ww.sciencedirect.com

j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 4 ( 2 0 1 3 ) 1 0 1e1 0 7

Available online at w

journal homepage: www.JournalofSurgicalResearch.com

Association for Academic Surgery

Vascular endothelial growth factor receptor-2 inhibitionin experimental murine colitis

Leslie Knod, MD, Eileen C. Donovan, BS, Artur Chernoguz, MD, Kelly M. Crawford, BS,Mary R. Dusing, BS, and Jason S. Frischer, MD*

Division of Pediatric General and Thoracic Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio

a r t i c l e i n f o

Article history:

Received 4 January 2013

Received in revised form

7 April 2013

Accepted 15 April 2013

Available online 6 May 2013

Keywords:

Inflammatory bowel disease

Angiogenesis

VEGFR2

Colitis

* Corresponding author. Colorectal CenterHospital Medical Center, 3333 Burnett Avenu

E-mail address: Jason.Frischer@cchmc.or0022-4804/$ e see front matter ª 2013 Elsevhttp://dx.doi.org/10.1016/j.jss.2013.04.026

a b s t r a c t

Background: In the setting of inflammatory bowel disease, inflammation is associated with

a simultaneous increase in angiogenesis; moreover, elevated vascular endothelial growth

factor (VEGF) levels implicate angiogenesis as a pathologic contributor to disease severity.We

hypothesize that selectively inhibiting vascular endothelial growth factor receptor-2 (VEGFR2)

in amodel ofmurine colitiswill reduceangiogenesis, resulting indecreased inflammationand

disease severity, providingmechanistic insight into the role of pathologic angiogenesis in IBD.

Materials and methods: In a dextran sodium sulfate model of murine colitis, anti-VEGFR2

monoclonal antibody (DC101) or placebo was administered. Clinical assessments fol-

lowed by histologic and molecular tissue analysis were performed to quantify inflamma-

tion, microvessel density (MVD), VEGF and VEGFR2 gene expression, and phosphorylated

mitogen-activated protein kinase protein expression.

Results: Weight loss began after d 6 with the treatment group demonstrating a more

favorable percent weight change. Inflammation and MVD were similar between cohorts,

both increasing in parallel toward a plateau. VEGF and VEGFR2 messenger RNA expression

were not significantly different, but phosphorylated mitogen-activated protein kinase was

elevated in the DC101 cohort (P ¼ 0.03).

Conclusions: Despite a more favorable weight change profile in the treated group, no differ-

ence was observed between cohorts regarding clinical disease severity. However, a parallel

rise in inflammation and MVD was observed coinciding with weight loss, suggesting their

relationship in IBD. VEGFR2 downstream signaling was significantly elevated in the treated

cohort, possibly by VEGF-independent signal transduction. Early and effective inhibition of

angiogenesis by limiting downstream VEGF signaling or targeting multiple angiogenic

pathways may block angiogenesis, thereby reducing disease severity and provide evidence

toward the mechanism and clinical benefit of antiangiogenics in the setting of IBD.

ª 2013 Elsevier Inc. All rights reserved.

for Children, Division of Pediatric General and Thoracic Surgery, Cincinnati Children’se, MLC-2023, Cincinnati, OH 45229. Tel.: þ1 513 636 3240; fax þ1 513 636 3248.g (J.S. Frischer).ier Inc. All rights reserved.

j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 4 ( 2 0 1 3 ) 1 0 1e1 0 7102

1. Introduction resulting in decreased inflammation and disease severity,

Ulcerative colitis and Crohn’s disease, the two manifestations

of inflammatory bowel disease (IBD), result from a complex

interplay of genetic, immunologic, and environmental factors

[1]. Recent human and animal studies have documented

increased angiogenesis as a pathophysiologic characteristic of

IBD. Elevated vascular endothelial growth factor (VEGF),

a primary angiogenic regulator, in clinical and experimental

IBD is demonstrated in systemic circulation and at the level of

genetic and protein expression, implicating the VEGF pathway

as a regulator of angiogenesis [2e7]. Angiogenesis serves as

a potential venue for molecular pharmacologic control to

decrease disease severity in IBD, similar to the antiangiogenic

benefits demonstrated in cancer management.

Angiogenesis, the formation of new vessels from preexist-

ing vasculature, is essential for physiologic growth and

development, but in disease, angiogenesis becomes pathologic

and a potential therapeutic target, as demonstrated in onco-

logic and chronic inflammatory processes [8e10]. In the field of

cancer therapy, antiangiogenics have been shown to decrease

tumor growth and has become part of many treatment regi-

mens for solid tumors [11,12]. Diseases of chronic inflamma-

tion, including rheumatoid arthritis, atherosclerosis, psoriasis,

and diabetic retinopathy [8,12e14], have upregulated angio-

genesis, which supports a relationship between angiogenesis

and inflammation.

In the VEGF ligandereceptor signaling pathway, the VEGF-

A mitogen is the strongest signal transducer and binds to the

endothelial cell tyrosine kinase receptor, vascular endothelial

growth factor receptor-2 (VEGFR2/Flk-1). VEGFR2 mediates

vascular endothelial cell growth and permeability including

DNA replication, cell proliferation, survival, and migration

[15e17]. Other receptors are less central to angiogenesis,

VEGFR1 exhibits weak tyrosine kinase activity, acting more

like a decoy receptor, and VEGFR3 is predominantly expressed

in lymphatic endothelium [18,19]. The major angiogenic

effects of VEGF are mediated through VEGFR2, making it an

ideal target to inhibit angiogenesis.

Decreased angiogenesis can result from inhibition of

the VEGF receptor or ligand. Selective receptor inhibition

of VEGFR2/Flk-1 by DC101 (anti-Flk-1 monoclonal antibody)

decreases tumor angiogenesis in vitro and in vivo and has

demonstrated tumor growth inhibition [20,21]. Neutralization

of the VEGF ligand with monoclonal antibody in murine IBD

has also shown a reduction in disease severity and inflam-

matory cell infiltrate [22]. Our model of murine colitis

is induced by dextran sodium sulfate (DSS), producing a che-

mically toxic disruption of colonic epithelial cells causing

a mucosal barrier dysfunction that allows bacterial trans-

location leading to an intense inflammatory response [23].

Inflammation and pathologic angiogenesis are likely in-

terrelated processes, possibly with bidirectional cell signaling

from one exacerbating the other [8,24]. Their independent

degree of contribution to disease severity is unknown,making

the angiogenic pathway an attractive target to elucidate the

complex pathophysiology of IBD. We hypothesize that selec-

tive VEGFR2 inhibition by systemic DC101 administration in

a DSS model of murine colitis will reduce angiogenesis,

which will better define the mechanism of pathologic angio-

genesis and its role in IBD.

2. Materials and methods

2.1. Murine colitis model

Experimental colitis was induced in 6- to 8-wk-old male

C57BL/6 mice (Harlan Laboratories, Indianapolis, IN) with a 7-

d course of 1.0%DSS (w/v) supplemented in ad libitum drinking

water. Colitis signs and symptoms were monitored daily as

weights were recorded. Mice were singly housed in the barrier

containment animal facility, and all experiments were per-

formed in accordance with the Institutional Animal Care and

Use Committee (IACUC) of Cincinnati Children’s Research

Foundation.

2.2. Administration of DC101

Subjects were divided into control or DC101-treated cohorts

(n ¼ 18 per group). Commencing on d 3 following the initiation

of DSS, mice in the treated cohort received intraperitoneal

injections of the VEGFR2 monoclonal antibody, DC101 (32

mg/kg in 0.1 mL), every 3 d, whereas control mice received

equivalent injections of phosphate-buffered saline (PBS) until

completion of the experiment on d 14. Optimal dosing was

established to be 400e800 mg per mouse based on published

data that demonstrated a maximal therapeutic response at

800 mg per mouse in C57BL/6 mice [20,25] and in colorectal

tumor xenografts [26]. Animals were killed at 2-d intervals

following the initiation of DSS. At necropsy, dissected colonic

specimens were irrigated with PBS and then longitudinally

subdivided and processed as either fixed or frozen tissues.

Fixed tissues were preserved in 4% paraformaldehyde over-

night followed by serial alcohol dehydrations and then

paraffin embedding for histologic evaluation.

2.3. Histopathologic colitis scoring

Colons were stained with hematoxylin and eosin and then

analyzed under light microscopy at �40 magnification scoring

four consecutive fields, starting at the anorectal junction.

Using an accepted colitis scoring method [27e29] with

amaximumpossible score of 18, two blinded observers graded

the tissues in the following categories: (1) percent area inv-

olved, (2) edema, (3) ulcerations, (4) crypt loss, and (5) immune

cell infiltration to generate a cumulative inflammation score.

2.4. Immunohistochemistry

Paraffin embedded colons were cut into 5 mm sections and

mounted. Following deparaffinization and rehydration,

citrate-based epitope enhancement was performed using

Antigen Retrieval Solution (DAKO, Carpinteria, CA). After a PBS

wash, Protein Blocker Solution (DAKO) was applied by the

manufacturer’s instructions, and slides were incubated with

MECA-32 primary monoclonal antibody (1:10; Developmental

j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 4 ( 2 0 1 3 ) 1 0 1e1 0 7 103

Studies Hybridoma Bank, Iowa City, IA). Alexa 488-tagged

fluorescent secondary antibody (1:200; Invitrogen, Carlsbad,

CA) was applied and followed by mounting with DAPI Vecta-

shield medium (Vector Laboratories, Burlingame, CA).

2.5. Microvessel density

Intestinal sections were immunostained with MECA-32 and

microphotographed for 10 consecutive, �200 fields starting

at the anorectal junction using fluorescent microscopy (Nikon

90i; Nikon Instruments, Inc, Melville, NY). In a blinded

manner, the area (mm2) of mucosal microvessel density (MVD)

per high-powered field was quantitated with NIS-Advanced

Elements software (Nikon Instruments, Inc) to determine the

mean microvascular area for each specimen.

2.6. Colonic VEGFR2 and VEGF-A genetic expressions

VEGF-A and VEGFR2 gene expressions were analyzed from

specimens stored in RNA later (Sigma Life Science, St. Louis,

MO). The tissues were homogenized followed by RNA extrac-

ted using the RNeasyMini Kit (Qiagen, Valencia, CA) according

to the manufacturer’s instruction. RNA was converted to

complementary DNA using the high-capacity complementary

DNA reverse transcription kit (Applied Biosystems, Carlsbad,

CA). Quantitative reverse transcriptionepolymerase chain

reaction was performed by StepOne Plus RT-PCR System

(Applied Biosystems) using SYBR Green detection with b-2-

microglobulin (RealTimePrimers.com; Elkins Park, Phila-

delphia, PA) as the internal reference gene. PCR primers

were used as follows: VEGF-A forward 50-TCTGCTCTCCTT-CTGTCGTG-30 and reverse 50-ACTGGACCCTGGCTTTACTG-30

and VEGFR2 forward 50-CGTTGTACAAATGTGAAGC-30 and

reverse 50-CACAGTAATTTCAGGACCC-30 (Integrated DNA

Technologies, Coralville, IA). Relative quantification of data

by the 2�DDCt method produced results as fold change in

expression of target gene relative to internal control.

Fig. 1 e Percent daily weight change was more favorable in

the DC101 treated cohort compared with controls. Both

groups exhibited a similar chronologic pattern of weight

change, declining after d 6 and reaching a nadir at d 10.

2.7. Phosphorylated mitogen-activated protein kinaseprotein expression

Tissues were homogenized in a buffer containing Halt

Protease and Phosphate Inhibitor Cocktail (Thermo Scientific,

Rockford, IL) and Nonidet-P40 substitute in PBS. Protein

concentrations were generated from a standard curve using

the Pierce BCA Protein Assay Kit (Thermo Scientific). Protein

extracts (10 or 30 mg per lane) were loaded with buffer and run

on a 12% TGX gel (Bio-Rad, Hercules, CA) for sodium dodecyl

sulfateepolyacrylamide gel electrophoresis and transferred

onto a nitrocellulose membrane. After application of

nonspecific blocker, the membrane was incubated with

primary antibody, either mitogen-activated protein kinase

(1:1000; MAPK) or phosphorylated MAPK (1:2000; pMAPK)

both from Cell Signaling Technology, Danvers, MA. The

secondary antibody (1:1000, IgG horseradish peroxidase-

linked; Cell Signaling Technology) was applied followed by

LumiGLO application. Densitometric analysis was performed

with ImageJ software (National Institutes of Health, Beth-

esda, MD) and normalized to glyceraldehyde-3-phosphate

dehydrogenase (1:200; Santa Cruz Biotechnology, Santa

Cruz, CA).

2.8. Statistical analysis

Statistical analysis was performed using GraphPad Prism

software (GraphPad Software, Inc, La Jolla, CA). Unpaired t-

test was used to compare bodyweights, inflammation scoring,

MVD, gene expression, and protein levels expressed as mean

� standard error of the mean, in which P � 0.05 represents

statistical significance. The rate of weight change was quan-

tified by linear regression analysis with a 95% confidence

interval (CI).

3. Results

3.1. Clinical assessment

Mice in the treated and control groups developed clinical signs

and symptoms of colitis induced by DSS including weight loss,

ruffled fur, indolence, diarrhea, and bloody stools. After d 6

of DSS administration, both groups began to lose weight. Daily

weights were recorded and percent weight change was cal-

culated based on the initial weight of each mouse. Starting

weights ranged from 20 to 25 g per mouse with amean weight

of 23 g in both cohorts. The weight change of four mice, fol-

lowed daily to d 14, was used to establishweight trends (Fig. 1).

The DC101-treated group exhibited an overall less severe

percent weight loss than controls (control �14.82% � 2.63%

versus treated �11.35% � 2.6%, P ¼ 0.39 at nadir, d 10). The

treated group also had a more clinically favorable rate of

weight loss compared with controls (slope of percent weight

loss in control �4.76 [95% CI �6.35 to �3.16] versus treated

group �2.99 [95% CI �4.17 to �1.8], P ¼ 0.07). A quicker return

toward initial weight in the DC101 group (slope of percent

weight gain in control þ1.23 [95% CI �0.46 to 2.92] versus

treated group þ2.7 [95% CI 1.37 to 4.04], P < 0.16) with a mean

Fig. 2 e Colonic inflammation scoring was similar between

treated and untreated groups, both rising after d 6 and

reaching a plateau at d 8.

j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 4 ( 2 0 1 3 ) 1 0 1e1 0 7104

percent weight change at d 14 of �10.86% � 3.19% control

versus �1.33% � 3.07% treated, P ¼ 0.07.

3.2. Colonic inflammation scoring and MVD

Colonic inflammation markedly increased after d 6 with

a similar trend in both DC101 and control groups, reaching

a plateau that continued through the end of the study (Figs. 2

and 3). Both groups demonstrated peak colonic inflammation

occurring at d 10, with amean inflammation score of 17� 0.01

Fig. 3 e Colons stained with hematoxylin and eosin viewed at

controls (A) and DC101 treated (C) colons, compared with DSS d

inflammatory infiltrate, crypt loss, edema, and ulceration but s

(D) groups. (Color version of figure is available online.)

(n ¼ 2 per group). No statistical difference exists between

treated and untreated groups.

Quantitative MVD increased from baseline after d 6 in

treated and untreated groups (Figs. 4 and 5). Both groups

exhibit elevated MVD between d 8e14 (control 7013 � 599 mm2

versus treated 7865 � 703 mm2, P ¼ 0.39, n ¼ 10 per group). The

apparent MVD difference between groups noted at d 10

(control 2914� 909 mm2 versus treated 8408� 608 mm2, n¼ 2 per

group) did not reach statistical significance (P ¼ 0.09) nor did

this separation in MVD between cohorts persist, as demon-

strated by similar d 14 MVD (control 6620 � 372 mm2 versus

treated 6265 � 358 mm2, P ¼ 0.52, n ¼ 4 per group).

3.3. Tissue genetic and protein analysis

Local colonic VEGF messenger RNA relative expression was

similar between treated and untreated groups at early time

points, before d 6, (8.70� 4.46 control versus 8.13� 2.05 DC101,

P ¼ 0.91, n ¼ 5 per group), but at later time points, after d 6,

VEGF suppression was observed in the DC101 group compared

with controls (7.94 � 5.68 control versus 2.68 � 0.67 DC101, P ¼0.44, n ¼ 4e5 per group), although not statistically significant

(Fig. 6). Between DC101 early and late groups, a decrease in

VEGF gene expression was noted but not significant (8.13 �2.05 early DC101 versus 2.68� 0.67 late DC101, P¼ 0.06, n¼ 4e5

per group). Relative expression of VEGFR2 messenger RNA in

control and DC101 colons was not significantly different at

early time points, before d 6, (3.61 � 0.83 control versus 2.42 �0.67 DC101, P ¼ 0.28, n ¼ 8 per group) or at later time points,

after d 6, (1.39 � 0.17 control versus 1.34 � 0.14 DC101, P ¼ 0.84,

n ¼ 8e10 per group).

340 magnification show limited inflammation in early

10 colons that were more inflamed, exhibiting worse

till appear similar between late control (B) and treated

Fig. 4 e Colonic MVD was similar between both cohorts.

j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 4 ( 2 0 1 3 ) 1 0 1e1 0 7 105

Downstream signaling of VEGFR2 was measured from DSS

initiation to d 8 bywestern blot analysis expressed as a ratio of

phosphorylated to total MAPK protein (Fig. 7). The percent of

phosphorylated MAPK was higher in the DC101-treated group

(22.5 � 6.4 untreated versus 57.6 � 13.5 treated, P ¼ 0.03, n ¼ 8

per group).

4. Discussion

Murine DSS colitis is a widely used animal model of experi-

mental IBD that produces clinical signs and symptoms similar

to humans including diarrhea, occult blood and weight loss as

Fig. 5 e Colonic MVD visualized with MECA-32 staining at 3200

(A) and DC101 treated (C) groups, but an appreciable increase in

DSS d 10 both control (B) and treated (D) colons. (Color version

well as similar histologic disease exhibiting ulceration,

immune cell infiltrate, and crypt abscesses [30,31]. When

applied to C57BL/6 mice, DSS colitis responds to conventional

IBD pharmacologic intervention demonstrating improve-

ments at the clinical, histologic, and molecular levels, which

demonstrates that this model is translatable into human

disease management [32]. In human subjects with IBD,

increased vascularity is noted in the acutely inflamed bowel,

as such angiogenesis is associated with the pathophysiology

of IBD [2]. The precise role of angiogenesis whether leading to

or exacerbating the inflammatory process remains unknown.

The principal angiogenic factor (VEGF) has its most potent

receptor (VEGFR2) predominantly expressed on vascular

endothelial cells [16,18]. Although no causal relationship is

defined between VEGF and disease activity, expression of

VEGF is elevated in IBD patients [7], and therefore, the VEGF/

VEGFR pathway is a potential target for molecular therapy.

In our model of acute colitis, the VEGFR2 monoclonal

antibodyetreated group exhibited an improved weight

change profile compared with controls, but no difference in

disease severity was present between groups as measured by

inflammation score or MVD. A temporal relationship was

noticed between angiogenesis and inflammation, with both

rising after d 6 of DSS administration, suggesting a linked

association. This parallel rise in vascularity and inflamma-

tion has previously been observed in experimental murine

colitis [33,34]. As weight loss commenced, a simultaneous

rise in histopathologic colitis ensued. Weight loss reached

a nadir by d 10, which corresponds to peak angiogenesis and

inflammation scores, identifying the period of most severe

clinical and histologic colitis. These findings show the

complex interplay that likely exists between angiogenesis

magnification showing similarities between early control

MVD when comparing early to late time points as seen in

of figure is available online.)

Fig. 6 e Local tissue gene expression of VEGF-A was

initially similar between control and DC101, but at later

time points (after d 6), the DC101-treated group

demonstrated decreased VEGF-A expression, but not

significant, when compared with either control groups or

early DC101.

j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 4 ( 2 0 1 3 ) 1 0 1e1 0 7106

and inflammation coinciding with the progression of disease

severity.

By applying DC101 to our DSS model of acute colitis,

we anticipated a decrease in VEGFR2 function by way of

competitive inhibition at the extracellular ligand-binding

domain, which would prevent receptor tyrosine kinasee

mediated signal transduction and thereby dampen the down-

stream angiogenic effect [35,36]. The expected angiogenesis

suppression was not achieved, without significant difference

in MVD between cohorts. Previous studies in our laboratory

demonstrated that VEGFR2 blockade with DC101 caused

elevated VEGF protein expression in wild-type colons

compared with placebo (E.C. Donovan, BS, A. Chernoguz, MD,

K.M. Crawford, BS, M.R. Dusing, BS, J.S. Frischer, MD, unpub-

lished data, 2011). Others have observed that following

Fig. 7 e Receptor tyrosine kinase activity measured by

phosphorylated MAPK (pMAPK) was elevated in the DC101

group (0.57 ± 0.14) compared with controls (0.22 ± 0.06); *P

[ 0.03.

administration of VEGFR2 monoclonal antibody, plasma levels

of VEGF increase suggesting that VEGF levels can serve as

a surrogate marker of effective dosing [26]. However, in this

experiment, gene expression of VEGF was unchanged at early

time points but diminished, though not significantly, in the

DC101 group at later time points. Given these unexpected

results and the clinical similarities between cohorts, additional

angiogenic pathway analysis was required to assess VEGFR2

downstream signaling.

To further evaluate effective receptor blockade, the PLC-

gamma/ERK pathway that leads to endothelial cell prolifera-

tion was assessed [18]. Phosphorylated MAPK expression was

greater in the treated group, giving us a paradoxical response

from the VEGFR2 blockade. This could explain the similarities

between treated and untreated mice regarding colonic angio-

genesis and inflammation. Alternatively, despite blocking

extracellular ligand binding, the downstream signaling of the

VEGF receptor can remain active by VEGF-independent path-

ways [37]. Other plausible explanations include that receptor

blockade may upregulate compensatory processes or alterna-

tive pathways such as platelet-derived growth factor,

epidermal growth factor, or fibroblast growth factor. IBD is

a heterogenous disease with many phenotypic presentations;

certainly, this variability is present among experimental

models of colitis. Angiogenesis upregulation can result from

increased proangiogenic mediators or loss of angiogenic inhi-

bition. Chidlow et al. [5] suggests that the pathologic angio-

genesis in the DSS model of IBD may result from loss of

angiogenic inhibition. Although DC101 was ineffective in

suppressing angiogenesis by means of VEGFR2 blockade, this

receptor remains a potential target given itswell-known role in

physiologic and pathologic angiogenesis. Perhaps combination

treatment targetingmultiple angiogenic pathways or restoring

antiangiogenic activity may be required to decipher the path-

ways contributing to IBD angiogenesis. This is certainly the

case in the oncologic antiangiogenic therapy paradigm [38].

The onset of inflammation and angiogenesis in the initial

disease phase of experimental colitis provides a therapeutic

window in which angiogenic suppression may exude its effect

on inflammation and overall disease severity. Multiple agents

are currently in use for treating the inflammatory components

of IBD [39], but no clinical studies have been designed based

on the contribution of angiogenesis in the disease process.

Treatment with the immunemodulator infliximab, a mono-

clonal antibody to tumornecrosis factor-alpha, has beenshown

to reduce vascularity in IBD but has unfavorable side effects

[39,40]. In future experiments, early and effective inhibition of

angiogenesis or inhibition of downstreamsignalingmay reduce

disease severity and provide evidenced toward the clinical

benefit of antiangiogenics in the setting of IBD.

r e f e r e n c e s

[1] Abraham C, Cho JH. Inflammatory bowel disease. N Engl JMed 2009;361:2066.

[2] Danese S, Sans M, de la Motte C, et al. Angiogenesis asa novel component of inflammatory bowel diseasepathogenesis. Gastroenterology 2006;130:2060.

j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 4 ( 2 0 1 3 ) 1 0 1e1 0 7 107

[3] Pousa ID, Mate J, Gisbert JP. Angiogenesis in inflammatorybowel disease. Eur J Clin Invest 2008;38:73.

[4] Spalinger J, Patriquin H, Miron MC, et al. Doppler US inpatients with crohn disease: vessel density in the diseasedbowel reflects disease activity. Radiology 2000;217:787.

[5] Chidlow JH Jr, Langston W, Greer JJ, et al. Differentialangiogenic regulation of experimental colitis. Am J Pathol2006;169:2014.

[6] Bousvaros A, Leichtner A, Zurakowski D, et al. Elevatedserum vascular endothelial growth factor in children andyoung adults with Crohn’s disease. Dig Dis Sci 1999;44:424.

[7] Scaldaferri F, Vetrano S, Sans M, et al. VEGF-A linksangiogenesis and inflammation in inflammatory boweldisease pathogenesis. Gastroenterology 2009;136:585.

[8] Carmeliet P. Angiogenesis in health and disease. Nat Med2003;9:653.

[9] Folkman J, Szabo S, Stovroff M, McNeil P, Li W, Shing Y.Duodenal ulcer. Discovery of a new mechanism anddevelopment of angiogenic therapy that accelerates healing.Ann Surg 1991;214:414. discussion 426e7.

[10] Tjwa M, Luttun A, Autiero M, Carmeliet P. VEGF and PlGF:two pleiotropic growth factors with distinct roles indevelopment and homeostasis. Cell Tissue Res 2003;314:5.

[11] Folkman J. Tumor angiogenesis: therapeutic implications.N Engl J Med 1971;285:1182.

[12] Folkman J. Angiogenesis in cancer, vascular, rheumatoid andother disease. Nat Med 1995;1:27.

[13] Sone H, Sakauchi M, Takahashi A, et al. Elevated levels ofvascular endothelial growth factor in the sera of patientswith rheumatoid arthritis correlation with disease activity.Life Sci 2001;69:1861.

[14] Creamer D, Allen M, Jaggar R, Stevens R, Bicknell R, Barker J.Mediation of systemic vascular hyperpermeability in severepsoriasis by circulating vascular endothelial growth factor.Arch Dermatol 2002;138:791.

[15] Ho QT, Kuo CJ. Vascular endothelial growth factor: biologyand therapeutic applications. Int J Biochem Cell Biol 2007;39:1349.

[16] Gille H, Kowalski J, Li B, et al. Analysis of biological effectsand signaling properties of Flt-1 (VEGFR-1) and KDR(VEGFR-2). A reassessment using novel receptor-specificvascular endothelial growth factor mutants. J Biol Chem2001;276:3222.

[17] Cross MJ, Dixelius J, Matsumoto T, Claesson-Welsh L.VEGF-receptor signal transduction. Trends Biochem Sci 2003;28:488.

[18] Koch S, Tugues S, Li X, Gualandi L, Claesson-Welsh L. Signaltransduction by vascular endothelial growth factorreceptors. Biochem J 2011;437:169.

[19] Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and itsreceptors. Nat Med 2003;9:669.

[20] Prewett M, Huber J, Li Y, et al. Antivascular endothelialgrowth factor receptor (fetal liver kinase 1) monoclonalantibody inhibits tumor angiogenesis and growth of severalmouse and human tumors. Cancer Res 1999;59:5209.

[21] Skobe M, Rockwell P, Goldstein N, Vosseler S, Fusenig NE.Halting angiogenesis suppresses carcinoma cell invasion.Nat Med 1997;3:1222.

[22] Tolstanova G, Khomenko T, Deng X, et al. Neutralizing anti-vascular endothelial growth factor (VEGF) antibody reducesseverity of experimental ulcerative colitis in rats: directevidence for the pathogenic role of VEGF. J Pharmacol ExpTher 2009;328:749.

[23] Wirtz S, Neurath MF. Mouse models of inflammatory boweldisease. Adv Drug Deliv Rev 2007;59:1073.

[24] Jackson JR, Seed MP, Kircher CH, Willoughby DA, Winkler JD.The codependence of angiogenesis and chronicinflammation. FASEB J 1997;11:457.

[25] Shaked Y, Bertolini F, Man S, et al. Genetic heterogeneity ofthe vasculogenic phenotype parallels angiogenesis;implications for cellular surrogate marker analysis ofantiangiogenesis. Cancer Cell 2005;7:101.

[26] Bocci G, Man S, Green SK, et al. Increased plasma vascularendothelial growth factor (VEGF) as a surrogate marker foroptimal therapeutic dosing of VEGF receptor-2 monoclonalantibodies. Cancer Res 2004;64:6616.

[27] Neimark E, Chen F, Li X, et al. c-Fos is a critical mediatorof inflammatory-mediated repression of the apicalsodium-dependent bile acid transporter. Gastroenterology2006;131:554.

[28] Dieleman LA, Palmen MJ, Akol H, et al. Chronic experimentalcolitis induced by dextran sulphate sodium (DSS) ischaracterized by Th1 and Th2 cytokines. Clin Exp Immunol1998;114:385.

[29] Chernoguz A, Crawford K, Vandersall A, et al.Pretreatment with anti-VEGF therapy may exacerbateinflammation in experimental acute colitis. J Pediatr Surg2012;47:347.

[30] Maxwell JR, Brown WA, Smith CL, Byrne FR, Viney JL.Methods of inducing inflammatory bowel disease in mice.Curr Protoc Pharmacol; 2009. Chapter 5: Unit5 58.

[31] Perse M, Cerar A. Dextran sodium sulphate colitis mousemodel: traps and tricks. J Biomed Biotechnol 2012;2012:718617.

[32] Melgar S, Karlsson L, Rehnstrom E, et al. Validation ofmurine dextran sulfate sodium-induced colitis using fourtherapeutic agents for human inflammatory bowel disease.Int Immunopharmacol 2008;8:836.

[33] Danese S, Sans M, Spencer DM, et al. Angiogenesis blockadeas a new therapeutic approach to experimental colitis. Gut2007;56:855.

[34] McLaren WJ, Anikijenko P, Thomas SG, Delaney PM,King RG. In vivo detection of morphological andmicrovascular changes of the colon in association withcolitis using fiberoptic confocal imaging (FOCI). Dig Dis Sci2002;47:2424.

[35] Witte L, Hicklin DJ, Zhu Z, et al. Monoclonal antibodiestargeting the VEGF receptor-2 (Flk1/KDR) as an anti-angiogenic therapeutic strategy. Cancer Metastasis Rev 1998;17:155.

[36] Shibuya M. Structure and function of VEGF/VEGF-receptorsystem involved in angiogenesis. Cell Struct Funct 2001;26:25.

[37] Jin ZG, Wong C, Wu J, Berk BC. Flow shear stress stimulatesGab1 tyrosine phosphorylation to mediate protein kinase Band endothelial nitric-oxide synthase activation inendothelial cells. J Biol Chem 2005;280:12305.

[38] Gerald D, Chintharlapalli S, Augustin HG, Benjamin LE.Angiopoietin-2: an attractive target for improvedantiangiogenic tumor therapy. Cancer Res 2013;73:1649.

[39] Baumgart DC, Sandborn WJ. Crohn’s disease. Lancet 2012;380:1590.

[40] Rutella S, Fiorino G, Vetrano S, et al. Infliximab therapyinhibits inflammation-induced angiogenesis in the mucosaof patients with Crohn’s disease. Am J Gastroenterol 2011;106:762.