The effect of glucagon-like Peptide-2 receptor agonists on colonic anastomotic wound healing.

Abstract

Background. Glucagon-like peptide 2 (GLP-2) is an intestinal specific trophic hormone, with therapeutic potential; the effects on intestinal healing are unknown. We used a rat model of colonic healing, under normoxic, and stress (hypoxic) conditions to examine the effect of GLP-2 on intestinal healing. Methods. Following colonic transection and reanastomosis, animals were randomized to one of six groups (n = 8/group): controls, native GLP-2, long-acting GLP-2 (GLP-2- MIMETIBODY, GLP-2-MMB), animals were housed under normoxic or hypoxic (11%  O(2)) conditions. Animals were studied five days post-operation for anastomotic strength and wound characteristics. Results. Anastomotic bursting pressure was unchanged by GLP-2 or GLP-2-MMB in normoxic or hypoxic animals; both treatments increased crypt cell proliferation. Wound IL-1β increased with GLP-2; IFNγ with GLP-2 and GLP-2-MMB. IL-10 and TGF-β were decreased; Type I collagen mRNA expression increased in hypoxic animals while Type III collagen was reduced with both GLP-2 agonists. GLP-2 MMB, but not native GLP-2 increased TIMP 1-3 mRNA levels in hypoxia. Conclusions. The effects on CCP, cytokines and wound healing were similar for both GLP-2 agonists under normoxic and hypoxic conditions; anastomotic strength was not affected. This suggests that GLP-2 (or agonists) could be safely used peri-operatively; direct studies will be required.

Full text

Page 1

Hindawi Publishing Corporation
Gastroenterology Research and Practice
Volume 2010, Article ID 672453, 12 pages
doi:10.1155/2010/672453
Research Article
TheEffectof Glucagon-LikePeptide-2 Receptor Agonists on
ColonicAnastomotic WoundHealing
HeatherA.Redstone,1WilliamD.Buie,1DavidA.Hart,2LaurieWallace,3PamelaJ.Hornby,4
SarahSague,4JenJ. Holst,5and DavidL.Sigalet1,3,6
1Division of Colorectal Surgery, Department of Surgery, Faculty of Medicine, University of Calgary, Calgary,
AB, Canada T2N 1N4
2McCaig Institute for Bone and Joint Health, Department of Surgery, Faculty of Medicine, University of Calgary, Calgary,
AB, Canada T2N 1N4
3Gastrointestinal Research Group and Division of Pediatric General Surgery, Department of Surgery, Faculty of Medicine,
University of Calgary, Calgary, AB, Canada T2N 1N4
4Clinical Research and Development, Centocor Inc., Radnor, PA 19087-4557, USA
5Panum Institute, University of Copenhagen, 1165 Copenhagen, Denmark
6Department of Surgery, Alberta Children’s Hospital, 2888 Shaganappi Trail NW, Calgary, AB, Canada T3B 6A8
Correspondence should be addressed to David L. Sigalet, sigalet@ucalgary.ca
Received 11 May 2010; Revised 23 July 2010; Accepted 29 July 2010
Academic Editor: Stanley Ashley
Copyright © 2010 Heather A. Redstone et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
Background. Glucagon-like peptide 2 (GLP-2) is an intestinal specific trophic hormone, with therapeutic potential; the effects
on intestinal healing are unknown. We used a rat model of colonic healing, under normoxic, and stress (hypoxic) conditions
to examine the effect of GLP-2 on intestinal healing. Methods. Following colonic transection and reanastomosis, animals were
randomized to one of six groups (n = 8/group): controls, native GLP-2, long-acting GLP-2 (GLP-2- MIMETIBODY, GLP-2-
MMB), animals were housed under normoxic or hypoxic (11% O2) conditions. Animals were studied five days post-operation
for anastomotic strength and wound characteristics. Results. Anastomotic bursting pressure was unchanged by GLP-2 or GLP-2-
MMB in normoxic or hypoxic animals; both treatments increased crypt cell proliferation. Wound IL-1β increased with GLP-2;
IFNγ with GLP-2 and GLP-2-MMB. IL-10 and TGF-β were decreased; Type I collagen mRNA expression increased in hypoxic
animals while Type III collagen was reduced with both GLP-2 agonists. GLP-2 MMB, but not native GLP-2 increased TIMP 1-3
mRNA levels in hypoxia. Conclusions. The effects on CCP, cytokines and wound healing were similar for both GLP-2 agonists
under normoxic and hypoxic conditions; anastomotic strength was not affected. This suggests that GLP-2 (or agonists) could be
safely used peri-operatively; direct studies will be required.
1.Introduction
Many patients require surgical resection of the small or
large bowel for a wide range of indications, including
colorectal cancer, inflammatory bowel disease, and trauma.
Leakage from such intestinal anastomoses is a major source
of morbidity and mortality in gastrointestinal surgery [1].
Anastomotic dehiscence not only leads to peritonitis and
sepsis but is also associated with increased rates of local
recurrence of carcinoma [2]. While the overall rate of
anastomotic leakage in colorectal surgery is between 3.4
and 6%, the mortality from anastomotic leaks following
low anterior resection ranges between 6.0 and 39.3% [3].
Thus any potential perioperative treatment must be closely
evaluated for its effect on healing.
Anastomotic healing is a complicated series of events
which proceeds along a course similar to that of the more
familiar skin wound, but at an accelerated rate [4–7];
intestinal wounds have regained their previous full strength
by 7–10 days. Alterations of cytokine or growth factor

Page 2

2 Gastroenterology Research and Practice
expression, quantitatively, qualitatively, or temporally, can
have an effect on subsequent wound healing [8, 9]. Agents
which have shown positive effects in experimental studies
include growth hormone [5, 10] and insulin-like growth
factor-1[11,12].Howeverthesystemiceffectsoftheseagents
limit their clinical usefulness for this indication.
Glucagon-like peptide-2 (GLP-2) is a 33-amino acid
peptide with structural similarity to glucagon. It is an
enteroendocrine hormone secreted by the L-cells in the
mucosa of the small intestine and colon and acts primarily as
a regulator of nutrient absorption and an intestinal specific
trophic factor [13–15]. Interestingly, the widespread effects
of GLP-2 in the intestine are mediated by indirect pathways;
thereceptorisnotexpressedontheepitheliumbutonenteric
neurons, scattered enteroendocrine cells, and the pericryptal
myofibroblast [16–20]. In the short term, GLP-2 increases
intestinal blood flow [18] while over the longer term in non-
inflamedtissueitspecificallyincreasescryptcellproliferation
(CCP), with subsequent increases in crypt depth and villus
height and a net increase in mucosal nutrient transport [14,
21].However,ininflamedmucosa,GLP-2isantiproliferative,
decreasing the expression of proinflammatory cytokines
while increasing the expression of IGF-1 with the global
effect of increasing the healing of inflamed mucosa [19,
20]. These effects are not mediated through IL-10 or TGF-
β expression [20]. Of interest, the intestinotrophic effects
of GLP-2 are proposed to be mediated through IGF-1
expression [22], and IGF-1 has been shown to enhance
anastomoticwoundhealingwithimprovedburstingstrength
and increased collagen deposition at the anastomosis [11,
12, 23, 24]. From this, we theorized that GLP-2 may have
the same wound healing benefits as IGF-1 but without
its systemic effects. Alternatively, GLP-2 could potentially
impair healing through its antiinflammatory effects, as has
been shown with other anti-inflammatory agents such as
corticosteroids and nonsteroidal anti-inflammatory drugs
[25–27].
Glucagon-likepeptide-2-mimetibodyconstruct(GLP-2-
MMB)isalong-actingformofGLP-2.ThehumanGLP-2(1-
33) long acting analogue (an arginine to glycine substitution
of the second amino acid at the N-terminus) is coupled to
a domain that includes the Fc portion of an antibody, in
this case a rat IgG2b Fc isotype with a human IgG4 hinge
[28]. The increased molecular weight and pharmacokinetic
properties of the Fc construct dramatically increase the half-
life compared to engineered native peptides which may be
useful therapeutically [28–30].
We hypothesized that GLP-2 would have a positive effect
on the healing anastomosis by increasing the production
of anti-inflammatory cytokines IL-10 and prohealing TGF-
B, with no change in the pro-inflammatory cytokines IL-
1βα and IFN-α. We compared the effect of the long-
acting receptor agonist GLP-2-MMB with native GLP-2
using different states of systemic oxygenation to mimic
clinical perioperative stressors [6, 31]. A human study has
shown that a perianastomotic oxygen tension of <20mmHg
was highly predictive of anastomotic leakage; in this study
ambient oxygenation levels were set so that tissue PO2levels
were 40–50mm Hg [6, 31]. The time at which leaks appear is
typically from days 5 to 7 post operation, which is the phase
of maximal collagenase activity in the wound; accordingly
the fifth day postoperation was chosen as the study endpoint
[3, 4, 32].
2.Methods
2.1. Animals and Surgical Procedure. Ethical approval for
the experimental protocol was granted by the Animal Care
Committee of the Faculty of Medicine, University of Calgary.
Male Sprague-Dawley rats (250–300g) were acclimatized for
a week and fasted for 24 hours prior to surgery with free
access to water. Prior to surgery, animals were randomized
to the GLP-2, the GLP-2-MMB, or control groups. Animals
were injected s.c. one hour prior to surgery with 100μg/kg
(1-33) GLP-2 (American Peptide, Sunnyvale California) in
1ml normal saline, 2 mg/kg GLP-2-MMB (Centocor Inc R.
and D., Radnor PA) diluted in 1ml of normal saline, or
1 ml of normal saline (controls). GLP-2-MMB dosing was
based on previous studies, establishing dose-response equiv-
alence for increasing CCP with native GLP-2 [28]. Animals
were weighed and anesthetized using 1-2% isoflurane. A
prophylactic dose of 25mg cefazolin (Novopharm Limited,
Toronto, ON, Canada) was given subcutaneously just prior
to surgical incision. The transection-anastomosis was done
in the midportion of the transverse colon, following our
previously described methods [6]. The marginal vessels
were ligated with 6-0 silk (Ethicon Inc., Somerville, NJ),
the bowel was transected and an anastomosis created with
10–14 interrupted 6-0 monofilament polypropylene sutures
(Proline, Ethicon Inc. Somerville, NJ). The abdomen was
irrigated with saline and closed in two layers (skin and
muscle) using 4-0 Vicryl running sutures (Ethicon). A
subcutaneousfluidbolusof10mlofnormalsalinewasgiven,
and a single intramuscular dose of 0.015mg buprenorphine
(Temgesic, Schering-Plough Ltd., Hertfordshire UK) was
given for analgesia.
2.2. Treatment Groups. Following surgery, the animals were
randomized a second time to be placed in a normoxic
(FiO2 = 21%) or hypoxic (FiO2 = 11%) environment,
following our previously described methods [6]. In brief, an
airtight lid was fitted to a standard polyethylene rat cage,
with appropriate connections for controlled air input. The
hypoxia generator (Hypoxico Inc, New York) was set on
50% which gave a hypoxic environment of approximately
11% FiO2, corresponding to a tissue oxygenation of 40–
50mmHg [6]. The oxygen content of the chamber was
measured daily using a MiniOx I oxygen analyzer (MSA
Medical Products, Pittsburg, PA). Normoxic animals were
maintained in standard rat cages. All rats were housed in
pairs, and food intake was recorded over the course of the
study period. Animals were pair-fed a standard rodent pellet
diet and allowed water ad libitum. A total of 48 animals were
used for the experiment and were divided into six treatment
groups: normoxic control, normoxic GLP-2, normoxic GLP-
2-MMB, hypoxic control, hypoxic GLP-2 and hypoxic GLP-
2-MMB. Animals were maintained in their normoxic or
hypoxic environments, for the duration of the experiment

Page 3

Gastroenterology Research and Practice3
(5 days). All animals were injected twice daily, from 0800 to
0900 and again at 1700 to 1800 controls with saline, GLP-
2, animals with 100μg/kg/day (1-33) GLP-2, GLP-2-MMB
animals with 2mg/kg of the GLP-2-MMB on days 0 and 3
and saline at the other time points. Each injection lasted less
than 90 seconds, and the animals were returned immediately
to the appropriate oxygen environment.
2.3.Endpoints. Onpostoperationdayfive,withcontinuation
of a normal feeding pattern between 9 and 11 am, coin-
cident with the treatment injection (saline/GLP-2 ligand),
animals were injected intraperitoneally with 100mg/kg of 5-
Bromo-2?-deoxyuridue(BrdU)(Sigma-AldrichInc,St.Louis
MO, USA). One hour post saline/GLP-2/MMB and BrdU
injection, they were anesthetized with 1-2% isoflurane and
weighed, and a laparotomy was performed. Approximately
5ml of blood was taken by cardiac puncture drawn into iced
EDTA vacutainers containing 10% volume of 5000KU tra-
sylol and 0.1mM diprotinin A, centrifuged within one half
hour, and the serum was frozen. GLP-2 (1-33) quantification
wasperformedusingaRIAspecificfortheactiveN-terminus
of 1-33 GLP-2 (J. J. Holst, Panum Institute, Copenhagen,
Denmark, antibody no. 92160) [33]. This does not capture
the GLP-2 activity of the MMB construct since the antibody
precipitates during the sample preparation [33].
The animals were then euthanized by exsanguination.
The entire length of bowel from pylorus to distal colon
was removed. The length and width of small and large
intestine were measured, under standardized tension (2g).
The colon was then divided 3cm proximally and distally to
theanastomoticsites.Anyadhesionstotheanastomosiswere
removed en bloc with the segment of colon to preserve the
integrity of the anastomosis.
2.4. Burst Pressure Measurements. The bursting pressure
was measured using an infusion pump coupled to an in-
line sphygmomanometer (Welch Allyn Tycos, Skaneateles
Falls, NY), as previously described in [6]. The bowel was
submerged in a saline bath and infused with air at a rate of
5ml/minute. Bursting pressure was recorded as the pressure
where visualization of bubbles was first observed.
2.5. Tissue Preparation. The ties were removed, and the
segment of bowel containing the anastomosis was opened
longitudinally. A segment containing the anastomosis and
5mm of colonic tissue on either side was removed. It was
dividedlongitudinallyintothreesegments:oneforhistology,
one for cytokine analysis, and one for RNA isolation for
RT-PCR and processed as described below. We have shown
previously that the process of determination of bursting
pressure does not change histological scores, cytokine or
mRNA content of the tissue [6].
2.6. Collagen, Tissue Inhibitors of Metalloproteinases, and Col-
lagenase Analysis. The primary strength within the intestinal
wound comes from collagen I and III; the increase in
collagenase 13 activity over days 4–8 is thought to contribute
to the development of leakage, especially with coexisting
inflammation [34, 35]. Tissue inhibitors of metallopro-
teinases (TIMPs) counteract this activity [36]. To examine
the effects of GLP-2 therapy on these factors, the mRNA
levels for each molecule were assessed in the perianastomotic
tissue. As described, a longitudinal segment of bowel tissue
containing the anastomosis and 5mm of bowel on either
side was flash frozen in liquid nitrogen and then stored
at −80◦C until processing. Total RNA was extracted using
the TRIspin method [37] and quantified using the SYBR
Green reagent (Molecular Probes, Eugene, OR) method.
Simultaneous reverse transcription (RT) reactions using 1ug
of RNA from all samples in each group were carried out with
the OmniScipt kit (Qiagen, Hilden, Germany). Polymerase
chain reaction (PCR) was used to assess mRNA levels.
Primers for collagen type I, type III, TIMP 1, 2, and 3, and
matrix metalloproteinase 13 (MMP-13) were synthesized on
site; sequences were as reported previously in [36, 38]. PCR
conditionswereoptimizedandrigorouslycontrolledforeach
molecule (temperatures, number of cycles) to ensure detec-
tion was in the linear phase of the amplification. Agarose gel
electrophoresis followed by staining with ethidium bromide
was used for separation and detection of the PCR-generated
cDNA amplicons (Gel Doc XR System; BioRad, Hercules,
CA). The results were normalized by dividing values for
each gene to the expression of β-actin in the same sample;
the expression of β-actin, a housekeeping gene, does not
change at the anastomosis [6]. Reanalysis for a subset of the
genes assessed with a second aliquot of RNA revealed nearly
identical results to those reported Furthermore, in other
systems, a direct comparisons between the mRNA analysis
using the indicated methodology and real-time qPCR has
revealed nearly identical results for the high copy number
genes assessed (Hart DA and Reno CR, unpublished).
2.7. Cytokine Analysis. Tissues were stored at −80◦C until
processing. Perianastomotic specimens as detailed above
were thawed, washed in saline, and then homogenized with
5 volumes of homogenizing solution. The homogenizing
solution was made of 50ml 100mM phosphate buffered
saline (PBS), pH 7.0, with one complete protease inhibitor
cocktail tablet (Roche Diagnostics, Mannheim, Germany).
The tissue was homogenized in a Polytron homogenizer
(Kinematic model 2100, Switzerland). Homogenates were
centrifugedat2000rpmfor20minutestoremovelargetissue
particles and then centrifuged at 13, 200rpm for 60 minutes
to obtain a clear supernatant. The supernatant was aliquoted
and stored at −80◦C until further use. Protein content
from the homogenate for each specimen was determined
according to Lowry’s method [39].
The supernatants were quantitatively assayed using
enzyme-linked immunosorbent assay (ELISA) kits for rat
tumor necrosis factor-α (TNF-α) (Assay Design, Ann Arbor,
Michigan, USA) interleukin 1-β (IL-1β), interleukin-10 (IL-
10), interleukin-13 (IL-13) transforming growth factor-β
(TGF-β),interferonγ (IFNγ)(Biosource,Camarillo,Califor-
nia, USA), each following the manufacturer’s directions. All
specimens were analyzed in duplicate, and then normalized
to the total protein content of the sample, and expressed as
pg/mg protein.

Page 4

4 Gastroenterology Research and Practice
0
0.2
0.4
0.6
0.8
1
Control GLP-2MMB
Proliferation index
Normoxic
∗
∗∗
(a)
0
0.2
0.4
0.6
0.8
1
Control
GLP-2
MMB
Proliferation index
Hypoxic
∗∗
(b)
Figure 1: Crypt cell proliferation index. Proliferating cells in the ileal crypts were detected by BrdU immunohistochemical labelling; the
index is the number of labelled cells per total cells in each half crypt, averaged from 5 or more crypts per animal, n = 5 or more per group.
Data: mean ± SEM,∗P < .05 versus the group indicated, †P < .05 versus similarly treated group under normoxic conditions. Comparisons
by ANOVA.
2.8. Crypt Cell Proliferation. As a measure of the biological
effect of each GLP-2 receptor ligand, crypt cell proliferation
was measured. At the time of euthanasia, a segment of
distal ileum, approximately 10cm from the ileocecal valve,
was harvested and preserved in 10% formalin for at least
48 hours. The tissue was then dehydrated and embedded
in paraffin blocks. Sections 6μm thick were cut using a
microtome,andimmunohistochemicalstainingwasdonefor
BrdU-(antibody 1:100 rabbit anti-BrdU; Serotec, London,
UK).Thenumber ofBrdUstained cellsthetotalcellsperhalf
crypt were then counted in five crypts for each animal, and
expressedastheratioofBrdU+cells/totalcellsperhalfcrypt,
and averaged to give the value for each animal [19].
2.9. Histology. A strip of tissue containing the anastomosis
was placed immediately into 10% formalin and preserved
for at least 48 hours and then dehydrated and embedded
in paraffin blocks. Sections 6μm thick were cut using a
microtome and stained with hematoxylin and eosin (Sigma,
St. Louis, MO, USA) and examined using standard light
microscopy. All observations were made by one examiner
(H.A. Redstone) who was blinded to the treatment group.
Sections were evaluated using a semiquantitative scoring
method modified from our previous report [6, 40]. In brief,
each of the parameters of healing, inflammatory reaction,
submucosal/muscularis bridging, and mucosal epithelial
healing, were assigned a score from 0 to 3, with 0 being
normal and the maximum total of nine with a lower score
indicating improved healing.
2.10. Statistical Analysis. Results are expressed as mean ±
standard error of the mean. Significance differences were
assessed using the one-way ANOVA with Tukey’s post test
analysis. P values of <.05 were considered statistically sig-
nificant. The computer software program GraphPad Prism
version 4.01 (Prism Corp, La Jolla, CA) was used for these
calculations.
3.Results
3.1. Clinical Outcome and Gross Morphology. All animals
tolerated surgery, with two perioperative deaths, one in the
normoxic control group on day one and one in the hypoxic
GLP-2-MMB group on the fourth postoperative day. The
anastomoses of these two animals were intact on necropsy,
and no specific cause could be identified; these results were
excluded from further analysis. There were no differences
in change in body weight, food consumption, and small or
large bowel length and width with GLP-2 and GLP-2-MMB
treatment as compared to controls in either the normoxic
group or hypoxic group (data not shown).
3.2. Crypt Cell Proliferation. Crypt cell proliferation (CCP)
was significantly increased in both the GLP-2 and GLP-2-
MMB-treatment groups (Figure 1) under normoxic condi-
tions. There was an expected increase in CCP in control
animals under hypoxic conditions; under these conditions,
the proliferation rate of the MMB-treated animals did
not increase (Figure 1) [6]. The native GLP-2 retained a
proliferative effect under hypoxic conditions.
3.3. Bursting Pressure Assessment. As expected, anastomotic
bursting pressures were significantly lower in the hypoxic
animals but there were no differences with either GLP-2
or GLP-2-MMB treatment within the hypoxic or normoxic
groups (Figure 2).

Page 5

Gastroenterology Research and Practice5
0
50
100
150
200
ControlGLP-2 MMB
Normoxic
(mmHg)
(a)
0
50
100
150
200
Control GLP-2 MMB
Hypoxic
(mmHg)
(b)
Figure 2: Anastomotic bursting pressures of normoxic and hypoxic animals. Groups treated with GLP-2, GLP-2-mimetibody (MMB) and
controls. Isolated segments of intestine insufflated with air under saline: pressure at which the first bubbles appear is recorded. (n = 7 or
more per group, Data: mean ± SEM,∗P < .05 versus similarly treated group under normoxic conditions. Comparisons by ANOVA).
3.4. Histological Analysis. The mean anastomotic healing
scores for the groups were not significantly different with
either GLP-2 or GLP-2 mimetibody treatment as compared
tocontrols,undereithernormoxicconditionorhypoxiccon-
dition (Figure 3). There were areas of considerable artifact
duetothedisruptionoftissueplanesbytheburstingpressure
studies; however, the status of the surface epithelium,
muscular layer, and inflammation in the subserosa could be
determined in all animals (Figure 3).
3.5. GLP-2 Serum Levels. Mean serum levels of GLP-2 were
35.0 ± 3.1 in control animals (similar in both normoxic and
hypoxic), 119 ± 32 in GLP-2-treated animals, and 49.1 ±
10.3pmol/l in GLP-2-MMB-treated animals. There were no
differences between the levels in hypoxic and normoxic
animals. The serum levels of GLP-2 were significantly higher
in animals receiving GLP-2 (1-33) (P < .05); however, the
levels in GLP-2-MMB-treated animals were not significantly
elevated compared to controls. Importantly, the levels in
GLP-2 MMB animals reflect only endogenous GLP-2, since
the assay does not detect antibody bound activity [33].
3.6. Cytokine Analysis. Levels of the pro-inflammatory
cytokines IL-1β, TNF-α, and IFN-γ in GLP-2-treated and
control animals are shown in Figure 4. Levels of IL-1β and
IFN-γ were both increased with GLP-2 under both normoxic
andhypoxicconditions.GLP-2-MMBtreatmentresultedina
decrease in IL-1β under both oxygen conditions and resulted
in a differential effect on IFNγ levels; these were decreased
under normoxic and increased under hypoxic conditions.
LevelsofTNF-αwerenotsignificantlyaffectedbyeitherform
of GLP-2. Levels of the anti-inflammatory cytokine IL-10 as
well as TGFβ and IL-13 are shown in Figure 5; GLP-2 and
GLP-2-MMB treatment resulted in a significant decrease in
mucosal IL-10, but no change in IL-13 levels. TGF-β levels
weredecreasedbybothGLP-2therapies,inbothhypoxicand
normoxic animals.
3.7. Collagen, Tissue Inhibitors of Metalloproteinases, and
Collagenase Expression. Results of RT-PCR assessment of
mRNA levels for collagen types I (cI) and III (cIII) and
MMP-13 are shown in Figure 6. Expression of cI mRNA in
hypoxic conditions was increased by both GLP-2 receptor
agonists. Levels of cIII mRNA were decreased under nor-
moxic conditions by both GLP-2 agonists, while the changes
under hypoxic conditions were not significant. There were
no major effects on MMP-13 mRNA levels by the GLP-2
agonists, under either ambient oxygen status; however, there
were interesting statistically significant increases in mRNA
levels for TIMP 1, 2, and 3 in hypoxic animals treated with
GLP-2-MMB, but not native GLP-2 (data not shown).
4.Discussion
This study is the first to examine the effects of GLP-2 on
intestinal anastomotic healing. It evaluates the effect of GLP-
2 on colonic anastomosis in an animal model of normal and
impaired wound healing (hypoxia) using both native GLP-2
as well as a long-acting form of the hormone, GLP-2-MMB.
Our original hypothesis was that healing would be improved
with the addition of GLP-2, especially in hypoxic animals;
this did not occur. However, this study did reveal several
important findings. First, the results indicate that exogenous
GLP-2 does not exert a negative effect on anastomotic
integrity. Secondly, exogenous GLP-2 affects both pro- and
anti-inflammatory pathways within the healing anastomosis
andalsoeffectscollagengeneexpression,alteringthetypesof