INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ADRENOMEDULLIN (ADM) and proadrenomedullin NH₂-terminal 20 peptide (PAMP) are distinct products encoded by the ADM gene (20-23). The human form of ADM consists of 52 amino acids and a six-membered ring structure that shares homology with CGRP and amylin (20-23). PAMP consists of 20 amino acids and does not share the six-membered ring structure (20-22). ADM was discovered in human pheochromocytoma cells and has been localized in many tissues, including vascular smooth muscle cells and the endothelium (11, 13, 15, 16, 20). ADM plasma levels are increased in disorders, such as hypertension, heart and renal failure, and hypoxia, and ADM induces a different spectrum of hemodynamic effects than do nitrovasodilators (8, 12-14, 18, 23, 26, 35). PAMP is found in the NH₂-terminal portion of the precursor for ADM and has a similar distribution, including vascular smooth muscle and the endothelium (17, 20, 24, 36). CGRP is a spliced variant of the calcitonin gene and this neuropeptide is present in perivascular sensory nerves (29). ADM, PAMP, and CGRP have vasodilator activity with different mechanisms of action (4, 6, 7, 16, 17, 27-29, 35). ADM has been reported to relax vascular muscle in an endothelium-dependent manner in some vascular beds studied, and those responses are species or vascular bed dependent, whereas PAMP has modest activity and has been reported to inhibit norepinephrine release or decrease vascular resistance by a nitric oxide-independent mechanism (4-6, 17, 27, 28, 33-35). CGRP is a potent vasodilator that has been reported to act as an endothelium-independent vasodilator agent in most arteries but has been reported to act by releasing nitric oxide in some arteries studied (7, 29, 30, 32).

The functional expression of receptors for CGRP and other vasoactive peptides has been described in the human thymus gland, and it has been postulated that these receptors may play a functional role in pathophysiological processes in this endocrine organ (25). Although CGRP receptors are present in the human thymus, little if anything is known about the actions of CGRP and related peptides on vascular function in small arteries from the thymus (25). Moreover, there is a paucity of information about the actions and mechanisms of action of CGRP, ADM, and PAMP in small human arteries (9, 27, 29, 30, 32, 36). The present study was, therefore, undertaken to investigate responses and the role of endothelium-derived nitric oxide in mediating responses to CGRP, ADM, and PAMP in small arteries from the human thymus.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Whole thymus tissue was obtained from pediatric patients undergoing cardiothoracic surgery at Tulane University Medical Center under a protocol for discarded tissue approved by the Institutional Committee on Human Tissue Research and follows the "Guiding Principles for Research Involving Animals and Human Beings" by the American Physiological Society (1). The tissue was immediately immersed into cold Krebs bicarbonate buffer (119 mM NaCl, 24 mM NaHCO₃, 4.7 mM KCl, 1.18 mM KH₂PO₄, 1.17 mM MgSO₄, 1.6 mM CaCl₂, and 5.5 mM glucose) aerated with 95% O₂-5% CO₂. Branched artery segments 1.5 to 2 mm in length were gently dissected from the thymus and cleared of adhering connective tissue. The arterial segments were transferred to the myograph chamber (Living Systems Instrumentation) containing Krebs bicarbonate buffer aerated with 95% O₂-5% CO₂, mounted on a glass micropipette of known internal diameter, and secured with strands of 10-0 nylon suture. Luminal cellular debris was gently flushed with buffer solution, and the distal lumen of the vessel was tied with suture to create a blind pouch configuration. The vessels were allowed to equilibrate for 45 min as temperature was slowly increased to 37°C and intraluminal pressure was gradually increased to 50 to 70 mmHg. Transmural pressure was maintained constant (±1 mmHg) with a microliter peristaltic pump with an automatic pressure servo-control.

The myograph chamber was placed on the mechanical stage of a TMS-F inverted stage microscope (Nikon Instruments), and video images were captured with a high-resolution CCD camera (Burle) and monitor (Panasonic TR-930B, Matsushita Electronic Industrial). Lumen diameter and wall thickness were monitored by a Video Dimension Analyzer (Living Systems Instrumentation) calibrated against a stage micrometer and recorded on a chart recorder (Western Graphtec).

All agonists and antagonists were added to the buffer superfusion reservoir and were allowed to recirculate through the myograph chamber (total volume 80 ml). The myograph chamber was continuously perfused at 37°C with Krebs bicarbonate buffer equilibrated with 21% O₂-5% CO₂- balance N₂ at pH 7.4. Both superfusion flow rate and gas flow rate were adjusted to maintain pH and temperature, which were continuously monitored with a YSI temperature probe placed in the myograph chamber and a pH probe in the reservoir. As an index of the maximal vasoconstrictor response, the arteries were initially challenged with 60 mM KCl solution. Arteries that demonstrated <50% decrease in resting diameter were not used. After washout of the KCl solution, the arteries were submaximally precontracted with the thromboxane mimic U-46619 at 30-100 nM or with endothelin-1 (ET-1) (30-100 nM).

Preparation of drugs. Human PAMP, PAMP(12-20), human synthetic ADM, CGRP(8-37) (Peptide Research Labs, Tulane Medical School), and human CGRP (Peptides International) were dissolved in 0.9% NaCl. Acetylcholine bromide, sodium nitroprusside, and ET-1 (Sigma) were dissolved in 0.9% NaCl. N-nitro-L-arginine (L-NA) (Sigma) was dissolved in acidified saline. 1H-[1,2,4]-oxadiazolo[4,3-a]quinoxalin-1-one (ODQ; Sigma) was dissolved in 95% ethanol and diluted with 0.9% NaCl. U-37883A (Upjohn), Rp-cyclic 3'-5'-hydrogen phosphorothiate adenosine (Rp-cAMPS), and 9-(tetrahydro-2-furanyl)-9H-purin-6-amine (SQ-22536; Sigma) were dissolved in 0.9% NaCl with sonication. U-46619 (Upjohn) was dissolved in 100% ethanol and diluted with 0.9% NaCl. Levcromakalim (SmithKline Beecham) was dissolved in a 20% ethanol-saline solution at a concentration of 1 mg/ml and was then diluted with 0.9% NaCl. The vehicles for these agents had no significant effect on resting vessel diameter or on responses to the vasoactive agents. The drug solutions were stored in a freezer in amber bottles, and working solutions were prepared on a frequent basis and kept on crushed ice.

Statistics. Data are expressed as percent relaxation of vessels preconstricted with U-46619 as means ± SE. Data were analyzed using a one-way analysis of variance with repeated measures and Scheffé's F-test or a paired t-test. A P value of <0.05 was used as the criterion for statistical significance.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Responses to the peptides. The effects of human CGRP, ADM, PAMP, and PAMP(12-20) on human thymic arteries 90 to 249 µm in diameter with intact endothelium preconstricted to 60 to 75% of the response to 60 mM KCl with U-46619 are shown in Fig. 1. Superfusion with CGRP, ADM, and PAMP at bath concentrations of 10¹⁰ to 10⁵ M produced concentration-dependent increases in arterial diameter with a similar maximal response (Fig. 1). In terms of relative vasodilator activity, CGRP and ADM had similar activity with an EC₅₀ of ~10⁸ M, whereas the dose-response curve for PAMP was 1 log unit to the right of the curves for CGRP and ADM with an approximate EC₅₀ of 10⁷ M (Fig. 1). PAMP(12-20) was a partial agonist, and the maximal response to PAMP(12-20) was significantly smaller than the maximal vasodilator response observed with CGRP, ADM, or PAMP (Fig. 1). The dose-response curve for the truncated PAMP analog had a shallower slope (Fig. 1).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Concentration-response curves comparing increases in arterial diameter (percent relaxation) in response to superfusion with human adrenomedullin (hADM), human calcitonin gene-related peptide (hCGRP), human proadrenomedullin NH2-terminal 20 peptide (hPAMP), and the truncated analog PAMP(12-20) in 3rd- and 4th-order arteries (90-249 µm in diameter) with intact endothelium from the human thymus gland. The arteries were preconstricted with the thromboxane A2 mimic U-46619; n = 10-13 experiments.

Role of vascular endothelium. The role of the vascular endothelium in mediating vasodilator responses to CGRP, ADM, and PAMP was investigated in small thymic arteries, and these data are shown in Fig. 2A. After removal of the endothelium with 1 ml of air perfused over a 20- to 25-min period, vasodilator responses to ADM and PAMP were significantly reduced when the arteries were preconstricted with U-46619 (Fig. 2A). Vasodilator responses to ACh were abolished after endothelial denudation, and a small but statistically significant decrease in arterial diameter (vasoconstrictor response) was observed (Fig. 2A). Vasodilator responses to sodium nitroprusside (10⁷ M) and levcromakalim (10⁶ M) were not altered, whereas the response to PAMP(12-20) (10⁶ M) was reduced significantly after endothelial denudation (data not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 2.   A: increases in arterial diameter in response to hADM, hPAMP, hCGRP, and ACh in thymic arteries with intact vascular endothelium (+ endothelium) and after the vascular endothelium was denuded ( endothelium). The thymic arteries were preconstricted with U-46619; n = 8-10 experiments. *P < 0.05. B: increases in arterial diameter (percent relaxation) in response to hADM, hPAMP, hCGRP, and ACh before and after treatment with the nitric oxide synthase inhibitor N-nitro-L-arginine (L-NA) (103 M) in endothelium-intact arteries preconstricted with U-46619; n = 8-9 experiments. *P < 0.05.

Role of nitric oxide formation. The role of nitric oxide release in mediating vasodilator responses to CGRP, ADM, and PAMP was investigated, and these data are summarized in Fig. 2B. L-NA decreased baseline diameter by 18 ± 2% from an average resting diameter of 173 ± 14 µM, and the vessels were then preconstricted with U-46619 to a similar level of tone as observed in endothelium-intact arteries with U-46619. After superfusion with the nitric oxide synthase inhibitor L-NA (10³ M), vasodilator responses to ADM, PAMP, and ACh were significantly reduced compared with control responses when the thymic arteries were preconstricted with U-46619 (Fig. 2B). In these experiments, vasodilator responses to CGRP were not changed by the administration of L-NA (Fig. 2B). Vasodilator responses to PAMP(12-20) (10⁶ M) were reduced significantly, whereas the response to levcromakalim (10⁷ M) was not altered and the response to sodium nitroprusside (10⁷ M) was enhanced significantly after administration of L-NA (10³ M) (data not shown).

Role of cGMP. The role of the activation of nitric oxide-sensitive guanylyl cyclase and cGMP in mediating vasodilator responses to ADM, PAMP, and CGRP was investigated in experiments with the soluble guanylyl cyclase inhibitor ODQ. ODQ decreased baseline diameter by an average of 15% from a resting diameter of 170 ± 12 µM, and after superfusion with ODQ (3 µM), vasodilator responses to ADM, PAMP, sodium nitroprusside, and ACh were reduced significantly when baseline diameter was decreased to similar values with U-46619 (Fig. 3A). Vasodilator responses to CGRP were not altered by the administration of ODQ (Fig. 3A). The administration of ODQ significantly decreased the vasodilator response to PAMP(12-20) (10⁶ M) but did not change the vasodilator response to levcromakalim (10⁷ M) (data not shown).

View larger version (35K): [in this window] [in a new window]   Fig. 3.   A: increases in arterial diameter (percent relaxation) in response to hADM, hPAMP, hCGRP, sodium nitroprusside (SNP), and ACh before and after treatment with the soluble guanylyl cyclase inhibitor (ODQ) (3 µM) in endothelium-intact preconstricted arteries; n = 8-9 experiments. *P < 0.05. B: increases in arterial diameter (percent relaxation) in response to hADM, hPAMP, hCGRP, and ACh before and after treatment with the CGRP1 receptor antagonist CGRP(8-37) (106 M) in endothelium-intact small arteries from the human thymus preconstricted with U-46619; n = 7-9 experiments. *P < 0.05.

Role of the CGRP1 receptor. The role of the CGRP1 receptor in mediating vasodilator responses to the peptides was investigated in experiments with the CGRP1 receptor antagonist CGRP(8-37), and these data are shown in Fig. 3B. Vasodilator responses to CGRP and ADM were significantly reduced after superfusion with CGRP(8-37) (10⁶ M) when the thymic arteries were preconstricted with U-46619 (Fig. 3B). Vasodilator responses to PAMP and ACh were not altered by the administration of CGRP(8-37) (10⁶ M) (Fig. 3B). This concentration of CGRP(8-37) did not alter resting vessel diameter. Vasodilator responses to PAMP(12-20) (10⁶ M) and levcromakalim (10⁷ M) were not altered by CGRP(8-37) (data not shown).

Role of the K channel. The role of K channel activation in mediating responses to the vasodilator peptides was investigated, and these data are summarized in Fig. 4A. Superfusion with the K channel antagonist U-37883A in a concentration of 10⁶ M did not alter vasodilator responses to CGRP, ADM, and PAMP, whereas the vasodilator response to the K channel opener levcromakalim was reduced significantly in preconstricted thymic arteries (Fig. 4A).

View larger version (34K): [in this window] [in a new window]   Fig. 4.   A: increases in arterial diameter (percent relaxation) in response to hADM, hPAMP, hCGRP, and levcromakalim (LK) before and after treatment with the vascular-selective K channel antagonist U-37883A. The endothelium-intact, small thymic arteries were preconstricted with U-46619; n = 7-8 experiments. *P < 0.05. B: increases in arterial diameter (percent relaxation) in response to hCGRP, isoproterenol, and ACh before and after treatment with the adenylyl cyclase inhibitor SQ-22536 (3 × 104 M) (left) and before and after treatment with the PKA inhibitor Rp-cAMPS (3 × 105 M) (right); n = 4-7 experiments. *P < 0.05.

Effect of SQ-22536 and Rp-cAMPS. The role of adenylyl cyclase activation and PKA in mediating the vasodilator response to CGRP was investigated, and these data are summarized in Fig. 4B. Superfusion with the adenylyl cyclase inhibitor SQ-22536 in a concentration of 3 × 10⁴ M significantly decreased vasodilator responses to CGRP and to isoproterenol without altering the response to ACh (Fig. 4B). Superfusion with the PKA inhibitor Rp-cAMPS in a concentration of 3 × 10⁵ M significantly attenuated vasodilator responses to CGRP and to isoproterenol without changing the response to ACh (Fig. 4B).

6

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of the present investigation show that human CGRP, ADM, and PAMP have significant vasodilator activity in small arteries from the human thymus. Vasodilator responses were concentration dependent, and the three peptides elicited similar maximal vasodilator responses when the vessels were preconstricted with U-46619, with similar vasodilator responses in vessels preconstricted with ET-1. In terms of relative potency, CGRP and ADM were similar with an approximate ED₅₀ of 10⁸ M and were 10-fold more potent than PAMP. The truncated PAMP analog PAMP(12-20) was a partial agonist that elicited ~50% of the maximal vasodilator response to the full-sequence peptide, suggesting that activity was retained when the first 11 amino acids were deleted, but that these residues may be necessary for the full expression of vasodilator activity. Although PAMP(12-20) is a partial agonist in human thymic arteries, this truncated peptide does not inhibit Ca²⁺-dependent agonist-stimulated aldosterone secretion (2). The reason for the difference in activity is uncertain but may suggest differences in activity of the truncated peptide on receptors in small thymic arteries and in the human adrenal cortex.

After endothelial removal, responses to ADM and PAMP were reduced, whereas the response to CGRP was not altered, and the vasodilator response to ACh was blocked, indicating that responses to ADM and PAMP are, in part, endothelium dependent in small human thymic arteries. Vasodilator responses to ADM, PAMP, and ACh were reduced by L-NA, whereas the nitric oxide synthase inhibitor did not reduce the response to CGRP, suggesting that responses to ADM and PAMP and to ACh are dependent, in part, on the release of nitric oxide from the endothelium. Nitric oxide activates soluble guanylyl cyclase and, following administration of ODQ, vasodilator responses to ADM and PAMP were significantly reduced, whereas the response to CGRP was not altered and responses to sodium nitroprusside and ACh were significantly decreased (3). These results suggest that endothelium-dependent vasodilator responses to ADM and PAMP, as well as to ACh, and the endothelium-independent response to nitroprusside are mediated by the activation of soluble guanylyl cyclase in small thymic arteries by nitric oxide released from the endothelium or from the nitric oxide donor sodium nitroprusside (3). The observation that the response to sodium nitroprusside was attenuated by ODQ is consistent with the observation that responses to this nitric oxide donor are inhibited by methylene blue and methemoglobin and provides support for the hypothesis that the response is mediated by activation of nitric oxide-sensitive guanylyl cyclase and increased cGMP formation in small thymic arteries (10).

The role of the CGRP1 receptor was investigated using the CGRP1 receptor antagonist CGRP(8-37) and shows that responses to CGRP and ADM are attenuated, whereas the response to PAMP was not altered. These data suggest that responses to CGRP and ADM are mediated by the activation of a CGRP(8-37)-sensitive receptor. However, the location of the CGRP(8-37)-sensitive receptor and signal transduction mechanism appears to be different for CGRP and ADM. Experiments with endothelial removal suggest that the ADM-sensitive CGRP(8-37) receptor is located on the vascular endothelium, whereas the CGRP-sensitive CGRP(8-37) receptor is not on the endothelium and is probably located on smooth muscle. The results with L-NA and ODQ indicate that the response to ADM stimulation of the endothelial CGRP(8-37)-sensitive receptor involves the release of nitric oxide and the activation of soluble guanylyl cyclase, whereas the response to CGRP activation of the CGRP(8-37)-sensitive receptor is independent of the vascular endothelium, nitric oxide release, or the activation of soluble guanylyl cyclase. The effects of selective ADM receptor antagonists on responses to ADM and CGRP would be of interest in future studies.

CGRP has been reported to act as a hyperpolarizing vasodilator by opening K channels in small arteries from the rabbit mesentery (30). However, the present results show that responses to CGRP, as well as to ADM and PAMP, are not altered by a K channel antagonist in a concentration that reduced vasodilator responses to the K channel opener levcromakalim. These data suggest that responses to CGRP are independent of K channel activation in small human thymic arteries and that a role for K channels may depend on the specific artery or species studied. It is possible that CGRP may hyperpolarize smooth muscle by activating a different potassium channel in small thymic arteries or by increasing cAMP levels, as has been reported in vascular smooth muscle (7, 16).

The role of a cAMP-dependent mechanism was examined in experiments with the adenylyl cyclase inhibitor SQ-22536 and the PKA inhibitor Rp-cAMPS (14, 31). These studies show that vasodilator responses to CGRP and to the -adrenergic agonist isoproterenol were decreased, whereas responses to ACh were unchanged and suggest that activation of adenylyl cyclase and PKA is involved in mediating the vasodilator response to CGRP in thymic arteries.

It has been reported that PAMP decreases arterial pressure and regional vascular resistance but has modest vasodilator activity compared with ADM (4-7, 17, 19, 33). It has also been reported that PAMP may induce vasodilation by inhibiting norepinephrine release or by a cAMP-dependent mechanism (5, 17, 19, 33, 34). In the present study, PAMP was shown to have full vasodilator activity in small thymic arteries but was 10-fold less potent than ADM or CGRP. Responses to PAMP were reduced by endothelial removal, L-NA, and ODQ, suggesting that in human thymic arteries this product of the ADM gene acts as an endothelium-dependent vasodilator agent signaling by a nitric oxide-guanylyl cyclase-dependent mechanism previously not reported in small human arteries. The present data show that responses to PAMP are not altered by CGRP(8-37), indicating that PAMP receptors on the vascular endothelium are distinct from ADM receptors and that PAMP receptors have a lower affinity for their ligand than ADM receptors for ADM in small thymic arteries.

In regard to a physiological role for CGRP and other vasoactive peptides in small thymic arteries, the molecular and functional expression of CGRP receptors has been characterized in the human thymus (25). CGRP receptor transcripts were identified, and CGRP increased cAMP levels in human thymic epithelial cell cultures (25). It has been suggested that CGRP may exert a functional role in pathophysiological processes, such as cancer and during production of antinicotinic receptor antibodies in the human thymus (25). The present data show that CGRP has significant vasodilator activity in small arteries from the human thymus, suggesting that CGRP could play a role in the regulation of vascular tone in this endocrine organ.

In summary, the present results show that CGRP, ADM, and PAMP dilate small arteries from the human thymus. Vasodilator responses to ADM and PAMP are endothelium dependent and involve the release of nitric oxide and the activation of soluble guanylyl cyclase. In contrast, vasodilator responses to CGRP are endothelium independent and are mediated by a cAMP-dependent mechanism. Vasodilator responses to ADM and CGRP are attenuated by CGRP(8-37), suggesting that distinct CGRP(8-37)-sensitive receptors recognizing ADM are located on the endothelium and that CGRP(8-37)-sensitive receptors recognizing CGRP are located on smooth muscle cells of small thymic arteries. The present data, along with results showing the presence of CGRP receptors in the human thymus, are consistent with the hypothesis that CGRP may have a role in regulating vascular function in the thymus gland in physiological and pathophysiological conditions. The results of the present study also indicate that ADM and PAMP could have a significant regulatory effect on vascular tone in the human thymus.