- Email: [email protected]
Partial and full agonism in endomorphin derivatives: Comparison by null and operational model
peptides 27 (2006) 1507–1513
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/peptides
Partial and full agonism in endomorphin derivatives: Comparison by null and operational model Andra´s Z. Ro´nai a,*, Mahmoud Al-Khrasani a, Sa´ndor Benyhe b, Imre Lengyel b, La´szlo´ Kocsis c, Gyo¨rgy Orosz c, Ge´za To´th b, Erzse´bet Kato´ a, La´szlo´ To´thfalusi d a
Department of Pharmacology and Pharmacotherapy, Semmelweis University, PO Box 370, H-1445 Budapest, Hungary Biological Research Centre, HAS, PO Box 521, H-6701 Szeged, Hungary c Research Group of Peptide Chemistry, Eo¨tvo¨s University/HAS, PO Box 32, H-1518 Budapest 112, Hungary d Department of Pharmacodynamics, Semmelweis University, Budapest, Hungary b
article info
abstract
Article history:
The partial m-opioid receptor pool inactivation strategy in isolated mouse vas deferens was
Received 24 August 2005
used to determine partial agonism of endomorphins and their analogs (endomorphin-1-ol,
Received in revised form
20 ,60 -dimethyltyrosine (Dmt)-endomorphin-1, endomorphin-2-ol and (D-Met2)-endomor-
7 December 2005
phin-2) using morphine, normorphine, morphiceptin, (D-Ala2,MePhe4,Gly5-ol)-enkephalin
Accepted 7 December 2005
(DAMGO) and its amide (DAMGA) as reference opioid agonists. Agonist affinities (KA) and
Published on line 18 January 2006
efficacies were assessed both by the ‘‘null’’ and the ‘‘operational’’ method. The KA values determined by the two methods correlated significantly with each other and also with the
Keywords:
displacing potencies against 3H-naloxone in the receptor binding assay in the presence of
Endomorphin derivatives
Na+. DAMGO and DAMGA were full agonist prototypes, morphine, endomorphin-1, endo-
Partial agonism
morphin-1-ol, Dmt-endomorphin-1, endomorphin-2-ol and (D-Met2)-endomorphin-2 were
Residual receptor fraction
found by both methods to be partial agonists whereas the parameters for normorphine,
b-Funaltrexamine
morphiceptin and endomorphin-2 were intermediate. # 2005 Elsevier Inc. All rights reserved.
Mouse vas deferens Null and operational model
1.
Introduction
Of the major opioid receptor types, i.e. d, k and m [28,29] dynorphins and neo-endorphins have a high preference for the k type, some proenkephalin-derived peptides for the d type, b-endorphin is equally active both at the m and d type and only some minor ones (e.g. the proenkephalin-derivative metorphamid) show a moderate m receptor type preference [14,18]. The discovery of endomorphins (1 and 2, Tyr-Pro-TrpPhe-NH2, EMO-1 and Tyr-Pro-Phe-Phe-NH2, EMO-2) in 1997 was a landmark in opioid peptide research [53]. The two tetrapeptides, isolated from bovine brain, were the first mammalian brain opioid peptides with high m-opioid receptor
selectivity and agonist potency. This selectivity of endomorphins is likely to be the consequence of their novel structure. The most common motif for an opioid agonist in the mammalian brain, and in other species, is the Tyr-Gly-GlyPhe-N-terminal core sequence. The Tyr-D-Ala/D-Met-Phemotifs were found only in amphibians (dermorphins, dermenkephalin and deltorphins, [32,37]), and the Tyr-Pro-Phe/ Trp- N-terminal sequences were identified in casomorphins, hemorphins and cytochrophin in milk and blood (most prominently in digests, [5–7]). Although some of these peptides, such as the b-casomorphin derivative morphiceptin (Tyr-Pro-Phe-Pro-NH2, [8]), have been reported to possess selective m-opioid receptor agonist properties, their potency
* Corresponding author. Tel.: +36 1 210 4416; fax: +36 1 210 4412. E-mail address: [email protected] (A.Z. Ro´nai). 0196-9781/$ – see front matter # 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2005.12.003
1508
peptides 27 (2006) 1507–1513
falls into the low or mid-micromolar range. Tyr-MIF-1 (TyrPro-Leu-Gly-NH2, [19]) and Tyr-W-MIF-1 (Tyr-Pro-Trp-GlyNH2, [10]) were the first Tyr-Pro-related brain opioid peptides, to which endomorphins belong, but these peptides were antagonists rather than agonists, acting in the mid-micromolar range. It has been suggested that endomorphins serve as peptide neurotransmitters/modulators [16] and they do, indeed, meet several of the requisite criteria. However they lack some important features (the mode of endogenous generation, the firm establishment of Ca2+-dependent release) which would contribute to substantiate this notion. Furthermore, endomorphins have been reported to display partial agonist properties in the [35S]GTPgS-binding assay [3,17,26,33,43]; an unusual feature for a first messenger candidate. Recently, we confirmed the partial agonism for endomorphin-1 and its-olderivative by using the fractional receptor pool inactivation strategy [12,13,15,47,48,52] using the field-stimulated mouse vas deferens (MVD, 2). Using the MVD method, in this paper we extended our analysis to Dmt-EMO-1, EMO-2, EMO-2-ol and (DMet2)-EMO-2 (met2-EMO-2). DAMGO, DAMGA, morphine, normorphine and morphiceptin were used as reference mopioid receptor agonists. Besides the partial receptor pool inactivation and the so-called double reciprocal plot [13,47], referred to as the ‘‘null’’ method [24,25], for our current analysis we used also the ‘‘operational’’ model devised by Leff and his coworkers [23–25]. Both methods have been devised to characterize agonist affinity and efficacy by pharmacological means on isolated organ preparations. The ‘‘null’’ method was developed avoiding the need to make assumptions about postreceptor events whereas the ‘‘operational’’ method calibrates first the parameters of receptorial transducer function by constructing an equation for the [E]/[A] relation of a full agonists then, using these parameters, extends the relationship to any other agonist.
2.
Materials and methods
The conditions of animal housing and experimentation followed the ethical guidelines set by the Ethical Board of Semmelweis University, based on EC Directive 86/609/EEC.
2.1.
Materials
b-Funaltrexamine hydrochloride (b-FNA) was obtained from Tocris Cockson Ltd. (Bristol, UK), normorphine base and morphine sulfate from ICN Alkaloida Ltd. (Tiszavasva´ri, Hungary). 3H-naloxone (57.0 Ci/mmol) was prepared by G. To´th of Biological Research Centre, Szeged, Hungary [49]. The syntheses of peptide amides and alcohols (-olderivatives) were performed by solid-phase. The details of synthetic and analytical procedures are described elsewhere [2,20,26,40]. 20 ,60 -dimethyl-Tyr1-endomorphin-1 was synthesized by G. To´th of Biological Research Centre, Szeged, Hungary [50,51]. According to the analytical proofsheets received with each peptide batch, the peptides had the correct amino acid composition and an at least 95% HPLC purity; batches of Dmt-EMO-1 were also routinely subjected to mass spectrometry. All the other substances used were of analy-
tical grade and were obtained either from Sigma–Aldrich Ltd. (St. Louis, USA) or Reanal Ltd. (Budapest, Hungary).
2.2.
Methods
2.2.1.
Rat brain receptor binding assay (RBA)
Data obtained in the receptor binding assay are used exclusively for correlation analytical purposes, and taken mostly from specified literature sources. An important aspect of data set selection was that the values be generated in the same laboratory and according to a uniform experimental paradigm; therefore, a brief methodological description is given below. Assays were carried out in crude brain membrane fraction prepared from Wistar rats as described previously [44]. Membranes were incubated in 50 mM Tris– HCl buffer (pH 7.4) supplemented with 100 mM NaCl at 0 8C for 60 min. The concentration of 3H-naloxone (57 Ci/mmol specific activity, 49) used in the displacement experiments was 1.0 nM. All experiments were carried out in duplicate assays and repeated at least three times.
2.2.2.
Isolated organ preparation
Vasa deferentia taken from CFLP mice weighing 35–45 g were prepared, mounted and stimulated as described previously [39]. In brief, a single vas per bath was mounted in Mg2+-free Krebs’ solution aerated with carbogen (O2:CO2 = 95:5) at 31 8C under an initial tension of 0.1 g. The stimulation parameters were as follows: field stimulation of pairs (100 ms pulse distance) of rectangular impulses (1 ms pulse width, 9 V/cm i.e. supramaximal intensity) were repeated every 10 s.
2.2.3.
Experimental paradigms
Following 30–40 min equilibration isolated vasa deferentia were exposed to agonists for up to 2 min with the exception of Dmt-endomorphin-1, where exposure was 10–25 min. Administration cycle was 12–18 min with three to four interim washes with the exception of Dmt-endomorphin-1 where it was 40–60 min with 8–12 washes. Incubation conditions with b-FNA and the construction of before–after dose–response curves with the agonists were the same as used previously [2]. In brief, the agonist dose– response curves were taken non-cumulatively at 4–6 concentration levels; the vasa were then exposed to 5 107 M b-FNA for 30 min. The exposure was followed by a 60-min washout period with 12 washes then the agonist dose– response curves were re-taken at 4–6 pre-set concentration levels. The proper concentration range for each agonist was determined in a pilot experiment.
2.2.4.
Evaluation
To obtain the parameters of agonist concentration–response (E/[A]) curves before and after b-FNA treatment non-linear curve fitting according to the Hill equation (three parameters, Eq. (1)) was used: E¼
Emax ½An ½A50 n þ ½An
(1)
where ‘‘E’’ is the biological response (percent inhibitory effect), ‘‘Emax’’ the maximal effect, ‘‘[A]’’ the molar concentration of agonist, ‘‘[A50]’’ the 50% effective concentration and ‘‘n’’ is the
peptides 27 (2006) 1507–1513
Hill slope. The ‘‘null’’ method, i.e. the double-reciprocal plot of equieffective concentrations of agonists before ([A], ordinate) and after ([A0 ], abscissa) b-FNA treatment [13,47] was carried out as described previously [2]. From the slope and ‘‘y’’ inter-
1509
cept the residual receptor fraction ‘‘q’’ and agonist dissociation constant ‘‘KA’’ were calculated. According to the operational model [23–25] the ‘‘Emax’’ and ‘‘n’’ parameters for the full agonist prototypes were determined by using Eq. (1) then these
Fig. 1 – The inhibitory dose–response curves of m-opioid receptor agonists in the field-stimulated mouse vas deferens before and after b-funaltrexamine treatment. The inhibitory dose–response curves of DAMGO (panel ‘‘A’’), DAMGA (panel ‘‘B’’), morphine (panel ‘‘C’’), normorphine (panel ‘‘D’’), morphiceptin (panel ‘‘E’’), endomorphin-1 (panel ‘‘F’’), endomorphin-1-ol (panel ‘‘G’’), Dmt-endomorphin-1 (panel ‘‘H’’), endomorphin-2 (panel ‘‘I’’) and endomorphin-2-ol (panel ‘‘J’’) before (dark circles) and after (open circles) 30-min incubation with 5 10S7 M b-funaltrexamine. Points represent the mean, vertical lines the S.E.M. of 4–6 independent determinations. Hill equation (three parameters) was used for curve fitting.
1510
peptides 27 (2006) 1507–1513
Table 1 – The dose–response curve parameters of m-opioid agonists in the mouse vas deferens before and after bfunaltrexamine treatment A50 (nM)a
Agonist (n)
DAMGO (5) DAMGA (4) Morphine (4) Normorphine (4) Morphiceptin (4) EMO-1 (6) EMO-1-ol (4) Dmt-EMO-1 (6) EMO-2 (4) EMO-2-ol (6) met2-EMO-2 (4)
Emax (%)b
Hill slopeb
Cc
Fc
Cc
Fc
Cc
Fc
46.9 (36.4–60.4) 17.0 (12.9–22.2) 140.0 (111.3.176.0) 181.2 (133.7–245.6) 2619 (2381–2882) 23.2 (16.4–32.9) 111.0 (82.1–150.1) 1.48 (1.11–1.99) 17.2 (14.5–20.5) 26.3 (20.3–34.1) 42.2 (28.0–63.5)
482.7 (263.0–641.8) 186.9 (153.3–227.9) 937.1 (671.1–1309) 4202 (3570–4945) 43342 (25490–73696) 242.1 (170.3–344.3) 1450 (818.4–2570) 4.54 (2.30–8.97) 127.1 (82.3–196.3) 212.1 (99.5–451.8) 192.5 (115.9–319.8)
96.3 2.2 96.7 5.6 81.6 0.7 94.3 3.5 99.2 1.7 96.1 2.8 99.0 0.5 99.8 0.2 99.5 2.7 98.5 3.3 89.5 5.3
97.7 1.9 98.5 1.9 34.4 3.4 96.3 3.0 91.9 0.6 95.2 3.3 67.9 9.1 63.2 7.7 79.5 3.1 84.7 2.9 53.9 8.6
1.62 0.10 1.37 0.11 1.52 0.15 1.40 0.16 1.35 0.06 1.39 0.13 1.17 0.07 1.57 0.10 1.61 0.13 1.52 0.11 1.19 0.07
1.02 0.04 1.04 0.06 1.01 0.09 0.60 0.04 0.83 0.08 0.48 0.06 0.46 0.10 1.01 0.24 1.11 0.04 0.84 0.10 0.76 0.10
The parameters were obtained from curve fitting by the Hill equation (three parameters). For the 50% effective concentration (A50) geometric mean and 95% confidence interval was given. b For Emax and Hill slope arithmetic mean S.E.M. was given. c C (control): before b-FNA; F: after 30 min incubation with 5 107 M b-FNA. a
parameters were used to determine the receptor constants for all the agonists by another equation E¼
Emax tn ½An ðKA þ ½AÞn þ tn ½An
(2)
where ‘‘t’’ is the model definition of efficacy, ‘‘KA’’ the agonist dissociation constant and other parameters as in Eq. (1). For convenience, agonist efficacies, obtained either by the null or the operational method, were characterized uniformly by the KA/[A50] ratios. For pooled [A50] and KA values geometric means and 95% confidence intervals [11] were calculated otherwise arithmetic mean S.E.M. values were listed. Statistical probes were applied to the logarithms of data sets. ANOVA followed by Dunnett’s test was used to compare efficacies, taking DAMGO as prototype full agonist for multiple comparisons.
Newman–Keuls post hoc test was applied in the case of residual receptor fraction (‘‘q’’, null method) values whereas Student’s ‘‘t’’-test was used to compare ‘‘before–after’’ data sets (paired arrangement). For correlation analysis, correlation coefficient (r) values were calculated [45].
3.
Results
The dose–response curves of DAMGO, DAMGA, morphine, normorphine, morphiceptin, EMO-1, EMO-1-ol, Dmt-EMO-1, EMO-2 and EMO-2-ol before and after b-FNA treatment are shown in Fig. 1. The changes can be sorted into three separate subgroups. (i) There was an almost parallel rightward shift with no Emax depression for DAMGO and DAMGA. (ii) There was a pronounced Emax depression ( p < 0.01) with
Table 2 – The receptor constants of m-opioid agonists in the mouse vas deferens KA (nM)a
Agonist (n) NULL DAMGO (5) DAMGA (4) Morphine (4) Normorphine (4) Morphiceptin (4) EMO-1 (6) EMO-1-ol (4) Dmt-EMO-1 (6) EMO-2 (4) EMO-2-ol (6) met2-EMO-2 (4)
727.7 (489.3–1220) 355.4 (274.3–460.5) 574.2 (257.5–1280) 1748 (1542–1981) 36 244 (28967–45350) 57.8 (24.0–139.6) 444.5 (220.5–895.9) 2.07 (0.53–8.04) 247.5 (188.1–325.6) 71.3 (35.0–145.3) 104.5 (70.2–155.1)
q (%)c
KA/A50b OPER
2412 (1281–4543) 1058 (583.3–1919) 528.7 (462.7–604.1) 3395 (2374–4855) 63 755 (49181–82649) 135.3 (64.1–285.5) 470.2 (325.9–678.4) 5.04 (3.23–7.86) 356.0 (275.5–460.0) 304.6 (229.3–404.5) 160.4 (110.4–233.1)
NULL
OPER
NULL
17.4 2.6 21.7 2.9 4.86 1.39 10.2 1.7 14.4 2.0 3.19 0.78 4.82 1.48 3.57 1.71 14.7 2.9 3.30 0.79 2.86 0.68
56.7 11.3 71.5 17.3 3.99 0.69 20.5 4.2 25.1 3.2 6.95 1.57 4.72 1.05 3.69 0.62 20.3 2.5 12.9 2.2 5.03 1.72
17.7 2.5 14.4 1.5 16.4 4.6 14.9 2.0 12.60.6 40.1 6.7 19.3 6.2 53.0 12.6 13.3 1.0 34.8 4.1 25.7 2.8
The parameters were determined by the ‘‘null’’ method (based on the double reciprocal plot of equieffective agonist concentrations before and after b-FNA) or the ‘‘operational’’ (OPER) method [24,25]. Statistics: for KA/A50 ratios: by ANOVA followed by Dunnett’s multiple comparison test DAMGO vs. morphine, EMO-1, EMO-1-ol, Dmt-EMO-1, EMO-2-ol and met2-EMO-2: p < 0.001 both by null and operational method; For residual receptor fraction ‘‘q’’: by ANOVA followed by Newman–Keuls test Dmt-EMO-1 vs. morphiceptin p < 0.01, Dmt-EMO-1 vs. DAMGO, DAMGA, morphine, normorphine and EMO-2 p < 0.05; EMO-1 vs. morphiceptin and EMO-2 p < 0.05. a For the agonist dissociation constant (KA) geometric mean and 95% confidence interval was given. b For the efficacy-related KA/A50 ratio arithmetic mean S.E.M. was given. c The residual receptor fraction ‘‘q’’ was expressed in percent; arithmetic mean S.E.M. was listed.
peptides 27 (2006) 1507–1513
moderate slope reduction for morphine, Dmt-EMO-1 and met2-EMO-2 (latter is not shown in the figure). (iii) There was a significant slope reduction with no or modest Emax depression for normorphine, EMO-1, and EMO-1-ol (see also Table 1). We determined the ‘‘Emax’’ and ‘‘n’’ values for DAMGO and DAMGA using the Hill equation (Eq. (1)); these were then used for calculations in the operational procedure. According to a previous analysis [2] both DAMGO and DAMGA appeared to fulfill the criteria of full agonism, therefore their Emax and ‘‘n’’ parameters were pooled. The pooled Emax value was 96.5 2.8 (n = 9) and the Hill slope 1.51 0.08 (n = 9). When calculating the agonist dissociation constants by either the null or the operational methods (Table 2), the KA values correlated significantly with each other (Pearson correlation coefficient r = 0.9998, p < 0.0001). Similarly, good correlation was observed between the Ki values determined in the rat brain receptor binding assay in the presence of Na+ (Table 3, r = 0.9900 for null and r = 0.9892 for operational method, respectively, p < 0.0001 for both). Likewise, there was a significant correlation (r = 0.9394) between the efficacies calculated by the two methods. Furthermore, when using Dunnett’s post hoc test for differentiating between full and partial agonism, using DAMGO as prototype full agonist, both methods gave a similar result. The test qualified morphine, EMO-1, EMO-1-ol, Dmt-EMO-1, EMO-2-ol and met2-EMO-2 as partial agonists, and left normorphine, morphiceptin and EMO-2 in the category of potential full agonism (DAMGA ranked as a full agonist).
Table 3 – The displacing potencies of some m opioid receptor ligands in the rat brain receptor binding assay (RBA) against 3H-naloxone in the presence of Na+ and their dissociation constants determined by pharmacological means in the mouse vas deferens (MVD) bioassay (table for correlation analysis) Agonist
DAMGO Morphine Morphiceptin Endomorphin-1 Endomorphin-1-ol Dmt-endomorphin-1 Endomorphin-2 Endomorphin-2-ol (D-Met2)-endomorphin-2 a
b c
d
e
f g
Ki (nM) (RBA)a 210d 40e 5047 126f 812 0.54g 152f 131d 133
KA (nM) (MVD)b NULLc
OPERc
728 574 36244 58 445 2.07 248 71 105
2412 529 63755 135 470 5.04 356 305 160
Calculated by displacement curve analysis or from the 50% displacing concentrations (IC50) according to the Cheng–Prusoff equation. Agonist dissociation constants (data taken from Table 2). Calculated by ‘‘null’’ and ‘‘operational’’ (OPER) method, respectively. Calculated from Lengyel et al. [26] according to the Cheng– Prusoff equation. Calculated from Benyhe et al. [4] according to the Cheng–Prusoff equation. Spetea et al. [46]. Calculated from To´th et al. [50,51] according to the Cheng– Prusoff equation.
1511
The residual receptor fraction (q) after b-FNA treatment was calculated by the null method (Table 2). There were significantly higher q values for Dmt-EMO-1 as compared to the majority of the other agonists; some comparisons yielded significant difference also for EMO-1.
4.
Discussion
While the isolated organ technique has been used to estimate the contribution of intrinsic activity to the action of opioids in vitro [30] and the receptor pool inactivation approach to determine agonist affinities to opioid receptors (e.g. of DAMGO, [23]) the affinities and efficacies of agonists have not been explored systematically by both these methods simultaneously. The isolated organ of our choice was the mouse deferens (MVD) where all three major opioid receptor types, d, m and k are present (for review see [27]) but the spare receptor pools are different for each of these three types [30]. The receptor reserve is high for the d type, moderate for m and minor for k. The inactivating agent b-FNA affects irreversibly mainly the m type, to a lesser extent, the d type and acts as a reversible agonist at k opioid receptors [15,52]. From a technical point of view, the moderate m and the minor k receptor reserve is an advantage, the high d receptor pool a disadvantage. It means that the receptor-related parameters determined by this method can only be claimed to reflect parameters characteristic of agonist-m-opioid receptor type interaction if the receptor type preference/selectivity of agonist is suitably high. This criterion has been shown to be met by the agonists we studied [2,20,38,40,51]. In addition, the agonist dose range where this selectivity has been documented must be specified. We use the naltrexone-agonist interaction to establish opioid receptor type preference by a given agonist. Naltrexone, in the concentration range used in these experiments (109 to 108 M), caused a 3–30-fold rightward shift of the agonist dose–response curves. Thus, the statement on the receptor type preference of an agonist can be extended unambiguously only for conditions where the rightward shift does not exceed the upper limit indicated above. We confirmed that the agonists studied met this criterion too. We used both the ‘‘null’’ and the ‘‘operational’’ evaluation/calculation models to obtain receptor constants for the m-opioid agonists we set out to study. The KA values determined by the two methods correlated significantly with each other. In addition, they correlated also with the displacing potencies against 3H-naloxone in the rat brain receptor binding assay in the presence of Na+. In spite of the apparently higher resolution power of the ‘‘operational’’ method to differentiate between full and partial agonism, when using Dunnett’s test for statistical evaluation and DAMGO as reference full agonist, both methods gave identical ranking patterns. Thus, we could confirm by both calculations the previously found full agonism for DAMGA and the partial agonism for EMO-1 and EMO-1-ol. Furthermore, we added a number of endomorphin derivatives to the list of potential partial agonists such as Dmt-EMO-1, EMO-2ol and met2-EMO-2. EMO-2 appears to have an efficacy much closer to full agonism. The fact that the two natural endomorphins may have distinct pharmacological proper-
1512
peptides 27 (2006) 1507–1513
ties is potentially very interesting but by no means completely unexpected [36,41]. It has been reported that the signal transduction patterns by EMO-1 and EMO-2 may have some distinctive features [42]. EMO-1 and particularly its Dmt-derivative have another distinctive property in the MVD assay. The apparent residual receptor fraction (q) available for these agonists after b-FNA treatment is significantly higher than for the other ligands. Taking into consideration the biochemical studies on the site of irreversible action of b-FNA with the m-opioid receptor and the efforts to identify the ligand binding sites on the receptor molecule, only tentative explanations can be offered for this observation. It has been reported that bFNA interacts covalently with a single amino acid within the m-opioid receptor molecule, Lys233 in the fifth transmembrane helix [9]. Studies in chimerical and mutational receptor constructs [21,22,31,34,35] failed to result in a consensus as to the exact alignment of binding sites in the receptor molecule. It is likely that the binding site combinations are assembled by amino acids located in different transmembrane segments and extracellular loops. There is consensus, however, that different molecular elements of mopioid receptors are responsible for the selective binding of DAMGO on the one hand and of some morphinan-based ligands on the other hand. The apparently higher residual receptor fraction available for EMO-1 and Dmt-EMO-1 after b-FNA treatment might indicate that the structural elements in the receptor molecule responsible for the selective binding of endomorphin-1 and some related analogues may be different from the binding site for DAMGO and morphinan-based ligands. The partial agonism among the endomorphin derivatives, and the theoretical residual receptor fraction (q) available for these ligands may also suggest some useful trends for medicinal chemistry. In a family of agents acting at the same receptors (in this case at opioid receptors), therapeutic target selectivity can be provided primarily by structural modifications aimed at altering the receptor type/subtype selectivity profile. However, partial agonism may offer a further factor for functional drug selectivity. All the functions where the spare receptor pool is high will be affected both by full and partial receptor agonists. On the other hand, the functions where the receptor reserve is low may remain unaffected by partial agonists. Therefore, m-opioid receptor-mediated functions should be systematically mapped both in the CNS and at the periphery in terms of relative receptor reserve [1]. Then, the structural requirements of partial agonism for a ligand (in this case for the endomorphin derivatives) should be determined and, based on these informations, peptidomimetics could be designed.
Acknowledgements Authors are indebted to Ms. Lilla Gabriel for her skilled technical assistance and to Mr. Jeno˜ Balogh for PC-editing. The work was supported by ETT grant 262/2001 (AZR), OTKA grant TO-32736 (SF), Ja´nos Bolyai Fellowship (IL, GO) and Semmelweis PhD School Fund (MAK). We wish to thank Ms. Minou Williams for the linguistic corrections.
references
[1] Al-Khrasani M, Elor G, Abbas MY, Ro´nai AZ. The effect of endomorphins on the release of 3H-norepinephrine from rat nucleus tractus solitarii slices. Regul Peptides 2003;111:97–101. [2] Al-Khrasani M, Orosz G, Kocsis L, Farkas V, Magyar A, Lengyel I, et al. Receptor constants for endomorphin-1 and endomorphin-1-ol indicate differences in efficacy and receptor occupancy. Eur J Pharmacol 2001;421: 61–7. [3] Alt A, Mansour A, Akil H, Medzihradsky F, Traynor JR, Woods JH. Stimulation of guanosine-50 -O-(3[35S]thio)triphosphate binding by endogenous opioids acting at a cloned m receptor. J Pharmacol Exp Ther 1998;286:282–8. [4] Benyhe S, Hepp J, Simon J, Borsodi A, Medzihradszky K, Wollemann M. Tyr-D-Ala-(Me)Phe-chloromethyl ketone: a m-specific affinity label for the opioid receptor. Neuropeptides 1987;9:225–35. [5] Brantl V, Gramsch C, Lottspeich F, Henschen A, Jaeger K-H, Herz A. Novel opioid peptides derived from mitochondrial cytochrome b: cytochrophins. Eur J Pharmacol 1985;111:293–4. [6] Brantl V, Gramsch C, Lottspeich F, Mertz R, Jaeger K-H, Herz A. Novel opioid peptides derived from hemoglobin: hemorphins. Eur J Pharmacol 1986;12:309–10. [7] Brantl V, Teschemacher H, Henschen A, Lottspeich F. Novel opioid peptides derived from casein (b-casomorpohins) I. Isolation from bovine casein peptone. Hoppe-Seyler’s Z Physiol Chem 1979;360:1211–6. [8] Chang K-J, Killian A, Hazum E, Cuatrecasas P, Chang J-K. Morphiceptin (Tyr-Pro-Phe-Pro-NH2): a potent and specific agonist for morphine (m) receptor. Science 1981;212:75–7. [9] Chen C, Yin J, de Rief JK, de Jarlais RL, Raveglia LF. Determination of the amino acid residue involved in [3H]bfunaltrexamine covalent binding in the cloned rat m opioid receptor. J Biol Chem 1996;271:2142–9. [10] Erchegyi J, Kastin AJ, Zadina JE. Isolation of a novel tetrapeptide with opiate and antiopiate activity from human cortex: Tyr-Pro-Trp-Gly-NH2 (Tyr-W-MIF-1). Peptides 1992;13:623–31. [11] Fleming WW, Westfall DP, De la Lande IS, Jellett LB. Log-normal distribution of equieffective doses of norepinephrine and acetylcholine in several tissues. J Pharmacol Exp Ther 1972;181:339–45. [12] Furchgott RF. Pharmacological characterization of receptors: its relation to radioligand-binding studies. Fed Proc 1978;37:115–20. [13] Furchgott RF, Bursztyn P. Comparison of dissociation constants and of relative efficacies of selected agonists acting on parasympathetic receptors. Ann NY Acad Sci 1967;144:882–99. [14] Goldstein A, Naidu A. Multiple opioid receptors: ligand selectivity profiles and binding site signatures. Mol Pharmacol 1989;36:265–72. [15] Hayes AG, Sheehan MJ, Tyers MB. Determination of receptor selectivity of opioid agonists in the guinea-pig ileum and mouse vas deferens by use of b-funaltrexamine. Br J Pharmacol 1985;86:899–904. [16] Horva´th G. Endomorphin-1 and endomorphin-2: pharmacology of the selective endogenous mu-opioid receptor agonists. Pharmacol Ther 2000;88:437–63. [17] Hosohata K, Burkey TH, Alfarp-Lopez J, Varga E, Hruby VJ, Roeske WR, et al. Endomorphin-1 and endomorphin-2 are partial agonists at the human mu-opioid receptor. Eur J Pharmacol 1998;346:111–4.
peptides 27 (2006) 1507–1513
[18] Ho¨llt V. Opioid peptide processing and receptor selectivity. Annu Rev Pharmacol Toxicol 1986;26:59–77. [19] Kastin AJ, Stephens E, Zadina JE, Coy DH, Fischman AJ. TyrMIF-1, identified in brain tissue, and its analogs are active in two models of antinociception. Pharmacol Biochem Behav 1985;23:1045–9. [20] Kocsis L, Benyhe S, Borsodi A, Ro´nai AZ, Al-Khrasani M, Magyar A, et al. Endomorphin related peptide alcohols: structure–activity studies. In: Martinez J, Fehrentz JA., editors. Peptides 2000. Paris: EDK; 2000. p. 823–4. [21] Law P-Y, Loh HH. Regulation of opioid receptor activities. J Pharmacol Exp Ther 1999;289:607–24. [22] Law P-Y, Wong YH, Loh HH. Mutational analysis of the structure and function of opioid receptors. Biopolymers (Peptide Sci) 1999;51:440–5. [23] Leff P, Dougall IG. Estimation of opioid receptor agonist dissociation constants with b-chlornaltrexamine, an irreversible ligand which also displays agonism. Br J Pharmacol 1988;95:234–40. [24] Leff P, Harper D, Dainty IA, Dougall IG. Pharmacological estimation of agonist affinity: detection of errors that may be caused by the operation of receptor isomerisation or ternary complex mechanisms. Br J Pharmacol 1990;101: 55–60. [25] Leff P, Prentice DJ, Giles H, Martin GR, Wood J. Estimation of agonist affinity and efficacy by direct, operational modelfitting. J Pharmacol Meth 1990;23:225–37. [26] Lengyel I, Orosz G, Biyashev D, Kocsis L, Al-Khrasani M, Ro´nai AZ, et al. Side chain modifications change the binding and agonist properties of endomorphin 2. Biochem Biophys Res Commun 2002;290:153–61. [27] Leslie FM. Methods used for the study of opioid receptors. Pharmacol Rev 1987;39:197–249. [28] Lord JAH, Waterfield AA, Hughes J, Kosterlitz HW. Endogenous opioid peptides: multiple agonists and receptors. Nature 1977;267:495–9. [29] Martin WR, Eades CG, Thompson JA, Huppler RE, Gilbert PE. The effect of morphine-and morphine-like drugs in the nondependent and morphine-dependent chronic spinal dog. J Pharmacol Exp Ther 1976;197:517–32. [30] Miller L, Shaw JS, Whiting EM. The contribution of intrinsic activity to the action of opioids in vitro. Br J Pharmacol 1986;87:595–601. [31] Minami M, Nakagawa T, Seki T, Onogi T, Aoki Y, Katao Y, et al. A single residue Lys 108 of the d-opioid receptor prevents the m-opioid-selective ligand (D-Ala2,NMePhe4,Gly-ol5)enkephalin from binding to the d-opioid receptor. Mol Pharmacol 1996;50:1413–22. [32] Montecucchi PC, de Castiglione R, Piani S, Gozzini L, Erspamer V. Amino acid composition and sequence of dermorphin, a novel opiate-like peptide from the skin of Phyllomedusa sauvagei. Int J Peptide Prot Res 1981;17:275–83. [33] Narita M, Mizoguchi H, Oji GS, Tseng EL, Suganuma C, Nagase H, et al. Characterization of endomorphin-1 and-2 on [35S]GTPgS binding in the mouse spinal cord. Eur J Pharmacol 1998;351:383–7. [34] Onogi T, Minami M, Katao Y, Nakagawa T, Aoki Y, Toya T, et al. DAMGO, a m-opioid receptor selective agonist, distinguishes between m-and d-opioid receptors around their first extracellular loops. FEBS Lett 1995;357:93–7. [35] Quock RM, Burkey TH, Varga E, Hosohata Y, Hosohata K, Cowell SM, et al. The d-opioid receptor: molecular pharmacology, signal transduction, and the determination of drug efficacy. Pharmacol Rev 1999;51:503–32. [36] Przewlocki R, Przewlocka B. Opioids in chronic pain. Eur J Pharmacol 2001;429:79–91.
1513
[37] Richter K, Egger R, Negri L, Corsi R, Severini C, Kreil G. cDNAs encoding [D-Ala2]deltorphin precursors from skin of Phyllomedusa bicolor also contain genetic information for three dermorphin-related opioid peptides. Proc Natl Acad Sci USA 1990;87:4836–9. [38] Ro´nai AZ, Berze´tei I. Similarities and differences of opioid receptors in different isolated organs. In: Gra´f L, Palkovits M, Ro´nai AZ., editors. Endorphins ’78. Budapest: Akade´miai Kiado´; 1979. p. 237–57. [39] Ro´nai AZ, Gra´f L, Sze´kely JI, Dunai-Kova´cs Zs, Bajusz S. Differential behaviour of LPH-(61–91)-peptide in different model systems: comparison of the opioid activities of LPH-(61–91)-peptide and its fragments. FEBS Lett 1977;74:182–4. [40] Ro´nai AZ, Tima´r J, Mako´ E´, Erdoˇˇ" F, Gyarmati Zs. To´th G, Fu¨rst S, Sze´kely JI. Diprotin A, an inhibitor of dipeptidyl aminopeptidase IV (EC 3.4.14.5) produces naloxonereversible analgesia in rats. Life Sci 1998/99;64:142–52. [41] Sakurada S, Hayahashi T, Yuhki M, Fujimura T, Murayama K, Yonezawa A, et al. Differential antagonism of endomorphin-1 and emdomorphin-2 spinal antinociception by naloxonazine and 3methoxynaltrexone. Brain Res 2000;881:1–8. [42] Sa´nchez-Bla´zquez P, Rodriguez-Diaz M, DeAntonio I, Garzo´n J. Endomorphin-1and endomorphin-2 show differences in their activation of m opioid receptorregulated G proteins in supraspinal antinociception in mice. J Pharmacol Exp Ther 1999;291:12–8. [43] Sim LJ, Liu Q, Childers SR, Selley DE. Endomorphinstimulated [35S]GTPgS binding in rat brain: evidence for partial agonist activity at m-opioid receptors. J Neurochem 1998;70:1567–76. [44] Simon J, Benyhe S, Abutidze K, Borsodi A, Szuˇˇ "cs M, To´th G, et al. Kinetics and physical parameters of rat brain opioid receptors solubilised by digitonin and CHAPS. J Neurochem 1986;46:695–701. [45] Snedecor GW, Cochran WG. Statistical methods. Ames: Iowa State University Press; 1994. [46] Spetea M, Monory K, To¨mbo¨ly Cs, To´th G, Tzavara E, Benyhe S, et al. In vitro binding and signaling profile of the novel m opioid receptor agonist endomorphin 2 in rat brain membranes. Biochem Biophys Res Commun 1998;250:720–5. [47] Tallarida RJ. The use of drug-receptor affinity measures in the differentation of receptors. Fed Proc 1982;41: 2323–7. [48] Tallarida RJ. Pharmacologic methods for the identification of receptors. Life Sci 1988;43:2169–76. [49] To´th G, Kramer M, Sirokman F, Borsodi A, Ro´nai A. Preparation of (7,8,19,20-3H)naloxone of high specific activity. J Label Comp Radiopharm 1982;19:1021–30. [50] To´th G, To¨mbo¨ly Cs. Biyashev D, Ko¨ve´r K, Pe´ter A, Borsodi A, Al-Khrasani M, Ro´nai A, Labuz D, Przewlocka B. New endomorphin analogues using Dmt in position 1. In: Proceedings of the International Narcotics Research Conference; 2001. [51] To´th G, To¨mbo¨ly Cs, Pe´ter A, Biyashev D, Borsodi A, Ro´nai A, Przewlocki R. New endomorphin analogues: design, synthesis and biological properties. Peptides, in press. [52] Ward SJ, Portoghese PS, Takemori AE. Pharmacological profiles of b-funaltrexamine (b-FNA) and bchlornaltrexamine (b-CNA) on the mouse vas deferens preparation. Eur J Pharmacol 1982;80:377–84. [53] Zadina JE, Hackler L, Ge LJ, Kastin AJ. A potent and selective endogenous agonist for the m-opiate receptor. Nature 1997;386:499–502.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/peptides
Partial and full agonism in endomorphin derivatives: Comparison by null and operational model Andra´s Z. Ro´nai a,*, Mahmoud Al-Khrasani a, Sa´ndor Benyhe b, Imre Lengyel b, La´szlo´ Kocsis c, Gyo¨rgy Orosz c, Ge´za To´th b, Erzse´bet Kato´ a, La´szlo´ To´thfalusi d a
Department of Pharmacology and Pharmacotherapy, Semmelweis University, PO Box 370, H-1445 Budapest, Hungary Biological Research Centre, HAS, PO Box 521, H-6701 Szeged, Hungary c Research Group of Peptide Chemistry, Eo¨tvo¨s University/HAS, PO Box 32, H-1518 Budapest 112, Hungary d Department of Pharmacodynamics, Semmelweis University, Budapest, Hungary b
article info
abstract
Article history:
The partial m-opioid receptor pool inactivation strategy in isolated mouse vas deferens was
Received 24 August 2005
used to determine partial agonism of endomorphins and their analogs (endomorphin-1-ol,
Received in revised form
20 ,60 -dimethyltyrosine (Dmt)-endomorphin-1, endomorphin-2-ol and (D-Met2)-endomor-
7 December 2005
phin-2) using morphine, normorphine, morphiceptin, (D-Ala2,MePhe4,Gly5-ol)-enkephalin
Accepted 7 December 2005
(DAMGO) and its amide (DAMGA) as reference opioid agonists. Agonist affinities (KA) and
Published on line 18 January 2006
efficacies were assessed both by the ‘‘null’’ and the ‘‘operational’’ method. The KA values determined by the two methods correlated significantly with each other and also with the
Keywords:
displacing potencies against 3H-naloxone in the receptor binding assay in the presence of
Endomorphin derivatives
Na+. DAMGO and DAMGA were full agonist prototypes, morphine, endomorphin-1, endo-
Partial agonism
morphin-1-ol, Dmt-endomorphin-1, endomorphin-2-ol and (D-Met2)-endomorphin-2 were
Residual receptor fraction
found by both methods to be partial agonists whereas the parameters for normorphine,
b-Funaltrexamine
morphiceptin and endomorphin-2 were intermediate. # 2005 Elsevier Inc. All rights reserved.
Mouse vas deferens Null and operational model
1.
Introduction
Of the major opioid receptor types, i.e. d, k and m [28,29] dynorphins and neo-endorphins have a high preference for the k type, some proenkephalin-derived peptides for the d type, b-endorphin is equally active both at the m and d type and only some minor ones (e.g. the proenkephalin-derivative metorphamid) show a moderate m receptor type preference [14,18]. The discovery of endomorphins (1 and 2, Tyr-Pro-TrpPhe-NH2, EMO-1 and Tyr-Pro-Phe-Phe-NH2, EMO-2) in 1997 was a landmark in opioid peptide research [53]. The two tetrapeptides, isolated from bovine brain, were the first mammalian brain opioid peptides with high m-opioid receptor
selectivity and agonist potency. This selectivity of endomorphins is likely to be the consequence of their novel structure. The most common motif for an opioid agonist in the mammalian brain, and in other species, is the Tyr-Gly-GlyPhe-N-terminal core sequence. The Tyr-D-Ala/D-Met-Phemotifs were found only in amphibians (dermorphins, dermenkephalin and deltorphins, [32,37]), and the Tyr-Pro-Phe/ Trp- N-terminal sequences were identified in casomorphins, hemorphins and cytochrophin in milk and blood (most prominently in digests, [5–7]). Although some of these peptides, such as the b-casomorphin derivative morphiceptin (Tyr-Pro-Phe-Pro-NH2, [8]), have been reported to possess selective m-opioid receptor agonist properties, their potency
* Corresponding author. Tel.: +36 1 210 4416; fax: +36 1 210 4412. E-mail address: [email protected] (A.Z. Ro´nai). 0196-9781/$ – see front matter # 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2005.12.003
1508
peptides 27 (2006) 1507–1513
falls into the low or mid-micromolar range. Tyr-MIF-1 (TyrPro-Leu-Gly-NH2, [19]) and Tyr-W-MIF-1 (Tyr-Pro-Trp-GlyNH2, [10]) were the first Tyr-Pro-related brain opioid peptides, to which endomorphins belong, but these peptides were antagonists rather than agonists, acting in the mid-micromolar range. It has been suggested that endomorphins serve as peptide neurotransmitters/modulators [16] and they do, indeed, meet several of the requisite criteria. However they lack some important features (the mode of endogenous generation, the firm establishment of Ca2+-dependent release) which would contribute to substantiate this notion. Furthermore, endomorphins have been reported to display partial agonist properties in the [35S]GTPgS-binding assay [3,17,26,33,43]; an unusual feature for a first messenger candidate. Recently, we confirmed the partial agonism for endomorphin-1 and its-olderivative by using the fractional receptor pool inactivation strategy [12,13,15,47,48,52] using the field-stimulated mouse vas deferens (MVD, 2). Using the MVD method, in this paper we extended our analysis to Dmt-EMO-1, EMO-2, EMO-2-ol and (DMet2)-EMO-2 (met2-EMO-2). DAMGO, DAMGA, morphine, normorphine and morphiceptin were used as reference mopioid receptor agonists. Besides the partial receptor pool inactivation and the so-called double reciprocal plot [13,47], referred to as the ‘‘null’’ method [24,25], for our current analysis we used also the ‘‘operational’’ model devised by Leff and his coworkers [23–25]. Both methods have been devised to characterize agonist affinity and efficacy by pharmacological means on isolated organ preparations. The ‘‘null’’ method was developed avoiding the need to make assumptions about postreceptor events whereas the ‘‘operational’’ method calibrates first the parameters of receptorial transducer function by constructing an equation for the [E]/[A] relation of a full agonists then, using these parameters, extends the relationship to any other agonist.
2.
Materials and methods
The conditions of animal housing and experimentation followed the ethical guidelines set by the Ethical Board of Semmelweis University, based on EC Directive 86/609/EEC.
2.1.
Materials
b-Funaltrexamine hydrochloride (b-FNA) was obtained from Tocris Cockson Ltd. (Bristol, UK), normorphine base and morphine sulfate from ICN Alkaloida Ltd. (Tiszavasva´ri, Hungary). 3H-naloxone (57.0 Ci/mmol) was prepared by G. To´th of Biological Research Centre, Szeged, Hungary [49]. The syntheses of peptide amides and alcohols (-olderivatives) were performed by solid-phase. The details of synthetic and analytical procedures are described elsewhere [2,20,26,40]. 20 ,60 -dimethyl-Tyr1-endomorphin-1 was synthesized by G. To´th of Biological Research Centre, Szeged, Hungary [50,51]. According to the analytical proofsheets received with each peptide batch, the peptides had the correct amino acid composition and an at least 95% HPLC purity; batches of Dmt-EMO-1 were also routinely subjected to mass spectrometry. All the other substances used were of analy-
tical grade and were obtained either from Sigma–Aldrich Ltd. (St. Louis, USA) or Reanal Ltd. (Budapest, Hungary).
2.2.
Methods
2.2.1.
Rat brain receptor binding assay (RBA)
Data obtained in the receptor binding assay are used exclusively for correlation analytical purposes, and taken mostly from specified literature sources. An important aspect of data set selection was that the values be generated in the same laboratory and according to a uniform experimental paradigm; therefore, a brief methodological description is given below. Assays were carried out in crude brain membrane fraction prepared from Wistar rats as described previously [44]. Membranes were incubated in 50 mM Tris– HCl buffer (pH 7.4) supplemented with 100 mM NaCl at 0 8C for 60 min. The concentration of 3H-naloxone (57 Ci/mmol specific activity, 49) used in the displacement experiments was 1.0 nM. All experiments were carried out in duplicate assays and repeated at least three times.
2.2.2.
Isolated organ preparation
Vasa deferentia taken from CFLP mice weighing 35–45 g were prepared, mounted and stimulated as described previously [39]. In brief, a single vas per bath was mounted in Mg2+-free Krebs’ solution aerated with carbogen (O2:CO2 = 95:5) at 31 8C under an initial tension of 0.1 g. The stimulation parameters were as follows: field stimulation of pairs (100 ms pulse distance) of rectangular impulses (1 ms pulse width, 9 V/cm i.e. supramaximal intensity) were repeated every 10 s.
2.2.3.
Experimental paradigms
Following 30–40 min equilibration isolated vasa deferentia were exposed to agonists for up to 2 min with the exception of Dmt-endomorphin-1, where exposure was 10–25 min. Administration cycle was 12–18 min with three to four interim washes with the exception of Dmt-endomorphin-1 where it was 40–60 min with 8–12 washes. Incubation conditions with b-FNA and the construction of before–after dose–response curves with the agonists were the same as used previously [2]. In brief, the agonist dose– response curves were taken non-cumulatively at 4–6 concentration levels; the vasa were then exposed to 5 107 M b-FNA for 30 min. The exposure was followed by a 60-min washout period with 12 washes then the agonist dose– response curves were re-taken at 4–6 pre-set concentration levels. The proper concentration range for each agonist was determined in a pilot experiment.
2.2.4.
Evaluation
To obtain the parameters of agonist concentration–response (E/[A]) curves before and after b-FNA treatment non-linear curve fitting according to the Hill equation (three parameters, Eq. (1)) was used: E¼
Emax ½An ½A50 n þ ½An
(1)
where ‘‘E’’ is the biological response (percent inhibitory effect), ‘‘Emax’’ the maximal effect, ‘‘[A]’’ the molar concentration of agonist, ‘‘[A50]’’ the 50% effective concentration and ‘‘n’’ is the
peptides 27 (2006) 1507–1513
Hill slope. The ‘‘null’’ method, i.e. the double-reciprocal plot of equieffective concentrations of agonists before ([A], ordinate) and after ([A0 ], abscissa) b-FNA treatment [13,47] was carried out as described previously [2]. From the slope and ‘‘y’’ inter-
1509
cept the residual receptor fraction ‘‘q’’ and agonist dissociation constant ‘‘KA’’ were calculated. According to the operational model [23–25] the ‘‘Emax’’ and ‘‘n’’ parameters for the full agonist prototypes were determined by using Eq. (1) then these
Fig. 1 – The inhibitory dose–response curves of m-opioid receptor agonists in the field-stimulated mouse vas deferens before and after b-funaltrexamine treatment. The inhibitory dose–response curves of DAMGO (panel ‘‘A’’), DAMGA (panel ‘‘B’’), morphine (panel ‘‘C’’), normorphine (panel ‘‘D’’), morphiceptin (panel ‘‘E’’), endomorphin-1 (panel ‘‘F’’), endomorphin-1-ol (panel ‘‘G’’), Dmt-endomorphin-1 (panel ‘‘H’’), endomorphin-2 (panel ‘‘I’’) and endomorphin-2-ol (panel ‘‘J’’) before (dark circles) and after (open circles) 30-min incubation with 5 10S7 M b-funaltrexamine. Points represent the mean, vertical lines the S.E.M. of 4–6 independent determinations. Hill equation (three parameters) was used for curve fitting.
1510
peptides 27 (2006) 1507–1513
Table 1 – The dose–response curve parameters of m-opioid agonists in the mouse vas deferens before and after bfunaltrexamine treatment A50 (nM)a
Agonist (n)
DAMGO (5) DAMGA (4) Morphine (4) Normorphine (4) Morphiceptin (4) EMO-1 (6) EMO-1-ol (4) Dmt-EMO-1 (6) EMO-2 (4) EMO-2-ol (6) met2-EMO-2 (4)
Emax (%)b
Hill slopeb
Cc
Fc
Cc
Fc
Cc
Fc
46.9 (36.4–60.4) 17.0 (12.9–22.2) 140.0 (111.3.176.0) 181.2 (133.7–245.6) 2619 (2381–2882) 23.2 (16.4–32.9) 111.0 (82.1–150.1) 1.48 (1.11–1.99) 17.2 (14.5–20.5) 26.3 (20.3–34.1) 42.2 (28.0–63.5)
482.7 (263.0–641.8) 186.9 (153.3–227.9) 937.1 (671.1–1309) 4202 (3570–4945) 43342 (25490–73696) 242.1 (170.3–344.3) 1450 (818.4–2570) 4.54 (2.30–8.97) 127.1 (82.3–196.3) 212.1 (99.5–451.8) 192.5 (115.9–319.8)
96.3 2.2 96.7 5.6 81.6 0.7 94.3 3.5 99.2 1.7 96.1 2.8 99.0 0.5 99.8 0.2 99.5 2.7 98.5 3.3 89.5 5.3
97.7 1.9 98.5 1.9 34.4 3.4 96.3 3.0 91.9 0.6 95.2 3.3 67.9 9.1 63.2 7.7 79.5 3.1 84.7 2.9 53.9 8.6
1.62 0.10 1.37 0.11 1.52 0.15 1.40 0.16 1.35 0.06 1.39 0.13 1.17 0.07 1.57 0.10 1.61 0.13 1.52 0.11 1.19 0.07
1.02 0.04 1.04 0.06 1.01 0.09 0.60 0.04 0.83 0.08 0.48 0.06 0.46 0.10 1.01 0.24 1.11 0.04 0.84 0.10 0.76 0.10
The parameters were obtained from curve fitting by the Hill equation (three parameters). For the 50% effective concentration (A50) geometric mean and 95% confidence interval was given. b For Emax and Hill slope arithmetic mean S.E.M. was given. c C (control): before b-FNA; F: after 30 min incubation with 5 107 M b-FNA. a
parameters were used to determine the receptor constants for all the agonists by another equation E¼
Emax tn ½An ðKA þ ½AÞn þ tn ½An
(2)
where ‘‘t’’ is the model definition of efficacy, ‘‘KA’’ the agonist dissociation constant and other parameters as in Eq. (1). For convenience, agonist efficacies, obtained either by the null or the operational method, were characterized uniformly by the KA/[A50] ratios. For pooled [A50] and KA values geometric means and 95% confidence intervals [11] were calculated otherwise arithmetic mean S.E.M. values were listed. Statistical probes were applied to the logarithms of data sets. ANOVA followed by Dunnett’s test was used to compare efficacies, taking DAMGO as prototype full agonist for multiple comparisons.
Newman–Keuls post hoc test was applied in the case of residual receptor fraction (‘‘q’’, null method) values whereas Student’s ‘‘t’’-test was used to compare ‘‘before–after’’ data sets (paired arrangement). For correlation analysis, correlation coefficient (r) values were calculated [45].
3.
Results
The dose–response curves of DAMGO, DAMGA, morphine, normorphine, morphiceptin, EMO-1, EMO-1-ol, Dmt-EMO-1, EMO-2 and EMO-2-ol before and after b-FNA treatment are shown in Fig. 1. The changes can be sorted into three separate subgroups. (i) There was an almost parallel rightward shift with no Emax depression for DAMGO and DAMGA. (ii) There was a pronounced Emax depression ( p < 0.01) with
Table 2 – The receptor constants of m-opioid agonists in the mouse vas deferens KA (nM)a
Agonist (n) NULL DAMGO (5) DAMGA (4) Morphine (4) Normorphine (4) Morphiceptin (4) EMO-1 (6) EMO-1-ol (4) Dmt-EMO-1 (6) EMO-2 (4) EMO-2-ol (6) met2-EMO-2 (4)
727.7 (489.3–1220) 355.4 (274.3–460.5) 574.2 (257.5–1280) 1748 (1542–1981) 36 244 (28967–45350) 57.8 (24.0–139.6) 444.5 (220.5–895.9) 2.07 (0.53–8.04) 247.5 (188.1–325.6) 71.3 (35.0–145.3) 104.5 (70.2–155.1)
q (%)c
KA/A50b OPER
2412 (1281–4543) 1058 (583.3–1919) 528.7 (462.7–604.1) 3395 (2374–4855) 63 755 (49181–82649) 135.3 (64.1–285.5) 470.2 (325.9–678.4) 5.04 (3.23–7.86) 356.0 (275.5–460.0) 304.6 (229.3–404.5) 160.4 (110.4–233.1)
NULL
OPER
NULL
17.4 2.6 21.7 2.9 4.86 1.39 10.2 1.7 14.4 2.0 3.19 0.78 4.82 1.48 3.57 1.71 14.7 2.9 3.30 0.79 2.86 0.68
56.7 11.3 71.5 17.3 3.99 0.69 20.5 4.2 25.1 3.2 6.95 1.57 4.72 1.05 3.69 0.62 20.3 2.5 12.9 2.2 5.03 1.72
17.7 2.5 14.4 1.5 16.4 4.6 14.9 2.0 12.60.6 40.1 6.7 19.3 6.2 53.0 12.6 13.3 1.0 34.8 4.1 25.7 2.8
The parameters were determined by the ‘‘null’’ method (based on the double reciprocal plot of equieffective agonist concentrations before and after b-FNA) or the ‘‘operational’’ (OPER) method [24,25]. Statistics: for KA/A50 ratios: by ANOVA followed by Dunnett’s multiple comparison test DAMGO vs. morphine, EMO-1, EMO-1-ol, Dmt-EMO-1, EMO-2-ol and met2-EMO-2: p < 0.001 both by null and operational method; For residual receptor fraction ‘‘q’’: by ANOVA followed by Newman–Keuls test Dmt-EMO-1 vs. morphiceptin p < 0.01, Dmt-EMO-1 vs. DAMGO, DAMGA, morphine, normorphine and EMO-2 p < 0.05; EMO-1 vs. morphiceptin and EMO-2 p < 0.05. a For the agonist dissociation constant (KA) geometric mean and 95% confidence interval was given. b For the efficacy-related KA/A50 ratio arithmetic mean S.E.M. was given. c The residual receptor fraction ‘‘q’’ was expressed in percent; arithmetic mean S.E.M. was listed.
peptides 27 (2006) 1507–1513
moderate slope reduction for morphine, Dmt-EMO-1 and met2-EMO-2 (latter is not shown in the figure). (iii) There was a significant slope reduction with no or modest Emax depression for normorphine, EMO-1, and EMO-1-ol (see also Table 1). We determined the ‘‘Emax’’ and ‘‘n’’ values for DAMGO and DAMGA using the Hill equation (Eq. (1)); these were then used for calculations in the operational procedure. According to a previous analysis [2] both DAMGO and DAMGA appeared to fulfill the criteria of full agonism, therefore their Emax and ‘‘n’’ parameters were pooled. The pooled Emax value was 96.5 2.8 (n = 9) and the Hill slope 1.51 0.08 (n = 9). When calculating the agonist dissociation constants by either the null or the operational methods (Table 2), the KA values correlated significantly with each other (Pearson correlation coefficient r = 0.9998, p < 0.0001). Similarly, good correlation was observed between the Ki values determined in the rat brain receptor binding assay in the presence of Na+ (Table 3, r = 0.9900 for null and r = 0.9892 for operational method, respectively, p < 0.0001 for both). Likewise, there was a significant correlation (r = 0.9394) between the efficacies calculated by the two methods. Furthermore, when using Dunnett’s post hoc test for differentiating between full and partial agonism, using DAMGO as prototype full agonist, both methods gave a similar result. The test qualified morphine, EMO-1, EMO-1-ol, Dmt-EMO-1, EMO-2-ol and met2-EMO-2 as partial agonists, and left normorphine, morphiceptin and EMO-2 in the category of potential full agonism (DAMGA ranked as a full agonist).
Table 3 – The displacing potencies of some m opioid receptor ligands in the rat brain receptor binding assay (RBA) against 3H-naloxone in the presence of Na+ and their dissociation constants determined by pharmacological means in the mouse vas deferens (MVD) bioassay (table for correlation analysis) Agonist
DAMGO Morphine Morphiceptin Endomorphin-1 Endomorphin-1-ol Dmt-endomorphin-1 Endomorphin-2 Endomorphin-2-ol (D-Met2)-endomorphin-2 a
b c
d
e
f g
Ki (nM) (RBA)a 210d 40e 5047 126f 812 0.54g 152f 131d 133
KA (nM) (MVD)b NULLc
OPERc
728 574 36244 58 445 2.07 248 71 105
2412 529 63755 135 470 5.04 356 305 160
Calculated by displacement curve analysis or from the 50% displacing concentrations (IC50) according to the Cheng–Prusoff equation. Agonist dissociation constants (data taken from Table 2). Calculated by ‘‘null’’ and ‘‘operational’’ (OPER) method, respectively. Calculated from Lengyel et al. [26] according to the Cheng– Prusoff equation. Calculated from Benyhe et al. [4] according to the Cheng–Prusoff equation. Spetea et al. [46]. Calculated from To´th et al. [50,51] according to the Cheng– Prusoff equation.
1511
The residual receptor fraction (q) after b-FNA treatment was calculated by the null method (Table 2). There were significantly higher q values for Dmt-EMO-1 as compared to the majority of the other agonists; some comparisons yielded significant difference also for EMO-1.
4.
Discussion
While the isolated organ technique has been used to estimate the contribution of intrinsic activity to the action of opioids in vitro [30] and the receptor pool inactivation approach to determine agonist affinities to opioid receptors (e.g. of DAMGO, [23]) the affinities and efficacies of agonists have not been explored systematically by both these methods simultaneously. The isolated organ of our choice was the mouse deferens (MVD) where all three major opioid receptor types, d, m and k are present (for review see [27]) but the spare receptor pools are different for each of these three types [30]. The receptor reserve is high for the d type, moderate for m and minor for k. The inactivating agent b-FNA affects irreversibly mainly the m type, to a lesser extent, the d type and acts as a reversible agonist at k opioid receptors [15,52]. From a technical point of view, the moderate m and the minor k receptor reserve is an advantage, the high d receptor pool a disadvantage. It means that the receptor-related parameters determined by this method can only be claimed to reflect parameters characteristic of agonist-m-opioid receptor type interaction if the receptor type preference/selectivity of agonist is suitably high. This criterion has been shown to be met by the agonists we studied [2,20,38,40,51]. In addition, the agonist dose range where this selectivity has been documented must be specified. We use the naltrexone-agonist interaction to establish opioid receptor type preference by a given agonist. Naltrexone, in the concentration range used in these experiments (109 to 108 M), caused a 3–30-fold rightward shift of the agonist dose–response curves. Thus, the statement on the receptor type preference of an agonist can be extended unambiguously only for conditions where the rightward shift does not exceed the upper limit indicated above. We confirmed that the agonists studied met this criterion too. We used both the ‘‘null’’ and the ‘‘operational’’ evaluation/calculation models to obtain receptor constants for the m-opioid agonists we set out to study. The KA values determined by the two methods correlated significantly with each other. In addition, they correlated also with the displacing potencies against 3H-naloxone in the rat brain receptor binding assay in the presence of Na+. In spite of the apparently higher resolution power of the ‘‘operational’’ method to differentiate between full and partial agonism, when using Dunnett’s test for statistical evaluation and DAMGO as reference full agonist, both methods gave identical ranking patterns. Thus, we could confirm by both calculations the previously found full agonism for DAMGA and the partial agonism for EMO-1 and EMO-1-ol. Furthermore, we added a number of endomorphin derivatives to the list of potential partial agonists such as Dmt-EMO-1, EMO-2ol and met2-EMO-2. EMO-2 appears to have an efficacy much closer to full agonism. The fact that the two natural endomorphins may have distinct pharmacological proper-
1512
peptides 27 (2006) 1507–1513
ties is potentially very interesting but by no means completely unexpected [36,41]. It has been reported that the signal transduction patterns by EMO-1 and EMO-2 may have some distinctive features [42]. EMO-1 and particularly its Dmt-derivative have another distinctive property in the MVD assay. The apparent residual receptor fraction (q) available for these agonists after b-FNA treatment is significantly higher than for the other ligands. Taking into consideration the biochemical studies on the site of irreversible action of b-FNA with the m-opioid receptor and the efforts to identify the ligand binding sites on the receptor molecule, only tentative explanations can be offered for this observation. It has been reported that bFNA interacts covalently with a single amino acid within the m-opioid receptor molecule, Lys233 in the fifth transmembrane helix [9]. Studies in chimerical and mutational receptor constructs [21,22,31,34,35] failed to result in a consensus as to the exact alignment of binding sites in the receptor molecule. It is likely that the binding site combinations are assembled by amino acids located in different transmembrane segments and extracellular loops. There is consensus, however, that different molecular elements of mopioid receptors are responsible for the selective binding of DAMGO on the one hand and of some morphinan-based ligands on the other hand. The apparently higher residual receptor fraction available for EMO-1 and Dmt-EMO-1 after b-FNA treatment might indicate that the structural elements in the receptor molecule responsible for the selective binding of endomorphin-1 and some related analogues may be different from the binding site for DAMGO and morphinan-based ligands. The partial agonism among the endomorphin derivatives, and the theoretical residual receptor fraction (q) available for these ligands may also suggest some useful trends for medicinal chemistry. In a family of agents acting at the same receptors (in this case at opioid receptors), therapeutic target selectivity can be provided primarily by structural modifications aimed at altering the receptor type/subtype selectivity profile. However, partial agonism may offer a further factor for functional drug selectivity. All the functions where the spare receptor pool is high will be affected both by full and partial receptor agonists. On the other hand, the functions where the receptor reserve is low may remain unaffected by partial agonists. Therefore, m-opioid receptor-mediated functions should be systematically mapped both in the CNS and at the periphery in terms of relative receptor reserve [1]. Then, the structural requirements of partial agonism for a ligand (in this case for the endomorphin derivatives) should be determined and, based on these informations, peptidomimetics could be designed.
Acknowledgements Authors are indebted to Ms. Lilla Gabriel for her skilled technical assistance and to Mr. Jeno˜ Balogh for PC-editing. The work was supported by ETT grant 262/2001 (AZR), OTKA grant TO-32736 (SF), Ja´nos Bolyai Fellowship (IL, GO) and Semmelweis PhD School Fund (MAK). We wish to thank Ms. Minou Williams for the linguistic corrections.
references
[1] Al-Khrasani M, Elor G, Abbas MY, Ro´nai AZ. The effect of endomorphins on the release of 3H-norepinephrine from rat nucleus tractus solitarii slices. Regul Peptides 2003;111:97–101. [2] Al-Khrasani M, Orosz G, Kocsis L, Farkas V, Magyar A, Lengyel I, et al. Receptor constants for endomorphin-1 and endomorphin-1-ol indicate differences in efficacy and receptor occupancy. Eur J Pharmacol 2001;421: 61–7. [3] Alt A, Mansour A, Akil H, Medzihradsky F, Traynor JR, Woods JH. Stimulation of guanosine-50 -O-(3[35S]thio)triphosphate binding by endogenous opioids acting at a cloned m receptor. J Pharmacol Exp Ther 1998;286:282–8. [4] Benyhe S, Hepp J, Simon J, Borsodi A, Medzihradszky K, Wollemann M. Tyr-D-Ala-(Me)Phe-chloromethyl ketone: a m-specific affinity label for the opioid receptor. Neuropeptides 1987;9:225–35. [5] Brantl V, Gramsch C, Lottspeich F, Henschen A, Jaeger K-H, Herz A. Novel opioid peptides derived from mitochondrial cytochrome b: cytochrophins. Eur J Pharmacol 1985;111:293–4. [6] Brantl V, Gramsch C, Lottspeich F, Mertz R, Jaeger K-H, Herz A. Novel opioid peptides derived from hemoglobin: hemorphins. Eur J Pharmacol 1986;12:309–10. [7] Brantl V, Teschemacher H, Henschen A, Lottspeich F. Novel opioid peptides derived from casein (b-casomorpohins) I. Isolation from bovine casein peptone. Hoppe-Seyler’s Z Physiol Chem 1979;360:1211–6. [8] Chang K-J, Killian A, Hazum E, Cuatrecasas P, Chang J-K. Morphiceptin (Tyr-Pro-Phe-Pro-NH2): a potent and specific agonist for morphine (m) receptor. Science 1981;212:75–7. [9] Chen C, Yin J, de Rief JK, de Jarlais RL, Raveglia LF. Determination of the amino acid residue involved in [3H]bfunaltrexamine covalent binding in the cloned rat m opioid receptor. J Biol Chem 1996;271:2142–9. [10] Erchegyi J, Kastin AJ, Zadina JE. Isolation of a novel tetrapeptide with opiate and antiopiate activity from human cortex: Tyr-Pro-Trp-Gly-NH2 (Tyr-W-MIF-1). Peptides 1992;13:623–31. [11] Fleming WW, Westfall DP, De la Lande IS, Jellett LB. Log-normal distribution of equieffective doses of norepinephrine and acetylcholine in several tissues. J Pharmacol Exp Ther 1972;181:339–45. [12] Furchgott RF. Pharmacological characterization of receptors: its relation to radioligand-binding studies. Fed Proc 1978;37:115–20. [13] Furchgott RF, Bursztyn P. Comparison of dissociation constants and of relative efficacies of selected agonists acting on parasympathetic receptors. Ann NY Acad Sci 1967;144:882–99. [14] Goldstein A, Naidu A. Multiple opioid receptors: ligand selectivity profiles and binding site signatures. Mol Pharmacol 1989;36:265–72. [15] Hayes AG, Sheehan MJ, Tyers MB. Determination of receptor selectivity of opioid agonists in the guinea-pig ileum and mouse vas deferens by use of b-funaltrexamine. Br J Pharmacol 1985;86:899–904. [16] Horva´th G. Endomorphin-1 and endomorphin-2: pharmacology of the selective endogenous mu-opioid receptor agonists. Pharmacol Ther 2000;88:437–63. [17] Hosohata K, Burkey TH, Alfarp-Lopez J, Varga E, Hruby VJ, Roeske WR, et al. Endomorphin-1 and endomorphin-2 are partial agonists at the human mu-opioid receptor. Eur J Pharmacol 1998;346:111–4.
peptides 27 (2006) 1507–1513
[18] Ho¨llt V. Opioid peptide processing and receptor selectivity. Annu Rev Pharmacol Toxicol 1986;26:59–77. [19] Kastin AJ, Stephens E, Zadina JE, Coy DH, Fischman AJ. TyrMIF-1, identified in brain tissue, and its analogs are active in two models of antinociception. Pharmacol Biochem Behav 1985;23:1045–9. [20] Kocsis L, Benyhe S, Borsodi A, Ro´nai AZ, Al-Khrasani M, Magyar A, et al. Endomorphin related peptide alcohols: structure–activity studies. In: Martinez J, Fehrentz JA., editors. Peptides 2000. Paris: EDK; 2000. p. 823–4. [21] Law P-Y, Loh HH. Regulation of opioid receptor activities. J Pharmacol Exp Ther 1999;289:607–24. [22] Law P-Y, Wong YH, Loh HH. Mutational analysis of the structure and function of opioid receptors. Biopolymers (Peptide Sci) 1999;51:440–5. [23] Leff P, Dougall IG. Estimation of opioid receptor agonist dissociation constants with b-chlornaltrexamine, an irreversible ligand which also displays agonism. Br J Pharmacol 1988;95:234–40. [24] Leff P, Harper D, Dainty IA, Dougall IG. Pharmacological estimation of agonist affinity: detection of errors that may be caused by the operation of receptor isomerisation or ternary complex mechanisms. Br J Pharmacol 1990;101: 55–60. [25] Leff P, Prentice DJ, Giles H, Martin GR, Wood J. Estimation of agonist affinity and efficacy by direct, operational modelfitting. J Pharmacol Meth 1990;23:225–37. [26] Lengyel I, Orosz G, Biyashev D, Kocsis L, Al-Khrasani M, Ro´nai AZ, et al. Side chain modifications change the binding and agonist properties of endomorphin 2. Biochem Biophys Res Commun 2002;290:153–61. [27] Leslie FM. Methods used for the study of opioid receptors. Pharmacol Rev 1987;39:197–249. [28] Lord JAH, Waterfield AA, Hughes J, Kosterlitz HW. Endogenous opioid peptides: multiple agonists and receptors. Nature 1977;267:495–9. [29] Martin WR, Eades CG, Thompson JA, Huppler RE, Gilbert PE. The effect of morphine-and morphine-like drugs in the nondependent and morphine-dependent chronic spinal dog. J Pharmacol Exp Ther 1976;197:517–32. [30] Miller L, Shaw JS, Whiting EM. The contribution of intrinsic activity to the action of opioids in vitro. Br J Pharmacol 1986;87:595–601. [31] Minami M, Nakagawa T, Seki T, Onogi T, Aoki Y, Katao Y, et al. A single residue Lys 108 of the d-opioid receptor prevents the m-opioid-selective ligand (D-Ala2,NMePhe4,Gly-ol5)enkephalin from binding to the d-opioid receptor. Mol Pharmacol 1996;50:1413–22. [32] Montecucchi PC, de Castiglione R, Piani S, Gozzini L, Erspamer V. Amino acid composition and sequence of dermorphin, a novel opiate-like peptide from the skin of Phyllomedusa sauvagei. Int J Peptide Prot Res 1981;17:275–83. [33] Narita M, Mizoguchi H, Oji GS, Tseng EL, Suganuma C, Nagase H, et al. Characterization of endomorphin-1 and-2 on [35S]GTPgS binding in the mouse spinal cord. Eur J Pharmacol 1998;351:383–7. [34] Onogi T, Minami M, Katao Y, Nakagawa T, Aoki Y, Toya T, et al. DAMGO, a m-opioid receptor selective agonist, distinguishes between m-and d-opioid receptors around their first extracellular loops. FEBS Lett 1995;357:93–7. [35] Quock RM, Burkey TH, Varga E, Hosohata Y, Hosohata K, Cowell SM, et al. The d-opioid receptor: molecular pharmacology, signal transduction, and the determination of drug efficacy. Pharmacol Rev 1999;51:503–32. [36] Przewlocki R, Przewlocka B. Opioids in chronic pain. Eur J Pharmacol 2001;429:79–91.
1513
[37] Richter K, Egger R, Negri L, Corsi R, Severini C, Kreil G. cDNAs encoding [D-Ala2]deltorphin precursors from skin of Phyllomedusa bicolor also contain genetic information for three dermorphin-related opioid peptides. Proc Natl Acad Sci USA 1990;87:4836–9. [38] Ro´nai AZ, Berze´tei I. Similarities and differences of opioid receptors in different isolated organs. In: Gra´f L, Palkovits M, Ro´nai AZ., editors. Endorphins ’78. Budapest: Akade´miai Kiado´; 1979. p. 237–57. [39] Ro´nai AZ, Gra´f L, Sze´kely JI, Dunai-Kova´cs Zs, Bajusz S. Differential behaviour of LPH-(61–91)-peptide in different model systems: comparison of the opioid activities of LPH-(61–91)-peptide and its fragments. FEBS Lett 1977;74:182–4. [40] Ro´nai AZ, Tima´r J, Mako´ E´, Erdoˇˇ" F, Gyarmati Zs. To´th G, Fu¨rst S, Sze´kely JI. Diprotin A, an inhibitor of dipeptidyl aminopeptidase IV (EC 3.4.14.5) produces naloxonereversible analgesia in rats. Life Sci 1998/99;64:142–52. [41] Sakurada S, Hayahashi T, Yuhki M, Fujimura T, Murayama K, Yonezawa A, et al. Differential antagonism of endomorphin-1 and emdomorphin-2 spinal antinociception by naloxonazine and 3methoxynaltrexone. Brain Res 2000;881:1–8. [42] Sa´nchez-Bla´zquez P, Rodriguez-Diaz M, DeAntonio I, Garzo´n J. Endomorphin-1and endomorphin-2 show differences in their activation of m opioid receptorregulated G proteins in supraspinal antinociception in mice. J Pharmacol Exp Ther 1999;291:12–8. [43] Sim LJ, Liu Q, Childers SR, Selley DE. Endomorphinstimulated [35S]GTPgS binding in rat brain: evidence for partial agonist activity at m-opioid receptors. J Neurochem 1998;70:1567–76. [44] Simon J, Benyhe S, Abutidze K, Borsodi A, Szuˇˇ "cs M, To´th G, et al. Kinetics and physical parameters of rat brain opioid receptors solubilised by digitonin and CHAPS. J Neurochem 1986;46:695–701. [45] Snedecor GW, Cochran WG. Statistical methods. Ames: Iowa State University Press; 1994. [46] Spetea M, Monory K, To¨mbo¨ly Cs, To´th G, Tzavara E, Benyhe S, et al. In vitro binding and signaling profile of the novel m opioid receptor agonist endomorphin 2 in rat brain membranes. Biochem Biophys Res Commun 1998;250:720–5. [47] Tallarida RJ. The use of drug-receptor affinity measures in the differentation of receptors. Fed Proc 1982;41: 2323–7. [48] Tallarida RJ. Pharmacologic methods for the identification of receptors. Life Sci 1988;43:2169–76. [49] To´th G, Kramer M, Sirokman F, Borsodi A, Ro´nai A. Preparation of (7,8,19,20-3H)naloxone of high specific activity. J Label Comp Radiopharm 1982;19:1021–30. [50] To´th G, To¨mbo¨ly Cs. Biyashev D, Ko¨ve´r K, Pe´ter A, Borsodi A, Al-Khrasani M, Ro´nai A, Labuz D, Przewlocka B. New endomorphin analogues using Dmt in position 1. In: Proceedings of the International Narcotics Research Conference; 2001. [51] To´th G, To¨mbo¨ly Cs, Pe´ter A, Biyashev D, Borsodi A, Ro´nai A, Przewlocki R. New endomorphin analogues: design, synthesis and biological properties. Peptides, in press. [52] Ward SJ, Portoghese PS, Takemori AE. Pharmacological profiles of b-funaltrexamine (b-FNA) and bchlornaltrexamine (b-CNA) on the mouse vas deferens preparation. Eur J Pharmacol 1982;80:377–84. [53] Zadina JE, Hackler L, Ge LJ, Kastin AJ. A potent and selective endogenous agonist for the m-opiate receptor. Nature 1997;386:499–502.