Correlations of the Expression of Fibroblast Growth Factor-2,Vascular Endothelial Growth Factor,and their Receptors with Angiogenesis in Synovial Tissues from Patients with Internal Derangement of the Temporomandibular Joint

Abstract

Synovitis in internal derangement of the temporomandibular joint (TMJ) is accompanied by the growth of new blood vessels. Fibroblast growth factor-2 (FGF-2) and vascular endothelial growth factor (VEGF) are well-characterized angiogenic factors. The objective of this study was to elucidate the correlation between the expression of FGF-2, VEGF, and their receptors—FGF receptor-1 (FGFR-1) and VEGF receptor-1 (Flt-1)—with microvessel density in synovial tissues of the TMJ. Using an immunohistochemical technique, we examined 47 joints (45 patients) with internal derangement. Individual microvessel density was evaluated by means of the CD34 antibody, a specific endothelial marker. The correlation between the percentage of immuno-positive cells and microvessel density was evaluated. In multiple logistic regression analysis, the correlation between the percentage of Flt-1-positive cells and microvessel density was significant [p = 0.005, odds ratio = 1.071, 95% confidence interval = 1.021-1.124]. These results suggest that the expression of the VEGF/Flt-1 system is involved in angiogenesis in inflamed synovial tissue in the TMJ.

INTRODUCTION

Internal derangement of the temporomandibular joint (TMJ) is often accompanied by synovitis, which, in turn, is accompanied by chronic inflammatory changes of the synovial tissue, including growth of small new blood vessels (Merrill et al., 1990; Gynther et al., 1994). In previous studies, pro-inflammatory cytokines, including interleukin 1β (IL-1β), IL-6, IL-8, and tumor necrosis factor-α, have been detected at significant levels in synovial fluids or synovial tissues of TMJs with internal derangement (Fu et al., 1995; Kubota et al., 1998; Takahashi et al., 1998; Suzuki et al., 1999), and these mediators are thought to contribute to both the clinical symptoms and the pathogenesis of the internal derangement.

Angiogenesis is promoted by angiogenic factors (Decaussin et al., 1999). Fibroblast growth factor-2 (FGF-2) (Galzie et al., 1997; Szebenyi and Fallon, 1999) and vascular endothelial growth factor (VEGF) (Folkman et al., 1989) are well-known to be mitogenic for endothelial cells and to induce angiogenesis in vivo, especially in solid tumors and inflammatory diseases, such as rheumatoid arthritis (Koch, 1998; Carmeliet and Jain, 2000).

In this study, we first assess the correlation between the expression of FGF-2, VEGF, fibroblast growth factor receptor-1 (FGFR-1), and vascular endothelial growth factor receptor-1 (VEGFR-1; Flt-1) with angiogenesis in the synovial tissues with internal derangement of the TMJ.

MATERIALS & METHODS

Patients

Forty-seven symptomatic joints in 45 patients with internal derangement of the TMJ were included in this study. Six of the patients were men, and 39 were women. Although 11 of the 39 women were menopausal, no one took estrogen. Their average age was 43 yrs (range, 17 to 84). Magnetic resonance imaging revealed that all of the patients had anterior disc displacement without reduction. Their mean maximum interincisal opening was 33 mm, and the mean duration of their symptoms was 5.9 mos. All patients reported pain in the TMJ region when they moved their jaws. All patients underwent arthroscopic surgery when appropriate non-surgical treatment failed to resolve the clinical symptoms. A control group was comprised of 7 joints in six patients, two men and four women (average age, 40 yrs), who had habitual dislocation without pain. All control patients underwent arthroscopic eminoplasty (Segami et al., 1999). Magnetic resonance imaging confirmed that the disc was in a normal position in all the control patients when their mouths were closed. The validity of adopting these patients as controls has been discussed in a previous article (Sato et al., 2002). Informed consent was obtained from all patients in this study. The protocol was approved by the Kanazawa Medical University Institutional Review Broad.

Synovial Biopsy and Immunohistochemical Staining

From each patient, synovial tissue specimens, about 2 mm in diameter, were obtained arthroscopically from the region of the posterior disc attachment by means of the triangular technique with direct arthroscopic visualization. Immediately after resection, the specimens were fixed in 4% paraformaldehyde for 8 hrs and embedded in paraffin. Consecutive sections were prepared and immunohistochemically stained by an avidin-biotin technique (Vector Laboratories, Burlingame, CA, USA) (HSU et al., 1981). We blocked the endogenous peroxidase by immersing the sections in 0.3% H₂O₂ in methanol for 10 min at room temperature. The sections were treated with 0.1% trypsin for 20 min at 37°C. After non-specific binding was blocked with 1.5% normal horse serum for 20 min at room temperature, the sections were treated with primary antibody. The primary antibodies used in this study were as follows: FGF-2 (bFM-1; 1 μg/mL; monoclonal) (Matsuzaki et al., 1989), VEGF (1 μg/mL, monoclonal; Santa Cruz Biotechnology, Santa Cruz, CA, USA) (Pufe et al., 2001), FGFR-1 (dilution rate, 1:200, polyclonal; Santa Cruz Biotechnology) (Ohta et al., 1995), Flt-1 (1 μg/mL, monoclonal; Santa Cruz Biotechnology) (Jin et al., 2000). The primary antibodies were applied for 1 hr at room temperature (VEGF and FGFR-1) or overnight at 4°C (FGF-2 and Flt-1). The specimens were left in a 1:200 dilution of anti-mouse or anti-rabbit biotinylated antibody (Dako, Carpinteria, CA, USA) for 60 min at room temperature. An avidin/biotinylated horseradish peroxidase complex was added, and the solution was incubated for a further 40 min at room temperature. The color was developed by 3-amino-9-ethyl carbazole, followed by counter-staining with hematoxylin. Negative controls in which the primary antibody was replaced with normal mouse or normal rabbit IgG were run with each specimen. The sections were viewed under a light microscope at 200X magnification. Cells whose cytoplasm was definitely stained red were considered to be immuno-positive. The percentage of immuno-positive cells was estimated in the same areas of the consecutive sections for each protein. For each section, the immuno-positive cells were counted in two regions, each containing from 200 to 500 cells, where the cell density was the highest. The cell count was made by two of the authors (J.S. and N.S.) who did not know from which patients the specimens came.

Vessel Staining and Counting

We determined the microvessel density of the tissues by staining endothelial cells using primary antibody for CD34 (dilution rate, 1:50, monoclonal; Nichirei, Tokyo, Japan), as specific endothelial markers, according to Weidner’s method, with minor modification (Weidner et al., 1993; Decaussin et al, 1999). The immunohistochemical method is described above. Red-stained endothelial cells with lumen formations were considered to be blood vessels. The microvessel density was evaluated as the total number of blood vessels in two areas of maximal vascularization under a light microscope (20X objective and 10X ocular, 0.74 mm² per field).

Statistical Analysis

We used the Spearman rank correlation coefficient to assess each correlation between the microvessel density and the percentage of the cells immuno-positive for FGF-2, VEGF, FGFR-1, and Flt-1. We used Student’s t test to check for differences in the degrees of expression of the 4 proteins between the internal derangement group and the control group. Moreover, we performed multiple logistic regression analysis to elucidate the independent contributions of the expressions of FGF-2, VEGF, FGFR-1, and Flt-1 to microvessel density. In multiple logistic regression analysis, microvessel density was considered to be "low" when under 20 (the median) and "high" when over 20. Stat View J-5.0 statistical software (Abacus Concepts, Berkeley, CA, USA) was used. Probabilities of less than 0.05 were considered to be significant.

RESULTS

The Expression of FGF-2, FGFR-1, VEGF, and Flt-1 in Synovial Tissues

In the internal derangement and control groups, FGF-2, FGFR-1, VEGF, and Flt-1 were immunohistochemically detected in synovial tissues (Table 1). In the internal derangement group, FGF-2, FGFR-1, and VEGF were present in the cells lining the synovium, fibroblasts beneath the lining and endothelial cells of the blood vessels (Figs. 1A, 1B, 1C ). Flt-1 was present in the endothelial cells of the blood vessels and in a few lining cells (Fig. 1D ). All negative control sections showed only background staining (data not shown). The distributions of the 4 proteins in the control specimens were the same as in the internal derangement specimens (Figs. 1F, 1G, 1H, 1I ). The cells lining the synovium, the fibroblast, and the endothelial cells of the blood vessels were counted, and the percentages of cells immuno-positive for FGF-2, FGFR-1, VEGF were determined. We determined the percentage of Flt-1-positive cells by counting the endothelial cells of the blood vessels. For all 4 proteins, the percentage of immuno-positive cells was significantly higher in the internal derangement group than in the control group (FGF-2, p = 0.005; FGFR-1, p < 0.001; VEGF, p = 0.019; Flt-1, p = 0.043) (Table 1 ). When the men and women in the internal derangement group were compared, there were no significant differences in the percentages of immuno-positive cells for any of the 4 proteins (data not shown).

Vessel Staining and Counting

In all the internal derangement and control specimens, endothelial cells were better visualized with CD34 staining (Figs. 1E, 1J ). The average microvessel density was 24 ± 18 (per two fields) in the internal derangement specimens vs. 11 ± 6 in the control specimens. The microvessel density was higher in the internal derangement group than in the control group (p = 0.049). In the internal derangement group, the microvessel density was the same for men and women (men, 23 ± 23; women, 25 ± 18). Moreover, in the female group, menstrual status did not affect microvessel density (data not shown).

Correlation of the Expression of FGF-2, FGFR-1, VEGF, and Flt-1 with Microvessel Density

Although the correlation between microvessel density and the percentage of FGF-2-positive cells was not significant, the correlations between microvessel density and the other 3 proteins were significant (Figs. 2A, 2B, 2C, 2D ) (FGF-2, p = 0.13, r = 0.23; FGFR-1, p = 0.025, r = 0.34; VEGF, p = 0.0003, r = 0.53; Flt-1, p < 0.0001, r = 0.68). In multiple logistic regression analysis, in the internal derangement group, the correlation between the percentage of Flt-1-positive cells and microvessel density was significant (p = 0.005, odds ratio = 1.071, 95% confidence interval = 1.021-1.124), after adjustment for the percentages of immuno-positive cells in the other proteins (Table 2 ).

DISCUSSION

Although our research group has previously demonstrated the immunohistochemical expression of VEGF in the synovial tissues of the TMJ with internal derangement (Sato et al., 2002), to our knowledge, this is the first report on the expression of FGF-2, FGFR-1, and Flt-1 in the synovial tissues of patients with internal derangement of the TMJ. Moreover, we found significant correlations of the expression of FGFR-1, VEGF, and Flt-1 with microvessel density in the internal derangement specimens.

The biologic activities of FGF-2 and VEGF are mediated by binding to specific cell-surface tyrosine kinase receptors. We chose to evaluate the receptors FGFR-1 and Flt-1. We chose FGFR-1 over the other three FGF-2 receptors (FGFR-2, FGFR-3, and FGFR-4), because it has the highest affinity for FGF-2 and because it is widely distributed in the human body (Galzie et al., 1997; Burke et al., 1998). Furthermore, FGFR-1 seems to be the most important receptor in the FGF-2/FGFR system (Dionne et al., 1990). Flt-1 was not our first choice as the receptor for VEGF. VEGF also binds to the receptor VEGFR-2 (Flk-1/KDR). KDR is mostly located on endothelial cells and is thought to be a more important receptor than Flt-1 for mitogenic activity of VEGF (Aiello et al., 1995). Our attempts to detect KDR immunohistochemically were unsuccessful (data not shown), so we tried to detect Flt-1. Coincidentally, recent evidence indicates that Flt-1 is the key receptor in the response to hypoxia-induced angiogenesis (Brogi et al., 1996).

In the present study, the correlation of the expression of VEGF with microvessel density was significant, as determined by the Spearman rank correlation coefficient (P = 0.003, r = 0.53), but the correlation of the expression of FGF-2 with microvessel density was not significant (P = 0.13, r = 0.23). The difference in the strength of correlation might result from a difference in their distributions in their normal states. Previous studies have indicated that FGF-2 is present in many normal tissues (Schulze-Osthoff et al., 1990), while VEGF is expressed in a limited number of sites in normal tissues (Berse et al., 1992). In fact, FGF-2 was expressed in all 7 of our control specimens, but VEGF was expressed in only 3 of the control specimens. Our finding that the expression of FGFR-1 correlated with microvessel density indicates that increased expression of FGFR-1, rather than increased expression of FGF-2, may contribute to pathological angiogenesis in the synovitic TMJ. Another possible explanation of the finding is that FGF-2 acts on many kinds of cells in addition to endothelial cells, whereas VEGF acts on mainly endothelial cells in addition to some monocytes (Koch, 1998). In our study, immunoreactive Flt-1 was found mainly in the endothelial cells, but FGFR-1 was found equally in the surface-lining cells, the fibroblasts, and the endothelial cells. These results may support our explanations, and FGF-2 may play an indirect role in angiogenesis by stimulating the lining cells and fibroblasts, which produce some inflammatory cytokines.

The role of FGF-2 and VEGF in human chronic inflammatory states has not been clarified. New blood vessels may maintain the chronic inflammatory state by transporting inflammatory cells and supplying nutrients and oxygen to the inflamed tissues (Jackson et al., 1997). It is suggested that small synovial blood vessels perform a role, not only in the inflammatory phase, but also in the late chronic inflammatory phase (Koch et al., 1994). Our study indicates that both FGF-2/FGFR-1 and VEGF/Flt-1 systems contribute to angiogenesis in the synovial tissues of the TMJ. The VEGF/Flt-1 system, however, may play a more important role than the FGF-2/FGFR-1 system in the angiogenesis. Analysis of the data obtained from multiple logistic regression analysis supports this investigation.

The normal vasculature is quiescent in adult mammals, except in the highly ordered processes of the female reproductive cycle, such as ovulation (Koch, 1998). In the present study, however, local microvessel density was not affected by sex difference or by menstrual status.

In conclusion, our study indicates that VEGF and its membrane receptor may be key regulators in angiogenesis in synovial tissues of the TMJ. In the near future, it may be possible to treat synovitis of the TMJ by anti-angiogenic therapy that tackles VEGF. It is true, however, that angiogenesis is likely to result from a delicate balance of many kinds of angiogenic factors and anti-angiogenic factors. Further studies are needed to elucidate the direct contribution of VEGF to vascularization in the synovial tissues of the TMJ.

Table 1.

Immunohistochemical Results of the 4 Proteins^e

	Patients with Internal Derangement		Control Patients
Protein	Patients with Immuno-positive Cells (%)	Percentage of Immuno-positive Cells (%) (mean ± SD)	Patients with Immuno-positive Cells (%)	Percentage of Immuno-positive Cells (%) (mean ± SD)
^a P = 0.005 (internal derangement vs. control).
^b P < 0.001.
^c P = 0.019.
^d P = 0.043.
^e Statistical analysis was performed by Student’s t test.
FGF-2	47/47 (100)	57 ± 22^a	7/7	31 ± 24^a
FGFR-1	47/47 (100)	51 ± 22^b	6/7	12 ± 14^b
VEGF	41/47 ( 87)	34 ± 29^c	3/7	7 ± 12^c
Flt-1	40/47 ( 85)	27 ± 24^d	4/7	7 ± 16^d

Table 2.

Statistical Parameters of the Logistic Regression Analysis of Microvessel Density

Factors^a	Coefficient	Standard Error	χ²	p Value	Odds Ratio (95% confidence interval)
^a FGF-2, FGFR-1, VEGF, Flt-1: the percentage of immuno-positive cells.
FGF-2	0.009	0.024	0.142	0.706	1.009 (0.963-1.057)
FGFR-1	0.015	0.022	0.491	0.483	1.015 (0.973-1.060)
VEGF	0.005	0.016	0.116	0.733	1.005 (0.975-1.037)
Flt-1	0.069	0.025	7.855	0.005	1.071 (1.021-1.124)

Figure 1.

Immunohistochemical overview for FGF-2, FGFR-1, VEGF, Flt-1, and CD34 of the synovial tissues. (A) FGF-2-stained sections obtained from a patient with internal derangement (insert, low power). (B) FGFR-1-stained sections obtained from the same patient (insert, low power). (C) VEGF-stained sections obtained from the same patient (insert, low power). Immunoreactivities for FGF-2, FGFR-1, and VEGF are observed in the synovial lining cells (large arrows), endothelial cells of the blood vessels (arrowheads), and fibroblasts (small arrows). (D) Flt-1-stained sections obtained from the same patient (insert, low power). Immunoreactivities for Flt-1 are observed in the endothelial cells of the blood vessels (arrowheads) and synovial lining cells (large arrows). (E) A specimen obtained from the same patient incubated with CD34 antibody (insert, low power). Beneath the synovial lining cells, lumen structures of various sizes are immuno-positive, making them easy to distinguish from other cells. (F) FGF-2-stained sections obtained from a control patient. Only a few immuno-positive cells were observed in the synovial lining cells (arrows). (G) FGFR-1, (H) VEGF, and (I) Flt-1-stained sections obtained from a control patient. No immunoreactivities are observed in the synovial tissue. (J) A control specimen incubated with CD34 antibody showing only a few immunopositive cells (arrows) (bar, 50 μm).

Figure 2.

Correlation between the percentage of immuno-positive cells and microvessel density. (A) (FGF-2) No significant correlation is observed by Spearman rank correlation coefficient (p = 0.13, r = 0.23). (B) FGFR-1, (C) VEGF, (D) Flt-1. A significant correlation is observed (FGFR-1, p = 0.025, r = 0.34; VEGF, p = 0.0003, r = 0.53; Flt-1, p < 0.0001, r = 0.68).

Footnotes

Acknowledgements

This study was supported by a Grant for Specially Promoted Research from Kanazawa Medical University. One of the primary antibodies used in this study (b-FM1) was supplied by Professor Katsuzo Nishikawa.

References

Aiello LP, Pierce EA, Foley ED, Takagi H, Chen H, Riddle L, et al. -1995-. Suppression of retinal neovascularization in vivo by inhibition of vascular endothelial growth factor -VEGF- using soluble VEGF-receptor chimeric proteins. Proc Natl Acad Sci USA 92:10457–10461.

Berse B, Brown LF, Van de Water L, Dvorak HF, Senger DR -1992-. Vascular permeability factor -vascular endothelial growth factor- gene is expressed differentially in normal tissues, macrophages, and tumors. Mol Biol Cell 3:211–220.

Brogi E, Schatteman G, Wu T, Kim EA, Varticovski L, Ket B, et al. -1996-. Hypoxia-induced paracrine regulation of vascular endothelial growth factor receptor expression. J Clin Invest 97:469–476.

Burke D, Wilkes D, Blundell TL, Malcolm S -1998-. Fibroblast growth factor receptors: lessons from the genes. Trends Biochem Sci 23:59–62.

Carmeliet P, Jain RK -2000-. Angiogenesis in cancer and other diseases. Nature 407:249–257.

Decaussin M, Sartelet H, Robert C, Moro D, Claraz C, Brambilla C, et al. -1999-. Expression of vascular endothelial growth factor -VEGF- and its two receptors -VEGF-R1-FLt-1 and VEGF-R2-FLk1/KDR- in non-small cell lung carcinomas -NSCLCs-: correlation with angiogenesis and survival. J Pathol 188:369–377.

Dionne CA, Crumley G, Bellot F, Kaplow JM, Searfoss G, Ruta M, et al. -1990-. Cloning and expression of two distinct high-affinity receptors cross reacting with acidic and basic fibroblast growth factors. EMBO J 9:2685–2692.

Folkman J, Watson K, Ingber D, Hanahan D -1989-. Induction of angiogenesis during the transition from hyperplasia to neoplasia. Nature 339:58–61.

Fu K, Ma X, Zhang Z, Pang X, Chen W -1995-. Interleukin-6 in synovial fluid and HLA-DR expression in synovium from patients with temporomandibular disorders. J Orofac Pain 9:131–137.

10.

Galzie Z, Kinsella AR, Smith JA -1997-. Fibroblast growth factors and their receptors. Biochem Cell Biol 75:669–685.

11.

Gynther GW, Holmlund AB, Reinholt FP -1994-. Synovitis in internal derangement of the temporomandibular joint: correlation between arthroscopic and histologic findings. J Oral Maxillofac Surg 52:913–917.

12.

Hsu SM, Raine L, Fanger H -1981-. Use of avidin-biotin-peroxidase complex -ABC- in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody -PAP- procedures. J Histochem Cytochem 29:577–580.

13.

Jackson JR, Minton JA, Ho ML, Wei N, Winkler JD -1997-. Expression of vascular endothelial growth factor in synovial fibroblast is induced by hypoxia and interleukin 1 beta. J Rheumatol 24:1253–1259.

14.

Jin KL, Mao XO, Greenberg DA -2000-. Vascular endothelial growth factor: direct neuroprotective effect in in vitro ischemia. Proc Natl Acad Sci USA 97:10242–10247.

15.

Koch AE -1998-. Angiogenesis: implication of rheumatoid arthritis. Arthritis Rheum 41:951–962.

16.

Koch AE, Harlow LA, Haines GK, Amento EP, Unemori EN, Wong WL, et al. -1994-. Vascular endothelial growth factor. A cytokine modulating endothelial function in rheumatoid arthritis. Am J Immunol 152:4149–4156.

17.

Kubota E, Kubota T, Matsumoto J, Shibata T, Murakami KI -1998-. Synovial fluid cytokines and proteinases as markers of temporomandibular joint disease. J Oral Maxillofac Surg 56:192–198.

18.

Matsuzaki K, Yoshitake Y, Matsuo Y, Sasaki H, Nishikawa K -1989-. Monoclonal antibodies against heparin-binding growth factor/basic fibroblast growth factor that block its biological activity: invalidity of the antibodies for tumor angiogenesis. Proc Natl Acad Sci USA 86:9911–9915.

19.

Merrill RG, Yih WY, Langan MJ -1990-. A histologic evaluation of the accuracy of TMJ diagnostic arthroscopy. Oral Surg 70:393–398.

20.

Ohta T, Yamamoto M, Numata M, Iseki S, Tsukioka Y, Miyashita T, et al. -1995-. Expression of basic fibroblast growth factor and its receptor in human pancreatic carcinomas. Br J Cancer 72:824–831.

21.

Pufe T, Petersen W, Tillmann B, Mentlein R -2001-. Splice variants VEGF 121 and VEGF 165 of the angiogenic peptide vascular endothelial cell growth factor are expressed in the synovial tissue of patients with rheumatoid arthritis. J Rheumatol 28:1482–1485.

22.

Sato J, Segami N, Suzuki T, Kaneyama K, Yoshitake Y, Nishikawa K -2002-. The expression of vascular endothelial growth factor in synovial tissues in patients with internal derangement of the temporomandibular joint. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 93:251–256.

23.

Schulze-Osthoff K, Risau W, Vollmer E, Sorg C -1990-. In situ detection of basic fibroblast growth factor by highly specific antibodies. Am J Pathol 137:85–92.

24.

Segami N, Kaneyama K, Tsurusako S, Suzuki T -1999-. Arthroscopic eminoplasty for habitual dislocation of the temporomandibular joint: preliminary study. J Craniomaxillofac Surg 27:390–397.

25.

Suzuki T, Segami N, Kaneyama K, Nishimura M, Nojima T -1999-. Specific expression of interleukin-1beta in temporomandibular joints with internal derangement: correlation with clinical findings. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 88:413–417.

26.

Szebenyi G, Fallon JF -1999-. Fibroblast growth factor as multifunctional signaling factors. Int Rev Cytol 185:45–106.

27.

Takahashi T, Kondoh T, Fukuda M, Yamazaki Y, Toyosaki T, Suzuki R -1998-. Proinflammatory cytokines detectable in synovial fluids from patients with temporomandibular disorders. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 85:135–141.

28.

Weidner N, Carroll PR, Flax J, Blumenfeld W, Folkman J -1993-. Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma. Am J Pathol 143:401–409.