SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Objective

To determine whether blood vessel growth at the onset of resolving synovitis leads to its subsequent persistence and whether inhibiting this angiogenesis at the onset of persistent inflammation leads to its subsequent resolution.

Methods

Inflammation and angiogenesis were induced by injection of 0.03% carrageenan and/or 6 pmoles of fibroblast growth factor 2 (FGF-2) into rat knees. A brief treatment with the angiogenesis inhibitor PPI-2458 (5 mg/kg orally on alternate days) was administered 1 day before and up to 3 days after synovitis induction. Controls comprised naive and vehicle-treated rats. Synovial angiogenesis was measured using the endothelial cell proliferation index, and inflammation was determined by measuring joint swelling and macrophage percentage area. Data are presented as the geometric mean (95% confidence interval).

Results

Intraarticular injection of 0.03% carrageenan into rat knees produced acute synovitis, which was not associated with synovial angiogenesis and which resolved within 29 days. Injection of FGF-2 (6 pmoles) induced synovial angiogenesis without significant synovitis. Stimulation of angiogenesis with FGF-2 at the time of carrageenan injection was followed by synovitis that persisted for 29 days. Persistence of carrageenan/FGF-2–induced synovitis was prevented by systemic administration of 3 doses of the angiogenesis inhibitor PPI-2458 during the acute phase.

Conclusion

Our findings indicate that conversion of acute inflammation to chronic inflammation may be due to the stimulation of angiogenesis, and brief antiangiogenic treatment during the acute phase of synovitis may prevent its subsequent progression. Clinical studies will be needed to determine whether brief antiangiogenic treatment may reduce the burden of inflammatory joint diseases such as rheumatoid arthritis by facilitating the resolution of early synovitis.

Inflammation is normally a protective response to pathogenic, traumatic, or toxic injury, reducing tissue damage and leading to resolution or repair (1). In chronic inflammatory diseases such as rheumatoid arthritis (RA), persistent inflammation itself causes tissue damage (2). The mechanisms by which inflammation persists rather than resolves remain incompletely understood, particularly as in RA, where a persisting injurious agent has not been definitively identified. Chronic inflammatory diseases are a major health problem worldwide, leading to chronic pain and disability (3–5). Identifying the factors that lead to persistent inflammation should lead to novel therapies, thus reducing the burden of chronic inflammatory diseases by facilitating resolution or repair.

Angiogenesis is tightly regulated during normal growth, wound healing, and the female reproductive cycle (6), whereas persistent angiogenesis is characteristic of chronic inflammatory disease. For example, blood vessel growth within the inflamed synovium is typical of both early and established RA (7–9). In chronic synovitis, a shift in the balance from antiangiogenic to angiogenic factors promotes endothelial cell (EC) proliferation and blood vessel growth (10). Chronic inflammation maintains blood vessel growth by the secretion of angiogenic factors by macrophages and other cells (11) while synovial angiogenesis can further facilitate inflammation by increasing plasma extravasation and enhancing inflammatory cell recruitment (12, 13).

The fibroblast growth factor (FGF) system plays an important role in angiogenesis and in the maintenance of vascular integrity (14, 15). FGF-2 promotes EC proliferation and the physical organization of ECs into tube-like structures. It also stimulates the proliferation of fibroblasts that give rise to granulation tissue in wound healing (16). FGF-2–induced angiogenesis occurs in the absence of inflammation (17, 18), a characteristic which distinguishes it from many other angiogenic factors such as vascular endothelial growth factor.

The 2 methionine aminopeptidase (MetAP) isoforms, MetAP-1 and MetAP-2, are cotranslational regulators of protein synthesis. MetAP remove the initiator methionine from growing polypeptide chains, a prerequisite for a variety of biologic processes such as protein stability (19, 20). Inhibition of MetAP-2 prevents the selective removal of the N-terminal methionine and potently inhibits EC proliferation, a prominent mechanism in angiogenesis. PPI-2458 is an antiangiogenic fumagillin analog that is an orally active, irreversible inhibitor of MetAP-2 via covalent modification of His-231 in the catalytic site of the enzyme. This triggers growth arrest of ECs in the late G1 phase of the cell cycle, inhibiting EC proliferation and angiogenesis without affecting inflammatory cytokine release (19, 21, 22) or inducing apoptosis in ECs (23). PPI-2458 has reduced toxicity and greater selectivity for angiogenesis inhibition versus inflammation as compared with another fumagillin analog, AGM-1470/TNP-470 (23, 24).

PPI-2458 has been shown to reduce synovitis and bone and cartilage damage in collagen-induced arthritis in mice and peptidoglycan–polysaccharide–induced arthritis in rats (21, 25). Those studies and studies of other antiangiogenic strategies (20, 26–29) indicate that angiogenesis inhibition may have therapeutic potential in RA in humans. However, treatment of chronic inflammatory diseases with sustained inhibition of angiogenesis may lead to adverse events due to the blockade of physiologic blood vessel growth. An antiangiogenic strategy targeted to key phases in the development of chronic inflammation may have a more favorable risk/benefit ratio.

Intraarticular injection of low doses (0.03%) of carrageenan induces resolving synovitis without angiogenesis in rats, whereas synovitis persists when it is accompanied by synovial angiogenesis (30). This led us to hypothesize that angiogenesis during the early phase of synovitis may be a key factor determining its persistence and, where this is the case, that brief treatment with an angiogenesis inhibitor at the onset of arthritis could prevent its subsequent persistence. We stimulated angiogenesis by intraarticular injection of FGF-2 in rats. Low-dose carrageenan alone did not cause angiogenesis, whereas combined injection of carrageenan and FGF-2 was used to model combined inflammation and angiogenesis, as seen in early synovitis in humans. We used this model to evaluate the possible contribution of angiogenesis to the transition from acute to persistent synovitis.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

All in vivo experiments were carried out in accordance with UK Home Office regulations and were performed on male Wistar rats (Charles River) weighing 150–180 gm.

Animal models.

Pilot dose- and time-ranging experiments were carried out on sample animals in accordance with good practice and the 3 Rs: reduction, refinement, and replacement.

Intraarticular FGF-2 and/or carrageenan injection.

To deduce the dose of FGF-2 that enables synovial angiogenesis without inflammation, on day 0 rats (n = 2 per treatment group) were given bilateral intraarticular injections in their knee joint cavities of either 6, 0.6, 0.06, 0.006, or 0.0006 pmoles of FGF-2 or 10 mM Tris (pH 7.6) containing 0.1% bovine serum albumin (BSA) vehicle control. To determine at what time point after injection of 6 pmoles of FGF-2 (on day 0) synovial angiogenesis returns to levels similar to those in naive controls, synovia were collected from controls and on days 1, 2, and 5 after bilateral intraarticular injection of 6 pmoles of FGF-2 into rat knee joint cavities (n = 3).

In order to confirm the previous findings that 0.03% carrageenan induces synovial inflammation without angiogenesis (30), on day 0 rats (n = 2 per treatment group) were given bilateral intraarticular injections in their knee joint cavities of either 0.03%, 0.01%, or 0.003% carrageenan or sterile 0.9% normal saline control (pH 7.4). To investigate the combined effects of FGF-2 and carrageenan, on day 0 rats (n = 6 per treatment group) were given intraarticular injections of either 0.03% carrageenan or 0.03% carrageenan and 6 pmoles of FGF-2 in their left knee joint cavities and appropriate control treatments (sterile 0.9% normal saline [pH 7.4] or 10 mM Tris [pH 7.6] containing 0.1% BSA) in their right knee joint cavities. FGF-2 was injected first, followed by carrageenan after 30 minutes. Control injections were given in a similar manner.

Dose-response effects of PPI-2458 on angiogenesis induced by 6 pmoles of FGF-2.

In order to select the appropriate dose of PPI-2458 that best inhibited angiogenesis induced by 6 pmoles of FGF-2, on day −1 (the day before administration of FGF-2) rats (n = 3 per treatment group) were given either a 500 μl single oral dose of the angiogenesis inhibitor PPI-2458 at 5 mg/kg, 0.5 mg/kg, or 0.05 mg/kg, or were given vehicle control (11% 2-hydroxypropyl-β-cyclodextrin [HPβCD] buffer in phosphate buffered saline [PBS]). Rats were given bilateral intraarticular injections of 6 pmoles of FGF-2 in the knee joint cavities on day 0, and synovia (n = 6 per treatment group) were harvested on day 1. Naive rats (n = 3) were used as the control group.

PPI-2458 and intraarticular injection of FGF-2 and carrageenan.

On days −1, 1, and 3, rats (n = 6) were given an oral dose of 500 μl of either 5 mg/kg of PPI-2458 or vehicle control (11% HPβCD buffer in PBS) (25). Intraarticular injections of 0.03% carrageenan and 6 pmoles of FGF-2 (in the left knee joint cavity) or the appropriate vehicle control (sterile 0.9% normal saline [pH 7.4] or 10 mM Tris [pH 7.6] containing 0.1% BSA) (in the right knee joint cavity) were administered on day 0.

Prior to receiving intraarticular injections, all rats were anesthetized with isoflurane (2% in O2), and their knee joint diameter was measured (in millimeters) with digital electronic calipers (Mitutoyo). The volumes of all intraarticular injections were 100 μl. The skin over the joint was shaved, swabbed with 70% ethanol, and injections were given using a 27-gauge 12.7-mm needle inserted through the suprapatellar ligament while the joint was held in 90° of flexion (30–32). Sequential injections were given in a similar manner, with the FGF-2 injection given first, followed by a delay of 30 minutes to allow the solution to disperse, after which the carrageenan injection was administered. Carrageenan was dissolved in sterile 0.9% normal saline (pH 7.4). FGF-2 was prepared in 10 mM Tris (pH 7.6) containing 0.1% BSA. PPI-2458 (500 μl; 5 mg/kg) was prepared in 11% HPβCD buffer in PBS.

Tissue collection and preparation.

At each time point after intraarticular injections (days 1, 2, and 5 for FGF-2; days 1, 14, and 29 for carrageenan; days 1, 7, 14, 28, and 29 for FGF-2 and carrageenan combined; day 1 for PPI-2458 and FGF-2; and days 1, 7, and 28 for PPI-2458 and FGF-2 and carrageenan combined), knee joint diameter was measured with digital electronic calipers, and rats were killed by asphyxiation in CO2. Synovia with patellae from the right and left knees were immediately harvested and embedded in Tissue-Tek OCT mounting medium (Miles) before being frozen in melting isopentane onto cork block mounts and stored at −80°C until used.

Immunohistochemistry.

Tissue sections (5 μm thick) from each block were cut with a motorized cryostat, mounted on slides, and briefly air dried. CD31-positive cells and proliferating cell nuclear antigen (PCNA)–immunoreactive CD31-positive cells were used to identify ECs and proliferating ECs, respectively, in the synovia as 2 separate measures of the extent of angiogenesis. Macrophage infiltration was identified by immunoreactivity for the monoclonal antibody clone ED1 (33). For multiple sequential staining, sections were first immunostained by the peroxidase-conjugated avidin–biotin–peroxidase complex (ABC) method for PCNA (34) and then by the ABC–alkaline phosphatase method using a monoclonal antibody directed against CD31 for endothelium (35). Nuclei were counterstained with fluorescent DNA ligand, 4′,6-diamidino-2-phenylindole (DAPI) (36). PCNA and ED1 were developed with diaminobenzidine using the glucose oxidase/nickel-enhanced method. Endothelium markers were developed using Sigma-Aldrich fast red. Staining procedures were similar to those described previously (30).

Image analysis and quantification.

Image analysis and quantification was performed by an observer (SA) who was blinded with regard to the experimental details, using a Zeiss microscope (Carl Zeiss) with a 20× objective lens. Transmitted light and fluorescence images of the same field were captured using a 3-CCD camera and analyzed using a KS300 image analysis system (Image Associates). The procedure was similar to that previously described (32). Briefly, within the synovial regions, the EC proliferation index was defined as the percentage of endothelial nuclei positive for PCNA. The macrophage fractional area was defined as the percentage of synovial area that was ED1-positive.

Reagents.

Monoclonal antibody to PCNA (clone PC10) was obtained from Dako. Biotinylated rat-adsorbed horse anti-mouse antibody and ABC kits were obtained from Vector. Monoclonal antibodies to rat CD31 (platelet EC adhesion molecule 1) antibody (clone TLD-3A12) and to macrophages (clone ED1) were from Serotec. FGF-2 was purchased from Invitrogen. Saline, λ-carrageenan, DAPI, BSA, Tris, EDTA, HPβCD, and other chemicals were obtained from Sigma-Aldrich. PPI-2458 was a kind donation from GlaxoSmithKline.

Statistical analysis.

Four fields per section and 1 section per rat knee were used for synovial macrophage infiltration and EC proliferation data. This number was determined in previous experiments (30) to be optimal for minimizing the coefficient of variation and SEM. For the endothelial fractional area (a measure of vascular density), 4 consecutive sections per rat knee with 4 fields of view per section were used to minimize the coefficient of variation and SEM during analysis. Data are presented for 6 knees per experimental group, except where stated otherwise, and were analyzed using GraphPad Prism software, version 4.03. EC PCNA indices and endothelial and macrophage fractional areas were logarithmically transformed. All data were analyzed using one-way analysis of variance. Univariate comparisons were made using Student's t-test with Bonferroni correction for multiple comparisons. Numerical data are presented as the geometric mean (95% confidence interval [95% CI]).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Synovial angiogenesis and inflammation after intraarticular injection of FGF-2 and/or carrageenan.

CD31-immunoreactive blood vessels were distributed throughout the depth of rat synovia, sometimes containing PCNA-immunoreactive nuclei (Figures 1A–C). ED1-immunoreactive macrophages were localized to the synovial lining and dispersed throughout the synovial sublining regions (Figures 1G–I). Synovia from animals given intraarticular injections of 6 pmoles of FGF-2 showed increased staining for proliferating ECs (Figure 1A), and those from animals injected with 0.03% carrageenan displayed increased macrophage infiltration (Figure 1G).

thumbnail image

Figure 1. Histologic evidence of angiogenesis and macrophage infiltration after coinjection of fibroblast growth factor 2 (FGF-2) and carrageenan into rat knees. A–C, Endothelial cell (EC) proliferation on day 1 in rats injected with carrageenan and FGF-2 (A), rats treated with PPI-2458 and injected with carrageenan and FGF-2 (B), and controls (C). ECs are delineated by immunoreactivity for CD31 (red), and proliferating nuclei are delineated by immunoreactivity for proliferating cell nuclear antigen (PCNA; black). Thin black arrow indicates proliferating ECs, which contain PCNA-immunoreactive nuclei; green arrows indicate nonproliferating ECs, which do not contain PCNA-immunoreactive nuclei; thick black arrows indicate cells other than ECs that display PCNA-immunoreactive nuclei. Proliferating ECs appeared more abundant on day 1 in rats injected with FGF-2 and carrageenan than in saline-injected controls. Administration of PPI-2458 was associated with a histologic appearance similar to that of controls without synovitis. D–F, Extent of ECs on day 7 in rats injected with carrageenan and FGF-2 (D), rats treated with PPI-2458 and injected with carrageenan and FGF-2 (E), and controls (F). Increased vascular density was observed on day 7. Administration of PPI-2458 reduced vascular density to levels seen in controls. G–I, Macrophage infiltration on day 28 in rats injected with carrageenan and FGF-2 (G), rats treated with PPI-2458 and injected with carrageenan and FGF-2 (H), and controls (I). Arrows show macrophages, which are delineated by immunoreactivity for ED1. Macrophage infiltration was increased 28 days after coadministration of FGF-2 and carrageenan and was reduced to control levels in groups given PPI-2458. Bar = 100 μm.

Download figure to PowerPoint

A significant increase in the EC PCNA index was observed 24 hours after intraarticular injection of FGF-2 (Figure 2A). Injection of FGF-2 (6 pmoles) was followed by increased synovial EC PCNA index at 24 hours (20% [95% CI 15–27]) compared with saline controls (3% [95% CI 1–6]; P < 0.01) (Figure 2A), which was reduced to levels similar to those found in naive controls by day 5 (Figure 2C). At 24 hours after injection of 6 pmoles of FGF-2, no significant effect on joint diameter (9.2 mm [95% CI 9–9.5] versus 9.2 mm [95% CI 8.7–9.7] in controls; P > 0.05), synovial macrophage infiltration (Figure 2B), or vascular density (EC fractional area 11% [95% CI 8–16] versus 11% [95% CI 9–14] in controls; P > 0.05) was seen.

thumbnail image

Figure 2. Synovial EC proliferation and macrophage infiltration after intraarticular injection of FGF-2 alone or carrageenan alone into rat knees. A, Dose-dependent increase in the EC proliferation index (EC PCNA index) 24 hours after intraarticular injection of FGF-2 but not carrageenan. B, Dose-dependent increase in macrophage infiltration 24 hours after intraarticular injection of carrageenan but not FGF-2. C, Reduction of the EC PCNA index to control levels 5 days after injection of FGF-2. Naive and vehicle-treated knees were similar for all parameters. Bars show the arithmetic mean and SEM of synovia from 4–6 knees per group. ∗∗ = P < 0.01. See Figure 1 for definitions.

Download figure to PowerPoint

There was a dose-dependent increase in synovial macrophage infiltration 24 hours after intraarticular carrageenan injection (Figure 2B). Synovial macrophage infiltration increased 24 hours after injection of 0.03% carrageenan compared with vehicle-injected controls and had returned to levels similar to those in controls by day 14 (5.2% [95% CI 2.7–10.1] in carrageenan-treated rats versus 2.7% [95% CI 1.4–5.2] in controls; P > 0.05]) (Figure 3A). No significant effects of intraarticular injection on joint swelling (Figure 3B), EC proliferation (Figures 2A and 3C), or vascular density (EC fractional area 20% [95% CI 17–23]) at 24 hours after injection of 0.03% carrageenan versus 16% [95% CI 12–21] in controls; P > 0.05) were observed.

thumbnail image

Figure 3. Persistent synovitis after coinjection of FGF-2 and carrageenan. Rat knees were injected intraarticularly with vehicle (▾), with 0.03% carrageenan alone (▴), or were coinjected with 6 pmoles of FGF-2 and 0.03% carrageenan (▪). A, Macrophage infiltration 1–29 days after injection. B, Joint swelling, as measured by knee diameter, 1–29 days after injection. C, EC proliferation 1–29 days after injection. Injection of carrageenan alone was followed by a brief increase in synovial macrophage infiltration and had no significant effect on joint swelling or endothelial proliferation as compared with control knees. Coinjection of 6 pmoles of FGF-2 and 0.03% carrageenan was followed by increased EC proliferation as well as increased inflammation (macrophage infiltration and joint swelling) after 24 hours and by persistent synovitis on days 14 and 29. No significant increase in any parameter was observed in knees from vehicle-injected controls as compared with knees from naive rats. Values are the arithmetic mean ± SEM of synovia from 6 knees per group. ∗ = P < 0.05; ∗∗ = P < 0.01; ∗∗∗ = P < 0.001, versus vehicle-injected controls. See Figure 1 for definitions.

Download figure to PowerPoint

Coinjection of 0.03% carrageenan and 6 pmoles of FGF-2 into rat knee joints was followed after 24 hours by increased synovial macrophage infiltration (9.4% [95% CI 6.4–13.8]) (Figures 3A and 4A) and joint swelling (9.6 mm [95% CI 9.2–10.0]) (Figures 3B and 4B) as compared with controls (1.6% [95% CI 0.5–4.8]; P < 0.05 for macrophage infiltration and 9.1 mm [95% CI 8.9–9.3]; P < 0.01 for joint swelling). Coinjection of carrageenan and FGF-2 was also followed after 24 hours by increased synovial EC proliferation (3% [95% CI 2–4.5]) (Figures 1A, 3C, and 4C) and EC fractional area (8.0% [95% CI 6.0–10.6]) (Figure 4D) as compared with controls (0.8% [95% CI 0.3–1.8]; P < 0.01 for synovial EC proliferation and 4.9% [95% CI 3.9–6.0]; P < 0.05 for EC fractional area). The increases in indices of inflammation and angiogenesis were maintained through day 28 after coinjection of carrageenan and FGF-2 (Figures 1D, 1G, 3, and 4), indicating the induction of chronic synovitis. The indices of angiogenesis and inflammation resolved within 28 days in rats injected with FGF-2 or carrageenan alone (Figures 2, 3, and 4).

thumbnail image

Figure 4. Inhibition of FGF-2–induced angiogenesis and promotion of the resolution of synovitis after PPI-2458 administration. Rats received an injection of vehicle (▾), an intraarticular injection of a combination of 0.03% carrageenan and 6 pmoles of FGF-2 (▪), or 3 oral doses of PPI-2458 (5 mg/kg) on alternate days from 1 day before until 3 days after intraarticular injection of 0.03% carrageenan and 6 pmoles of FGF-2 (▴). A, Macrophage infiltration 1–28 days after injection. B, Joint swelling, as measured by knee diameter, 1–28 days after injection. C, EC proliferation 1–28 days after injection. D, EC fractional area 1–28 days after injection. Following injection of 0.03% carrageenan and 6 pmoles of FGF-2, there were sustained increases in joint inflammation (synovial macrophage infiltration and joint swelling) and synovial angiogenesis indices (EC proliferation and vascular density), compared with vehicle-injected controls. Animals given a total of 3 oral doses of PPI-2458 (5 mg/kg) on alternate days from 1 day before until 3 days after intraarticular injection of 0.03% carrageenan and 6 pmoles of FGF-2 displayed synovial angiogenesis indices similar to those in vehicle-injected controls throughout the experiment. PPI-2458 administration did not significantly inhibit joint inflammation after 24 hours, and macrophage infiltration was only partially reduced on day 7. However, the brief treatment with PPI-2458 completely inhibited joint inflammation to control levels by day 28, thereby converting the model from one of persistent inflammation to one of transient synovitis. The broken line indicates the time of last dose of PPI-2458. Values are the arithmetic mean ± SEM of synovia from 6 knees per group. ∗ = P < 0.05; ∗∗ = P < 0.01; ∗∗∗ = P < 0.001, versus vehicle-injected controls. + = P < 0.05; ++ = P < 0.01; +++ = P < 0.001, versus PPI-2458–treated animals. See Figure 1 for definitions.

Download figure to PowerPoint

Dose-dependent inhibition of FGF-2–induced angiogenesis by PPI-2458.

Intraarticular injection of FGF-2 alone without PPI-2458 administration was followed 24 hours later by a significant increase in synovial EC proliferation (6.9% [95% CI 4.0–11.7]) as compared with naive animals (1.1% [95% CI 0.6–1.9]; P < 0.001) (Figure 5). A dose-dependent reduction in FGF-2–induced synovial angiogenesis was observed in groups treated with increasing doses of PPI-2458 administered orally 1 day prior to FGF-2 injection, with complete inhibition of the EC PCNA index to the levels in naive controls in rats treated with 5 mg/kg of PPI-2458 (1.1% [95% CI 0.6–2.3]; P > 0.05 as compared with naive controls) (Figure 5). PPI-2458 did not alter synovial macrophage infiltration. Values remained similar to those observed in naive controls within all groups treated with FGF-2.

thumbnail image

Figure 5. Dose-dependent reduction in FGF-2–induced synovial EC proliferation after treatment with PPI-2458. Oral doses of PPI-2458 were given 1 day before bilateral intraarticular injection of 6 pmoles of FGF-2 into rat knee joints. The EC proliferation index (EC PCNA index) was reduced in a dose-dependent manner 24 hours after FGF-2 injection, with the highest dose of PPI-2458 reducing the EC PCNA index to levels found in naive controls. Bars show the arithmetic mean and SEM of synovia from 6 knees per group. ∗∗∗ = P < 0.001; ∗∗ = P < 0.01, versus naive controls. See Figure 1 for definitions.

Download figure to PowerPoint

Synovial angiogenesis and inflammation after brief administration of PPI-2458 during the early phase of FGF-2/carrageenan–induced synovitis.

Animals were given 3 doses of PPI-2458 between the day before and 3 days after the induction of chronic synovitis by coinjection of carrageenan and FGF-2. This dose regimen was designed to cover the period of FGF-2–enhanced angiogenesis, and was determined as described above. PPI-2458 administration was associated with a reduction of angiogenesis indices to control levels 24 hours after carrageenan/FGF-2 injection (Figures 1B, 4C, and 4D), whereas inflammation indices were still increased 24 hours after injection of carrageenan/FGF-2 in PPI-2458–treated animals when compared with control animals (Figures 4A and B). Furthermore, synovial macrophage infiltration was only partially reduced 7 days after carrageenan/FGF-2 injection, remaining elevated in animals given PPI-2458 (9.3% [95% CI 7.0–12.4]), as compared with controls without synovitis (2.8% [95% CI 1.9–4.2]; P < 0.001) (Figure 4A). However, inflammation indices were normalized to levels observed in controls without synovitis by day 28 in animals briefly treated with PPI-2458 up to 3 days after intraarticular injection of carrageenan/FGF-2 (Figures 1H and 4A–D).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

In the present study, we examined whether angiogenesis may be a key factor in the transition from acute to persistent inflammation in an animal model of synovitis. We have shown that at a dose of 0.03%, carrageenan alone did not induce angiogenesis. Injection of carrageenan alone was associated with a transient synovitis that resolved after 14 days. When angiogenesis was stimulated with 6 pmoles of FGF-2 at the time that synovitis was induced by intraarticular injection of carrageenan, the synovitis persisted for at least 4 weeks. Inhibition of this effect of FGF-2 by PPI-2458 suggests that this chronicity is mediated by angiogenesis rather than by other actions of FGF-2.

The mechanisms leading to persistent inflammation in human diseases remain unclear. Autoimmunity contributes to inflammation and tissue damage in diseases such as RA, although its contribution to persistence in early RA remains incompletely understood. Evidence of autoimmunity, such as circulating anti–citrullinated peptide antibodies, has limited sensitivity in predicting persistence in early synovitis, and immunosuppression does not prevent persistence in the majority of patients. Many other chronic inflammatory diseases (e.g., psoriasis and inflammatory bowel disease) display little evidence of autoimmunity. Angiogenesis inhibition may reduce arthritis severity in animal models of autoimmunity (21, 23, 25, 27, 28, 37, 38). Our data indicate that angiogenesis inhibition may prevent the progression from acute to chronic inflammation in situations where new blood vessel growth is the key switch to persistence.

In the present study, we used FGF-2 and PPI-2458 and specific outcome measures of angiogenesis and inflammation to determine the effects of angiogenesis inhibition on the persistence of inflammation. FGFs are implicated in the initiation and development of synovial hyperplasia and inflammation in RA (39). FGF-2 is up-regulated at the cartilage–pannus interface in RA (40), and increased synovial fluid levels of FGF-2 are associated with greater disease severity (41, 42). FGF-2 was also shown to augment antigen-induced arthritis in rabbits (27), and FGF-2 overexpression worsened inflammation and joint damage in adjuvant-induced arthritis in rats (43).

Our data support previous evidence that FGF-2 is proangiogenic but not directly proinflammatory (44–46). FGF-2 is mitogenic for ECs and acts via specific FGF-2 receptors displayed on these cells. In some experimental systems, FGF-2 can additionally enhance vascular integrity (14) and increase adhesion molecule expression by cytokine-activated ECs, thereby synergistically enhancing the recruitment of inflammatory cells (18). However, we found that a single intraarticular injection of FGF-2, either alone or together with carrageenan, was not sufficient to increase joint swelling or macrophage infiltration at 24 hours, despite its ability to increase EC proliferation. The FGF-2–induced increase in chronic synovitis was inhibited both by the brief administration of PPI-2458 in our study and by repeated administration of an antiangiogenic inhibitor of αvβ3 integrin in antigen-induced arthritis in a previous study (27). FGF-2, therefore, may contribute to arthritis by increasing synovial angiogenesis, rather than through any direct effect on inflammation.

PPI-2458 inhibits human EC proliferation by inhibiting MetAP-2. It does not reduce angiogenic growth factor production (19, 21, 22) or induce apoptosis in ECs (23). However, high doses of PPI-2458 may have effects in addition to angiogenesis inhibition, and mediation of its therapeutic effects by angiogenesis inhibition has been uncertain in previous studies (23). We found that PPI-2458 administration abolished synovial angiogenesis prior to reduction in synovitis, whereas synovitis reduction followed the discontinuation of PPI-2458. This indicates that PPI-2458 may exert its effects not through the direct inhibition of inflammation, but rather through the inhibition of angiogenesis.

Synovial angiogenesis is an early feature of RA, and occurs very early in the development of persistent synovitis in animal models (30, 37). Synovial angiogenesis is also a feature of chronic synovitis in both humans and animals (30, 47, 48). In the present study, we demonstrated increased EC proliferation during chronic synovitis 29 days after coinjection of carrageenan and FGF-2. It is likely that angiogenesis on day 29 was mediated by factors other than the injected FGF-2, since FGF-2 alone induced only a transient increase in EC proliferation that had normalized after 5 days. Angiogenesis during chronic synovitis may be mediated by an imbalance between a variety of angiogenic and antiangiogenic factors (12). Further studies are needed to determine which molecules sustain synovial angiogenesis in human or animal models.

Mechanisms by which acute synovitis may lead to chronic arthritis remain incompletely understood. Clinical trials aiming to induce remission in early synovitis by intensive immunosuppression have raised the prospect that it may eventually be possible to modify the course of, for example, RA, without the need for lifelong medication. Immunosuppression alone, however, has not yet achieved this goal, and other factors may be important in the transition from acute to chronic synovitis. Angiogenesis is a feature of early, as well as persistent, synovitis in humans (7–9), and our findings indicate that synovial angiogenesis may be a factor leading to persistent synovitis. Synovitis typically progresses more rapidly in animal models than does arthritis in humans, and further work is needed to determine the duration of any window of opportunity in early synovitis during which inhibition of angiogenesis may facilitate its resolution. Furthermore, since there may be multiple factors that lead to persistence, angiogenesis inhibition may have greatest therapeutic potential in combination with other pharmacologic agents. Similar combined approaches with angiogenesis inhibitors have proven successful in oncology (49, 50).

Sustained inhibition of angiogenesis that is begun during the induction, or at the onset, of synovitis can reduce inflammation and joint damage (21, 23, 27, 37, 38). However, adverse effects of long-term angiogenesis inhibition on physiologic blood vessel growth may outweigh the therapeutic benefit in chronic inflammatory diseases. Our data indicate that short-term angiogenesis inhibition targeted to a key period in the development of inflammation has the potential to achieve sustained clinical benefit. Clinical studies are needed to determine whether brief antiangiogenic treatments can facilitate resolution in early synovitis, at the onset of synovitis in a previously uninvolved joint, or at reactivation of previously quiescent inflammation. A greater understanding of the mechanisms underlying persistence in different clinical scenarios will be necessary in order to realize this potential.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Ms Ashraf had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design.Ashraf, Mapp, Walsh.

Acquisition of data.Ashraf, Mapp, Walsh.

Analysis and interpretation of data.Ashraf, Mapp, Walsh.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

The authors thank Dr. Inma Rioja (GlaxoSmithKline) for providing the angiogenesis inhibitor PPI-2458.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES