e at SciVerse ScienceDirect

Toxicon 75 (2013) 160–167

Contents lists availabl

Toxicon

journal homepage: www.elsevier .com/locate/ toxicon

New a-adrenergic property for synthetic MTb and CM-3three-finger fold toxins from black mamba

Guillaume Blanchet a,b, Gregory Upert a, Gilles Mourier a, Bernard Gilquin c,Nicolas Gilles a, Denis Servent a,*aCEA, iBiTec-S, Service d’Ingénierie Moléculaire des Protéines (SIMOPRO), F-91191 Gif sur Yvette, FrancebUFR Sciences de la Vie, Université Pierre et Marie Curie (UPMC), 4 place Jussieu, Paris, FrancecCEA, iBiTec-S, Service de Bioénergétique, Biologie Structurale et Mécanismes (SBSM), F-91191 Gif sur Yvette, France

a r t i c l e i n f o

Article history:Received 30 January 2013Received in revised form 15 April 2013Accepted 19 April 2013Available online 3 May 2013

Keywords:Muscarinic toxinMuscarinic receptorAdrenergic toxinAdrenoceptorThree-finger fold toxinGPCRs

Abbreviations: GPCR, G-protein coupled receptor;acetylcholine receptors; MTs, muscarinic toxins; Namine; HCTU, 2-(6-Chloro-1H-benzotriazole-1-ylaminium hexafluorophosphate; NMM, N-methylmmethy-2-lpyrrolidone.* Corresponding author.

E-mail address: denis.servent@cea.fr (D. Servent

0041-0101/$ – see front matter � 2013 Elsevier Ltdhttp://dx.doi.org/10.1016/j.toxicon.2013.04.017

a b s t r a c t

Despite their isolation more than fifteen years ago from the venom of the African mambaDendroaspis polylepis, very few data are known on the functional activity of MTb and CM-3toxins. MTb was initially classified as a muscarinic toxin interacting non-selectively andwith low affinity with the five muscarinic receptor subtypes while no biological functionwas determined for CM-3. Recent results highlight the multifunctional activity of three-finger fold toxins for muscarinic and adrenergic receptors and reveal some discrepanciesin the pharmacological profiles of their venom-purified and synthetic forms. Here, wereport the pharmacological characterization of chemically-synthesized MTb and CM-3toxins on nine subtypes of muscarinic and adrenergic receptors and demonstrate theirhigh potency for a-adrenoceptors and in particular a sub-nanomolar affinity for the a1A-subtype. Strikingly, no or very weak affinity were found for muscarinic receptors, high-lighting that pharmacological characterizations of venom-purified peptides may be riskydue to possible contaminations. The biological profile of these two homologous toxinslooks like that one previously reported for the Dendroaspis angusticeps r-Da1a toxin.Nevertheless, MTb and CM-3 interact more potently than r-Da1a with a1B- and a1D-ARsubtypes. A computational analysis of the stability of the MTb structure suggests thatmutation S38I, could be involved in this gain in function.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Four species of the Elapidae snakes of the genus Den-droaspis have been identified so far: Dendroaspis angusti-ceps,Dendroaspis polylepis,D. jamesoni andD. viridis and thecomposition of these African mambas’ venoms is quite

mAChRs, muscarinicMS, N-methylscopol-)-1,1,3,3-tetramethyl-orpholine; NMP, N-

).

. All rights reserved.

atypical. Indeed, these venoms display very low enzymaticactivities (Tan et al.,1991) but a strong neurotoxic effect dueto the presence of several low molecular weight peptides(6–8000 Da), mainly belonging to the Kunitz and three-finger folds structural families (Schweitz and Moinier,1999). In the first structural family, dendrotoxins and cal-cicludine interact with potassium and calcium channels,while in the second structural family, a-neurotoxins, fas-ciculins, calciseptin andGPCRs toxins interactwith nicotinicreceptors, acetylcholinesterase, L-type calcium channelsand muscarinic or adrenergic GPCRs, respectively.

Muscarinic toxins (MTs) were first identified 25 yearsago (Adem et al., 1988) and until now less than ten differentMTs have been isolated from mamba venoms and

G. Blanchet et al. / Toxicon 75 (2013) 160–167 161

characterized for their interaction with the five differentmAChRs subtypes (Bradley, 2000; Karlsson et al., 2000;Servent et al., 2011; Servent and Fruchart-Gaillard, 2009).These toxins contain 65–66 residues, display more than60% in sequence identity, belong to the three-finger foldstructural family but nonetheless possess quite variouspharmacological profiles. For example, MT3 was describedas a selective antagonist of the M4 receptor subtype(Jerusalinsky et al., 1998; Olianas et al., 1996), MT1 is a non-selectiveM1–M4 ligands (Kornisiuk et al., 1995;Waelbroecket al., 1996) and MT7 is a highly selective allosteric modu-lator of the M1 receptor (Fruchart-Gaillard et al., 2006,2008; Marquer et al., 2011; Max et al., 1993; Mourier et al.,2003; Olianas et al., 2000). If MTs from the D. angusticepswere thoroughly studied and their mode of action charac-terized in detail, this is much less the case for toxins issuefrom D. polylepis and even less for D. jamesoni and D. viridis.Only two toxins from the black mamba venom (D. poly-lepis), MTa and MTb, were initially studied and character-ized as non-selective ligands with respectively nanomolar(3–45 nM) and micromolar range (120 nM–mM) affinity forthe five mAChRs subtypes (Jolkkonen et al., 2001, 1995;Karlsson et al., 2000). The specificity of MTs for mAChRswhich was initially postulated was challenged first by thecapacity of MT1 and MT2 toxins to displace the binding ofan adrenergic ligand to vas deferens or cerebral cortexmembranes (Harvey et al., 2002), secondly by the identi-fication from mamba’s venom of selective and potentadrenergic toxins with high sequence identity with MTs(Quinton et al., 2010; Rouget et al., 2010) and finally by therecent demonstration that different MTs initially classifiedas selective muscarinic ligands, interact very efficientlywith various adrenoceptors (Nareoja et al., 2011). Moresurprisingly, MTa was shown to interact very efficientlyand selectively with the a2B-adrenoceptor subtypewhereasno muscarinic activity was detected (Koivula et al., 2010;Nareoja et al., 2011). This last result highlights major dif-ferences in the pharmacological profiles establish withvenom-purified MTa (Jolkkonen et al., 1995) and its syn-thetic analog (Koivula et al., 2010), suggesting the presenceof low-abundant contaminant in the venom fraction asso-ciated with the muscarinic function.

In this context, based on high sequence identity betweenMTs and recently identified r-Da1a adrenergic toxin, wedecided to reinvestigate the pharmacological profile of MTbfor the various muscarinic and a-adrenergic receptor sub-types and characterize that one of CM-3 toxin, a blackmamba toxin sequenced a long time ago (Joubert, 1985) butnever studied pharmacologically. Here we describe thesolid-phase synthesis, mass and circular dichroism analysesof MTb and CM-3 and report that they display similar phar-macological properties as the r-Da1a toxin, a high affinity offor a1-adrenoceptors, particularly the a1A-subtype, and verylow or no affinity at all for the various mAChRs subtypes.

2. Materials and methods

2.1. Materials

[3H]N-Methylscopolamine, [3H]-Prazosin and [3H]-Rauwolscine were purchased from PerkinElmer Life

Sciences (Courtaboeuf, France). N-methylscopolamine,prazosin and yohimbine were from Sigma–Aldrich (StQuentin-Fallavier, France).

2.2. Peptide synthesis, disulfide bond formation and proteinpurification

The toxins were synthesized on a 50 mmol scale usingFmoc chemistry on a Prelude synthesizer (Protein Tech-nologies�) using HMPB ChemMatrix� resin functionalizedwith the appropriate protected amino acid (AA). Eachcoupling step was carried out twice 5 min using a mixture10 eq./10 eq./20 eq. Fmoc-amino acid/HCTU/NMM in NMPfollowed by a capping step with a solution of Ac2O/NMM1/1 v/v in NMP. Fmoc deprotection was performed using a20% piperidine solution in NMP. After concomitant cleav-age from the resin and amino acid side chains depro-tection using a TFA/H2O/TIS 90/5/5 solution, the crudepeptides were precipitated in cold Et2O, lyophilized from a10% acetic acid solution and purified by reverse phasesemi-preparative HPLC using a Vydac C18 column(250 � 10 mm) with a gradient of 40–60% of solvent B in40 min (A: 0.1% TFA in H2O, B: 60% acetonitrile and 0.1%TFA in H2O). The peptides were then subjected to foldingin a buffer Tris–HCL 100 mM pH 8 containing 1 mM ofEDTA, 1 mM of GSH and GSSG and 20% v/v of glycerol. Thefinal reticulated peptides were purified using the samemethod described above and characterized by ESI-MS(Bruker Esquire HCT 3000plus).

2.3. Circular dichroism analysis

CD spectra were recorded on a Jasco J-815 CD spec-tropolarimeter. Measurements were routinely performedat 20 �C in 0.1 cm path length quartz cells (Hellma) with apeptide concentration of 5.10�6 M in 5 mM sodium phos-phate buffer pH 7.4. Spectra were recorded in the 190–260 nm wavelength range. Each graph represents theaverage of four spectra.

2.4. [3H]-NMS and [3H]-Prazosin binding assays

Membrane preparations were prepared from transientor stable cell lines as previously described for muscarinicmembrane preparations in the paper of Marquer et al.(2011) and for alpha adrenergic membrane preparationsin Quinton et al. (2010) and Rouget et al. (2010). The af-finity constants of the radioactive ligands were deter-mined by equilibrium saturation experiments (n � 2).Competition experiments on the 5 muscarinic subtypes,the 3 a1 adrenoceptor subtypes and a2C adrenoceptorsubtype were carried out with various concentrations ofCM-3 or MTb toxins. For muscarinic receptor subtypes,binding experiments were carried out in 10 mM sodiumphosphate, pH 7.2, 135 mM NaCl, 2.5 mM KCl, pH 7.4, 0.1%bovine serum albumin. [3H]-NMS was used at 1 nM. Thealpha1 membranes were incubated in a 50 mM TrisHCl,pH 7.4, 10 mM MgCl2, 0.1% bovine serum albumin, with1.75 nM of [3H]-Prazosin as radioactive ligand. For a2Cadrenoceptor binding, the buffer was composed by 50 mMHEPES buffer, pH 7.4, 5 mM MgCl2, 1 mM CaCl2, 0.1%

Fig. 1. Sequence alignment of selected muscarinic and adrenergic toxins from D. angusticeps and D. polylepis. Loops I, II and III correspond to the three loops of thethree-finger fold and the four disulfide bridges are indicated. The four residues that differ between MTb, CM-3 and r-Da1a are indicated by the star. The % ofsequence identity as compared to MTb was indicated.

Fig. 2. Reverse phase chromatography and mass analyses profiles of synthetic CM-3. A: Chromatogram of the crude of synthesis after TFA deprotection. Thearrow corresponds to the peak of the reduced toxin. B: HPLC profiles of the glutathione-mediated oxidation of the synthetic CM-3. The arrow corresponds to thepeak of the refolded toxin. Peptide detectionwas followed at 278 nm. Insert, the ESI-MS profile of the reduced and oxidized CM-3 with 7343.7 and 7336.2 masses,respectively.

G. Blanchet et al. / Toxicon 75 (2013) 160–167162

G. Blanchet et al. / Toxicon 75 (2013) 160–167 163

bovine serum albumin and [3H]-Rauwolscine was used at2 nM as radioactive ligands.

The binding data from individual experiments (n ¼ 2)were analyzed by non-linear regression analysis usingKaleidagraph 4.0 (Synergy software, Reading, PA, USA).After subtraction of the non-specific binding andnormalization, a one-site inhibition curve was fitted toinhibition binding data, IC50 values were converted to Ki

using the Cheng–Prusoff equation (Cheng and Prusoff,1973).

2.5. MTb structural model and FoldX analysis

A model of MTb was generated with MODELLER (Saliand Blundell, 1993) using r-Da1a structure (A. Maiga, E.Stura, personal communication) as a template. A quanti-tative estimation of the effect of mutation (Ser/Ile38 andAla/Val43) on the stability of r-Da1a was obtained by usingFoldX (Guerois et al., 2002; Schymkowitz et al., 2005).

3. Results and discussion

Recently published results suggested that minor con-taminants present in purified fractions from snake’svenom may affect dramatically the pharmacological pro-file of the studied toxin. The most evident demonstrationconcerns the synthetic MTa which possesses high a2B-adrenoceptor selectivity (Koivula et al., 2010) as comparedto the venom-purified fraction that was initially charac-terized as a non-selective muscarinic toxin (Jolkkonenet al., 1995). In addition, this result highlights that fewsequence modifications (four between MTa and MT1) mayaffect drastically the receptor-selectivity profile of thesetoxins, as also shown with non-natural toxins obtained bya loop-shuffling approach (Fruchart-Gaillard et al., 2012).In this context, it appears crucial to determine the phar-macological property of chemically-synthesized toxins,initially classified as MTs and to studied uncharacterizedtoxins, even with a high sequence identity. Thus, weperformed the solid-phase chemical synthesis and phar-macological study of MTb and CM-3 toxins described inFig. 1.

Fig. 3. Far-ultraviolet CD spectra of CM-3 and MTb toxins. Measurementswere performed at 20 �C in 0.1 cm path length quartz cells with a peptideconcentration of 5 � 10�6 M in 5 mM sodium phosphate buffer pH 7.4.Spectra were recorded in the 190–260 nm wavelength range. Each spectrumrepresents the average of four spectra.

3.1. Fast parallel chemical synthesis of CM-3 and MTb toxins

CM-3 and MTb were synthesized automatically usingfast coupling and deprotection times using HCTU ascoupling reagent (Hood et al., 2008). In comparison withour previous report (Mourier et al., 2003), the total syn-thesis times of these long toxins was dramatically reducedfrom 5 to 2 days. So far, this approach was applied withsuccess and efficiency to several muscarinic and adren-ergic toxins (personal data). Fig. 2A shows the HPLC pro-file of the CM-3 toxin after the final TFA cleavage of theresin. The two major peaks, at 24 and 33 min retentiontimes correspond respectively to, a truncated and acety-lated peptide resulting to a known difficult coupling ofVal33 on Pro34 residue (Carpino and Elfaham, 1994;Takuma et al., 1982) and to the reduced form of the toxin,as indicated by the mass analysis (theoretical and

measured values for reduced CM-3 are 7344.4 and7343.7). Similar profile was observed for MTb (data notshown). HCTU is a very efficient coupling agent mainlyused to perform fast synthesis of small to medium sizepeptides. Our results demonstrate that after optimizationof the synthesis conditions, this coupling agent can alsobe used for three-finger fold peptides of more than 65residues.

After purification by reverse phase HPLC, the reducedtoxins were submitted to oxidation conditions in order tofavor the formation of disulfide bridges. Optimal condi-tions for disulfide formation required the addition of 20%of glycerol that ensure a good solubilization avoiding theformation of aggregates, and the presence of 1 mMreduced and 1 mM oxidized glutathione for improving theformation of oxidized species that are eluted with aretention time of 31 min (Fig. 2B). Mass analyses of thepurified oxidized peaks confirm the formation of the fourdisulfide bridges (theoretical and measured values foroxidized CM-3 and MTb: 7336.4; 7336.2 and 7337.4;7336.5 respectively).

3.2. CD spectra

CD spectra of the synthetic forms of CM-3 and MTbtoxins exhibit typical b-sheet signature with pronouncedmaxima and minima around 195 nm and 214 nm,respectively (Fig. 3). The CD spectrum of CM-3 toxin dif-fers slightly from that one of MTb with a minimumcentered at 210 nm in place of 214 nm for MTb. Never-theless, these spectra are similar to those previously ob-tained for various three-finger fold muscarinic toxins(Mourier et al., 2003).

G. Blanchet et al. / Toxicon 75 (2013) 160–167164

3.3. Radioligand binding experiments on muscarinic and a-adrenergic receptors

Based on the high sequence homology between CM-3,MTb and r-Da1a, up to 95%, and to the previously pro-posed muscarinic affinity of the venom-purified MTb(Jolkkonen et al., 2001, 1995), the pharmacological profilesof chemically-synthesized CM-3 and MTb were charac-terized by binding experiments on the a1-ARs, a2C-ARand the five mAChRs subtypes. These assays were per-formed at equilibrium (overnight incubation) usingmembrane preparations from cell lines expressing thedifferent muscarinic and a-adrenergic receptor subtypesand [3H]-NMS, [3H]-prazosin or [3H]-rauwolscine asradiotracers.

Fig. 4. Inhibition of radiotracer binding to muscarinic and a-adrenergic receptorsadrenergic receptors, respectively. C and D: CM-3 interaction with muscarinic andincubating membrane preparations of each receptor subtype with [3H]-NMS (M1concentrations of toxin at room temperature, overnight. The results are expressed(n ¼ 3 for each toxin).

In competition binding experiments, CM-3 and MTbdisplay similar binding profileswith a clear selectivity for a-adrenoceptors as compared tomAChRs (Fig. 4). As shownonFig. 4A and C, 10 mM of CM-3 and MTb induce no or a veryweak displacement of [3H]-NMS from the five mAChR sub-types, associated with pKi values in the high micromolarrange (Table 1). On the opposite, these two toxins werehighly potent on various a-adrenoceptors with sub-nanomolar affinity for the a1A-subtype and a similar orderof potency: a1A > a1B > a1D > a2C (Fig. 4B and D, Table 1).Thus, these two toxins, such as the r-Da1a (Quinton et al.,2010), can be classified as adrenergic toxins as their affin-ities for a-ARs were one to four order of magnitude higherthan those obtainedwithmAChRs. Concerning theMTb, ourresults highlight major differences, from one to two orders

by MTb and CM-3 toxins. A and B: MTb interaction with muscarinic andadrenergic receptors, respectively. Binding experiments were performed by-M5), [3H]-prazosin (a1A, a1B, a1D) or [3H]-rauwolscine (a2C) and varyingas the ratio of the specific binding measured with (B) or without toxin (Bo)

Table 1Affinity constants of CM-3, MTb and r-Da1a for muscarinic and alpha-adrenoceptors.

Receptor subtype Kd (nM) CM-3 (r-Dp1b) pKi MTb (r-Dp1a) pKi v-MTb pKi r-Da1a pKi

a1A 0.24 9.43 � 0.15 9.30 � 0.22 9.44a

a1B 0.12 7.98 � 0.09 7.99 � 0.05 7.27a

a1D 0.40 6.98 � 0.18 6.87 � 0.14 5.95a

a2C 0.45 6.78 � 0.05 6.77 � 0.03 6.80a

M1 0.10 z5.75 � 0.05 �5 5.94 � 0.11b <5a

M2 1.73 z4.75 � 0.06 �5 5.68 � 0.06b <5a

M3 0.86 �5 �5 6.85 � 0.03b <5a

M4 0.10 z5.35 � 0.03 z5.35 � 0.02 6.90 � 0.06b z5.5a

M5 1.10 �5 �5 6.46 � 0.01b <5a

Kd were determined with 3H-Prazosin on a1-AR, 3H-Rauwolscine on a2-AR and 3H-NMS on mAChRs.a (Quinton et al., 2010); Maïga et al. personal communication.b (Jolkkonen et al., 1995). v- MTb for venom-purified MTb.

Fig. 5. Structural model of MTb based on the r-Da1a crystallographicstructure (A. Maiga, E. Stura, personal communication) using MODELLER(Sali and Blundell, 1993). The Ile38 is colored in red and its structurally-related residues in blue. (For interpretation of the references to color inthis figure legend, the reader is referred to the web version of this article.)

G. Blanchet et al. / Toxicon 75 (2013) 160–167 165

of magnitude, in the affinity constants of the previouslydescribed venom-purified toxin and the chemically-synthesized toxin for mAChRs and especially at the M3and M4 subtypes (Table 1). Therefore, as already reportedfor MTa, another D. polylepis toxin (Koivula et al., 2010), themuscarinic activity of the venom-purified toxins could berelated to the presence of lowamountof contaminant toxinsthat would be interesting to identify. Nevertheless, wecannot exclude a sequencing error of the natural toxin.

As particularly evident on the a1A competition bindingcurve, CM-3 and MTb toxins were not able to totallydisplace [3H]-radiotracer from the targeted receptor. Thisproperty was commonly described for similar toxin–re-ceptor interactions, such as r-Da1a on a1A-AR (Quintonet al., 2010) or MT1 and MTa on a2B-AR (Koivula et al.,2010; Nareoja et al., 2011) and suggests that these in-teractions could be non-competitive or that part of thereceptor accessible to small organic and lipophilic radio-ligands was not accessible to peptide toxins.

Interestingly, comparison of the adrenergic andmuscarinic affinity constants in one hand and amino-acidssequences of CM-3, MTb and r-Da1a on the other handsuggests some structure-activity relationships. First, theidentical affinities of CM-3 and MTb for the nine receptorsstudied highlights that the two sequence differences be-tween these toxins, the conservative Asp61/Glu61 substi-tution and the Glu65/Asn65 substitution, does not affecttheir binding property, excluding a major role of thesepositions in the toxin–receptor interactions. More strik-ingly, despitemore than 95% of sequence identity, the pKi ofCM-3 and MTb were not totally identical than those pre-viously calculated for r-Da1a (Table 1), suggesting that theSer/Ile38 and Ala/Val43 substitutions (Fig. 1) could beresponsible for the five-fold and ten-fold increases in af-finity observed on a1B-AR and a1D-AR, respectively (Table1). These two residues were not located at the tip of thetoxin’s loops, regions previously identified as crucial in theinteraction of three-finger fold toxins with their moleculartargets: nAChRs (Antil-Delbeke et al., 2000; Bourne et al.,2005; Fruchart-Gaillard et al., 2002); AChE (Marchotet al., 1997) or mAChRs (Fruchart-Gaillard et al., 2008;Marquer et al., 2011; Mourier et al., 2003), but in themiddleof loop 2 (Ser/Ile38) or at the top of loop 3 (Ala/Val43)(Fig. 5). In order to evaluate the potential effect of thesemodifications on the general stability of the toxin, we haveused the computer algorithm FoldX (Guerois et al., 2002)

which provide a quantitative estimation of the importanceof point mutations to the stability of proteins and proteincomplexes. Starting from the crystallographic structure ofr-Da1a (A. Maiga, E. Stura, personal communication), thetwo substitutions (Ser/Ile38 and Ala/Val43) present in theMTb sequence were introduced and five iterative calcula-tions were performed with FoldX leading to the calculationof the variation in free energy (DDG in kCal/mol) betweenthe wild-type and modified structures. If FoldX cannotconclude on a significant effect of the A43V substitution onthe structural stability of the toxin, the S38I mutationseems induce a significant stabilizing effect associated witha DDG ¼ �1.47 � 0.07 kCal/mol. This effect seems to be dueto additional hydrophobic interactions between Ile38 andthe aliphatic side chains of Lys26, Arg28, Arg40 and Tyr36(Fig. 5) that may induce slight displacement of the sidechains surrounding Ser/Ile38. The global stabilizing effectof this S38I mutationmay explain theweak gain in functionof MTb, and by analogy of CM-3, as compared to r-Da1a ona1B-AR and a1D-AR. The selectivity of this effect beingrelated to the involvement of the stability of the top of loop2 of MTb with specific residues of these receptor subtypes.

G. Blanchet et al. / Toxicon 75 (2013) 160–167166

4. Conclusion

In this study, we have performed the pharmacologicalcharacterization of chemically-synthesized MTb and CM-3 toxins originally purified from the black mambavenom. We demonstrated by equilibrium binding ex-periments on cell lines expressing muscarinic andadrenergic receptors that these toxins interact very effi-ciently on a-adrenoceptors, particularly the a1A-subtypewhile no or high micromolar range affinity weremeasured for the five different mAChRs subtypes. Thus,according to the new nomenclature applied to animaltoxins (King et al., 2008), MTb and CM-3 should berenamed r-Dp1a and r-Dp1b, respectively. These phar-macological profiles look like the previously character-ized r-Da1a toxin (Quinton et al., 2010) and suggests apossible correlation between punctual sequence varia-tion, a local structural stabilization and weak differencesin the toxins selectivity profiles. Finally, this study con-firms the need to use synthetic toxins in order to avoidany misinterpretation of the pharmacological property ofvenom-purified toxin that could be associated with low-abundant contaminants.

Ethical statement

International ethical guidelines for scientific paperswere followed in the preparation of this manuscript.

Conflict of interest statement

The author declares that there are no conflicts of inter-est regarding the preparation of this manuscript.

References

Adem, A., Asblom, A., Johansson, G., Mbugua, P.M., Karlsson, E., 1988.Toxins from the venom of the green mamba Dendroaspis angus-ticeps that inhibit the binding of quinuclidinyl benzilate tomuscarinic acetylcholine receptors. Biochim. Biophys. Acta 968,340–345.

Antil-Delbeke, S., Gaillard, C., Tamiya, T., Corringer, P.J., Changeux, J.P.,Servent, D., Ménez, A., 2000. Molecular determinants by which along chain toxin from snake venom interacts with the neuronal{alpha}7 nicotinic acetylcholine receptor. J. Biol. Chem. 275, 29594–29601.

Bourne, Y., Talley, T.T., Hansen, S.B., Taylor, P., Marchot, P., 2005. Crystalstructure of a Cbtx-AChBP complex reveals essential interactionsbetween snake alpha-neurotoxins and nicotinic receptors. Embo J. 24,1512–1522.

Bradley, K.N., 2000. Muscarinic toxins from the green mamba. Pharmacol.Ther. 85, 87–109.

Carpino, L.A., Elfaham, A., 1994. Effect of tertiary bases on O-benzo-triazolyluronium salt-induced peptide segment coupling. J. Org.Chem. 59, 695–698.

Cheng, Y.-C., Prusoff, W.H., 1973. Relationship between the inhibitionconstant (Ki) and the concentration of inhibitor which causes 50 percent inhibition (I50) of an enzymatic reaction. Biochem. Pharmacol.22, 3099–3108.

Fruchart-Gaillard, C., Gilquin, B., Antil-Delbeke, S., Le Novère, N.,Tamiya, T., Corringer, P.J., Changeux, J.P., Ménez, A., Servent, D., 2002.Experimentally-based model of a complex between a snake toxin andthe a7 nicotinic acetylcholine receptor. Proc. Natl. Acad. Sci. U.S.A. 99,3216–3221.

Fruchart-Gaillard, C., Mourier, G., Blanchet, G., Vera, L., Gilles, N.,Menez, R., Marcon, E., Stura, E.A., Servent, D., 2012. Engineering ofthree-finger fold toxins creates ligands with original

pharmacological profiles for muscarinic and adrenergic receptors.PLoS One 7, e39166.

Fruchart-Gaillard, C., Mourier, G., Marquer, C., Menez, A., Servent, D.,2006. Identification of various allosteric interaction sites on M1muscarinic receptor using 125I-Met35-oxidized muscarinic toxin 7.Mol. Pharmacol. 69, 1641–1651.

Fruchart-Gaillard, C., Mourier, G., Marquer, C., Stura, E., Birdsall, N.J.,Servent, D., 2008. Different interactions between MT7 toxin and thehuman muscarinic M1 receptor in its free and NMS-occupied states.Mol. Pharmacol. 74, 1554–1563.

Guerois, R., Nielsen, J.E., Serrano, L., 2002. Predicting changes in the sta-bility of proteins and protein complexes: a study of more than 1000mutations. J. Mol. Biol. 320, 369–387.

Harvey, A.L., Kornisiuk, E., Bradley, K.N., Cervenansky, C., Duran, R.,Adrover, M., Sanchez, G., Jerusalinsky, D., 2002. Effects of muscarinictoxins MT1 and MT2 from green mamba on different muscariniccholinoceptors. Neurochem. Res. 27, 1543–1554.

Hood, C.A., Fuentes, G., Patel, H., Page, K., Menakuru, M., Park, J.H., 2008.Fast conventional Fmoc solid-phase peptide synthesis with HCTU. J.Pept. Sci. 14, 97–101.

Jerusalinsky, D., Kornisiuk, E., Alfaro, P., Quillfeldt, J., Alonso, M.,Verde, E.R., Cervenansky, C., Harvey, A., 1998. Muscarinic toxin se-lective for m4 receptors impairs memory in the rat. Neuroreport 9,1407–1411.

Jolkkonen, M., Oras, A., Toomela, T., Karlsson, E., Jarv, J., Akerman, K.E.,2001. Kinetic evidence for different mechanisms of interaction ofblack mamba toxins MT alpha and MT beta with muscarinic re-ceptors. Toxicon 39, 377–382.

Jolkkonen, M., Van Giersbergen, P.L., Hellman, U., Wernstedt, C.,Oras, A., Satyapan, N., Adem, A., Karlsson, E., 1995. Muscarinictoxins from the black mamba Dendroaspis polylepis. Eur. J. Bio-chem. 234, 579–585.

Joubert, F.J., 1985. The amino acid sequence of protein CM-3 from Den-droaspis polylepis polylepis (black mamba) venom. Int. J. Biochem. 17,695–699.

Karlsson, E., Jolkonen, M., Mulugeta, E., Onali, P., Adem, A., 2000. Snaketoxins with high selectivity for subtypes of muscarinic acetylcholinereceptors. Biochimie 82, 793–806.

King, G.F., Gentz, M.C., Escoubas, P., Nicholson, G.M., 2008. A rationalnomenclature for naming peptide toxins from spiders and othervenomous animals. Toxicon 52, 264–276.

Koivula, K., Rondinelli, S., Nasman, J., 2010. The three-finger toxin MTal-pha is a selective alpha(2B)-adrenoceptor antagonist. Toxicon 56,440–447.

Kornisiuk, E., Jerusalinsky, D., Cervenansky, C., Harvey, A.L., 1995. Bindingof muscarinic toxins MTx1 and MTx2 from the venom of the greenmamba Dendroaspis angusticeps to cloned human muscarinic chol-inoceptors. Toxicon 33, 11–18.

Marchot, P., Prowse, C.N., Kanter, J., Camp, S., Ackermann, E.J., Radic, Z.,Bougis, P.E., Taylor, P., 1997. Expression and activity of mutants offasciculin, a peptidic acetylcholinesterase inhibitor from mambavenom. J. Biol. Chem. 272, 3502–3510.

Marquer, C., Fruchart-Gaillard, C., Letellier, G., Marcon, E., Mourier, G.,Zinn-Justin, S., Menez, A., Servent, D., Gilquin, B., 2011. A structuralmodel of a ligand/GPCR complex based on experimental double-mutant cycle data: the MT7 snake toxin bound to a dimeric hM1muscarinic receptor. J. Biol. Chem. 286, 31661–31675.

Max, S.I., Liang, J.S., Potter, L.T., 1993. Stable allosteric binding of m1-toxinto m1 muscarinic receptors. Mol. Pharmacol. 44, 1171–1175.

Mourier, G., Dutertre, S., Fruchart-Gaillard, C., Ménez, A., Servent, D.,2003. Chemical synthesis of MT1 and MT7 muscarinic toxins: criticalrole of Arg-34 in their interaction with M1 muscarinic receptor. Mol.Pharmacol. 63, 26–35.

Nareoja, K., Kukkonen, J.P., Rondinelli, S., Toivola, D.M., Meriluoto, J.,Nasman, J., 2011. Adrenoceptor activity of muscarinic toxins identifiedfrom mamba venoms. Br. J. Pharmacol. 164, 538–550.

Olianas, M.C., Adem, A., Karlsson, E., Onali, P., 1996. Rat striatal muscarinicreceptors coupled to the inhibition of adenylyl cyclase activity: potentblock by the selective m4 ligand muscarinic toxin 3 (MT3). Br. J.Pharmacol. 118, 283–288.

Olianas, M.C., Maullu, C., Adem, A., Mulugeta, E., Karlsson, E., Onali, P.,2000. Inhibition of acetylcholine muscarinic M(1) receptor functionby the M(1)-selective ligand muscarinic toxin 7 (MT-7). Br. J. Phar-macol. 131, 447–452.

Quinton, L., Girard, E., Maiga, A., Rekik, M., Lluel, P., Masuyer, G.,Larregola, M., Marquer, C., Ciolek, J., Magnin, T., Wagner, R., Molgo, J.,Thai, R., Fruchart-Gaillard, C., Mourier, G., Chamot-Rooke, J.,Menez, A., Palea, S., Servent, D., Gilles, N., 2010. Isolation and phar-macological characterization of AdTx1, a natural peptide displaying

G. Blanchet et al. / Toxicon 75 (2013) 160–167 167

specific insurmountable antagonism of the alpha(1A)-adrenoceptor.Br. J. Pharmacol. 159, 316–325.

Rouget, C., Quinton, L., Maiga, A., Gales, C., Masuyer, G., Malosse, C.,Chamot-Rooke, J., Thai, R., Mourier, G., dePauw, E., Gilles, N.,Servent, D., 2010. Specific inhibition of the alpha2-adrenoceptors by anatural antagonist peptide toxin. Br. J. Pharmacol. 161, 1361–1374.

Sali, A., Blundell, T.L., 1993. Comparative protein modelling by satisfactionof spatial restraints. J. Mol. Biol. 234, 779–815.

Schweitz, H., Moinier, D., 1999. Mamba toxins. Perspect. Drug Discov. Des.16 (16), 83–110.

Schymkowitz, J., Borg, J., Stricher, F., Nys, R., Rousseau, F., Serrano, L., 2005.The FoldX web server: an online force field. Nucleic Acids Res. 33,W382–W388.

Servent, D., Blanchet, G., Mourier, G., Marquer, C., Marcon, E., Fruchart-Gaillard, C., 2011. Muscarinic toxins. Toxicon 58, 455–463.

Servent, D., Fruchart-Gaillard, C., 2009. Muscarinic toxins: tools for thestudy of the pharmacological and functional properties of muscarinicreceptors. J. Neurochem. 109, 1193–1202.

Takuma, S., Hamada, Y., Shioiri, T., 1982. New methods and reagents inorganic-synthesis .27. An extensive survey by the use of high-performance liquid-chromatography on racemization during thecoupling of benzyloxycarbonyl-L-phenylalanyl-L-valine with L-prolinetert-butyl ester. Chem. Pharm. Bull. 30, 3147–3153.

Tan, N.H., Arunmozhiarasi, A., Ponnudurai, G., 1991. A comparative studyof the biological properties of Dendroaspis (mamba) snake venoms.Comp. Biochem. Physiol. C 99, 463–466.

Waelbroeck, M., De Neef, P., Domenach, V., Vandermeers-Piret, M.C.,Vandermeers, A., 1996. Binding of the labelled muscarinic toxin 125I-MT1 to rat brain muscarinic M1 receptors. Eur. J. Pharmacol. 305,187–192.