Supraphysiologic extracellular pressure inhibits intestinal

T

Sen

TM

DS

mpplrp

1

0d

The American Journal of Surgery (2008) 196, 683–689

he Association of VA Surgeons

upraphysiologic extracellular pressure inhibits intestinalpithelial wound healing independently of luminalutrient flow

homas L. Flanigan, M.D., Cheri R. Owen, Ph.D., Christopher Gayer, M.D.,arc D. Basson, M.D., Ph.D., M.B.A.*

epartments of Surgery, Anatomy and Cell Biology, John D. Dingell VA Medical Center and Wayne State University,
urgical Service (11S), 4646 John R. St., Detroit, MI 48201, USA
AbstractBACKGROUND: Luminal pressure may injure the gut mucosa in obstruction, ileus, or inflammatory

bowel disease.METHODS: We formed Roux-en-Y anastomoses in 19 mice, creating proximal and defunctionalized

partially obstructed limbs and a distal limb to vary luminal pressure and flow. We induced mucosalulcers by serosal acetic acid, and assessed proliferation (proliferating cell nuclear antigen) and ERK(immunoblotting). Parallel studies compared Caco-2 enterocyte migration and proliferation after pres-sure and/or ERK blockade.

RESULTS: At 3 days, anastomoses were probe-patent, proximal and distal limbs contained chyme,and defunctionalized limbs were empty. The proximal and defunctionalized limbs showed increasedpressure and slower healing despite increased proliferation, ERK protein, and ERK activation. In vitro,pressure decreased Caco-2 migration across collagen or fibronectin, stimulated proliferation, andactivated ERK. However, ERK blockade did not prevent pressure effects.

CONCLUSIONS: Luminal pressure during obstruction or ileus may impair mucosal healing inde-pendently of luminal flow despite increased mitosis and ERK activation.Published by Elsevier Inc.

KEYWORDS:Force;Mechan transduction;Migration;Pressure;Proliferation;Roux-en-Y

ps

ppratpdc

The intestinal mucosa experiences diverse forces in nor-al and diseased states. The small intestine mixes and

ropels the chyme by peristaltic and segmental contractions,endular movements and villous motility’s.1,2 The liquiduminal contents are largely noncompressible3 and interactepetitively with the gut mucosa in complex ways as theyass along the villi.4 These forces all alter intraluminal

* Corresponding author: Tel.: �1-313-576-3598; fax: �1-313-576-002.

E-mail address: [email protected]
mManuscript received May 9, 2008; revised manuscript July 2, 2008
002-9610/$ - see front matter Published by Elsevier Inc.oi:10.1016/j.amjsurg.2008.07.016

ressure, which along with other physiologic forces mayupport the normal gut mucosal cytoarchitecture.5

However, supraphysiologic forces, such as increasedressure, caused by diet or illness, may adversely impact guthysiology and mucosal healing.6 Luminal jejunal pressureeaches 50 mm Hg in irritable bowel syndrome.7 Intra-bdominal pressure also may increase after surgery owingo tissue edema.8 Inflammation and injury increase luminalressure in chronic inflammatory states such as Crohn’sisease or ulcerative colitis.9–11 Such increases in pressureould affect wound healing.5,12

The gut mucosa is constantly subjected to injuries that it
ust heal to maintain normal function.13 Biophysical forces
mailto:[email protected]

imaicddmsmmcetAb

tapfiflspttmcdfcEbtbmocgEc

M

I

RRtjag(w

PWsdtorcTtwtiagcibcip

MwcsUSt.

IffmatpUa

W4PfsTcwnt

I

C

684 The American Journal of Surgery, Vol 196, No 5, November 2008

n the gut stimulate intestinal epithelial proliferation andodulate intestinal epithelial differentiation in vitro,14 and

ctivate mucosal tyrosine kinases in vivo.15 Mucosal repairs required for recovery from pathologic injury such ashronic ulceration and inflammation in inflammatory bowelisease, and is likely deficient when the mucosal barriereteriorates in sepsis.16,17 Mucosal healing is affected inany pathophysiologic states that show altered luminal pres-

ure. Sepsis, ileus, fasting, and inflammatory bowel diseaseay be associated with altered contractile rhythms, villousotility, and mucosal deformation from luminal contents with

onsequent changes in luminal pressure. Ulcerative colitis, forxample, shows decreased contractility, increased low-ampli-ude propagating contractions, and variable transit times.18

nastomoses rupture at lower pressures in a rat inflammatoryowel disease model, suggesting impaired healing.19

In vitro, repetitive deformation promotes intestinal epi-helial proliferation and differentiation when the enterocytesre cultured on collagen or laminin substrates,20 but inhibitsroliferation and promotes epithelial sheet migration onbronectin substrates.21 However, the effects of physicalorces on the biology of the intestinal mucosa in vivo areess clear, although repetitive deformation stimulates muco-al tyrosine kinase activity in anesthetized rats.15 Moreover,ressure may affect intestinal epithelial cells differentlyhan repetitive deformation.12 It therefore becomes impor-ant to understand how increased pressure might modulateucosal healing during altered intestinal homeostasis. We

reated a murine model of partial bowel obstruction in aefunctionalized jejunal Roux-en-Y limb in which the ef-ects of luminal pressure were dissociated from luminalontents, and compared mucosal healing, proliferation, andRK signaling among the proximal partially obstructedowel, the defunctionalized partially obstructed bowel, andhe bowel distal to the partial obstruction. We studied ERKecause it critically mediates the effects of repetitive defor-ation on intestinal epithelial cells in vitro.21 We validated

ur observations in vitro by studying the effects of extra-ellular pressure on human Caco-2 intestinal epithelial mi-ration, proliferation, and ERK activation, and blockedRK with the MEK inhibitor PD98059. Caco-2 cells are aommon model for human intestinal epithelial biology.22

aterials and Methods

n vivo studies

oux-en-Y anastomosis. We created defunctionalizedoux limbs in C57 Black mice (Chas River Labs, Wilming-

on, MA, USA) by an approved protocol. We divided theejunum 1 cm from the ligament of Treitz with 5–0 silk,nastomosed the proximal jejunum side to side to the distalut 2 cm distal to the original transection with 9–0 VicrylEthicon, Somerville, NJ, USA), and closed the abdomen
ith running 5–0 silk. c
ressure determination within the murine small bowel.e assessed luminal pressures with a Stryker 295–1 Pres-

ure Monitor (Kalamazoo, MI) before and 3 days afterefunctionalized limb creation. The pressures measured inhe animal model using the Stryker needle were on the orderf 0 to 6 mm Hg. Preliminary calibration showed the accu-acy of the needle and pressure monitor, measured against alinical intensive care unit pressure transducer (not shown).hese pressures were reproducible, but notably lower than

hose reported in human unobstructed or obstructed bo-el.9–11 These differences may reflect species differences or

he use of the Stryker needle technique in which the fluid isnjected and the pressure peaks briefly before the fluid runsway into the measured space. However, the constraintsenerated by the closed and intact abdominal wall also mayontribute substantially to measurements of bowel pressuresn intact human beings, and were absent when we measuredowel pressure in the mouse at necropsy. We thereforehose to apply a pressure of 80 mm Hg to the human Caco-2ntestinal epithelial cells based on literature reports of suchressures within the lumen of obstructed human bowel.

ucosal ulcers. Paper disks pretreated with 70% acetic acidere applied to the small-bowel serosa for 15 seconds to

reate circumscribed ischemic mucosal ulcers of predictableize, by a modification of a method for gastric ulcers.23

lcers were photographed and measured on a Kodak Imagetation (PerkinElmer, Boston, MA). Initial studies showed

hat this routinely produces jejunal mucosal ulcers of 3.1 �3 mm2 in normal jejunum at ulcer induction.

mmunohistochemical studies of proliferation. Tissuerom the proximal, distal, and Roux limbs was fixed in 10%ormalin for 24 hours and embedded in paraffin. Five-icron sections were stained for proliferating cell nuclear

ntigen (PCNA) (Zymed, San Francisco, CA, USA), hema-oxylin-counterstained for orientation, and visualized andhotographed on a Nikon Microphot-FXA (Melville, NY,SA). These results were confirmed by using antibody

gainst Ki-67 (not shown).

estern blotting. Samples were immersed immediately in°C lysis buffer, agitated by a Tissue Tearor (BioSpecroducts, Bartlesville, OK) at 10,000g to 15,000g, centri-uged at 12,000g for 30 minutes at 2°C, assayed for protein,ubjected to Western blot for phospho-ERK Thr 202,yr204, or total ERK, and visualized using enhancedhemiluminescence as previously described.21 Membranesere reprobed for glyceraldehyde-3-phosphate-dehydroge-ase and appropriate secondary antibody as a loading con-rol. All exposures were within the linear range.

n vitro studies

ell culture. We studied Caco-2 BBE intestinal epithelial
ells, a subclone of the original Caco-2 cell line selected for

ic

Pwot

Mc(wrgbJaw

Mtcfa(c

Pstspcw

Wi1pbtdohu

Deuv

R

s

e(bgpt7atPstd2rs

Pt

R

Fttapucpf

685T.L. Flanigan et al. Pressure inhibits gut mucosal healing

ts ability to differentiate in culture.24 Caco-2 cells wereultured as previously described.25

ressure regulation. Pressure was controlled using a pre-armed airtight box with an inlet valve for gassing and anutlet connected to a manometer.5 Temperature was main-ained within 2°C, and pressure within 1.5 mm Hg.

atrix and medium modulation. Six-well dishes were pre-oated with 12.5 �g/mL collagen I or tissue fibronectinSigma, St. Louis, MO) at saturating concentrations. Cellsere seeded 50,000/well for proliferation studies and se-

um-starved for 24 hours. Cells seeded at 500,000/well wererown to confluence for migration assays. ERK was blockedy 20 mmol/L MEK antagonist PD98059 (Calbiochem, Laolla, CA) for 30 minutes before application of pressure tossess wound closure with inhibited ERK. Control cellsere treated with a .1% dimethyl sulfoxide vehicle control.

otility measurement. Caco-2 monolayers on 6-well ma-rix-coated dishes were subjected to 0 to 80 mm Hg in-reased pressure for 24 hours after induction of small uni-orm circular wounds as described.26 We measured woundreas at 0 and 24 hours on a Kodak Image StationPerkinElmer), and calculated the percentage of woundlosure.

roliferation. Subconfluent (30%–40%) cells were serum-tarved for 24 hours. We reserved one 6-well plate for aime 0 measurement, and incubated the remaining serum-tarved cells in normal growth medium under ambient orressure conditions for 24 hours before counting cells by arystal violet absorbance assay over the linear assay range,ith interpolation against a standard curve, as described.27

estern blotting. After pressure, Caco-2 cells were lysedn buffer with protease inhibitors, centrifuged at 10,000g for5 minutes at 4°C, and resolved by sodium dodecyl sulfate–olyacrylamide gel electrophoresis as described.21 Mem-ranes were blotted for phospho-ERK Thr 202, Tyr204, orotal ERK, reprobed with antibodies specific for glyceral-ehyde-3-phosphate-dehydrogenase and appropriate sec-ndary antibody as a loading control, and detected by en-anced chemiluminescence as described.28 All exposuressed for densitometric analysis were within the linear range.

ata analysis. All data are expressed as mean � standardrror. Statistical analysis was performed using paired ornpaired t tests or analysis of variance as appropriate. A Palue of less than .05 was considered significant.

esults

Murine roux limb formation affects pressure within
mall-bowel lumen, as well as mucosal healing and prolif- 1
ration. Creation of a defunctionalized roux limb in miceFig. 1A) yielded probe-patent but partially obstructed smallowel. We induced ulcers on the Roux, proximal, and distalut (Fig. 1B). At 72 hours, the roux and proximal limbressures were 3.3 � .1- and 5.0 � .1-fold normal, respec-ively (n � 10, P � 0.0001). Ulcer size was measured after2 hours, at necropsy. Ulcers in proximal and defunction-lized limbs were 224% � 14.9% and 55.7% � 7.8% largerhan those in the distal limbs, respectively (Fig. 1C; n � 15;

� .000,1 each). Proliferation was assessed using a PCNAtain for proliferating cells (Fig. 2A). PCNA immunoreac-ivity within the mucosa of the proximal (Fig. 2A) andefunctionalized limbs (Fig. 2B) was 186% � 15.8% and56% � 9.9% greater than in the distal limb (Fig. 2C),espectively (n � 4; P � .05 each Fig. 2D). We obtainedimilar results using Ki-67 staining (not shown).

ressure increases phosphorylated andotal ERK in vivo

Phosphorylated ERK was increased in the proximal andoux limbs compared with the distal limb (Fig. 3, top blot;

igure 1 (A) The roux limb ex vivo is shown with the defunc-ionalized roux limb to the left, the proximal limb at the top, andhe distal limb at the bottom. (B) Gross appearance of an aceticcid–induced jejunal mucosal ulcer after 3 days. (C) At 3 days, theroximal limb shows the highest luminal pressure and the slowestlcer healing (n � 10 and 15, respectively; P � .05 for each),ompared with the distal limb. The roux limb also shows higherressure and slower healing (n � 10 and 15, respectively; P � .05or each) than the distal limb. �, Pressure; p, ulcer size.

71.0% � 37.5% and 140.6% � 39%; n � 19; P � .05).

Tl9eapawbc

P

Wpcsia

Ps

fisiCHFiMmPps

P

owPp

FPrpa*

FolP


otal ERK also was increased in the Roux and proximalimbs compared with the distal limb (Fig. 3, middle blot;7.9% � 33.7% and 60.9% � 30.9%; n � 19; P � .05). Tonsure that the increase in phosphorylated ERK was not anrtifact of the increase in total ERK, we calculated thehosphorylated to total ERK ratio (Fig. 3A). The ratio ofctivated to total ERK in the proximal and Roux limb alsoas increased in the proximal and defunctionalized limbsy 184.6% � 83.6% and 243.3% � 127.5%, respectively,ompared with the distal limb (n � 19; P � .05).

igure 2 Mucosal PCNA staining. Typical images are shown ofCNA staining of the mucosa from the (A) proximal limb, (B)oux limb, and (C) distal limb. Areas of brown stain indicateroliferating cells. (D) The proliferative index is increased in rouxnd proximal limb mucosa compared with the distal limb (n � 4;P � .001 for each).

igure 3 (A) Phosphorylated and total ERK are each increased sif the anastomoses. The phospho ERK/total ERK ratio shows a truimbs compared with the distal gut (n � 19; P � .05 for each
hosphorylated ERK is increased in pressurized caco-2 cells compared
ressure activates ERK in Caco-2 cells

ERK is activated in Caco-2 cells by repetitive strain.22

e asked whether pressure also activates ERK in vitro inarallel to ERK activation in the partially obstructed mu-osa in vivo. When confluent Caco-2 cells on collagenubstrates were subjected to 30 minutes of 80 mm Hgncreased pressure on collagen, pressure increased ERKctivation by 35.5% � 6.6% (Fig. 3B; n � 18; P � .05).

ressure inhibits Caco-2 epithelialheet migration

Basal migration was more rapid on collagen I than onbronectin, consistent with previous findings.25 On collagenubstrates, migration was reduced at 20 to 80 mm Hgncreased pressure (n � 18; P � .001 each) (Fig. 4A).aco-2 motility was similarly inhibited by 20 and 80 mmg pressures across fibronectin (n � 12; P � .001 each;ig. 4B). It was thought that Caco-2 migration may be

nfluenced by ERK.21 However, blocking ERK with theEK inhibitor PD98059 did not prevent the inhibition ofigration by pressure at either 20 or 80 mm Hg (n � 6 each;� .05 each; Fig. 4C). Additional studies conducted in

arallel confirmed that pressure inhibits IEC-6 cell motilityimilarly to its effects on Caco-2 cells (not shown).

ressure stimulates Caco-2 cell proliferation

Caco-2 cells exposed to pressure at 20 to 80 mm Hg,ver 24 hours, showed increased cell numbers comparedith control cells at ambient pressures. Blocking ERK withD98059 did not prevent the mitogenic effect of increasedressure (Fig. 5; n � 32; P � .05 for all).

ntly in the proximal and roux limbs compared with the distal limbsase in the proportion of activated ERK in both proximal and rouxarison of phosphorylated and total ERK and for the ratio). (B)

gnificae increcomp

with controls (n � 18; P � .05).

C

wwmpmdpcommtOmoErce

ddTlamcrcbcaersupmtw

Fafisa O; p,

Fb


omments

Peristalsis and the interactions of the luminal contentsith the mucosa expose the small intestinal epithelium to aide range of luminal pressures. Nocturnal luminal pressureanometry in healthy volunteers29 showed a mean average

ressure of 15 to 20 mm Hg during phase II of the migratingotor complex. Luminal pressures decreased to almost zero

uring phase I of the migrating motor complex, beforehase II restarts. Thus, the gut mucosa normally experienceshanges in luminal pressure ranging from 0 to 30 mm Hgver 85 to 110 minutes with an average pressure of 15 to 21m Hg. We studied 0 to 80 mm Hg pressures to approxi-ate the normal variance of the intraluminal pressure and

he high supraphysiologic pressures in pathologic states.ur results suggest that supraphysiologic pressure inhibitsucosal wound healing in vivo and in vitro independently

f both extracellular matrix regulation and intracellularRK activation. The effects on wound healing are likely to

eflect true motility rather than effects on proliferation be-ause proliferation actually is stimulated by pressure, anffect that also was ERK-independent.

igure 4 (A) Wound closure on a collagen substrate is reducedmbient pressure control (n � 36 each; P � .005 for each compbronectin versus ambient pressure control (n � 24 each; P � .0imilarly after treatment with a dimethyl sulfoxide vehicle controlmbient pressure control (n � 36 each; P � .005 each). �, DMS

igure 5 (A) Proliferation is increased on collagen by a 20 or 80
y ERK blockade by PD98059 (gray bars) compared with ambient cont
In vivo wound healing was decreased significantly in theefunctionalized and proximal bowel as compared with theistal gut. Ulcer healing was slowest in the proximal limb.he magnitude of inhibition of restitution correlated with

uminal pressure. The presence or absence of luminal chymend its own chemical and physical interactions with theucosa is an additional variable in vivo. However, luminal

hyme was present within the proximal limb whereas theoux limb was empty and had no significant intraluminalontents at necropsy, being distended only by gas. Becauseoth limbs generally showed the same tendencies to de-reased wound healing, increased proliferation, and ERKctivation, and because the in vitro model reproduced thisffect simply by manipulating extracellular pressure, ouresults suggest that the effects of pathologic luminal pres-ure dominates any effect of the chyme itself. The murinelcers were created simultaneously with the Roux-en-Yreparation. It is conceivable that the edema of the anasto-osis contributed to the proximal and Roux obstruction. If

he anastomosis had been allowed to heal for a period ofeeks, it is possible that the obstruction would have cleared,

cells are exposed to 20 to 80 mm Hg increased pressure versus). (B) Twenty and 80 mm Hg pressures reduce migration acrossh). (C) Pressure decreases wound closure on a collagen substrater ERK blockade by the upstream MEK inhibitor PD98059 versusPD98059.

g increase in pressure (black bars). (B) This effect is not prevented

whenarison05 eacor afte

mm H
rols (n � 22; P � .05 for each).

itipt

moicaordeiCpwIv

seomtwnecnpuWlitmsbic

iatwcflcpaisb

fhnspi

A

(DT

R

1

1

1

1

1

1

1


ntraluminal pressure would have normalized, and the epi-helial migration and mucosal healing in the defunctional-zed and proximal limbs would no longer have shown aressure effect. However, this awaits further study beyondhe scope of the current article.

The magnitude of wound closure in these studies appearsodest, but is highly statistically significant. We14,30 and

thers31–34 have described similar magnitude changes inntestinal epithelial migration, proliferation, or signalingaused by other stimuli. For instance, strain induces anverage 22% change in proliferation in Caco-2 cells platedn various matrices.14 However, the effects of physiologicepetitive strain on intestinal epithelial cells are matrix-ependent, whereas those of pathologic pressures seem toscape regulation by the extracellular matrix just as they arendependent of luminal chyme. Repetitive strain acceleratesaco-2 or IEC-6 monolayer wound closure but inhibitsroliferation on fibronectin, whereas deformation inhibitsound closure but stimulates proliferation on collagen.14,21

n contrast, increased pressure inhibited wound closure initro on either collagen or fibronectin.

ERK stimulates proliferation and migration in the ab-ence of physical forces35,36 and mediates the mitogenicffects of repetitive deformation in intestinal epithelial cellsn collagen and the motogenic effects of repetitive defor-ation in intestinal epithelial cells on fibronectin.21,22 It was

herefore somewhat surprising that pressure activates ERKhile inhibiting motility in vivo and that blocking ERK didot prevent inhibition of motility by pressure in vitro. How-ver, not all physical force effects on intestinal epithelialells involve ERK. Repetitive deformation inhibits intesti-al epithelial cell motility across collagen substrates inde-endently of ERK.21 Moreover, extracellular pressure stim-lates colon cancer cell adhesion independently of ERK.12

e hypothesize that pressure may inhibit intestinal epithe-ial motility by an ERK-independent signal mechanism sim-lar to that responsible for the inhibition of intestinal epi-helial migration across collagen. The precise signals thatediate these effects await further study. Moreover, it

eems likely that the activation of intestinal epithelial ERKy pressure in vivo and in vitro reflects some other yet to bedentified ERK-mediated effect of pressure on the gut mu-osa.

Intestinal mucosal healing is a complex process involv-ng lamellipodial extension of epithelial cells, migration,nd proliferation. Physical forces owing to intestinal con-raction as in peristalsis, villous motility, and interactionsith luminal contents likely interact with the effects of

ell-matrix interactions and soluble factor signaling to in-uence these processes. These effects are interesting be-ause the intestinal lumen experiences greatly increasedressure during diverse pathologic states, including the ex-cerbation of inflammatory bowel disease, postoperativeleus, and partial or complete bowel obstruction.37–39 Inome pathologic situations that increase pressure within the
owel, this pressure may itself have a deleterious effect that
urther worsens the patient’s condition. When the normalomeostasis of the small bowel is disturbed by disease, theormal physical forces of the gut are altered and the re-ponse to these forces also changes. In particular, increasedressures within the gut may impair mucosal healing inntestinal disease or injury.

cknowledgment

Supported in part by a VA Merit Research AwardM.D.B.), National Institutes of Health grants RO 1K067257 (M.D.B.) and 2 T32 GM008420 (M.D.B., C.G.,.F.).

eferences

1. Sarna SK, Otterson MF. Small intestinal physiology and pathophysi-ology. Gastroenterol Clin North Am 1989;18:375–404.

2. Mailman D, Womack WA, Kvietys PR, et al. Villous motility andunstirred water layers in canine intestine. Am J Physiol 1990;258:G238–46.

3. Miftakhov R, Wingate D. Numerical simulation of the peristaltic reflexof the small bowel. Biorheology 1994;31:309–25.

4. McNeil PL, Ito S. Gastrointestinal cell plasma membrane woundingand resealing in vivo. Gastroenterology 1989:96:1238–48.

5. Basson MD, Yu CF, Herden-Kirchoff O, et al. Effects of increasedambient pressure on colon cancer cell adhesion. J Cell Biochem2000;78:47–61.

6. Brodribb AJ, Condon RE, Cowles V, et al. Effect of dietary fiber onintraluminal pressure and myoelectrical activity of left colon in mon-keys. Gastroenterology 1979;77:70–4.

7. Kellow JE, Phillips SF. Altered small bowel motility in irritable bowelsyndrome is correlated with symptoms. Gastroenterology 1987;92:1885–93.

8. Granger DN, Barrowman JA. Microcirculation of the alimentary tractI. Physiology of transcapillary fluid and solute exchange. Gastroenter-ology 1983;84:846–68.

9. Allan A, Wyke J, Allan RN, et al. Plasma fibronectin in Crohn’sdisease. Gut 1989;30:627–33.

0. Fallingborg J, Nielsen D, Pedersen JO. Decreased gelatin-bindingfibronectin in patients with chronic inflammatory bowel diseases.Scand J Gastroenterol 1985;20:1062–4.

1. Verspaget HW, Biemond I, Allaart CF, et al. Assessment of plasmafibronectin in Crohn’s disease. Hepatogastroenterology 1991;38:231– 4.

2. Thamilselvan V, Basson MD. Pressure activates colon cancer celladhesion by inside-out focal adhesion complex and actin cytoskeletalsignaling. Gastroenterology 2004;126:8–18.

3. McNeil PL, Ito S. Molecular traffic through plasma membrane disrup-tions of cells in vivo. J Cell Sci 1990;96:549–56.

4. Zhang J, Li W, Sanders MA, et al. Regulation of the intestinal epi-thelial response to cyclic strain by extracellular matrix proteins.FASEB J 2003;17:926–8.

5. Basson MD, Coppola CP. Repetitive deformation and pressure activatesmall bowel and colonic mucosal tyrosine kinase activity in vivo.Metabolism 2002;51:1525–7.

6. Kirsner JB, Shorter RG. Recent developments in nonspecific inflam-matory bowel disease (second of two parts). N Engl J Med 1982;306:
837–48.

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3


7. Kirsner JB, Shorter RG. Recent developments in “nonspecific” inflam-matory bowel disease (first of two parts). N Engl J Med 1982;306:775–85.

8. Reddy SN, Bazzocchi G, Chan S, et al. Colonic motility and transit inhealth and ulcerative colitis. Gastroenterology 1991;101:1289–97.

9. Rabauq M, Eyal Aw, Kluger Y, et al. Bursting pressure in anastomotichealing in experimentally induced colitis in rats. Dis Colon Rectum1998;41:468–72.

0. Basson MD, Li GD, Hong F, et al. Amplitude-dependent modulationof brush border enzymes and proliferation by cyclic strain in humanintestinal Caco-2 monolayers. J Cell Physiol 1996;168:476–88.

1. Zhang J, Owen CR, Sanders MA, et al. The motogenic effects of cyclicmechanical strain on intestinal epithelial monolayer wound closure arematrix dependent. Gastroenterology 2006;131:1179–89.

2. Chaturvedi LS, Marsh HM, Shang X, et al. Repetitive deformationactivates focal adhesion kinase and ERK mitogenic signals in humanCaco-2 intestinal epithelial cells through Src and Rac1. J Biol Chem2007;282:14–28.

3. Araki H, Komoike Y, Matsumoto M, et al. Healing of duodenal ulcersis not impaired by indomethacin or rofecoxib, the selective COX-2inhibitor, in rats. Digestion 2002;66:145–53.

4. Peterson MD, Mooseker MS. Characterization of the enterocyte-likebrush border cytoskeleton of the C2BBe clones of the human intestinalcell line, caco-2. J Cell Sci 1992;102:581–600.

5. Basson MD, Modlin IM, Madri JA. Human enterocyte (caco-2) mi-gration is modulated in vitro by extracellular matrix composition andepidermal growth factor. J Clin Invest 1992;90:15–23.

6. Providence KM, White LA, Tang J, et al. Epithelial monolayer wound-ing stimulates binding of USF-1 to an E-box motif in the plasminogenactivator inhibitor type 1 gene. J Cell Sci 2002;115:3767–77.

7. Chaturvedi LS, Marsh HM, Basson MD. Src and focal adhesion kinasemediate mechanical strain-induced proliferation and ERK1/2 phos-phorylation in human H441 pulmonary epithelial cells. Am J PhysiolCell Physiol 2007;292:C1701–13.

8. Craig DH, Zhang J, Basson MD. Cytoskeletal signaling by way ofalpha-actinin-1 mediates ERK1/2 activation by repetitive deformationin human CaCO intestinal epithelial cells. Am J Surg 2007;194:
2
618–22.

9. Scott SM, Knowles CH, Wang D, et al. The nocturnal jejunal migrat-ing motor complex: defining normal ranges by study of 51 healthyadult volunteers and meta-analysis. Neurogastroenterol Motil 2006;18:927–35.

0. Han O, Sumpio BE, Basson MD. Mechanical strain rapidly redistrib-utes tyrosine phosphorylated proteins in human intestinal caco-2 cells.Biochem Biophys Res Commun 1998;250:668–73.

1. Zhong J, Gencay MM, Bubendorf L, et al. ERK1/2 and p38 MAPkinase control MMP-2, MT1-MMP, and TIMP action and affect cellmigration: a comparison between mesothelioma and mesothelial cells.J Cell Physiol 2006;207:540–52.

2. Von Offenberg Sweeney N, Cummins PM, et al. Cyclic strain-medi-ated regulation of vascular endothelial cell migration and tube forma-tion. Biochem Biophys Res Commun 2005;329:573–82.

3. Kobayashi M, Nishita M, Mishima T, et al. MAPKAPK-2-mediatedLim-kinase activation is critical for VEGF-induced actin remodelingand cell migration. EMBO J 2006;25:713–26.

4. Antico G, Lingen MW, Sassano A, et al. Recombinant human utero-globin/CC10 inhibits the adhesion and migration of primary humanendothelial cells via specific and saturable binding to fibronectin.J Cell Physiol 2006;207:553–61.

5. Gu J, Chen L, Shatos MA, et al. Presence of EGF growth factor ligandsand their effects on cultured rat conjunctival goblet cell proliferation.Exp Eye Res 2008;86:322–34.

6. Buffin-Meyer B, Crassous PA, Delage C, et al. EGF receptor transac-tivation and PI3-kinase mediate stimulation of ERK by alpha(2 A)-adrenoreceptor in intestinal epithelial cells: a role in wound healing.Eur J Pharmacol 2007;574:85–93.

7. Summers RW, Anuras S, Green J. Jejunal manometry patterns inhealth, partial intestinal obstruction, and pseudoobstruction. Gastroen-terology 1983;85:1290–300.

8. Dabareiner RM, White NA, Donaldson LL. Effects of intraluminaldistention and decompression on microvascular permeability and he-modynamics of the equine jejunum. Am J Vet Res 2001;62:225–36.

9. Camilleri M. Jejunal manometry in distal subacute mechanical ob-struction: Significance of prolonged simultaneous contractions. Gut
1989;30:468–75.

Documents

Supraphysiologic extracellular pressure inhibits intestinal