reviewed in detail elsewhere [45].
5. THE CONTROL OF CELL MIGRATION
In the past few years, a large number of factors
have been found to modulate migration in the
various models of gastrointestinal epithelium, and
many of these are listed in Table 1. These include
factors external to the cell which interact with membrane-bound receptors, such as soluble factors and
extracellular matrix components, and pharmacological agents which have provided clues about the
mechanisms by which cells migrate.
5. I Soluble factors
Soluble factors are likely to be derived from multiple cellular sources. Epithelial cells, endothelial
cells, bacteria and lamina propria cells adjacent to,
within, or migrating through, the epithelium are all
likely candidates. These factors may influence cell
A. J. Wilson and P. R. Gibson
I02
migration either in solution or when bound to
matrix components. For example, growth factors
such as basic fibroblast growth factor (bFGF) and
transforming growth factor-/3 (TGF-P) bind to heparin.
5.1.1 Growth factors. The importance of growth
factors in regulating the. response to wounding is
suggested, first, by their upregulation in wounded
monolayers [30] and, secondly, by the fact that
exogenous application of bFGF and acidic FGF
(aFGF), TGF-P, platelet-derived growth factor, epidermal growth factor (EGF), hepatocyte growth
factor (HGF), transforming growth factor-a (TGFa), insulin-like growth factor-1 (IGF-1) and keratinocyte growth factor (KGF) stimulate the migration
of a wide range of gastrointestinal epithelial cells,
both normal and neoplastic. The effects are, however, dependent upon the cells being studied and
the nature of the substratum. In a limited number of
undifferentiated small intestinal cell lines, TGF-P
has been identified as a key regulator of migration.
In IEC-6 cells, the stimulatory effects of EGF, TGFa, HGF and FGF on migration over a plastic substratum are all dependent upon TGF-P [46, 48, 491,
while it also mediates the stimulatory effect of
phorbol esters, which activate protein kinase C, over
I
a range of substrates in IEC-18 cells [57]. However,
work in other models of gastrointestinal epithelium
utilizing differentiated cells has raised significant
doubts about the actual role that TGF-P does play.
In colon cancer cell lines, EGF is the most potent of
known motogenic growth factors, while TGF-/3 has
only a mild stimulatory effect [29]. Furthermore, in
cultured gastric mucosal cells, TGF-j3 has no effect
on migration [47] and HGF has been identified as
their most potent motogenic growth factor [35].
5.1.2 Trefoil peptides. The trefoil peptides are a
family of polypeptides that are characterized by the
so-called 'trefoil' domain: a triple-loop structure
dependent upon three disulphide bonds. Members
of the family differ in the number of trefoil domains
and in their primary location within the gastrointestinal tract. pS2 and spasmolytic polypeptide (SP) are
predominantly found in the gastric mucosa, whereas
intestinal trefoil factor is localized to the small and
large intestines [70]. Trefoil peptides are upregulated in response to gastric mucosal injury [71]. Their
function remains uncertain, but this finding suggests
a role in the reestablishment of mucosal integrity
after wounding. This has been supported by a
number of studies in vitro. SP and intestinal trefoil
factor stimulate the migration of IEC-6 cells by a
Soluble factors
eg GF. TP. SCFAs, neuropeptides,
I
cytokines
n
Proteolysis
Adhesion
7
Fig. 3. Factors known to modulate migration (soluble factors and matrix components) activate a variety of signalling molecules either by binding to membrane receptors or by unknown mechanisms. These signalling molecules can
either directly or indirectly (via modulation of gene expression) control processes such as: adhesion to, and detachment from,
the substratum; loosening of contacts from neighbouring cells; secretion of motogenic factors, especially growth factors, and
actin reorganization to allow lamellipodial extrusion. This allows directed cell movement, as indicated by the mow. Key: u-PA
(I); uPAR (2); integrin (3); matrix metalloproteinase (4); plasminogen activator inhibitor-I (5); HA receptors (6). Abbreviations: GF, g r o h factors; TP, trefoil peptides.
Epithelial migration in the colon
TGF-j-independent mechanism, an effect which is
synergistic with mucus glycoproteins [52]; HT-29
cells, where they synergize with EGF [53,72]; and in
other colon cancer cell lines, such as LIM1215 [29].
The cell-surface molecules with which they interact
to exert these effects have, to date, been poorly
characterized.
5.1.3 Nutrients. A number of factors utilized for
the generation of energy, or for other key metabolic
functions, stimulate migration. These include the
following.
Short-chain fatty acids. In the large bowel, short
chain fatty acids (SCFAs) are particularly relevant.
Acetate, propionate and butyrate are the major
SCFAs produced in the colonic lumen after bacterial fermentation of carbohydrate. Produced
in high amounts, they exert several physiological
effects on the colonic epithelium, such as mediating
ion absorption, while butyrate is their preferred
energy source [73]. It has recently been established
that acetate, propionate and butyrate stimulate
migration in colon cancer cell lines of varying differ-
I03
entiation, independently of TGF-j [29]. Preliminary
experiments have revealed that this effect is more
likely to be mediated through the upregulation of a
motogenic factor, rather than simply by improving
the cellular energy state. Butyrate is able to stimulate migration even when its j-oxidation is inhibited
by sulphur-containing compounds [29]. Whether this
effect of SCFAs also applies in vivo remains to be
established. A study using rat colonic mucosa
wounded with hydrochloric acid failed to demonstrate a stimulatory effect of butyrate on restitution
[lo]. However, the duration of exposure to butyrate
was less than that used in vitro and was applied at a
considerably higher, possibly toxic, concentration.
Glutamine. Glutamine is a major energy source
for small intestinal epithelial cells and stimulates
restitution in wounded rat colonic mucosa by
unknown mechanisms [lo].
Polyamines. The polyamines are thought to play
an important role in cell migration. The suppression
of IEC-6 migration mediated by inhibition of ornithine decarboxylase, the enzyme controlling the
Table I. Factors known t o modulate the migration of gastrointestinal epithelial cells. Abbreviations: CGRP, calcitonin gene-related peptide; DFMO,
cc-difluoromethylornithine; IFN-1, interferon-y; ITF, intestinal trefoil factor; KGF, keratinocyte growth factor; ODC, ornithine decarboxylase; PDGF, platelet-derived
growth factor; PKA, protein kinase A PKC, protein kinase C s/s, synthesis; TNF-cc, tumour ecrosis factor-cc.
In vitro
Stimulate
No effect
Growth factors
EGF [29, 3 I, 461
TGF-g [46,4T]
TGF-j [29, 30, 311
HGF [29, 33, 35, 481
bFGF, aFGF [49]
KGF [49] PDGF [SO]
IGF-I [51]
PDGF [46]
TGF-j [47l
Trefoil peptides
Cytokines
ITF, SP [29. 52, 531
IL-lj, IFN-y [46]
Nutrients
11-2 [54]
SCFAr [29]
Others (physiologically
relevant)
Matrix components
11-6, TNF-u [46]
Ex vivo
Inhibit
Endotoxin [46]
Collagen I*, laminin*
Collagen VIP [29]
Pharmacological agents
Cytoskeletal inhibitors
Inhibit
11-4, 11-10 [58]
Glutamine [lo]
Polyamines [55]
ATP, ADP [56]
HA [40]
No effect
EGF [6 I]
TGF-j [62]
bFGF [63]
Laminin* [29]
[311
Protein s/s inhibitors
RNA s h inhibitors
DNA s/s inhibitors
NaC/Ht inhibitors
ODC inhibitors
PKC activators
PKA activators
Stimulate
Cytochalasin [32,40]
Colchicine [36]
Cycloheximide [29,40]
Actinomycin D [29, 401
Butyrate [lo]
Nitric oxide [64]
CGRP [65]
L-Arginine [66]
Collagen lll/lV [67l
Laminin [68]
Cytochalasin [27l
Cycloheximide [27,28]
Cycloheximidet [28]
Mitomycin C [29-3 I]
Amiloride [59]
DFMO [55]
Phorbol esters [57l
*Relative to tissue-culture plastic.
+Delayed restitution.
Forskolin [60]
Amiloride [69]
I04
A. J. Wilson and P. R. Gibson
rate-limiting step in polyamine biosynthesis, is reversible by the exogenous application of polyamines
such as spermine and putrescine [55]. Possible roles
that polyamines may play include their function as
physiological substrates for the transglutaminases,
which may augment migration through the crosslinking of cytoskeletal proteins and/or extracellular
matrix components, or via specific modulation of
gene expression. For example, polyamines regulate
the expression of TGF-P in IEC-6 cells [74].
Adenine nucleotides. The adenine nucleotides,
ATP and ADP, stimulate the migration of IEC-6
cells through a TGF-P-dependent mechanism [56].
5.1.4 InJlarnrnatoly cytokines. Traditionally, the
part that these cytokines play in intestinal repair has
been considered to be the modulation of immune
and inflammatory responses. However, their possible
direct role in the regulation of epithelial migration
has recently been appreciated by virtue of several
studies in vitro. Epithelial cells are known to express
receptors for a number of cytokines and secrete
them in response to the appropriate stimulation
(reviewed in [18]). Interleukin-8 (IL-8) is a potent
neutrophil chemoattractant and may play a similar
role in epithelial cells [75]. Isolated colonic epithelial crypts are known to produce IL-8 as a response
to their isolation and culture [76], which is a potentially significant finding given its effects on the promotion of cell motility. In wounded IEC-6
monolayers, IL-2, interferon-y and IL-1P stimulate
IEC-6 migration through TGF-P, but IL-6 and
tumour necrosis factor-a have no effect [46, 541. In
contrast, IL-4 and IL-10 inhibit the migration of
colon cancer cells, suggesting that these cytokines
may play a role in the perpetuation of epithelial
injury [58].
5.1.5 Others. Nitric oxide [64] and the neuropeptide, calcitonin gene-related peptide [65] improve
gastric restitution after injury. Their direct effects on
intestinal epithelial restitution have not been established, although a possible role for nitric oxide is
suggested by one study. Its precurser, L-arginine,
improves intestinal epithelial migration after ischaemic damage in rats [66]. A possible mechanism of
action of neurotransmitters may be the upregulation
of trefoil peptide secretion, as has been demonstrated in HT-29 colon cancer cells [77].
Whether all these factors play a role in regulating
migration in vivo is unclear. It is possible that the
effects of some of these factors are epiphenomenological, due to their unimpeded access to cellular
membranes in vitro. Access to epithelial cell membranes in vivo may be limited by, for example, the
rapid time frame in which restitution occurs and the
possible blocking effects of mucus. From a teleological viewpoint, the interaction of a vast number of
factors to regulate restitution would not seem the
most efficient way to mediate repair, unless they
were acting to upregulate the activity of single
factors that control migration. TGF-P has emerged
as a putative unifying factor by virtue of the work
described earlier. However, this is not supported by
the contradictory findings in other cell types in vitro
and the lack of direct evidence about its effects on
migration in vivo.
5.2 The regulation of cell adhesion
The regulation of cell adhesion during migration
involves the tightly controlled spatial and temporal
interaction of factors that promote cell adhesion and
those which cause cell detachment from the substratum. Multiple lines of evidence suggest this
interaction does not occur in isolation, but rather is
a part of the integrated response to cell injury. Key
components include the following.
Integrins. The involvement of integrins, which are
transmembrane aP heterodimers that recognize
specific components of the substratum, in cell adhesion and migration has recently been reviewed [78].
Integrins localize at focal adhesions, along with cytoskeletal proteins and protein kinases such as focal
adhesion kinase (FAK) [79]. Such an arrangement
allows integrins to be the pivotal mediators by which
the cell can detect, and respond to, changes in its
environment. Motogens such as EGF [31] and TGFP [62] are known to modulate integrin function.
Cadherins. Cadherins are calcium-dependent
adhesion molecules that regulate homotypic cellcell adhesion, whose structure and association with
cytoskeletal components via a-, P- and y-catenin has
recently been reviewed [80]. Changing normal cadherin function has been associated with altered cell
function. For example, overexpression of E-cadherin
in the small intestinal epithelium of transgenic mice
leads to a reduction in migration along the cryptvillus axis [Sl]. The motogenic effects of growth
factors such as EGF [82] and also human SP [83]
may involve an inhibition of normal cadherin function (see section 5.3).
Hyaluronic acid receptors. There are two receptors
for the extracellular matrix constituent, hyaluronic
acid (HA), the receptor for HA-mediated motility
(RHAMM) and CD44. This interaction may be of
relevance to gastrointestinal epithelial cells, since
HA stimulates the migration of IEC-6 cells [40]. The
ability of EGF to modulate cell adhesion and motility via the regulation of CD44 expression has been
demonstrated recently [84].
Proteolytic enzymes. The best characterized mediators of cell detachment are u-PA and matrix
metalloproteinases. The involvement of these and
other molecules involved in the regulation of
proteolysis has been reviewed recently [85].
Although the proteolytic activity of u-PA has long
been appreciated to be localized to the cell surface,
through interactions with its receptor (u-PAR) [86],
recent evidence suggests that interaction with integrins may also localize matrix metalloproteinase
activity in this way [87]. Recent studies suggest that
alteration of integrin function and proteolytic
Epithelial migration in the colon
activity may both be mechanisms by which components of the u-PA system mediate migration. Plasminogen activator inhibitor-1 inhibits the migration
of smooth muscle cells by blocking the binding of
vitronectin to the c@3 integrin, an effect independent of its inhibition of u-PA-mediated proteolysis
[88]. Furthermore, the u-PAR associates with integrins to change their binding characteristics [89]. A
variety of motogens, such as phorbol esters, EGF
(A. Wilson, unpublished work) and butyrate [90],
markedly induce u-PAR expression in LIM1215
colon cancer cells.
There is a clear association between the strength
of adhesive contacts and the process of migration
[91]. It is therefore not surprising that changing the
composition of the cell's substratum would alter its
migration. Restitution in intact mucosa is dependent
upon basement membrane proteins such as type IV
collagen [67] and laminin [68], while migration in
vitro also varies with the substratum used [31]. Furthermore, the migratory response to exogenous
factors, such as TGF-P, EGF [31] and phorbol esters
[57], but not SCFAs [29], can be altered by the composition of the substratum. The cell itself has also
been demonstrated to change the composition of its
environment, which may underlie a mechanism by
which soluble factors influence migration. A recent
study has demonstrated that IEC-6 cells lay down
their own matrix components, such as fibronectin
and type IV collagen, and that the blocking of their
interaction with the cells inhibited migration [34].
TGF-P, which stimulates IEC-6 migration as
described earlier, upregulates the expression of
these matrix proteins after wounding [34], and is
thought to promote gastric restitution through the
deposition of basement membrane collagens [62].
5.3 lntracellular mechanisms controlling migration
The intracellular events that control cell migration
are highly complex and as yet remain incompletely
understood. A detailed analysis of current knowledge about such events is beyond the scope of this
review. At least five major classes of cell surface
receptors are activated by factors known to modulate cell migration, as shown in Table 2. The signalling events following the activation of integrins [92],
I05
G-protein-coupled receptors [93, 941, cytokine
receptors (which lack intrinsic kinase activity) [95]
and receptors with tyrosine kinase [95, 961 and serinehhreonine kinase activity [97, 981 have recently
been detailed elsewhere. Most of the work described
in the references above has been performed in cells
of non-gastrointestinal origin. Although it would be
expected that many important pathways are conserved, the identification of factors that specifically
regulate signalling in intestinal epithelial cells is vital
in increasing knowledge of all facets of their biology.
A number of intracellular molecules have emerged
as likely candidates in the regulation of the initial
response to cell injury and/or in the resultant migration (Table 2). These include the following.
Protein kinases. Better established examples of
protein kinases which influence migration include:
the mitogen-activated protein kinases, which are
activated by various cellular stressors and by many
of the factors described above (reviewed in [39]);
FAK, a cytoplasmic tyrosine kinase which is
recruited to focal adhesions in response to growthfactor binding and the activation of integrins [79],
and protein kinases A and C, whose activation
respectively inhibits [60] and stimulates [57] the
migration of IEC-18 cells.
Rho and Rac. These are members of the Rho
GTPase family, which are involved in the remodelling of actin, the formation of lamellipodia (Rac)
and the assembly of focal adhesions by regulating
FAK function (Rho) (reviewed in [99]). Selective
inhibition of Rho induces an inhibition of the migration of IEC-6 cells [loo].
Phospholipases. The regulation of phospholipid
metabolism by phospholipases appears to be important in regulating migration. The products of arachidonic acid turnover, which is regulated by
cytoplasmic phospholipase A2,mediate the spreading response in HeLa cells [loll, while the specific
inhibition of phospholipase Cy suppresses the stimulatory effect of EGF on the migration of IEC-6 cells
[102].
P-Catenin. There has been a considerable recent
advancement in knowledge about the role p-catenin
plays in the regulation of gene expression after activation of E-cadherin, and its role in regulating cell
migration through interactions with the APC gene
product (summarized in [103]). Additionally, phos-
Table 2. Examples of cell-surface and intracellular molecules activated by various factors which modulate migration. Abbreviations: ]A& ]anus-associated kinase;
KGF, keratinocyte growth factor; Mad, Mothers against dpp; MAPK, mitogen-activated protein kinase; cPLA2, cytoplasmic phospholipase A2; PLCy, phospholipase Cy;
PDGF, platelet-derived nrowth factor; PKA, protein kinase A; PKC, protein kinase C; STAT, signal transducer and activator of transcription.
Receptor type
Examples of ligands
Examples of signalling molecules activated
References
lntegrins
G-protein coupled
Cytokine
Tyrosine kinase
Matrix components
11-8, neurotransmitters
11-1,11-2, 114, 11-10, TNF-a
EGF, HGF, FGF, TGF-c~,
PDGF, IGF, KGF
~921
Serinelthreonine kinase
TGF-fi
MAPK, Rac/Rho, PKC
PKA, MAPK
I N S T A T , MAPK
MAPK, Rac/Rho, PKC, PLCy, cPL42, JAK
,%Catenin
Mad, MAPK
[93,941
"W
[95,961
P
I
[97,981
I06
A. J.Wilson and P. R. Gibson
phorylation of p-catenin by the EGF receptor has
been implicated in increased cell motility via downregulation of E-cadherin function [82].
Ion channels. Electrophysiological events, such as
the activation of Na+/H+ exchange, have been
implicated in mediating basal and EGF-stimulated
migration in cultured gastric epithelial cells [59] and
restitution of guinea pig gastric epithelium [69].
Janus-associated kinaselsignal transducer and activator of transcription. Although the Janus-associated
kinasehignal transducer and activator of transcription (JAKETAT) pathway is known to be activated
by cytokines and certain growth factors [95], there is
no direct evidence at present that its components
regulate migration.
6. IMPLICATIONS FOR THE PREVENTIONOR
TREATMENT OF DISEASE STATES
There are several diseases in which epithelial
injury and consequent erosion or ulceration occur.
Some are relatively common, such as inflammatory
bowel disease (Crohn’s disease and ulcerative
colitis), while others are of considerable clinical concern, such as post-chemotherapy ‘mucositis’.
Impaired barrier function may lead to mucosal
inflammation, depending upon the duration, extent
and severity of injury. It is likely, then, that efficient
and rapid repair of epithelial breeches is paramount
in the prevention or limitation of mucosal inflammation and in the healing of ulcerated lesions. Interventions that hasten epithelial restitution and
regeneration are likely therefore to have therapeutic
and preventive benefit.
The identification of proteins that regulate the
migratory process may ultimately lead to the
pharmacological enhancement of restitution. The
administration of specific activators of either the
protein itself, or the gene controlling it, is theoretically possible, given that issues such as tissue specificity, stability in vivo and cell uptake are resolved. In
the meantime, however, the search for novel ways to
enhance restitution in a clinical setting will concentrate on the identification of physiologically relevant
factors that promote it. It is reasonable to expect
that many of the factors which stimulate epithelial
migration in vitro and in isolated mucosa exert similar effects in vivo. Several of these factors (e.g.
butyrate, EGF, TGF-/3, IGF-1, KGF) are known to
reduce inflammation in experimental injury models,
such as chemically induced or radiation colitis. However, it is not possible to distinguish between effects
on migration from those on regeneration, since most
of these factors augment the regenerative response.
7. CONCLUSIONS
The importance of cell migration in the reestablishment of epithelial integrity after physiological
and pathological insult suggests that its enhance-
ment in a clinical setting is of considerable importance in gastrointestinal diseases characterized by
epithelial erosion or ulceration, such as inflammatory bowel disease. However, before this is possible,
much work remains in order to identify physiologically relevant compounds which enhance migration in vivo, to fully understand all relevant aspects
of their biology and to define the molecular mechanisms by which they modulate migration. At
present, however, most of the current knowledge
about migration has emanated from cell lines,
results which need to be viewed with caution. Issues
such as the suitability of the cells as models of
gastrointestinal epithelium, the substratum over
which they migrate and technical problems with the
models themselves have yet to be satisfactorily
resolved. Notwithstanding this, considerable progress has been made in defining key mediators of
migration, which has led to the recognition that it
requires the tightly controlled interaction of multiple
cellular components. Allied with the continued use
of more sophisticated techniques, such as antisense
oligonucleotides, transfection and transgenic animal
technology, as well as the development of specific
modulators of key intracellular molecules, these
models of restitution will provide further clues into
unlocking the mechanisms by which cells migrate.
REFERENCES
I. Modlin IM, Hunt RH. Critical reappraisal of mucosal repair mechanisms. Scand
J Gastroenterol 1995; 30 (Suppl. 210): 28-31.
2. Lacy ER Rapid epithelial restitution in the stomach: an updated perspective.
Scand J Gastroenterol 1995; 30 (Suppl. 210): 6-8.
3. Paimela H, Goddard PJ, Silen W. Present views on restitution of gastrointestinal
epithelium. Dig Dis Sci 1995; 40 2495-6.
4. Feil W, Lacy ER, Wong YM, Burger D, Wenzl E, Starlinger M, Schiessel R. Rapid
epithelial restitution of human and rabbit colonic mucosa. Gastroenterology
1989; 97: 685-70 I.
5. Lipkin M, Sherlock P, Bell 8. Cell proliferationkinetics in the gastrointestinal
tract II: cell renewal in stomach, ileum, colon and rectum. Gastroenterology
1963; 45: 721-9.
6. Modlin IM, Basson MD, Soroka CJ, Wright NA, Poulsom R, Chinery R, Hanby
A, Patel K, Rogers L, Alison M. Ulcer-induced alterations in cell phenotype and
matrix and growth factor expression. Eur J Gastroenterol Hepatol 1993; 5
(Suppl3): S59467.
7. Silen W, It0 S. Mechanisms for rapid re-epithelialintion of the gastric mucosal
surface. Ann Rev Physiol 1985; 47:217-29.
8. Yeomans ND, Saint-JohnDJ, Boer WD. Regeneration of gastric mucosa after
aspirin-induced injury in the rat. Am J Dig Dis 1973: 18:773-80.
9. Lacy ER, It0 S. Rapid epithelial restitution of the rat gastric mucosa after ethanol
injury. Lab Invest 1984; 51: 573-583.
10. Scheppach W, Dusel G, Kuhn T, Loges C, Karch H, Bartram H-P, Richter F,
Christ1 FU, Kasper H. Effect of L-glutamine and n-butyrate on the restitution of
rat colonic mucosal cells after acid-induced injury. Gut 1996; 3 8 878-85.
I I. Lacy ER Epithelial restitution in the gastrointestinal tract. J Clin Gastroenterol
1988; 10 S72-7.
12. Wallace JL, Whittle BJ. The role of extracellular mucus as a protective cap over
gastric mucosal damage. Scand J Gastroenterol (Suppl) 1986; 125 (Suppl. I):
79-85.
13. Moore R, Carlson S, Madara JL. Villus contraction aids repair of intestinal
epithelium after injury. Am J Physiol 1989; 257: G274-83.
14. Olson AD. Contractionof collagen gels by intestinal epithelial cells depends on
microfilament function. Dig Dis Sci 1993; 3 8 388-95.
15. Moore R, Carlson S, Madara JL. Rapid barrier restitution in an in vitro model of
intestinalepithelial injury. Lab Invest 1989; 60 237-244.
Epithelial migration in the colon
16. Lacy ER, Morris GP, Cohen MM. Rapid repair of the surface epithelium in
human gastric mucosa after acute superficial injury. J Clin Gastroenterol 1993;
17:s125-3s.
17. Feil W, Wenzl E, Vattay P, Starlinger M, Sogukoglu T, Schiessel R Repair of
rabbit duodenal mucosa after acid injury in vivo and in vitro. Gastroenterology
1987; 9 2 I973- 1986.
18. Gibson PR, Anderson RP, Mariadason JM, Wilson A]. Protective role of the
epithelium of the small intestine and colon. lntlamm Bowel Dis 1996; 2
279-302.
19. Sanderson IR,Walker WA. Uptake and transport of macromolecules by the
intestine: possible role in clinid disorders (an update) Gastroenterology 1993;
104 622-39.
20. McNeil PL, It0 S. Gastrointestinalcell plasma membrane wounding and resealing
in vivo. Gastroenterology 1989; 9 6 1238-48.
21. Barkla DH, Gibson PR, Nov R Cell death in human colonic epithelium
(Abstract). J Gastroenterol Hepatol 1994; 9 A77.
22. GavrieliY, Sherman Y, Ben Sason SA. Identification of programmed cell death
in situ via specific labelling of nuclear DNA fragments. J Cell Biol 1992; I1 9
493-501.
23. Cook GA,Elliott SL, Skeljo MV, Giraud AS,Yeomans ND. Correlation between
transmucosal potential difference and morphologicaldamage during aspirin
injury of gastric mucosa in rats. J Gastroenterol Hepatol 1996; I I :264-69.
24. Endo K, Watanabe S, NagaharaA, Mikami H, Yamaguchi Y, Ohkura R, Ogihara
T, Hirose M, Sat0 N. Expression of gap junction and tight junction during the
maturation of regenerativegastric mucosa induced by ethanol. Gastroenterology
1995: 108 A89.
25. Dieckgaefe BK, Wang JL, Alpers DH. Rapid induction of c-fos and c-jun
following colonic epithelial i n j u y identification of signaling pathways potentially
involved in the regulation of restitution (Abstract). Gastroenterology 1996; I10
A9S.
26. Buck RC. Ultrastructuralfeatures of rectal epithelium of the mouse during the
early phases of migration to repair a defect. Virchows Arch B Cell Pathol 1986;
51: 331-40.
27. Critchlow J, Magee D, It0 S, Takeuchi K, Silen W. Requirements for restitution
of the surface epithelium of frog stomach after mucosal injury. Gastroenterology
1985; 88: 237-49.
28. Albers T, Lomakina I, Moore R Protractedrestitution in contrast to rapid
restitution is aided by protein synthesis in an in vitro model using native
intestinal epithelium. Gastroenterology 1996; I1 0 A308.
29. Wilson AJ, Gibson PR Short chain fatty acids promote the migration of colonic
epithelial cells in vitro. Gastroenterology 1997; in the press.
30. Ciacci C, Lind SE, Podolsky DK. Transforming growth factor fi regulation of
migration in wounded rat intestinal epithelial monolayers. Gastroenterology
1993; 105:93-101.
3 I. Basson MD, Modlin IM, Flynn SD, Jena BP, Madri JA Independent modulation
of enterocyte migration and proliferationby growth factors, matrix proteins,
and pharmacologic agents in an in vitro model of mucosal healing. Surgery I992
I I2 299-307.
32. Watanabe S, Hirose M, Yasuda T, Miyazaki A, Sat0 N. Role of actin and
calmodulin in migration and proliferationof rabbit gastric mucosal cells in
culture. J Gastroenterol Hepatol 1994; 9: 325-33.
33. Nusrat A, Parkos CA,B a r n AE,Godowski PJ, Delp-Archer C, Rosen EM,
Madara JL. Hepatocytegrowth factorhcatter factor effects on epithelia. J Clin
Invest 1994; 93: 2056-5.
34. Goke MN, Zuk A, Podolsky DK. Regulation and function of extracellular matrix
in intestinal epithelial restitution in vitro. Am J Physiol. 1996; 271: G729-40.
35. Takahashi M, Ota S,Shimada T, Hamada E, Kawabe T, Okudaira T, Matsumura
M, Kaneko N, Terano A, Nakamura T, Omata M. Hepatocyte growth factor is
the most potent endogenous stimulant of rabbit gastric epithelial cell
proliferationand migration in primary culture. J Clin Invest 1995; 95:
1994-2003.
36. Dieckgraefe BK, Santoro SA, Alpers DH. Rapid in vitro restitution of human
37.
38.
39.
40.
colonic epithelium: a new model for identification of important signaling
pathways (Abstract). Gastroenterology 1995; 108 A809.
McNeil PL, It0 S. Molecular traffic through plasma membrane disruptions of
cells in vivo. J Cell Sci 1990; 9 6 549-56.
Bayraktutan U, Jones P. Expression of the human gene encoding urokinase
plasminogen activator receptor is activated by disruption of the cytoskeleton.
Exp Cell Res 1995; 221: 486-95.
Dieckgraefe BK, Stenson WF, Alpers DH. Gastrointestinalepithelial response to
injury. Curr Opin Gastroenterol 1996; I2 109-14.
McCormack SA, Viar MJ, Johnson LR Migration of IEC-6 cells: a model for
mucosal healing. Am J Physiol 1992; 263: G426-35.
I07
41. Lauffenberger DA, Howitz AF. Cell migration: a physically integrated molecular
process. Cell 1996; 8 4 359-69.
42. Oberleithner H, Giebixh G, Geibel J.Imagingthe lamellipodiumof migrating
epithelial cells in vNo by atomic force microscopy. PtlugenArch 1993; 425
506- 10.
43. Bretscher MS. Moving membrane up to the front of migratingcells. Cell 1996;
85: 465-7.
44. Felsenfeld DP, Choquet D, Sheee MP. Ligand binding regulates the directed
movement of j I integrins on fibroblasts. Nature (London) 1996; 383 438-40.
45. Mithchison TJ, Crarner LP. Actin-based cell motility and cell locomotion. Cell
1996; 84: 37 I-79.
46. Dignass AU, Podolsky DK. Cytokine modulation of intestinal epithelial cell
restitution: central role of transforming growth factor /3. Gastroenterology
1993; 105 1323-32.
47. Kobayashi 0, Watanabe S, Hirose M, Sat0 N. Effects of transforminggrowth
factors on the wound repair of cultured rabbit gastric mucosal cells.
J Gastroenterol Hepatol 1996; I I: 129-36.
48. Dignass AU, Lynch-Devaney K, Podolsky DK. Hepatocyte growth factorlscatter
factor modulates intestinal epithelial cell proliferationand migration. Biochem
Biophys Res Commun 1994; 2 0 2 70 1-9.
49. Dignass AU, Tsunekawa S, Podolsky DK. Fibroblastgrowth factors modulate
intestinal epithelial cell growth and migration. Gastroenterology 1994; 106
1254-62.
50. Watanabe S, Wang X-E, Hirose M, Oide H, KitamuraT, Mwa H, Myazaki A,
Sat0 N. Platelet-derivedgrowth factor accelerates gastric epithelial restoration
in a rabbit cultured cell model. Gastroenterology 1996; I 1 0 775-9.
5 I. Kato K, Chen MC, Lehman F, Nguyen M, Nakajima N, Kuwayama H, Arakawa
Y, Sol1AH. Trefoil peptides, IGF-I and basic FGF stimulate restitution in
primary cultures of canine oxyntic mucosal cells (Abstract). Gastroenterology
1995; 108A130.
52. Dignass A, Lynch-Devaney K, Kindon H, Thim L, Podolsky DK. Trefoil peptides
promote epithelial migration through a transforming growth factor
Bindependent mechanism. J Clin Invest 1994; 9 4 376-83.
53. Playford RJ, Marchbank T, Chinery R, Thim L, Hanby AH. Human spasmolytic
polypeptide is a cytoprotectiveagent which stimulates cell migration.
Gastroenterology 1995; I08 108- I6.
54. Dignass AU, Podolsky DK. Interleukin-2 modulates intestinal epithelial cell
function in vitro. Exp Cell Res 1996; 225: 422-9.
55. McCormack SA, Viar MJ, Johnson LR Poiyamines are necessary for cell
migration by a small intestinal crypt cell line. Am J Physiol 1993; 264 G367-74.
56. Dignass AU, 8ecker A, Spiegler S,Layer P, Goebell H. Adenine nucleotides
stimulate intestinal epithelial restitution in an in vitro wounding model (Abstract).
Gastroenterology 1996; I1 0 A895.
57. Ma N,Hoa N, Bui T. Protein kinase C and extracellular matrix modulation of
IEC- I8 intestinal epithelial migration (Abstract). Gastroenterology 1996; I 10
A132.
58. JiangWG, Puntis MCA, Hallett MB. 11-4 and IL-I0 inhibit monocyte stimulated
growth and motility of human colon cancer cells. Gut 1994; 35 (Suppl. 5): F286.
59. Puolakkainen P, Kiviluoto T, Lehtonen E, Paimela H, Watanabe S,Sat0 N,
Kivilaakso E. The intracellular signaling pathway for gastric epithelial restitution
in monolayer cell culture includes activation of tyrosine kinase and Na+/H+
antiport (Abstract). Gastroenterology 1996; I 1 0 A235
60. Ma N,Kim K, Kim K.Protein kinase A regulation of IEC-I 8 intestinal epithelial
migration (Abstract). Gastroenterology 1995: I08 Al5S.
61. Riegler M, Sedivy R Sogukoglu T, Cosentini E, Bischof G, Teleky B, Feil W,
Schiessel R, Hamilton G, Wenzl E. Epidermal growth factor promotes rapid
response to epithelial injury in rabbit duodenum in vitro. Gastroenterology 1996;
I I I: 28-36.
62. Yanaka A, Muto H, Fukotomi H, Ito S, Silen W. Role of transforming growth
factor-fi in the restitution of injured guinea pig gastric mucosa in vitro. Am
J Physiol 1996; 27 I:G7S-G85.
63. Paimela H, Goddard PJ, Carter K, Khakee R, McNeil PL, Ito S,Silen W.
Restitution of frog gastric mucosa in vitro: effect of basic fibroblast growth
factor. Gastroenterology 1993; 104 1337-4s.
64. Yanaka A, Muto H, Fukutomi H, It0 S,Silen W. Role of nitric oxide in
restitution of injured guinea pig gastric mucosa in vitro. Am J Physiol 1995; 268
G933-42.
65. Miller CA,Nakashima M, Gittes GK, Debas HT. Calcitonin gene-related peptide
(CGRP) upregulates the restitution of rat gastric mucosa in vitro. J Surg Res
1995; 5 0 421-4.
66. Raul F, Galluser M, Schleiffer R, Gosse F, Hasselman M, Seiler N. Beneficial
effects of L-arginine on intestinal epithelial restitution after ischemic damage in
rats. Digestion 1995; 5 6 400-5.
67. Moore R, Madri J, Carlson S, MadataJL. Collagens facilitate epithelial migration
I08
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
A. J. Wilson and P. R. Gibson
in restitution of native guinea pig intestinalepithelium. Gastroenterology 1992;
102 119-30.
Riegler M, Sedivy R, Feil W, Hamilton G, Teleky 8,Bischof G, Cosentini E,
Sogukoglu T, Schiessel R, Wenzl E. Laminin stimulates rapid epithelial restitution
of rabbit duodenal mucosa in vitro. Scand J Gastroenterol 1996; 3 I: I 167-75.
Paimela H, JoutsiT, Bhowmik A, Kiviluoto T, Kivilaakso E. Persistent activity of
Na+/H+ antiport in restitution: a reflection of ligand binding?(Abstract).
Gastroenterology 1995; 108 A8 16.
Poulsom R, Chinery R, Sarraf C, Lalani E-N, Stamp G, Ella G, Wright N. Trefoil
peptide expression in intestinal adaptation and renewal. Scand J Gastroenterol
1992; 27 (Suppl. 192): 17-28.
Alison MR, Chinery R, Poulsom R, Ashwood P, Longcroft JM, Wright NA
Experimental ulceration leads to sequential expression of spasmolytic
polypeptide, intestinal trefoil factor, epidermal growth factor and transforming
growth factor c( mRNAs in rat stomach. J Pathol 1995; 175 405-14.
Chinery R, Playford RJ. Combined intestinal trefoil factor and epidermal growth
factor is prophylactic against indomethacin-induced gastric damage in the rat.
Clin Sci 1995; 8 8 401-3.
Roediger WE. Role of anaerobic bacteria in the metabolic welfare of the colonic
mucosa in man. Gut 1980 21: 793-8.
Wang J-Y, Viar MJ, McCormack SA, Johnson LR Polyamines are necessary for
the gene expression of TGFP during epithelial cell migration (Abstract).
Gastroenterology 1995; I 0 8 A55 I.
Michel G, Kemeny L, Peter RU, Beetz A, Ried C, Arenberger P, Ruzicka T.
Interleukind receptor-mediated chemotaxis of normal human epidermal cells.
Febs Lett 1992; 305 24 1-43,
Gibson PR, Rosella 0. Interleukin-8 secretion by colonic crypt cells in vitro: a
response to injury that is suppressed by butyrate and enhanced in inflammatory
bowel disease. Gut 1995; 37: 536-43.
Ogata H, Podolsky DK. Neurotransmitters regulate human intestinal trefoil
peptide expression in intestinal epithelial cells (Abstract). Gastroenterology
1996; I10 A983.
Huttenlocher A, Sandborg RR, Horwitz AF. Adhesion in cell migration. Curr
Opin Cell Biol 1995; 7: 697-706.
Craig SW, Johnson RP. Assembly of focal adhesions: progress, paradigms, and
portents. Curr Opin Cell Biol 1996; 8 74-85.
Gumbiner BM. Signal transduction by p-catenin. Curr Opin Cell Biol 1995; 7:
634-40.
Hermiston ML, Wong MH, Gordon ]I. Forced expression of E-cadherin in the
mouse intestinal epithelium slows cell migration and provides evidence for
nonautonomous regulation of cell fate in a self-renewingsystem. Genes Dev
1996; 1 0 985-96.
Fuiii K, Furukuwa F, Matsuyoshi N. Ligand activation of overexpressed
epidermal growth factor receptor results in colony dissociation and disturbed
E-cadherin function in HSC- I human cutaneous squamous carcinoma cells. Exp
Cell Res 1996; 223 50-62.
Offori T, Karayiannakis A, Playford R, Pignatelli M, Krniot W. Human
spasmolytic polypeptide stimulation of colorectal cancer cell migration involves
modulation of epithelial cadherin function: preliminary observations. Gut 1996;
38 (Suppl. I): T124.
84. Zhang M, Singh RK, Wang MH, Wells A, Siegal GP. Epidermal growth factor
modulates cell attachment to hyaluronic acid by the cell surface glycoprotein
CD44. Clin Exp Metastasis 1996; 1 4 268-76.
85. Basbaum CB, Werb 2. Focalized proteolysis: spatial and temporal regulation of
extracellular matrix degradation at the cell surface. Curr Biol 1996; 8: 73 1-8.
86. Blasi F. Urokinase and urokinase receptor: a paracrinelautocrinesystem
regulatingcell migration and invasiveness. Bioessays 1993; 15 105-1 I.
87. Brooks PC, Stromblad S, Sanders LC, von Schalscha TL, Aims RT, StetlerStevenson WG, Quigley JP, Cheresh D A localization of matrix
metalloproteinaseMMP-2 to the surface of invasive cells by interaction with
Cell 1996; 85: 683-93.
integrin ~4.
88. Stefansson S, Lawrence DA. The serpin PAL I inhibits cell migration by blocking
integrin a& binding to vitronectin. Nature (London) 1996; 383 441-3.
89. Wei Y, Lukashev M, Simon DI, Bodary SC, Rosenberg S, Doyle MV. Chapman
HA. Regulation of integrin function by the urokinase receptor. Science 1996;
273: I55 1-5.
90. Wilson AJ, Gibson PR. Butyrate promotes interleukin-8 secretion and urokinase
receptor expression in in vitro models of colonic epithelium (Abstract).
Gastroenterology 1996; I 10 A1045.
91. Palecek SP. Loftus JC, Ginsberg MH, Lauffenburger DA, Horwia AF. Integrinligand binding propertiesgovern cell migration speed through cell-substratum
adhesiveness. Nature (London) 1997; 385 537-40.
92. Dedhar S, Hannigan GE. lntegrin cytoplasmic interactions and bidirectional
transmembrane signalling. Curr Biol 1996; 8 657-69.
93. Dessauer CW, Posner BA, Gilman AG. Visualizing signal transduction: receptors,
G-proteins, and adenylate cyclases. Clin Sci 1996; 91: 527-37.
94. Knall C, Young S, Nick JA, Buhl AM, Worthen GS, Johnson GL. Interleukin-8
regulation of the RaslRaflmitogen-activatedprotein kinase pathway in human
neutrophils. J Biol Chem 1996; 271: 2832-8.
95. Winston LA, Hunter T.lntracellular signalling: puttingJAKs on the kinase map.
Curr Biol 1996; 6 668-71.
96. Malarkey K, Belharn CM, Paul A, Graham A, McLees A, Scott PH, Plevin R The
regulation of tyrosine kinase signalling pathways by growth factor and
G-protein-coupled receptors. Biochem J 1995; 3 0 9 36 1-75.
97. Massague J. TGFj signalling: receptors, transducers and Mad proteins. Cell
1996; 85: 947-50.
98. Schrum LW, Brenner D A TNFE and TGFP induce ERK and JNK activity in
activated hepatic stellate cells (Abstract). Gastroenterology 1996; I 10 A l I 16.
99. Ridley AJ. Rho: theme and variations. Curr Biol 1996; 1 0 1256-64.
100. Santos MF, McCormack SA, Oblicany J, Tigyi G, Johnson LR The smdl
molecular weight G-protein Rho is important for migration of small intestinal
crypt cells (Abstract). Gastroenterology 1996; I10 A l I 14.
101. Chun IS, Jacobson BS. Spreading of HeLa cells on a collagen Substratum
requires a second messenger formed by the lipoxygenase metabolism of
arachidonic acid and released by collagen receptor clustering. Mol Cell Biol
1992; 3: 481 -92.
102. Polk DB. Phospholipase C and tyrosine kinase activities are required for EGFmediated intestinal cell migration and adhesion (Abstract). Gastroenterology
1996; 1 1 0 A832.
103. Peifer M. /?-cateninas oncogene: the smoking gun. Science 1997; 275 1752-3.