A role for the sensory neuropeptide calcitonin gene-related peptide in endothelial cell proliferation in vivo

Abstract

Background and purpose: We have tested the hypothesis that calcitonin gene-related peptide (CGRP) is a mediator of capsaicin-induced angiogenesis in vivo.

Experimental approach: In a series of experiments, the knee joints of rats were injected with CGRP, capsaicin or vehicle control. Groups of animals (n=6) were treated with the CGRP receptor antagonist BIBN4096BS and/or the NK₁ receptor antagonist SR140333. Endothelium, proliferating endothelial cell nuclei and macrophages were identified 24 h later in the synovium by immunohistochemistry and quantified by image analysis. mRNA for the receptors for CGRP and adrenomedullin were sought in normal and inflamed rat and human synovia using RT-PCR.

Key results: Intra-articular CGRP injection increased the endothelial cell proliferation index, whereas macrophage infiltration and knee joint diameters were similar to saline-injected controls. CGRP-induced endothelial cell proliferation was dose-dependently inhibited by BIBN4096BS. mRNA for adrenomedullin and the CGRP receptor subunits were detected in normal and inflamed human and rat synovia. In capsaicin-induced synovitis, the increased endothelial cell proliferation index was partially blocked by administration of NK₁ or CGRP antagonists individually and was reduced to the level of saline controls by coadministration of both receptor antagonists.

Conclusions and implications: These data support the hypothesis that CGRP stimulates angiogenesis in vivo directly by activating CGRP receptors. Capsaicin-induced endothelial cell proliferation was completely blocked by coadministration of CGRP and NK₁ receptor antagonists, indicating that both CGRP and substance P may contribute to angiogenesis in this model of synovitis.

Figure 1

Intra-articular administration of exogenous CGRP causes a dose-dependent increase in the endothelial cell proliferation index in the synovium. (A). PCNA indices in saline-injected controls and 24 h after intra-articular injection of CGRP from 0.025 to 2.5 nmol per knee. The two higher doses of CGRP caused an increase in the PCNA index (P < 0.01 and P < 0.001, respectively) compared with saline-injected controls. Data shown are means ± SD; n = 6 rats per group. (B). Main panel: Synovium from a rat 24 h after intra-articular injection of 2.5 nmol CGRP, showing CD31-positive vascular cells and PCNA-positive proliferating nuclei. Counterstaining with DAPI shows all other nonstained nuclei. Arrows indicate PCNA-positive endothelial cells. This allows determination of the PCNA index as the percentage of vascular nuclei, which were positive for the proliferation marker. Inset shows a blood vessel from a control saline-injected synovium not displaying PCNA-positive endothelial cell nuclei. Bar = 100 µm.

Figure 2

Blockade of the effect of exogenous intra-articular CGRP administration on the endothelial cell proliferation index using the specific CGRP receptor antagonist BIBN4096BS. Bar 1 shows the baseline PCNA index in naïve animals, negative control. Bar 2 shows the response to saline alone injection, which was not significantly different from naïve control. Bar 3 shows the PCNA index in animals intra-articularly injected with saline and i.v. injected with the CGRP receptor antagonist BIBN 4096BS. Intra-articular saline did not significantly increase the PCNA index, negative control. Bar 4 shows the effect of intra-articular administration of 2.5 nmol CGRP (positive control) on the PCNA index. The effect of CGRP on the PCNA index was diminished by the CGRP receptor antagonist BIBN406BS (bars 5 and 6) when compared with CGRP injection alone. Rats were given CGRP 2.5 nmol alone or together with BIBN406BS at either 0.07 µmol per rat (bar 5) 0.7 µmol per rat (bar 6). P < 0.001 for both treatments. Data shown are means ± SD; n = 6 rats per group.

Figure 3

Capsaicin-induced angiogenesis was partially inhibited by the administration of each receptor antagonist, and coadministration of BIBN1096BS and SR140333 completely blocked capsaicin-enhanced angiogenesis. (A) Endothelial PCNA indices of rat synovia from knees injected with 0.5% capsaicin were significantly reduced by either the SP receptor antagonist SR140333 (bar 3) or by the CGRP receptor antagonist BIBN4096BS (bar 4) when compared with the positive control capsaicin alone, bar 2; P < 0.001 for both treatments. When given in combination, the two receptor antagonists (bar 5) reduced the rise in PCNA index attributable to capsaicin to the same level as saline controls; P > 0.05. PCNA index was defined as the percentage of endothelial cell nuclei positive for PCNA. Data shown are means ± SD; n = 6 rats per group. (B). Knee joint diameters corresponding to the same groups in (A), showing significant inhibition of joint swelling by SR140333, either alone (bar 2, P < 0.01) or in combination with BIBN4096BS (bar 5, P < 0.01). The change in knee diameter from baseline to 24 h is shown in mm, as means ± SD; n = 6 rats per group.. In each panel, the negative control (bar 1) represents data from synovia 24 h after intra-articular saline injection accompanied by i.v. injection of the CGRP receptor antagonist BIBN4096BS and i.p. injection of the SP receptor antagonist SR140333. The positive control (bar 2) represents data from synovia 24 h after intra-articular injection of capsaicin in the absence of either antagonist. Capsaicin is believed to release endogenous CGRP and SP from sensory nerves in the synovium. All groups of rats (bars 2–5), with the exception of saline controls (bar 1), were treated with capsaicin.

Figure 4

Detection of macrophages in the synovium. (A). Low levels of macrophage infiltration in all groups. There were no statistically significant differences in macrophage fractional areas for each group of rats, indicating that any small cellular inflammatory response present was below the detection threshold. Data shown are means ± SD; n = 6 rats per group Macrophages in the synovium were detected by immunohistochemistry using the primary antibody ED1 and were stained with nickel/DAB and appear as black. (B) Synovium from a capsaicin-treated animal. (C) Synovium from a saline-injected knee of an animal treated with the CGRP receptor antagonist BIBN4096BS i.v. and the SP receptor antagonist SR140333 i.p. The number of positively staining cells in each panel appears similar. When quantified, there were no statistically significant differences in the macrophage fractional area between any of the groups. Bar = 100 µm.

Figure 5

Components of the CGRP and adrenomedullin receptor-ligand complex are expressed in rat and human synovia. (A). Products of RT-PCR for GAPDH, adrenomedullin (AdM), CRLR, RAMP-1, RAMP-2 and RAMP-3 are shown after separation on agarose gels. Samples from synovia taken 24 h after intra-articular injection with capsaicin (C1–C6, lanes 1–6) and naïve knees (N1–N6, lanes 7–12) were analysed. Positive controls (+, lane 13) were rat brain cDNA. Negative controls (–, lane 14) were PCR's without any template DNA. (B). Examples of PCR products from RT-PCR on human synovia for CRLR, RAMP-1, RAMP-2, RAMP-3 and adrenomedullin are shown after separation on agarose gels. Samples from OA without inflammation (OA(n), lanes 1, 2), severely inflamed OA (OA(i), lanes 3, 4), normal post mortem (PM; lanes 5, 6) and RA (lane 7) are shown. Positive controls (+, lane 9) were HUVEC cells for CRLR and DU145 cells for RAMPs-1, -2 and -3. Negative controls (–, lane 10) were PCRs without any template DNA. (C). Table showing detection rates for PCR products in human synovia.