Inappropriate Angiogenic Response as a Novel Mechanism of Duodenal Ulceration and Impaired Healing

1 23

Digestive Diseases and Sciences ISSN 0163-2116Volume 56Number 10 Dig Dis Sci (2011) 56:2792-2801DOI 10.1007/s10620-011-1753-4

Inappropriate Angiogenic Response as aNovel Mechanism of Duodenal Ulcerationand Impaired Healing

Xiaoming Deng, Ximing Xiong, TetyanaKhomenko, Zsuzsanna Sandor, KlaraOsapay, Ganna Tolstanova, JosephShiloach, Longchuan Chen, et al.

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer

Science+Business Media, LLC (Outside the

USA). This e-offprint is for personal use only

and shall not be self-archived in electronic

repositories. If you wish to self-archive your

work, please use the accepted author’s

version for posting to your own website or

your institution’s repository. You may further

deposit the accepted author’s version on

a funder’s repository at a funder’s request,

provided it is not made publicly available until

12 months after publication.

ORIGINAL ARTICLE

Inappropriate Angiogenic Response as a Novel Mechanismof Duodenal Ulceration and Impaired Healing

Xiaoming Deng • Ximing Xiong • Tetyana Khomenko •

Zsuzsanna Sandor • Klara Osapay • Ganna Tolstanova •

Joseph Shiloach • Longchuan Chen • Judah Folkman •

Sandor Szabo

Received: 21 December 2010 / Accepted: 9 May 2011 / Published online: 7 July 2011

� Springer Science+Business Media, LLC (Outside the USA) 2011

Abstract

Background Despite recent advances and better under-

standing of the etiology and the pathogenesis of gastroin-

testinal ulcer diseases, e.g., duodenal ulcer, the molecular

events leading to ulcer development, delayed healing, and

recurrence remain poorly elucidated.

Aims After we found that duodenal ulcers did not heal

despite increased levels of vascular endothelial growth

factor (VEGF), we tested the hypothesis that an imbalance

in angiogenic VEGF and anti-angiogenic endostatin and

angiostatin might be important in the development and

delayed healing of experimental duodenal ulcers.

Methods Levels of VEGF, endostatin, and angiostatin,

and the expression and activity of related matrix metallo-

proteinases (MMP) 2 and 9 were measured in scrapings of

rat proximal duodenal mucosa in the early and late stages

of chemically induced duodenal ulceration. Furthermore,

animals were treated with recombinant endostatin and

MMP 2 inhibitor to test the relationship between MMP2

and endostatin and their involvement in healing of exper-

imental duodenal ulcers.

Results A concurrent increase of duodenal VEGF, endo-

statin, and angiostatin was noted during duodenal ulcera-

tion. Endostatin treatment aggravated duodenal ulcer.

Levels of MMP2, but not MMP9, were increased. Inhibi-

tion of MMP2 reduced levels of endostatin and angiostatin,

and attenuated duodenal ulcers.

Conclusions Increased levels of endostatin and angio-

statin induced by MMP2 delayed healing of duodenal

ulcers despite concurrently increased VEGF. Thus, an

inappropriate angiogenic response or ‘‘angiogenic imbal-

ance’’ may be an important new mechanism in ulcer

development and impaired healing.

Keywords Duodenal ulcer � Angiogenic imbalance �VEGF � Endostatin � Angiostatin � MMP2 � TIMP-1

Introduction

Despite advances in H. pylori-related basic and clinical

research, duodenal ulcers remain the most prevalent form

of ‘‘peptic ulcer’’ with major public health and economic

effects [1]. Part of the problem is the increasing proportion

of H. pylori-negative duodenal ulcers, which has reached

X. Deng � X. Xiong � T. Khomenko � K. Osapay �G. Tolstanova � L. Chen � S. Szabo (&)

Diagnostic & Molecular Medicine, VA Medical Center, 5901 E.

7th Street, Long Beach, CA 90822, USA

e-mail: [email protected]

X. Deng � T. Khomenko � G. Tolstanova

Department of Pathology, University of California-Irvine, Irvine,

CA 92697, USA

Z. Sandor

Medical Health Care Groups, VA Medical Center, 5901 E. 7th

Street, Long Beach, CA 90822, USA

Z. Sandor

Department of Medicine, University of California-Irvine, Irvine,

CA 92697, USA

J. Shiloach

Biotechnology Unit, NIDDK, NIH Bldg 14A, Room 173,

Bethesda, MD 20892, USA

J. Folkman

Departments of Pediatric Surgery and Cell Biology, Children’s

Hospital, Harvard Medical School, Boston, MA 02115, USA

S. Szabo

Departments of Pathology and Pharmacology, University of

California-Irvine, Irvine, CA 92697, USA

123

Dig Dis Sci (2011) 56:2792–2801

DOI 10.1007/s10620-011-1753-4

Author's personal copy

20–30% in clinical studies [2, 3], and the fact that

‘‘increasing eradication of H. pylori infection in the US has

not resulted in fewer hospital admissions for peptic ulcer

disease-related complications,’’ for example hemorrhage

and perforations [4]. These data also emphasize the role of

other etiologic factors such as stress, non-steroidal anti-

inflammatory drugs (NSAID), and other chemicals in ulcer

pathogenesis [5, 6] and the need for more mechanistic

studies related to duodenal ulceration. In rodents NSAID

and stress induce gastric ulcers only, and specific duodenal

ulcerogens (e.g., cysteamine, propionitrile) are needed to

reproduce the most prevalent form of ‘‘peptic ulcers’’ in

rats [7–9]. For this purpose animal models, for example the

cysteamine rat model [7–9], provide unique opportunity to

gain insights into the molecular pathogenesis of early pre-

ulcer lesions and the pathways leading to poor healing.

Angiogenesis, i.e., endothelial cell proliferation and tube

formation in postembryonic tissue, is a crucial element in

external (e.g., skin) and internal (e.g., gastrointestinal)

wound/ulcer healing that needs granulation tissue which

forms the basis of proliferating and migrating epithelial cells

to complete the healing process [10]. The process is gov-

erned by the balance between angiogenic factors such as

vascular endothelial growth factor (VEGF), basic fibroblast

growth factor (bFGF), and platelet-derived growth factor

(PDGF), and anti-angiogenic factors such as endostatin and

angiostatin [10]. The critical switch to angiogenesis involves

a change in the local equilibrium between these positive and

negative regulators of microvessels [11, 12].

Our previous studies revealed elevated levels of angio-

genic factors bFGF, PDGF, and VEGF in chemically

induced acute duodenal ulceration [13, 14]. Because of the

perceived protective roles of these growth factors, these

results were surprising. For example, we could not under-

stand and explain why the healing of duodenal ulcers would

be impaired or delayed despite elevated local tissue

expression and concentration of the angiogenic peptides

[10]. We also observed accelerated healing of chronic duo-

denal ulcers in rats treated with peptides or genes of bFGF,

PDGF, or VEGF [15–17]. New biochemical, molecular

biological, and immunohistochemical studies indicate that

bFGF, PDGF, and VEGF play a pathophysiologic role in the

natural history of ulcer healing [18]. Because angiogenesis is

governed by a balance of pro and anti-angiogenic factors, it is

possible that, as in cancerogenesis [19], the inhibitory

activity on endothelial cells is because of a net excess of anti-

angiogenic factors in the wound/ulcer environment.

Angiostatin and endostatin are endogenously produced

peptides which specifically target endothelial cells, result-

ing in potent inhibition of angiogenesis. These proteins

may be involved, leading to ulcer aggravation and delayed

healing. Angiostatin and endostatin are 50 and 20-kDa

fragments cleaved from plasminogen and collagen XVIII,

respectively, by proteinases such as MMP2 and MMP9

[20–23]. Because they have anti-angiogenic roles, by

inhibiting endothelial cell proliferation and migration, and

inducing apoptosis in proliferating endothelial cells [24,

25], increased levels of endostatin and angiostatin may

explain why the healing of duodenal ulcer is poor, despite

the increased level of VEGF, which promotes mucosal

angiogenesis and healing of injury. This may provide a

novel and mechanistic insight into duodenal ulceration.

MMPs, involved in angiogenesis, are released from endo-

thelial cells in response to cytokines during wound healing.

MMPs have been implicated as among the main factors

contributing to mucosal ulceration. The ratio of MMPs to

their natural inhibitors TIMPs may be important in angio-

genesis. The most widely investigated MMPs and TIMPs

are MMP2, MMP9, TIMP-1, and TIMP-2, because they are

closely involved in angiogenesis. TIMPs might participate

in the repair process [26].

In this study, we examined levels of VEGF, angiostatin,

endostatin, MMP2, MMP9, TIMP-1, and TIMP-2 in duo-

denal mucosa, and tested the mechanistic effects of

over-expression of endostatin and inhibition of MMP2 on

duodenal ulceration induced by cysteamine or propionitrile

in rats.

Methods

Animal Experiment

Adult Sprague–Dawley female rats (180–210 g) with

unlimited access to food and water were allowed to accli-

matize for 3–5 days in stainless-steel mesh cages (three

rats/cage) in a room with a 12:12 h light–dark cycle at a

constant temperature of 22�C. All experiments were carried

out in compliance with our Institutional Regulations for

Animal Use and Care. Three animal experiments were

performed in this study (Table 1).

Experiment 1

Randomized groups (n = 6) of unfasted rats were given

cysteamine-HCl (Aldrich, Milwaukee, WI, USA; 25 mg/

100 g, p.o.) or propionitrile (Aldrich; 5 mg/100 g, s.c.) to

induce duodenal ulcers, and euthanized 0.5 or 2 h after a

single dose or 12 or 24 h after three doses (4 h intervals) of

cysteamine or propionitrile. Scrapings of the 2.5 cm

proximal duodenal mucosa were harvested for examination

of VEGF, endostatin, angiostatin, MMP2, MMP9, TIMP-1,

and TIMP-2. Gene and protein expression was detected by

real-time PCR, Western blotting, and ELISA. Proteolytic

activity of MMP2 and MMP9 was detected by

zymography.

Dig Dis Sci (2011) 56:2792–2801 2793

123


Experiment 2

Randomized groups (n = 8–12) of unfasted rats were

injected s.c. with saline or 2 mg/100 g of rat endostatin

(provided by Dr Joseph Shiloach as a gift) either 91,

30 min before the 3rd dose of cysteamine on the first day

and 92 on the 2nd day, and euthanized on the 3rd day, or

92 daily on the 2nd–6th days and euthanized on the 7th

day after cysteamine. Duodenal ulcer diameters were

measured at autopsy and calculated by use of the ellipsoid

formula. The opened stomachs with 2.5 cm duodenum

were fixed in 10% formalin for light microscopic histology.

Experiment 3

Groups of unfasted rats were given saline or a selective

inhibitor of MMP2 (MMP2 Inhibitor I; EMD Chemicals,

Gibbstown, NJ, USA) at 1 mg/rat, s.c. 0.5 h before and

24 h after cysteamine. The rats were euthanized 48 h after

cysteamine. Duodenal ulcer diameters were measured at

autopsy and ulcer areas were calculated by use of the

ellipsoid formula. Mucosal scrapings of the 2 cm proximal

duodenum obtained at autopsy were homogenized and

tested for expression of MMP2, endostatin, and angiostatin

by zymography and Western blotting.

Total RNA Extraction and Purification

An RNeasy Mini kit (Qiagen, San Diego, CA, USA) was

used for total RNA extraction and purification. RNA

was extracted twice with saturated phenol–chloroform and

was cleaned using Spin Columns (Qiagen). The quality and

quantity of extracted total RNA was determined by spec-

trophotometry (Beckman, Fullerton, CA, USA; 610) fol-

lowed by agarose–formaldehyde gel electrophoresis.

Real-Time PCR

Real-time PCR was performed by use of TaqMan gene

expression assays (Applied Biosystems). The following

cycles were used on a BioRad iCycler Real-time PCR

machine: 2 min at 50�C, 10 min at 95�C, and 40 cycles of

two steps: 15 s at 95�C and 1 min at 60�C. The level of

target gene mRNA measured by threshold cycle number

was compared with GAPDH, which is used as an internal

control to correct for variability in starting mRNA con-

centration. Amounts for treated groups, as multiples of the

amount for the control group, were calculated.

Total Protein Extraction

Duodenal mucosal scrapings (200–250 mg/each) were

homogenized in lytic buffer with proteinase inhibitors and

centrifuged at 15,000 rpm. Protein concentrations of the

supernatants were determined by use of a Bio-Rad protein

assay, using bovine serum albumin as standard (Bradford).

Tissue samples were loaded at 1:1 ratio with sample buffer

(Sigma).

Western Blot Analysis

Total proteins (100 lg) were separated by 12% SDS–

PAGE. The blot was blocked with a 5% solution of dry

milk for 2 h, and incubated with antibodies against endo-

statin (Lab Vision, Fremont, CA, USA) and angiostatin

(Novus Biologicals, Littleton, CO, USA) at 1:200 dilutions,

VEGF (Santa Cruz Biotechnology, Santa Cruz, CA, USA)

at 1:200 dilution, and MMP2, MMP9, TIMP-1 and TIMP-2

(Santa Cruz Biotechnology) at 1:500 dilution. The blots

were incubated with anti-mouse or rabbit IgG (Santa Cruz

Biotechnology) at 1:4000 dilution. The membrane was

exposed to Hyper film ECL (Amersham). The density of

the bands was determined by scanning densitometry Eagle

Eye II (Stratagene, Cedar Creek, TX, USA).

Enzyme-Linked Immunosorbent Assay (ELISA)

Rat MMP2 and MMP9 immunoassay kits (R&D Systems,

Minneapolis, MN, USA), mouse endostatin immunoassay

kits (CytImmuno, Minneapolis, MN, USA), and human

VEGF immunoassay kits (R&D Systems, Minneapolis,

MN, USA) were used for measurement of endostatin,

Table 1 Animal experiment designs

Experiment Ulcerogen Treatment Euthanasia

Experiment 1 Cysteamine, 91 – 0.5 or 2 h after cysteamine

Cysteamine, 93 – 12 or 24 h after cysteamine

Propionitrile, 91 – 2 h after propionitrile

Propionitrile, 93 – 12 h after propionitrile

Experiment 2 Cysteamine, 93 Endostatin, 2 mg/rat, 30 min before and 2nd day after C 3rd day after cysteamine

Cysteamine, 93 Endostatin, 2 mg/rat, 92/day, 2nd–6th days after C 7th day after cysteamine

Experiment 3 Cysteamine, 93 MMP2 Inhibitor I, 1 mg/rat, 0.5 h before and 24 h after C 3rd day after cysteamine

2794 Dig Dis Sci (2011) 56:2792–2801

123


angiostatin and VEGF concentration in the duodenal

mucosa after administration of the chemicals according to

the manufacturer’s directions. We calculated the concen-

trations as the ratio (ng/mg) of endogenous endostatin,

angiostatin, VEGF, MMP2, or MMP9 to total protein.

Gelatin Zymography

Enzymatic activity of MMP2 and MMP9 in duodenal

mucosa was measured by electrophoretic zymography.

Briefly, 50 lg total protein was electrophoresed in 8%

SDS–PAGE gel containing gelatin (1 mg/ml). Gel was

incubated in Triton X-100 (2.5%) for 30 min followed by

incubation at 37�C over night in Zymogram developing

buffer (50 mM Tris base, 50 mM Tris acid, 0.2 mM NaCl,

5 mM CaCl2, and 0.02 mM Brij). Substrate gels were

stained with Coomassie brilliant blue (0.25%) in methanol–

acetic acid–water (50:10:40). Proteolytic activity was

visualized as clear bands of lysis on a blue background of

undigested gelatin.

Histology

To assess histologic damage, full-thickness duodenal tissue

samples were embedded in paraffin, sectioned, and stained

with hematoxylin and eosin (H&E) or periodic acid Schiff

(PAS, to assess the regeneration of mucosa, especially

mucus-secreting epithelial cells).

Data Analysis

The statistical significance of differences among groups was

calculated by use of the non-parametric Mann–Whitney

U test. Data are reported as mean ± SEM. For statistical

significance, P \ 0.05 or smaller values were accepted.

Results

Concurrently Increased Levels of Angiogenic Factor

VEGF and Anti-angiogenic Factors Endostatin

and Angiostatin

Because angiogenesis is governed by a balance between

pro and anti-angiogenic factors, we examined the levels of

both the angiogenic VEGF and anti-angiogenic factors

endostatin and angiostatin in duodenal mucosa during

chemically induced duodenal ulceration. Western blotting

showed that levels of VEGF were significantly increased in

both early (0.5–2 h) and late (12–24 h) stages of duodenal

ulceration after administration of the ulcerogenic chemicals

(Fig. 1a). ELISA showed significantly increased VEGF

concentration in the duodenal mucosa in both early and late

stages of the ulceration, which was similar to the changes

seen in Western blotting (Fig. 1b).

Western blotting demonstrated that endostatin was

increased in the late stage and angiostatin was enhanced at

all time points after administration of the ulcerogenic

chemicals (Fig. 2a). We also detected significantly

increased concentrations of endostatin and angiostatin by

ELISA in the late stages of duodenal ulceration (Fig. 2b).

These confirmed the findings by Western blot.

Increased MMP2 and Reduced TIMP-1 in Duodenal

Mucosa During Duodenal Ulceration

MMP2 and MMP9 are important enzymes which cleave

collagen XVIII and plasminogen to generate endostatin and

angiostatin, respectively. Hence, we investigated the gene

expression of MMP2 and MMP9 by real-time PCR.

MMP2, but not MMP9, mRNA levels had increased by

approximately threefold to fourfold 2 h after administra-

tion of cysteamine or propionitrile (Fig. 3a). Furthermore,

Western blotting showed that levels of both precursor and

active forms of MMP2 were also increased after the

increased gene expression in duodenal mucosa, whereas

MMP9 did not change in duodenal mucosa during chemi-

cally induced duodenal ulceration (Fig. 3b). ELISA

revealed that active MMP2 concentrations were signifi-

cantly enhanced in duodenal mucosa 2–24 h after admin-

istration of cysteamine or propionitrile (Fig. 3c). Western

blotting also revealed significantly reduced levels of

TIMP-1 from 2 to 24 h after administration of the ulcero-

genic cysteamine and propionitrile, whereas levels of

TIMP-2 did not change (Fig. 3d).

Up-regulated Enzyme Activity of MMP2 in Duodenal

Ulceration

The proteolytic activities of these MMPs were further

measured by gelatin zymography in duodenal mucosa after

administration of the ulcerogenic chemicals. Increased

gelatinolytic activity (white bands) at a molecular weight

of approximately 63 kDa, corresponding to the active

MMP2, was observed in the both early and late stages of

duodenal ulceration induced by cysteamine or propionitrile

(Fig. 4a). There were no marked changes of MMP9

(92 kDa) levels in duodenal mucosa after administration of

any of the chemicals (Fig. 4b).

Aggravation of Duodenal Ulcers After Endostatin

Administration

Pooled results from several experiments showed that

average ulcer size was significantly increased from

15.8 ± 2.7 mm2 in control to 23.0 ± 5.2 mm2 on the 3rd

Dig Dis Sci (2011) 56:2792–2801 2795

123


day, and to 27.8 ± 5.7 mm2 on the 7th day in endostatin-

treated rats (P \ 0.05 for both) (Fig. 5a).

Light microscopic histologic examination revealed that

typical cysteamine-induced duodenal lesions were super-

ficial ulcers in control rats whereas deep ulcer craters were

observed in endostatin-treated rats 7 days after cysteamine

administration (Fig. 5b). High-magnification light micros-

copy revealed a sharply demarcated transmucosal necrosis,

i.e., a superficial ulcer in a control rat whereas a deep

perforated or penetrated ulcer was seen in endostatin-

treated rats (Fig. 5c). PAS-stained ulcer sections confirmed

the superficial necrosis in control rats, whereas in endo-

statin-treated rats no mucus-secreting epithelial cells were

observed near the necrotic ulcer crater, which often pene-

trated into the liver or pancreas (Fig. 5d).

Attenuated Duodenal Ulceration and Reduced Levels

of MMP2, Endostatin, and Angiostatin, and Increased

Levels of TIMP-1 by Inhibition of MMP2

Because MMP2, not MMP9, was markedly increased

during duodenal ulceration, we further investigated

Fig. 1 Increased levels of

angiogenic factor VEGF. Levels

of angiogenic factor VEGF

(23 kDa) measured by Western

blot (a) and ELISA (b) in the

duodenal mucosa after

administration of cysteamine or

propionitrile to rats. Ctrl:control. *P \ 0.05

Fig. 2 Increased levels of anti-

angiogenic factors endostatin

and angiostatin. Levels of anti-

angiogenic factors endostatin

(19 kDa) and angiostatin

(50 kDa) measured by Western

blot (a) and ELISA (b) in the



propionitrile to rats. Ctrl:control. *P \ 0.05

2796 Dig Dis Sci (2011) 56:2792–2801

123


inhibition of MMP2, by use of a selective inhibitor, to test

whether MMP2 plays a pathogenic role in generation of

endostatin and angiostatin in duodenal ulceration. The

results showed that duodenal ulcers were significantly

smaller in MMP2 inhibitor-treated rats than in controls

(5.9 ± 1.5 vs. 18.1 ± 5.2 mm2) (Fig. 6a) but were no

different in MMP9 inhibitor-treated animals (results not

shown). Both Western blotting and zymography revealed

that levels of MMP2 expression and proteolytic activity in

the MMP2 inhibitor-treated rats were significantly lower

than in the controls (P \ 0.01) (Fig. 6b), which was fol-

lowed by significantly reduced levels of endostatin and

angiostatin in duodenal ulceration induced by cysteamine

(Fig. 6c). Interestingly, levels of VEGF (Fig. 6d) and

TIMP-1 (Fig. 6e) were significantly increased by inhibition

of MMP2.

Discussion

Angiogenesis is an essential component of wound/ulcer

healing, which is modulated by a balance between angio-

genic and anti-angiogenic factors [15, 19, 20]. In this study

we demonstrated a concurrent increase of angiogenic

VEGF and anti-angiogenic endostatin and angiostatin in

the duodenal mucosa during chemically induced duodenal

Fig. 3 Increased MMP2 and

reduced TIMP-1 in duodenal

mucosa during duodenal

ulceration. Gene expression (a),

protein expression (b),

concentrations of MMP2 and

MMP9 (c), and expression of

endogenous MMP inhibitors

TIMP-1 and TIMP-2 (d) in rat



propionitrile. Ctrl: control.

*P \ 0.05

Fig. 4 Proteolytic activity of

MMP2 and MMP9 in duodenal

ulceration. Up-regulated

enzyme activity of MMP2

(a) but not MMP9 (b) measured

by zymography in rat duodenal

mucosa after administration of

cysteamine or propionitrile.

Ctrl: control. *P \ 0.05

Dig Dis Sci (2011) 56:2792–2801 2797

123


ulceration, indicating, for the first time, an altered balance

between pro- and anti-angiogenic factors in duodenal

ulceration. We also detected a similar interaction between

angiogenic and anti-angiogenic factors in rats with exper-

imental ulcerative colitis [27, 28]. These unexpected find-

ings may actually explain, for the first time, the initially

surprising results demonstrating increased levels of

angiogenic growth factors (e.g., bFGF, PDGF, VEGF) in

the early stages of experimental duodenal ulceration.

Namely, increased local concentrations of angiogenic

stimulators should lead to rapid healing, but we actually

see acute ulcer development. We speculate that the

potentially beneficial effect of VEGF is apparently antag-

onized by the simultaneously elevated levels of endostatin

and angiostatin. Thus, these results indicate an inappro-

priate angiogenic response or ‘‘angiogenic imbalance,’’ and

Fig. 5 Effect of endostatin on

development and healing of

duodenal ulcer induced by

cysteamine. a Aggravation of

duodenal ulcers after endostatin

administration (*P \ 0.05

compared with controls). Light

microscopic analysis of

duodenal ulcers 3 and 7 days

after administration of

endostatin to rats. b Ulcer

craters in a control rat and in an

endostatin-treated rat under low

power view (920). c Control

ulcer and endostatin-treated

ulcer under high power view

(940). d Ulcer areas with PAS

staining under high power view

(940)

Fig. 6 Effect of MMP2

inhibition on cysteamine-

induced duodenal ulcer and

expression of MMP2, TIMP-1,

endostatin, and angiostatin.

Inhibition of MMP2 by

selective MMP2 inhibitor I

attenuated cysteamine-induced

duodenal ulceration (a), reduced

protein expression and

proteolytic activity of MMP2

(b), reduced levels of endostatin

and angiostatin (c), and

increased levels of VEGF

(d) and TIMP-1 (e) in duodenal

mucosa. *P \ 0.05

2798 Dig Dis Sci (2011) 56:2792–2801

123


suggest a novel mechanism of duodenal ulceration. These

studies also demonstrate the value of reproducible animal

models of human diseases that enable easy mechanistic

validation of new molecular biologic results before the

morphologic appearance of organ lesions.

Simultaneous increase of angiogenic and anti-angio-

genic factors in ulceration is a novel finding but the precise

mechanisms are unclear. Several studies have shown that

fluid from venous leg ulcers, particularly those that heal

slowly, contains endostatin and angiostatin which inhibit

in-vitro angiogenesis [29–31]. Drinkwater et al. [32]

demonstrated that the increased levels of anti-angiogenic

factors, e.g., endostatin and angiostatin, were not associ-

ated with VEGF downregulation and, on the contrary, leg

ulcers express elevated levels of VEGF relative to levels

found in normal skin, whereas in tumors the increased anti-

angiogenic activity of endostatin and angiostatin was

through down-regulation of VEGF expression at both

mRNA and protein levels. Ma et al. [33] demonstrated that

gastric ulcer healing is associated with a balance between

VEGF and endostatin released by platelets. A possible

mechanism of increased endostatin and angiostatin in

duodenal ulcer might be that the ulcer environment alters

the expression of various components that could potentially

modify anti-angiogenic activity compared with normally

healing wounds. For example, chronic wounds have altered

expression of the proteoglycan glypican that binds and

antagonizes bFGF in the wound environment [34]. In

addition, elevated levels of several proteinases may

degrade angiogenic factors such as VEGF [35–37].

Biologic confirmation of our molecular biochemical

results came from our pharmacologic experiments in which

markedly aggravated duodenal ulcers induced by cyste-

amine were observed in rats receiving daily injections of

endostatin peptide. Most rats after endostatin administra-

tion had a very extensive necrotic duodenal mucosa with

large and deep ulcer craters that often perforated or pene-

trated, as in patients, adjacent organs such as the liver or

pancreas. Bloch et al. [38] demonstrated that systemic

administration of endostatin impairs blood vessel matura-

tion and delays the healing of full-thickness skin wounds in

mice during wound healing. Other in-vitro studies have

revealed that endostatin induced endothelial cell apoptosis

and inhibited the proliferation and migration of some types

of endothelial cell [39]. It has been suggested that inter-

actions of endostatin with tropomyosin result in disruption

of the integrity of microfilaments and might thereby con-

tribute to the angiogenic effect of endostatin by inhibiting

cell motility [40, 41].

Because angiostatin and endostatin are generated by

proteinases, for example MMP2 and MMP9, through

cleavage of collagen XVIII and plasminogen [22, 23], we

further examined whether MMP2 and MMP9 were

important in the generation of endostatin and angiostatin in

duodenal ulceration. We demonstrated markedly increased

expression of the MMP2 (but not MMP9) gene, which

increased 2.5 to 3.5-fold in duodenal mucosa in duodenal

ulceration induced by cysteamine or propionitrile. The

increased gene expression of MMP2 was followed by

elevated levels of both precursor and active forms of

MMP2 protein in duodenal mucosa after administration of

the ulcerogenic chemicals. We also found that endogenous

MMP2 inhibitor TIMP-1, but not TIMP-2, was reduced in

the duodenal mucosa after administration of the ulcero-

genic chemicals. It has been implied that the ratio between

MMPs and TIMPs is important in wound healing and

TIMPs might participate in the tissue-repair process [26].

Two recent studies have demonstrated that high levels of

MMP2 and low levels of TIMP-1 and TIMP-2 are associ-

ated with non-healing venous leg ulcers [42] and with

gastric ulceration induced by ethanol [43]. Taken together,

these data and our findings indicate another imbalance of

MMP2 and TIMPs in ulceration and impaired healing.

Furthermore, we demonstrated that inhibition of MMP2

resulted in decreased levels of both endostatin and angio-

statin in duodenal mucosa during the ulceration, and duo-

denal ulcers were markedly attenuated by the inhibition of

MMP2. This implied that MMP2 played a mechanistic role

in generation of endostatin and angiostatin during duodenal

ulceration. Recent studies have highlighted the function of

MMPs as negative regulators of angiogenesis by their

release of anti-angiogenic fragments, for example endo-

statin and angiostatin [35, 44–47], and two other studies

have suggested that increased MMP2 may be important in

the process of healing of rat gastric ulcer induced by acetic

acid [48, 49].

In summary, this study demonstrated, for the first time,

that a concurrent increase of the angiogenic factor VEGF

and of anti-angiogenic factors endostatin and angiostatin

occurs in duodenal mucosa during duodenal ulceration

induced by cysteamine or propionitrile. Because the effect

was reproducible after use of two structurally different

chemicals, their action may be ascribed to their duodenal

ulcerogenic property. These findings were confirmed by

ELISA. Daily administration of endostatin confirmed the

biologic effect of endostatin which aggravated the cyste-

amine-induced duodenal ulcers and delayed ulcer healing.

Increased MMP2 levels were associated with over-

production of endostatin and angiostatin in duodenal

ulceration, which was identified by inhibition of MMP2.

We conclude that:

1. early increase of angiostatin and late upregulation of

endostatin, accompanying the increase of VEGF, are

involved in duodenal ulceration;

2. endostatin and angiostatin may play a pathologic role

in impaired and delayed duodenal ulcer healing;

Dig Dis Sci (2011) 56:2792–2801 2799

123


3. MMP2 seems to be of major importance in the

generation of endostatin and angiostatin in duodenal

ulceration; and

4. inhibition of angiogenesis is one of the important

mechanisms of duodenal ulceration and delayed

healing.

Thus, an inappropriate angiogenic response or ‘‘angio-

genic imbalance’’ created by the simultaneous upregulation

of both pro and anti-angiogenic factors seems to be a novel

mechanism of duodenal ulceration.

Acknowledgments This study was supported by a Department of

Veterans Affairs, Veterans Health Administration Merit Review grant

and by contributions from CPRC, Inc.

Conflict of interest No conflicts of interest exist.

References

1. Sonnenberg A, Everhart JE. Health impact of peptic ulcer in the

United States. Am J Gastroenterol. 1997;92:614–620.

2. Bytzer P, Teglbjaerg PS. Danish Ulcer Study Group. Helico-bacter pylori-negative duodenal ulcers: prevalence, clinical

characteristics, and prognosis—results from a randomized trial

with 2-year follow-up. Am J Gastroenterol. 2001;96:1409–1416.

3. Kong SX, Hatoum HT, Zhao SZ, Agrawal NM, Geis SG. Prev-

alence and cost of hospitalization for gastrointestinal complica-

tions related to peptic ulcers with bleeding or perforation:

comparison of two national databases. Am J Manag Care.

1998;4:399–409.

4. Manuel D, Cutler A, Goldstein J, Fennerty MB, Brown K.

Decreasing prevalence combined with increasing eradication of

Helicobacter pylori infection in the United States has not resulted

in fewer hospital admissions for peptic ulcer disease-related

complications. Aliment Pharmacol Ther. 2007;25:1423–1427.

5. Chan FK, Leung WK. Peptic-ulcer disease. Lancet. 2002;360:

933–941.

6. Jayaraj AP, Tovey FI, Lewin MR, Clark CG. Duodenal ulcer

prevalence: experimental evidence for the possible role of dietary

lipids. J Gastroenterol Hepatol. 2000;15:610–616.

7. Selye H, Szabo S. Experimental model for production of perfo-

rating duodenal ulcers by cysteamine in the rat. Nature. 1973;244:

458–459.

8. Szabo S, Reynolds ES, Unger SH. Structure-activity relations

between alkyl nucleophilic chemicals causing duodenal ulcer and

adrenocortical necrosis. J Pharmacol Exp Ther. 1982;223:68–76.

9. Fukuhara S, Suzuki H, Masaoka T, et al. Enhanced ghrelin

secretion in rats with cysteamine-induced duodenal ulcers. Am JPhysiol Gastrointest Liver Physiol. 2005;289:G138–G145.

10. Szabo S, Shing Y, Folkman MJ, et al. Angiogenesis and growth

factors in ulcer healing. In: Fan Tai-Ping D, Kohn EC, eds. NewAngiogenesis. Totowa, NJ: Humana Press, 2001;119–211.

11. Folkman J. Angiogenesis. Annu Rev Med. 2006;57:1–18.

12. Pandya NM, Dhalla NS, Santani DD. Angiogenesis—a new tar-

get for future therapy. Vascul Pharmacol. 2006;44:265–274.

13. Vincze A, Nagata M, Sandor Zs, Szabo S. ELISA and western

blot studies with basic fibroblast growth factor (bFGF) and

platelet-derived growth factor (PDGF) in experimental duodenal

ulceration and healing. Inflammopharmacology. 1996;4:261–265.

14. Szabo S, Vincze A, Sandor Zs, et al. Vascular approach to gas-

troduodenal ulceration: new studies with endothelins and VEGF.

Dig Dis Sci. 1998;43(9 Suppl):40S–45S.

15. Folkman J, Szabo S, Stovroff M, McNeil P, Li W, Shing Y.

Duodenal ulcer: discovery of a new mechanism and development

of angiogenic therapy that accelerates healing. Ann Surg.

1991;214:414–426.

16. Szabo S, Kusstatscher S, Sakoulas G, Sandor Z, Vincze A, Jadus

M. Growth factors: new ‘endogenous drugs’ for ulcer healing.

Scand J Gastroenterol. 1995;30:15–18.

17. Deng X, Szabo S, Khomenko T, Jadus MR, Yoshida M. Gene

therapy with naked DNA or adenoviral vector of VEGF or PDGF

increases endogenous VEGF, PDGF and bFGF expression and

accelerates chronic duodenal ulcer healing in rats. J PharmacolExp Ther. 2004;311:982–988.

18. Szabo S, Vincze A. Growth factors in ulcer healing: lessons from

recent studies. J Physiol (Paris). 2000;94:77–81.

19. Hanahan D, Folkman J. Patterns and emerging mechanisms of the

angiogenic switch during tumorigenesis. Cell. 1996;86:353–364.

20. O’Reilly MS, Holmgren L, Shing Y, et al. Angiostatin: a novel

angiogenesis inhibitor that mediates the suppression of metasta-

ses by a Lewis lung carcinoma. Cell. 1994;79:315–328.

21. O’Reilly MS, Boehm T, Shing Y, et al. Endostatin: an endoge-

nous inhibitor of angiogenesis and tumor growth. Cell. 1997;88:

277–285.

22. Distler JH, Hirth A, Kurowska-Stolarska M, Gay RE, Gay S,

Distler O. Angiogenic and angiostatic factors in the molecular

control of angiogenesis. Q J Nucl Med. 2003;47:149–161.

23. Heljasvaara R, Nyberg P, Luostarinen J, et al. Generation of

biologically active endostatin fragments from human collagen

XVIII by distinct matrix metalloproteases. Exp Cell Res. 2005;

307:292–304.

24. Claesson-Welsh L, Welsh M, Ito N, et al. Angiostatin induces

endothelial cell apoptosis and activation of focal adhesion kinase

independently of the integrin-binding motif RGD. Proc NatlAcad Sci USA. 1998;95:5579–5583.

25. Taddei L, Chiarugi P, Brogelli L, et al. Inhibitory effect of full-

length human endostatin on in vitro angiogenesis. Biochem Bio-phys Res Commun. 1999;263:340–345.

26. Rundhaug JE. Matrix metalloproteinases and angiogenesis. J CellMol Med. 2005;9:267–285.

27. Sandor Z, Deng XM, Khomenko T, Tarnawski AS, Szabo S.

Altered angiogenic balance in ulcerative colitis: a key to impaired

healing? Biochem Biophys Res Commun. 2006;350:147–150.

28. Deng X, Tolstanova G, Khomenko T, et al. Mesalamine restores

angiogenic balance in experimental ulcerative colitis by reducing

expression of endostatin and angiostatin: novel molecular

mechanism for therapeutic action of mesalamine. J PharmacolExp Ther. 2009;331:1071–1078.

29. Smith E, Hoffman R. Multiple fragments related to angiostatin

and endostatin in fluid from venous leg ulcers. Wound Rep Reg.

2005;13:148–157.

30. Bucalo B, Eaglstein WH, Falanga V. Inhibition of cell prolifer-

ation by chronic wound fluid. Wound Rep Reg. 1993;1:181–186.

31. Drinkwater SL, Smith A, Sawyer BM, Burnand KG. Effect of venous

ulcer exudates on angiogenesis in vitro. Br J Surg. 2002;89:709–713.

32. Drinkwater SL, Burnand KG, Ding R, Smith A. Increased but

ineffectual angiogenic drive in non-healing venous leg ulcers.

J Vasc Surg. 2003;38:1106–1112.

33. Ma L, Elliott SN, Cirino G, Buret A, Ignarro LJ, Wallace JL.

Platelets modulate gastric ulcer healing: role of endostatin and

vascular endothelial growth factor release. Proc Natl Acad SciUSA. 2001;98:6470–6475.

34. Kato M, Wang H, Kainulainen V, et al. Physiological degradation

converts the soluble syndecan-1 ectodomain from an inhibitor to

a potent activator of FGF-2. Nat Med. 1998;4:691–697.

2800 Dig Dis Sci (2011) 56:2792–2801

123


35. Sottile J. Regulation of angiogenesis by extracellular matrix.

Biochim Biophys Acta. 2004;1654:13–22.

36. Weckroth M, Vaheri A, Lauharanta J, Sorsa T, Konttinen YT.

Matrix metalloproteinases, gelatinase and collagenase, in chronic

leg ulcers. J Invest Dermatol. 1996;106:1119–1124.

37. Lauer G, Sollberg S, Cole M, et al. Expression and proteolysis of

vascular endothelial growth factor is increased in chronic

wounds. J Invest Dermatol. 2000;115:12–18.

38. Bloch W, Huggel K, Sasaki T, et al. The angiogenesis inhibitor

endostatin impairs blood vessel maturation during wound heal-

ing. FASEB J. 2000;14:2373–2376.

39. Boehm T, Folkman J, Browder T, O’Reilly MS. Antiangiogenic

therapy of experimental cancer does not induce acquired drug

resistance. Nature. 1997;390:404–407.

40. Yamaguchi N, Anand-Apte B, Lee M, et al. Endostatin inhibits

VEGF-induced endothelial cell migration and tumor growth

independently of zinc binding. EMBO J. 1999;18:4414–4423.

41. McDonald NJ, Shivers WY, Narum DL, et al. Endostatin binds

tropomyosin: a potential modulator of the antitumor activity of

endostatin. J Biol Chem. 2001;276:25190–25196.

42. Mwaura B, Mahendran B, Hynes N, et al. The impact of differ-

ential expression of extracellular matrix metalloproteinase indu-

cer, matrix metalloproteinase-2, tissue inhibitor of matrix

metalloproteinase-2 and PDGF-AA on the chronicity of venous

leg ulcers. Eur J Vasc Endovasc Surg. 2006;31:306–310.

43. Swarnakar S, Mishra A, Ganguly K, Sharma AV. Matrix

metalloproteinase-9 activity and expression is reduced by mela-

tonin during prevention of ethanol-induced gastric ulcer in mice.

J Pineal Res. 2007;43:56–64.

44. Mott JD, Werb Z. Regulation of matrix biology by matrix

metalloproteinases. Curr Opin Cell Biol. 2004;16:558–564.

45. Cornelius LA, Nehring LC, Harding E, et al. Matrix metallo-

proteinases generate angiostatin: effects on neovascularization.

J Immunol. 1998;161:6845–6852.

46. Ferreras M, Felbor U, Lenhard T, Olsen BR, Delaisse J. Gener-

ation and degradation of human endostatin proteins by various

proteinases. FEBS Lett. 2000;486:247–251.

47. Nilsson UW, Dabrosin C. Estradiol and tamoxifen regulate

endostatin generation via matrix metalloproteinase activity in

breast cancer in vivo. Cancer Res. 2006;66:4789–4794.

48. Ito A, Nagase H, Mori Y. Characterization of metalloproteinases

in rat gastric tissues with acetic acid-induced ulcers. Scand JGastroenterol Suppl. 1989;162:146–149.

49. Kobayashi S, Nakajima N, Ito Y, Moriyama M. Effects of lan-

soprazole on the expression of VEGF and cellular proliferation in

a rat model of acetic acid-induced gastric ulcer. J Gastroenterol.2010;45:846–858.

Dig Dis Sci (2011) 56:2792–2801 2801

123


Documents

Inappropriate Angiogenic Response as a Novel Mechanism of Duodenal Ulceration and Impaired Healing