Characterisation of CGRP receptors in human and porcine isolated coronaryarteries: Evidence for CGRP receptor heterogeneity
Saurabh Gupta a, Suneet Mehrotra a, Carlos M. Villalón b, Ingrid M. Garrelds a, René de Vries a,Jorge P. van Kats c, Hari S. Sharma a, Pramod R. Saxena a, Antoinette MaassenVanDenBrink a,⁎
a Department of Pharmacology, University Medical Center Rotterdam, P.O. Box 1738, 3000 DR Rotterdam, The Netherlandsb Departamento de Farmacobiología, Cinvestav-Coapa, Czda. de los Tenorios 235, Col. Granjas-Coapa, Deleg. Tlalpan, C.P. 14330, México D.F., México
c Thoracic Surgery and Heart Valve Bank, Erasmus MC, University Medical Center Rotterdam, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands
Received 12 May 2005; received in revised form 1 November 2005; accepted 8 November 2005
Abstract
This study sets out to characterise calcitonin gene-related peptide (CGRP) receptors in human and porcine isolated proximal and distalcoronary arteries using BIBN4096BS. Human (h)-αCGRP induced relaxations that were blocked by BIBN4096BS in all arteries studied. Incontrast to the other vessels, the Schild plot slope in the human distal coronary artery segments (0.68±0.07) was significantly less than unity andBIBN4096BS potently blocked these responses (pKb (10 nM): 9.29±0.34, n=5). In the same preparation, h-αCGRP8-37 behaved as a weakantagonist of h-αCGRP-induced relaxations (pKb (3 μM): 6.28±0.17, n=4), with also a Schild plot slope smaller than unity. The linear agonists,[ethylamide-Cys2,7]-h-αCGRP ([Cys(Et)2,7]-h-αCGRP) and [acetimidomethyl-Cys2,7]-h-αCGRP ([Cys(Acm)2,7]-h-αCGRP), had a high potency(pEC50: 8.21±0.25 and 7.25±0.14, respectively), suggesting the presence of CGRP2 receptors, while the potent blockade by BIBN4096BS (pKb
The human calcitonin gene-related peptide (h-αCGRP), a37-amino acid peptide, is one of the most potent endogenousvasodilators known. It exists in two forms, h-αCGRP and h-βCGRP, which differ from each other by three amino acids. Inhumans, these forms have almost similar biological actions(Quirion et al., 1992; Van Rossum et al., 1997), but are encodedby separate genes. Although CGRP was first described in 1982(Amara et al., 1982) the classification of CGRP receptors has
been painstakingly slow because of the lack of selective andpotent ligands.
Presently, CGRP receptors are functionally classified intoCGRP1 and CGRP2 types. The CGRP1 receptor has been clonedand consists of at least three main entities, namely, thecalcitonin receptor like receptor (CLR), receptor activitymodifying protein-1 (RAMP-1) and receptor component protein(RCP) (Luebke et al., 1996; Poyner et al., 2002); in contrast, theCGRP2 receptor, described in some animal tissues (Dumont etal., 1997; Quirion et al., 1992), has not yet been reported inhumans or deciphered molecularly. Therefore, the classificationand characterisation of CGRP receptors are primarily based ondifferent functional pharmacological responses. In this respect,the C-terminal fragment h-αCGRP8-37 has a higher antagonistpotency at the prototypic CGRP1 receptor described in guinea-
108 S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
pig atrium (pKb: 7–8) than at the CGRP2 receptor described inrat vas deferens (pKb: 5.5–6.5) (Dennis et al., 1989; Quirion etal., 1992). However, the range of antagonist affinities reportedfor h-αCGRP8-37 within and between different species (Poyneret al., 2002) is too wide to be explained by the existence of justtwo receptors. Further, the CGRP2 receptor seems to be moresensitive to the linear agonists [ethylamide-Cys2,7]-h-αCGRP([Cys(Et)2,7]-h-αCGRP) and [acetimidomethyl-Cys2,7]-h-αCGRP ([Cys(Acm)2,7]-h-αCGRP) (pEC50: ≈7) (Dumont etal., 1997) than the CGRP1 receptor (Dennis et al., 1990;Mimeault et al., 1991). It should be noted that the selectivity ofthese linear agonists is ambiguous; in porcine large coronaryartery [Cys(Acm)2,7]-h-αCGRP acts like a partial agonist(Waugh et al., 1999), while [Cys(Et)2,7]-h-αCGRP is knownto activate the CGRP1 receptor in cell lines (Choksi et al., 2002).However, these linear agonists still are important for the studyof CGRP receptors in the absence of more selective CGRPreceptor agonists and antagonists.
Interestingly, the responses to CGRP receptor agonists varydepending on the location and the size of blood vessels (Foulkeset al., 1991; Sheykhzade and Nyborg, 1998). Moreover, therelaxation induced by CGRP is endothelium-dependent in therat aorta (Wisskirchen et al., 1999), while it is endothelium-independent in human and porcine arteries (Luu et al., 1995;Wisskirchen et al., 1999). Therefore, the extrapolation ofaffinities from one species to another to classify CGRPreceptors should be done with caution.
BIBN4096BS, a lys-tyr dipeptide derivative, (1-piperidine-carboxamide, N-[2-[[5-amino-1-[[4-(4-pyridinyl)-1-piperazinyl]carbonyl] pentyl] amino]-1-[(3,5-dibromo-4-hydroxyphenyl)methyl]-2-oxoethyl]-4-(1,4-dihydro-2-oxo-3(2H)-quinazoli-nyl), displays a high antagonist potency and selectivity forhuman CGRP receptors (Doods et al., 2000; Durham, 2004).Moreover, this antagonist is reported to have a 10-fold higheraffinity for the CGRP1 receptor (Wu et al., 2000) as comparedto the CGRP2 receptor. These properties of BIBN4096BSprovide the possibility for an in-depth characterisation of CGRPreceptors. Although the antagonist potency of BIBN4096BS inthe human coronary artery has been studied earlier (Edvinssonet al., 2002), a detailed investigation of CGRP receptors inhuman coronary arteries has not yet been performed. The factthat BIBN4096BS is effective in the acute treatment ofmigraine (Olesen et al., 2004) underlines the need for adetailed investigation of CGRP receptors in human bloodvessels. Therefore, we used BIBN4096BS and other availableconventional CGRP receptor ligands to characterise CGRPreceptors in human and porcine isolated coronary arteries. Apart of this study has been published as an abstract (Gupta et al.,2004).
2. Materials and methods
2.1. Functional studies
Human hearts were obtained from ‘heart-beating’ donors (22male, 28 female; 48±2 years, range 3–66 years) who died dueto non-cardiac causes. The hearts were provided by the Heart
Valve Bank, Rotterdam after donor mediation by Bio ImplantServices Foundation/Euro Transplant Foundation (Leiden, TheNetherlands). Porcine hearts (pigs of either sex; 6–12 months ofage) were obtained from a local slaughterhouse. In both cases,proximal (internal diameter: 2–3 mm) and distal (internaldiameter: 250–600 μm) portions of the right coronary arterywere dissected, placed in oxygenated Krebs bicarbonatesolution (NaCl 118, KCl 4.7, CaCl2 2.5, MgSO4 1.2, KH2PO4
1.2, NaHCO3 25 and glucose 11.1 mM; pH 7.4) and storedovernight at 4 °C. Vessel segments containing distinct, macro-scopically visible atherosclerotic lesions were not used.
Proximal coronary artery segments (3–4 mm length) weresuspended with the help of stainless-steel hooks in 15-ml organbaths with 15 mN pretension (optimal tension shown in earlierexperiments). Distal artery segments (1–2 mm length) wereplaced in Mulvany myographs between two parallel titaniumwires with a tension normalised to 90% of l100 (distance whentransmural pressure equals 100 mm Hg), thus achieving optimalconditions for active force development (Nyborg et al., 1987). Theorgan baths and myograph chambers, containing Krebs bicar-bonate solution (for composition, see above) at 37 °C, werecontinuously bubbled with 95% O2 and 5% CO2. Sincespontaneous contractions frequently occurred in human distalcoronary arteries, these experiments were performed in thepresence of the cyclo-oxygenase inhibitor indomethacin (0.1μM). Unless mentioned otherwise, no attempt was made toremove the endothelium.
After an initial equilibration period of 30 min, all segmentswere challenged twice with 30 mM KCl at 30-min intervals toverify reproducibility of the responses. The integrity of theendothelium was assessed by observing relaxations to substanceP (1–10 nM) after precontraction with prostaglandin F2α (1 μM,proximal coronary arteries) or U46619 (9,11-dideoxy-11α, 9α-epoxymethano-prostaglandin F2α, 10–100 nM, distal coronaryarteries). In 4 human and 5 porcine distal coronary arterysegments, the endothelium was denuded using a human hair.After 30 min, 100 mM KCl was added to determine thereference contractile response in each segment. Then, a stablecontraction plateau of around 60% of the maximal contractionwas obtained with KCl (30 mM) and, subsequently, the CGRPreceptor agonists (i.e. h-αCGRP, h-βCGRP, [Cys(Acm)2,7]-h-αCGRP and [Cys(Et)2,7]-h-αCGRP; range: 0.1 nM-3 μM) wereadded in a cumulative manner in 0.5 log steps every 5 min orwhen the maximum effect of a given concentration was reached.In human proximal coronary artery, it was difficult to obtain astable precontraction with 30 mM KCl. In fact, othercompounds, such as U46619, showed even more unstableprecontraction in pilot experiments, and the responses toconcentrations of CGRP b30 nM were small and difficult todiscriminate from the spontaneous decline in the baselinetension. Therefore, we chose to start the concentration–response curves to h-αCGRP at a higher concentration, sothat quantification was less sensitive to artefacts. Furthermore,in three human proximal vessels, but none of the distal vessels,the control segment did not respond to h-αCGRP and,consequently, all segments from these non-responding vesselswere excluded from the study. Only one concentration–
109S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
response curve was constructed in the arterial segments, eitherin the absence or presence of the antagonists (BIBN4096BS,0.1 nM–10 μM or h-αCGRP8-37, 0.1–10 μM) that wereincubated for a period of 30 min, unless mentioned otherwise.
2.2. Isolation of total RNA and reverse transcriptase-polymerase chain reaction (RT-PCR) studies
After isolation from the heart, the segments of proximal anddistal human coronary arteries were snap frozen in liquidnitrogen and stored at −80 °C until use. The tissues weretransferred to guanidium thiocyanate buffer, homogenised(Ultra-Turrax homogeniser, model T8; Janke and KunkelGmbh, Staufen, Germany) and the total RNA was extracted(Chomczynski and Sacchi, 1987; Sharma et al., 1996). TheRNA concentration was measured by UV absorbance at awavelength of 260 nm using a Gene Quant RNA/DNAcalculator (Pharmacia-LKB, Uppsala, Sweden) and the qualityof RNAwas assessed by formaldehyde agarose gel electropho-resis and a DNA/protein ratio (OD260 /OD280) of N1.8. TotalRNA was denatured at 65 °C and the first strand of cDNAwassynthesised in a reaction volume of 20 μl by adding thefollowing reagents: reverse transcription buffer (25 mM Tris–HCl; pH 8.3, 50 mM KCl; 5 mM MgCl2, 2 mM dithiothreitol),1 mM deoxy nucleotide triphosphate (dNTPs), ribonucleaseinhibitor (1 U/μl), random hexamer (150 ng/μg total RNA) andfinally moloney-murine leukaemia virus-reverse transcriptase(MMLV-RT) (Pro-omega, Benelux b.v., Leiden, The Nether-lands). A control reaction with all of the above ingredients,except MMLV-RT, was prepared. The reactions were carriedout for 90 min at 42 °C, extended for another 10 min at 75 °Cand then cooled at 4 °C. The cDNA thus synthesised wasstored at −20 °C until used as a PCR template.
The quality of cDNA was verified by PCR amplification ofglyceraldehyde-3-phosphate-dehydrogenase (GAPDH) usingspecific oligonucleotide primers (see Results section). ForPCR amplification, a 20 μl reaction mixture containing thefollowing components was prepared: 1.25 mM of each dNTP,3 mM MgCl2, PCR buffer (1×PCR buffer: 10 mM Tris–HCl,pH 8.3, 50mMKCl) Ampli TaqGold™ enzyme (0.5 U), 0.5 μMof each of the forward and reverse primers and 2 μl of the cDNAtemplate. After a brief centrifugation the enzyme was firstactivated for 5 min at 94 °C in a PCR thermocycler (model PTC-100™, MJ Research Inc., Watertown, MA, U.S.A.). PCR wascarried out as: 45 s at 94 °C, 45 s at 55 °C and 90 s at 72 °C with atotal of 40 cycles. Finally, the reaction was extended for anadditional 10 min. For PCR amplification of RAMP-3 cDNAtemplate obtained from the proximal coronary artery, wereduced the concentration of MgCl2 from 3 mM to 2 mM andperformed 35 instead of 40 cycles to eliminate an unspecificband, while the remaining part of the conditions remainedunchanged. An aliquot of PCR reaction product was analysed on2% agarose gel, visualised under UV light and digitallyphotographed. The different CGRP receptor components weresemi-quantitatively assessed using 1D-analysis software (KodakDigital Science 1D Image Analysis, Version 1.6, ScientificImaging System, Rochester, NY, U.S.A.).
2.3. cAMP measurements
Human proximal and distal, as well as porcine distalcoronary artery segments were incubated in a medium contain-ing isobutylmethylxanthine (IBMX, 0.5 mM) for 30 min in theabsence or presence of BIBN4096BS (1 μM). The arterialsegments were exposed to KCl (30 mM), challenged with h-αCGRP (100 nM) or forskolin (10 μM) for 5 min and then snapfrozen. Forskolin, which increases intracellular cAMP concen-trations by activating adenylyl cyclase, was used to assess thespecificity of BIBN4096BS for CGRP-mediated increases incAMP concentrations. The samples were stored at −80 °C untilcAMP assay using the ELISA kit and manual (R and D SystemsEurope Ltd., Abingdom, U.K.).
2.4. Data presentation and statistical analysis
The relaxant responses elicited by agonists are expressed aspercentage relaxation of the tone induced by 30 mM KCl. Alldata are presented as means±S.E.M and n represents thenumber of blood vessels obtained from different donors. Theconcentration–response curves for all agonists were analysedusing non-linear regression analysis and the potency of agonistswas expressed as pEC50 using Graph Pad Prism 3.01 (GraphPad Software Inc., San Diego, CA, U.S.A.). The blockingpotency of the antagonists was estimated by calculating EC50
ratios and plotting a Schild-plot (Arunlakshana and Schild,1959) using linear regression to get the slope value. Only theEC50 ratios N2 were used for Schild plot analysis and forcalculation of respective pKb values. In certain cases, the slopevalue was significantly different from one, thus prohibiting us tocalculate pA2 values. Therefore, to enable a uniform comparisonbetween groups, “apparent pKb” values were calculated for theantagonists at each given concentration, constraining the slopeto unity. Correlation analyses were carried out using Pearson'scoefficient of correlation using SPSS 11.01 (SPSS Inc.,Chicago, Illinois, U.S.A.). Statistical significance was deter-mined by unpaired Student t-test, with differences consideredsignificant at P≤0.05. The ethical committee of Erasmus MC,Rotterdam, approved the study.
2.5. Compounds
The compounds used in the present study (obtained from thesources indicated) were: h-αCGRP, h-βCGRP, [Cys(Acm)2,7]-h-αCGRP, [Cys (Et)2,7]-h-αCGRP and h-αCGRP8-37 (Poly-peptide, Wolfenbüttel Germany); BIBN4096BS (gift from Dr.Henri Doods, Boehringer Ingelheim Pharma K.G., Biberach,Germany); U46619, isobutylmethylxanthine (IBMX), forsko-lin and substance P (Sigma Chemicals Co., Steinheim,Germany); and KCl (Merck, Darmstad, Germany). All com-pounds were dissolved in distilled water except forforskolin, which was dissolved in dimethylsulfoxide andBIBN4096BS, which was dissolved in 4% HCl (1 N) to obtaina 0.01 M stock solution; the latter was then diluted withdistilled water. All peptides and antagonists were stored inaliquots at −80 °C.
110 S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
3. Results
3.1. Functional studies
3.1.1. Human coronary arteriesThe contraction induced by 100 mM KCl was significantly
higher in the proximal (56±3 mN) than in the distal (8±1 mN)coronary arteries. In contrast, the endothelium-dependentrelaxant response to substance P (10 nM) was less in theproximal than in the distal coronary segments (26±13 and76±4%, respectively).
h-αCGRP induced a concentration-dependent relaxation inprecontracted human coronary arteries (Fig. 1). The maximalresponse to h-αCGRP was significantly less in the proximal(Emax: 43±7% of contraction to 30 mM KCl) than in the distal(Emax: 92±4%) segments. In distal segments where no attemptwas made to remove the endothelium, we assessed the cor-relation between the pEC50 and Emax of h-αCGRP and therelaxation to substance P, as determined in each experiment.These correlations were not significant (Pearson's r=0.066and 0.088, respectively; P=0.76 and 0.69, respectively,n=23–27), indicating that the endothelium is not involvedin relaxations to h-αCGRP. Accordingly, in distal coronarysegments, we also performed an additional series of ex-periments where the endothelium was denuded (relaxation tosubstance P: 1.0±1.4% and 64±22% of the precontraction to
Fig. 1. Effect of increasing concentrations of BIBN4096BS on the relaxant responseshuman right coronary artery (n=6–11) and average Schild plot slope (panel D) obtainA, C and D) or 2 h (panel B).
U46619 in endothelium-denuded and -intact segments,respectively). In these segments, both the Emax (96±2.3%and 96±2.1%, respectively) and pEC50 (9.23±0.44 and9.18±0.86, respectively, n=4) of h-αCGRP were similar.BIBN4096BS did not change the basal tone or the contraction toKCl (30 mM) in either the proximal or distal human coronaryartery. In proximal segments, the antagonism seemed to benon-competitive (Fig. 1A) as demonstrated by a significantsuppression of the Emax of h-αCGRP (43±7%), even at lowconcentrations of BIBN4096BS (Emax: 24±5% and 18±8%and in the presence of 0.1 and 1 nM of BIBN4096BS,respectively). However, in view of the large variability inherentto this preparation, as described in Materials and methodssection, these data should be interpreted with caution.
In distal segments BIBN4096BS concentration-dependent-ly antagonised the responses to h-αCGRP, but the Schild plotslope was significantly less than one (0.68±0.07; Table 1).Further experiments were carried out in the human distalcoronary artery segments to investigate the anomalous natureof the slope. To exclude that this finding was due to theinability of BIBN4096BS to reach equilibrium conditionsafter 30 min, we performed experiments in distal coronaryarteries with BIBN4096BS using a longer incubation period(2 h; Fig. 1B). In addition, we conducted experiments using awider range of concentrations of BIBN4096BS (Fig. 1C) toobtain an estimate of the Schild slope based on more data
to h-αCGRP in proximal (panel A) and distal (panels B and C) segments of theed from panel C. BIBN4096BS (0.1 nM–1 μM) was incubated for 0.5 h (panels
Table 1Effect of BIBN4096BS on relaxations to CGRP receptor agonists in human distal coronary artery segments
BIBN4096BS (0.1 nM–1 μM) was incubated for 0.5 h with concentrations increasing in one-log steps, except in the case of the human coronary artery and h-αCGRPwhere two additional groups (+more concentrations or #2-h incubation) were used.Emax was expressed as % of the response to 30 mM KCl. All data are means±S.E.M (n). aSlopes of the Schild regression significantly different (P≤0.05) from unity.[Cys(Acm)2,7], [Cys(Acm)2,7]h-αCGRP; [Cys(Et)2,7], [Cys(Et)2,7]h-αCGRP. §pKb at 0.3 nM BIBN4096BS.
111S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
points than in our initial experiments (Fig. 1D). In both cases,the slope remained significantly less than one (0.65±0.09 and0.64±0.05, respectively), but the longer incubation period didincrease the potency of BIBN4096BS (see apparent pKb
values in Table 1).h-αCGRP8-37 also antagonised the responses to h-αCGRP in
the human distal coronary artery (Fig. 2), but with a low potency(pKb at 3 μM: 6.28±0.17 n=4), and again the Schild plot slopewas significantly less than one (0.36±0.05, n=4).
Both linear agonists induced concentration-dependentrelaxations and BIBN4096BS, similarly as with h-αCGRP,antagonised the responses to [Cys(Acm)2,7]-h-αCGRP in aseemingly competitive manner with a slope significantly lessthan one (0.49±0.13; Fig. 3). Although the antagonism ofBIBN4096BS appeared competitive in nature, it should bekept in mind that in view of the low potency of [Cys(Acm)2,7]-h-αCGRP the values had to be extrapolated forcalculation of the pharmacological parameters. In contrast, theslope obtained with [Cys(Et)2,7]-h-αCGRP (1.13±0.25) wasnot different from unity. h-βCGRP relaxed the distal coronaryartery with a potency not different from that of h-αCGRP(Fig. 3; Table 1), and this response was potently antagonisedby BIBN4096BS with a Schild plot slope (0.70±0.08)significantly less than one.
Fig. 2. Effect of increasing concentrations of h-αCGRP8-37 on the relaxantresponses to h-αCGRP in distal segments of the human right coronary artery(n=4–12). h-αCGRP8-37 (100 nM–10 μM) was incubated for 0.5 h.
3.1.2. Porcine coronary arteriesThe contractile response to 100 mM KCl in proximal
coronary arteries was 69±6 mN as compared to 14±4 mN inthe distal segments. h-αCGRP induced concentration-depen-dent relaxations in both proximal and distal coronary arterysegments (Fig. 4). Endothelium denudation in the porcine distalcoronary segments resulted in significantly decreased relaxa-tions to substance P (7±4% and 78±7% of the precontraction toU46619 for endothelium-denuded and -intact artery segments,respectively). In contrast, the relaxations to h-(CGRP in endo-thelium-denuded segments (Emax: 97±2%; pEC50: 8.80±0.17)were not different from those in control segments (Emax:95±2%; pEC50: 8.77±0.16, n=5). Similar to the humancoronary artery, the Emax of h-αCGRP was significantly lessin porcine proximal segments (72±3%) than in distalsegments (95±2%). Moreover, h-αCGRP-induced relaxationswere blocked with equal potency by BIBN4096BS (1 μM)in the proximal and distal segments (pKb: 7.44±0.10 and7.63±0.14, respectively), while the Schild plot slope obtainedin both cases were not different from unity (0.80±0.18 and0.87±0.10, respectively).
3.2. RT-PCR studies in human coronary arteries
Using human specific forward and reverse primers,designed on the basis of nucleotide sequences of severalcomponents of the CGRP receptor family reported in theNCBI Genebank (Table 2), PCR products corresponding insize to CLR, RAMP-1, RAMP-2, RAMP-3, RDC and RCPwere consistently amplified in both proximal and distalhuman coronary arteries (Fig. 5). Semi-quantitative analysisrevealed no difference in the expression levels of these CGRPreceptor components, except for a two-fold higher expressionof RAMP-2 in the distal as compared to the proximalsegments.
3.3. cAMP measurements in human and porcine coronaryarteries
h-αCGRP (100 nM) increased cAMP concentrationssignificantly in the human proximal and distal arterysegments as compared to the segments incubated with
Fig. 3. Effect of increasing concentrations of BIBN4096BS on the relaxant responses to [Cys(Acm)2,7]-h-αCGRP, [Cys (Et)2,7]-h-αCGRP and h-βCGRP in distalsegments of the human right coronary artery (n=5–9). BIBN4096BS (0.1 nM–1 μM) was incubated for 0.5 h.
112 S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
vehicle (42±7 versus 107±21 and 14±9 versus 106±69 pMmg protein−1 mg tissue−1, respectively). This increase wassignificantly blocked by BIBN4096BS in both the humanartery segments (Fig. 6). Similarly in porcine distal coronaryarteries also there was a significant increase in cAMP levelsin response to h-αCGRP and this effect was antagonised byBIBN4096BS (Fig. 6). BIBN4096BS did not significantlyinhibit the forskolin-induced increase in the cAMP levels,either in human (389±134 and 292±110 pM mg protein−1 mgtissue−1 in the absence or presence of antagonist, respectively,P=0.47, n=5) or porcine (200±40 and 141±72 pM mgprotein−1 mg tissue−1, respectively, P=0.61, n=4) distalcoronary arteries.
Fig. 4. Effect of increasing concentrations of BIBN4096BS on the relaxant responses(n=6–8). BIBN4096BS (0.1 nM–1 μM) was incubated for 0.5 h.
4. Discussion
4.1. Functional interaction between CGRP receptor agonistsand antagonists
In both human and porcine proximal (internal diameter:2–3 mm) as well as distal (internal diameter: 250–600 μm)segments of the right coronary artery, h-αCGRP elicited aconcentration-dependent relaxation that was antagonised byBIBN4096BS, which proved more potent in the humans thanin the pig. Moreover, our results demonstrate that theserelaxations were endothelium-independent in both human andporcine distal coronary artery. Our findings, showing that the
to h-αCGRP in proximal and distal segments of the porcine right coronary artery
Table 2Detection of molecular components relevant for CGRP receptors
Components NCBI genebank accession no. Forward primer (5′-3′) Reverse primer (5′-3′) Amplified sequence Size (bp)
Each set of forward and reverse primers was designed from the nucleotide sequences reported in the NCBI genebank. CLR, Calcitonin receptor like receptor; RAMP,receptor activity modifying protein; RCP, receptor component protein; RDC-1, the orphan receptor (originally cloned from canine thyroid cDNA); GAPDH,glyceraldehyde-3-phosphate dehydrogenase.
113S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
responses to h-αCGRP in all four segment types are mediated bythe CGRP1 receptor, are consistent with those reported recentlyin the human (Edvinsson et al., 2002; Hasbak et al., 2003) andporcine (Wu et al., 2002) left anterior descending coronaryartery, where the CGRP-induced relaxation was also antag-onised by BIBN4096BS with similar potencies.
As reported earlier (Sheykhzade and Nyborg, 1998, Foulkeset al., 1991), we also found that the efficacy of CGRP inverselyrelated to the vessel diameter, Emax values in both human andporcine arteries were lower in the proximal (43±7% and72±3%, respectively) than in the distal (92±4% and 95±2%,respectively) coronary segments. The antagonism in the case ofhuman proximal arteries seems to be non-competitive, asdemonstrated by the suppression of the Emax. Although thesedata should be interpreted with caution, it is worth noting that asimilar non-competitive behaviour of BIBN4096BS has beenreported in human subcutaneous arteries at similar concentra-tions (0.1 and 1 nM) (Sheykhzade et al., 2004).
The Schild plot slope in the human distal coronary artery(0.68±0.07) was significantly lower than unity and, therefore,
Fig. 5. Agarose gel electrophoresis of PCR amplified products derived fromcDNA obtained from proximal (upper panel) and distal (lower panel) segmentsof the human right coronary artery. The different lanes marked on top of thepanels denote:ϕx174 DNA/Hae III marker (M), RAMP-1 (298 bp; 1), RAMP-2(227 bp; 2), RAMP-3 (189 bp; 3), CLR (560 bp; 4), RCP (344 bp; 5) and RDC(220 bp; 6). The size of 3 marker bands is indicated in the left margins.
additional experiments were undertaken with human distalcoronary arteries. To rule out that the Schild plot slope was lessthan unity because BIBN4096BS had not yet reached theequilibrium conditions in the distal segments, we employed alonger incubation period (2 h) with the antagonist. Notwith-standing, the slope (0.65±0.09) remained significantly less thanone, although the potency of the antagonist increased withlonger incubation period. Within this framework, if there werea non-equilibrium state between an antagonist and its re-ceptor, the Schild plot slope would tend to be greater thanunity (Kenakin, 1993) instead of less than one, as was thecase in our experiments. The fact that a longer incubationperiod increased the potency of BIBN4096BS (apparent pKb at10 nM: 10.12±0.18) is in line with previous observations onslow on and off kinetics of BIBN4096BS (Schindler and Doods,2002). Similarly, as with the experiments employing a longerincubation period for BIBN5096BS, the slope of the Schildplot obtained with more concentrations of BIBN4096BS re-mained less than one (0.64±0.05).
In subsequent studies, we used the conventional antagonisth-αCGRP8-37 as a tool to distinguish between CGRP1 (pKb: 7–8) and CGRP2 (pKb: 5.5–6.5) receptors (Dennis et al., 1989;Quirion et al., 1992; Wisskirchen et al., 1998). The fact that h-
Fig. 6. Effect of BIBN4096BS (1 μM) on the increases in cAMP concentrationby h-αCGRP (100 nM) in the human (proximal and distal segments) and porcine(distal segments) right coronary artery (n=5–17). *Significantly (Pb0.05)different from vehicle-treated segments.
114 S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
αCGRP8-37 antagonised h-αCGRP with a low potency (pKb:6.28±0.17) supports the possible involvement of CGRP2receptors (Juaneda et al., 2000). Again, the Schild plot slopewas less than unity, demonstrating that the smaller Schild slopevalues should most likely be attributed to a heterogeneousreceptor population in the human distal coronary artery and isnot due to a BIBN4096BS-specific phenomenon. Further, ourfinding showing that the relatively selective CGRP2 receptoragonists, [Cys(Acm)2,7]-h-αCGRP and [Cys(Et)2,7]-h-αCGRP(Dennis et al., 1989; Juaneda et al., 2000; Quirion et al., 1992),also induced concentration-dependent relaxations indicate thepresence of putative CGRP2 receptors. However, in similarexperiments performed in left anterior descending coronaryartery segments (1–2 mm external diameter) obtained frompatients with cardiac myopathy, Hasbak et al. (2003) observedonly a very weak relaxant effect of [Cys(Acm)2,7]-h-αCGRP(1 μM) amounting to less than 5% of the effect we observed.This difference may be due to the fact that we used donorswithout cardiac pathology and, moreover, used smaller sizesegments (0.25–0.6 mm internal diameter) of the rightcoronary artery, which may contain different subtypes ofCGRP-sensitive receptors. The selectivity of the linear agonistsis not unequivocal; Waugh et al. (1999), who observed thatwith increasing concentrations of KCl (8–15 mM) used asprecontraction, the relaxation induced by [Cys(Acm)2,7]-h-αCGRP in pig coronary arteries decreased from 90% to almostzero, demonstrating that [Cys(Acm)2,7]-h-αCGRP may act as apartial agonist. However, in our preparation this was not thecase, since even with a precontraction induced by 30 mM KClthe Emax of [Cys(Acm)2,7]-h-αCGRP was as high as 93±2%.Further, the other linear agonist [Cys(Et)2,7]-h-αCGRP acti-vated CGRP1-like receptor in SK-N-MC and Col-29 cells,while the same agonist acted on CGRP2-like receptors in ratvas deferens with a pEC50 ≈8 (Wu et al., 2000). In the presentstudy, [Cys(Et)2,7]-h-αCGRP also relaxed arteries with asimilar potency (pEC50: 8.21±0.25). According to Kenakin(1993), different slopes obtained with Schild plots for differentagonists with the same antagonist indicate a heterogeneousreceptor population. Indeed, in contrast to our experimentswith h-αCGRP and [Cys(Acm)2,7]-h-αCGRP, the Schildslopes obtained with BIBN4096BS antagonising [Cys(Et)2,7]-h-αCGRP was equal to unity, suggesting that these agonists donot stimulate the same receptor population or activate samereceptor but with different potencies. Further, the observationthat the apparent pKb values decrease with increasingconcentrations of antagonist in case the slope was less thanone again points to a heterogeneous CGRP receptor population(Table 1).
Intriguingly, the fact that the apparent pKb of BIBN4096BSwas around 10 for both linear agonists points to the presence ofCGRP1 receptors as BIBN4096BS is supposed to be at least 10times less potent for CGRP2 receptors (Wu et al., 2000). Hence,it is reasonable to assume that the receptors in the human distalcoronary artery do not completely comply with the pharmaco-logical profile of the presently accepted CGRP receptorclassification (Juaneda et al., 2000); notwithstanding, it wouldbe prudent to consider several lines of arguments within this
framework. Firstly, Sheykhzade et al. (2004) have recentlysuggested a non-competitive nature of antagonism ofBIBN4096BS at concentrations higher than 10 pM against h-αCGRP-induced relaxations in human subcutaneous arteries.Although, as mentioned above, this might be the case in ourresults with the proximal coronary artery, it does not apply to thedistal coronary artery, where even at high concentrations ofBIBN4096BS the Emax to h-αCGRP or other agonist was notsuppressed showing antagonism is competitive in nature.Secondly, the interaction between BIBN4096BS and theadrenomedullin receptor might be responsible for the slope ofthe Schild plot being less than one. However, the affinity ofBIBN4096BS for adrenomedullin receptors is too low (IC50:10.3 μM) for such an interaction (Doods et al., 2000). Thirdly,metabolism of h-αCGRP or CGRP receptor ligands by tissuepeptidases can affect the data obtained. Nevertheless, inhibitionof these peptidases has shown that the contribution ofmetabolism in determining the affinity of CGRP ligands isnegligible (Wu et al., 2000). Fourthly, the linear agonists mightnot be selective enough to form the basis for the classification ofCGRP receptors. In view of our findings, we also have ourreservation towards the purported selectivity of linear agonisttowards CGRP2 receptors, as both agonists under similarexperimental conditions yielded significantly different Schildplot slopes with BIBN4096BS, suggesting that these agonistsdo not activate the same receptor subtypes. BIBN4096BS issupposed to be selective for the CGRP1 receptor, but it showed asimilar or even higher antagonist potency with the linearagonists in human distal coronary artery as compared to h-αCGRP.Finally, the slope being different from one cannot beattributed only to BIBN4096BS as in the case of porcineproximal and distal coronary arteries the slope was notsignificantly different from one, and on the other hand h-αCGRP8-37 with h-αCGRP also produced a slope less than one,thus underlining a heterogeneous receptor population in thehuman distal coronary artery.
4.2. Molecular components of the CGRP receptor family
Our RT-PCR results showed the expression of various CGRPreceptor components like CLR, RCP, RDC-1, RAMP-1, RAMP-2 and RAMP-3 in both proximal and distal human coronaryarteries; these molecular studies confirmed the presence of allthe essential components (CLR+RAMP-1+RCP) required forfunctional CGRP1 receptors in the above coronary segments. Inthis context, CLR is a G-protein coupled receptor and thepresence of RAMP-1 ensures intracellular trafficking andmaturation of the receptor (McLatchie et al., 1998). The thirdentity, RCP, is required for the formation of a high-affinity G-protein-coupled receptor, there by ensuring the signal transduc-tion of CLR (Evans et al., 2000). Unfortunately, the molecularcounterpart for the CGRP2 receptor is yet to be determined.
4.3. Transduction mechanisms of the CGRP receptors
Both in human and porcine coronary arteries cAMP levelsincreased after a challenge with h-αCGRP. The fact that this
115S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
increase in cAMP was abolished after incubation withBIBN4096BS clearly demonstrates that the relaxations to h-αCGRP in the above blood vessels are, at least partly, mediatedvia this second messenger system.
In conclusion, the relaxation to h-αCGRP in the humandistal coronary arteries is mediated by CGRP1 receptors asdemonstrated by the high apparent pKb values (≈9–10)obtained with BIBN4096BS. In addition, another CGRPreceptor subtype, possibly also acting via an increase incAMP, seems to be present in the human distal coronary artery.Although the nature of this receptor is not entirely clear, itshows some functional characteristics of CGRP2 receptors, suchas the low antagonist potency of h-αCGRP8-37 and the highpotency of the linear agonists, whereas the high apparent pKb
values of BIBN4096BS point to the presence of CGRP1receptors. A prerequisite to further characterise these receptorswould be the advent of more selective CGRP receptor agonistsand antagonists along with the molecular characterisation of theputative CGRP2 receptor.
Acknowledgements
We would like to acknowledge Dr. H. Doods (BoehringerIngelheim Pharma, Biberach, Germany) for the generous gift ofBIBN4096BS and other CGRP receptor-ligands and Dr. P.G.H.Mulder (Department of Epidemiology and Biostatistics,Erasmus MC, Rotterdam, The Netherlands) for his kind helpin the statistical analyses. This study was partly supported fromfunds obtained from Boehringer Ingelheim Pharma KG(Biberach, Germany) for a contract research project withErasmus Pharma B.V.
Dennis, T., Fournier, A., St. Pierre, S., Quirion, R., 1989. Structure–activityprofile of calcitonin gene-related peptide in peripheral and brain tissues.Evidence for receptor multiplicity. J. Pharmacol. Exp. Ther. 251, 718–725.
Dennis, T., Fournier, A., Cadieux, A., Pomerleau, F., Jolicoeur, F.B., St. Pierre,S., Quirion, R., 1990. h-CGRP8-37, a calcitonin gene-related peptideantagonist revealing calcitonin gene-related peptide receptor heterogeneityin brain and periphery. J. Pharmacol. Exp. Ther. 254, 123–128.
Doods, H., Hallermayer, G., Wu, D., Entzeroth, M., Rudolf, K., Engel, W.,Eberlein, W., 2000. Pharmacological profile of BIBN4096BS, the firstselective small molecule CGRP antagonist. Br. J. Pharmacol. 129, 420–423.
Dumont, Y., Fournier, A., St.-Pierre, S., Quirion, R., 1997. A potent andselective CGRP2 agonist, [Cys(Et)2,7]h-CGRP alpha: comparison inprototypical CGRP1 and CGRP2 in vitro bioassays. Can. J. Physiol.Pharmacol. 75, 671–676.
Durham, P.L., 2004. CGRP-receptor antagonists—a fresh approach to migrainetherapy? N. Engl. J. Med. 350, 1073–1075.
Edvinsson, L., Alm, R., Shaw, D., Rutledge, R.Z., Koblan, K.S., Longmore, J.,Kane, S.A., 2002. Effect of the CGRP receptor antagonist BIBN4096BS inhuman cerebral, coronary and omental arteries and in SK-N-MC cells. Eur. J.Pharmacol. 434, 49–53.
Evans, B.N., Rosenblatt, M.I., Mnayer, L.O., Oliver, K.R., Dickerson, I.M.,2000. CGRP-RCP, a novel protein required for signal transduction atcalcitonin gene-related peptide and adrenomedullin receptors. J. Biol. Chem.275, 31438–31443.
Foulkes, R., Shaw, N., Bose, C., Hughes, B., 1991. Differential vasodilatorprofile of calcitonin gene-related peptide in porcine large and small diametercoronary artery rings. Eur. J. Pharmacol. 201, 143–149.
Gupta, S., Mehrotra, S., Garrelds, I.M., Sharma, H.S., Saxena, P.R.,MaassenVanDenBrink, A., 2004. Evidence for a heterogeneous CGRPreceptor population in the human coronary artery. Neuropeptides 38,122–123.
Hasbak, P., Saetrum Opgaard, O., Eskesen, K., Schifter, S., Arendrup, H.,Longmore, J., Edvinsson, L., 2003. Investigation of CGRP receptors andpeptide pharmacology in human coronary arteries. Characterization with anonpeptide antagonist. J. Pharmacol. Exp. Ther. 304, 326–333.
Juaneda, C., Dumont, Y., Quirion, R., 2000. The molecular pharmacology ofCGRP and related peptide receptor subtypes. Trends Pharmacol. Sci. 21,432–438.
Kenakin, T., 1993. Competitive antagonism. Pharmcologic Analysis of Drug-receptor Interaction. Raven Press, Ltd, New York, pp. 278–322.
Luebke, A.E., Dahl, G.P., Roos, B.A., Dickerson, I.M., 1996. Identification of aprotein that confers calcitonin gene-related peptide responsiveness tooocytes by using a cystic fibrosis transmembrane conductance regulatorassay. Proc. Natl. Acad. Sci. U. S. A. 93, 3455–3460.
Luu, T.N., Dashwood, M.R., Chester, A.H., Muddle, J.R., Yacoub, M.H.,1995. Calcitonin gene-related peptide in healthy and atheromatoushuman epicardial coronary arteries. Function and receptor characteriza-tion. J. Vasc. Res. 32, 93–99.
McLatchie, L.M., Fraser, N.J., Main, M.J., Wise, A., Brown, J., Thompson, N.,Solari, R., Lee, M.G., Foord, S.M., 1998. RAMPs regulate the transport andligand specificity of the calcitonin-receptor-like receptor. Nature 393,333–339.
Mimeault, M., Fournier, A., Dumont, Y., St-Pierre, S., Quirion, R., 1991.Comparative affinities and antagonistic potencies of various humancalcitonin gene-related peptide fragments on calcitonin gene-relatedpeptide receptors in brain and periphery. J. Pharmacol. Exp. Ther. 258,1084–1090.
Nyborg, N.C., Baandrup, U., Mikkelsen, E.O., Mulvany, M.J., 1987. Active,passive and myogenic characteristics of isolated rat intramural coronaryresistance arteries. Pflügers Arch. 410, 664–670.
Olesen, J., Diener, H.C., Husstedt, I.W., Goadsby, P.J., Hall, D., Meier, U.,Pollentier, S., Lesko, L.M., 2004. Calcitonin gene-related peptide receptorantagonist BIBN 4096 BS for the acute treatment of migraine. N. Engl. J.Med. 350, 1104–1110.
Poyner, D.R., Sexton, P.M., Marshall, I., Smith, D.M., Quirion, R., Born, W.,Muff, R., Fischer, J.A., Foord, S.M., 2002. International Union ofPharmacology: XXXII. The mammalian calcitonin gene-related peptides,adrenomedullin, amylin, and calcitonin receptors. Pharmacol. Rev. 54,233–246.
Quirion, R., Van Rossum, D., Dumont, Y., St-Pierre, S., Fournier, A., 1992.Characterization of CGRP1 and CGRP2 receptor subtypes. Ann. N. Y. Acad.Sci. 657, 88–105.
Schindler, M., Doods, H.N., 2002. Binding properties of the novel, non-peptideCGRP receptor antagonist radioligand, [3H]BIBN4096BS. Eur. J. Pharma-col. 442, 187–193.
Sharma, H.S., Maulik, N., Gho, B.C., Das, D.K., Verdouw, P.D., 1996.Coordinated expression of heme oxygenase-1 and ubiquitin in the porcineheart subjected to ischemia and reperfusion. Mol. Cell. Biochem. 157,111–116.
Sheykhzade, M., Nyborg, N.C., 1998. Caliber dependent calcitonin gene-relatedpeptide-induced relaxation in rat coronary arteries: effect of K+ on thetachyphylaxis. Eur. J. Pharmacol. 351, 53–59.
116 S. Gupta et al. / European Journal of Pharmacology 530 (2006) 107–116
Sheykhzade, M., Lind, H., Edvinsson, L., 2004. Noncompetitive antagonism ofBIBN4096BS on CGRP-induced responses in human subcutaneous arteries.Br. J. Pharmacol. 143, 1066–1073.
Van Rossum, D., Hanisch, U.K., Quirion, R., 1997. Neuroanatomicallocalization, pharmacological characterization and functions of CGRP,related peptides and their receptors. Neurosci. Biobehav. Rev. 21, 649–678.
Wisskirchen, F.M., Burt, R.P., Marshall, I., 1998. Pharmacological character-ization of CGRP receptors mediating relaxation of the rat pulmonary arteryand inhibition of twitch responses of the rat vas deferens. Br. J. Pharmacol.123, 1673–1683.
Wisskirchen, F.M., Gray, D.W., Marshall, I., 1999. Receptors mediating CGRP-induced relaxation in the rat isolated thoracic aorta and porcine isolatedcoronary artery differentiated by h-alpha-CGRP8-37. Br. J. Pharmacol. 128,283–292.
Wu, D., Eberlein, W., Rudolf, K., Engel, W., Hallermayer, G., Doods, H., 2000.Characterisation of calcitonin gene-related peptide receptors in rat atriumand vas deferens: evidence for a [Cys(Et)2,7]h-CGRP-preferring receptor.Eur. J. Pharmacol. 400, 313–319.
Wu, D., Doods, H., Arndt, K., Schindler, M., 2002. Development and potentialof non-peptide antagonists for calcitonin-gene-related peptide (CGRP)receptors: evidence for CGRP receptor heterogeneity. Biochem. Soc. Trans.30, 468–473.
