VEGF-A isoform modulation in an preclinical TNBS model of ulcerative colitis: protective effects of a VEGF164b therapy

Abstract

Background: Ulcerative colitis (UC) is the most common form of inflammatory bowel disease in the USA. A key component of UC is the increase in inflammatory angiogenesis of the colon during active disease. This increase is driven to a great extent by the over expression of VEGF-A. Recently, VEGF165(b) (VEGF164(b) in mouse), an anti-angiogenic form of VEGF-A was described and its regulation was determined to be disturbed in many pathologies such as cancer and pre-eclampsia.

Results: The aims of this study were to examine the role of this inhibitory VEGF by expressing this molecule in a model of intestinal inflammation, and to evaluate its expression as a potential new therapeutic approach for treating UC. A modified model of TNBS colitis was used to determine the effects of rVEGF164(b) expression on colon inflammation. Expansion of the vascular system was assessed by immunhistochemical methods and macro- and microscopic measurements of inflammation in the colon were measured. Leukocyte invasion of the tissue was measured by myeloperoxidase assay and identification and counting of lymphoid follicles. Both angio- and lymphangiogenesis were reduced by expression of rVEGF164(b), which correlated with reduction in both gross and microscopic inflammatory scores. Leukocyte invasion of the tissue was also reduced by rVEGF164(b) expression.

Conclusions: This is the first report using an endogenous inhibitory VEGF-A isoform for therapy in a model of experimental colitis. Inhibitory VEGF molecules play an important role in maintenance of gut homeostasis and may be dysregulated in UC. The results of this study suggest that restoration of rVEGF164(b) expression has anti-inflammatory activity in a TNBS model and warrants further examination as a possible therapeutic for UC.

Figure 1

Measurement of serum levels of VEGF in the TNBS and DSS models of UC. The level of total serum VEGF-A isoforms was measured by ELISA in serum from control, vehicle, and TNBS treated mice (A); n = 4. The level of total serum VEGF-A isoforms was measured by ELISA in serum from control and DSS treated mice (B); n = 4. Serum protein was measured at either the acute (C) or chronic stages of the model (D); n = 5. Levels of total serum VEGF-A isoforms was measured in serum of animals treated with Ad5-CMV-rVEGF164b at 7 days after injection of Ad5-CMV-rVEGF164_b using doses of 0 (control), 1 × 10⁷, 1 × 10⁹, and 1 × 10¹¹ viral particles (E). An estimate of the persistence of expression of rVEGF164_b was generated by measuring total serum VEGF-A isoforms in the serum of animals injected with 1 × 10¹¹ viral particles of Ad5-CMV-rVEGF164_b at days 0, 10, and 21 (F).

Figure 2

Assessment of disease progression in the TNBS model of UC. Disease activity index (DAI) was determined in control, Ad5-CMV-rVEGF164_b, TNBS, and TNBS + Ad5-CMV-rVEGF164_b treated mice. * denotes significant differences between treated and untreated TNBS mice on that day. (A); n = 5. Colon length to weight ratio was determined in control, Ad5-CMV-rVEGF164_b, TNBS, and TNBS + Ad5-CMV-rVEGF164_b mice at day 3 (B) and day 21 (C) of disease. MPO assay results at day 3 (D) and day 21 (E) in disease were determined from control, Ad5-CMV-rVEGF164_b, TNBS, and TNBS + Ad5-CMV-rVEGF164b colons; n = 5. Shown are *p < 0.05, **p < 0.005.

Figure 3

Determination of lymphoid follicle aggregates in the TNBS model of UC. Representative image (10× magnification) shows detailed composition of follicle aggregate in TNBS treated colon (A). Representative image (40× magnification) of follicle aggregate showing staining for Mac-2 positive cells (arrows) (B). The number of follicles were quantified at day 3 (C) and 21 days (D) of disease; n = 5.

Figure 4

Evaluation of blood vessel density in the TNBS model of UC. Representative images of colon sections (20× magnification) are shown from control (A), TNBS (B), Ad5-CMV-rVEGF164_b(C) and TNBS + Ad5-CMV-rVEGF164_b(D) mice immunostained with stained with MECA-32. Angiogenesis was quantified at day 3 (E) and day 21 (F) of disease; n = 5.

Figure 5

Evaluation of lymphatic vessel density in the TNBS model of UC. Representative images of colon sections (20× magnification) are shown from control (A), TNBS (B), Ad5-CMV-rVEGF164_b(C) and TNBS + Ad5-CMV-rVEGF164_b(D) mice immunostained with VEGFR-3. Lymphangiogenesis was quantified at day 3 (E) and day 21 (F) of disease; n = 5.

Figure 6

Analysis of histopathology changes in the TNBS model of UC. Histopathology was scored by 2 blinded scorers on 6 parameters. Representative images of hematoxylin and eosin stained sections (20× magnification) are shown from control (A), TNBS (B), Ad5-CMV-rVEGF164_b(C) and TNBS + Ad5-CMV-rVEGF164_b(D) treated mice. Histopathology scores for colons on day 3 (E) and day 21 (F) of disease were combined from both blinded scorers; n = 5.