Conopressin-T from Conus tulipa Reveals an Antagonist Switch in Vasopressin-like Peptides

We report the discovery of conopressin-T, a novel bioactive peptide isolated from Conus tulipa venom. Conopressin-T belongs to the vasopressin-like peptide family and displays high sequence homology to the mammalian hormone oxytocin (OT) and to vasotocin, the endogenous vasopressin analogue found in teleost fish, the cone snail’s prey. Conopressin-T was found to act as a selective antagonist at the human V 1a receptor. All peptides in this family contain two conserved amino acids within the exocyclic tripeptide (Pro 7 and Gly 9 ), which are replaced with Leu 7 and Val 9 in conopressin-T. Whereas conopressin-T binds only to OT and V 1a receptors, an L7P analogue had increased affinity for the V 1a receptor and weak V 2 receptor binding. Sur-prisingly, replacing Gly 9 with Val 9 in OT and vasopressin revealed that this position can function as an agonist/antagonist

The vasopressin (AVP) 3 and oxytocin (OT) peptides were originally discovered and identified as neurohypophysial hormones in mammals (1). In humans, AVP acts via three vasopressin receptors (vascular V 1a R, pituitary V 1b R, and renal V 2 R), whereas OT acts via one OT receptor (OTR). All targets are members of the G protein-coupled receptor family (2). Peripherally, they regulate water balance, the control of blood pressure, and contraction of uterine smooth muscle and mammary myoepithelium (3). Centrally, these peptides affect levels of aggression, depression, and young parent bonding (4 -6). Endogenous analogues of OT and AVP have been reported in nonmammalian vertebrates, annelids, molluscs, and insects, suggesting an old lineage for these peptides (7). Surprisingly, two variants were also found in the venom of predatory cone snails. The original discovery of these two AVP analogues, named conopressins, was based on the characteristic "scratching" effect observed upon intracerebral injection into mice (8). Although the sequences of conopressins are similar to vasopressin itself, they have an additional positive charge in position 4, which is only found in two other endogenous vasopressin analogues, cephalotocin (Octopus vulgaris) and annetocin (Eisenia foetida). Conopressin-S was isolated from Conus striatus, whereas conopressin-G was first isolated from Conus geographus venom but later found in Conus imperialis venom as well as in tissue extracts of the nonvenomous snails Lymnea stagnalis and Aplysia californica and the leech Erpobdella octoculata (9 -11).
Molluscs of the genus Conus produce bioactive peptides in a combinatorial fashion. As demonstrated for the snake toxins (12), most conotoxins or conopeptides are believed to be derived from an endogenous structural template (13). Because conopressin-G is widely distributed, it may represent the endogenous hormone in Gastropods and Annelids. However, a role in prey capture has also been proposed (14).
In this study, we report the discovery of conopressin-T isolated from Conus tulipa venom. Pharmacological characterizations of Con-T across human receptors revealed that it is a selective V 1a antagonist, with partial agonist activity at the OT receptor and no detectable activity at V 1b and V 2 receptors. The exocyclic tripeptide segment of conopressin-T shows unusual sequence divergence. L7P-Con-T had increased affinity at the V 1a receptor but minimal effect on the selectivity profile across all human receptors compared with Con-T. Interestingly, replacing Gly 9 with Val 9 in OT and AVP converted these peptides from full agonist to full antagonist at the V 1a receptor, demonstrating the role of position 9 as an antagonist switch in these peptides. Finally, the NMR structures of conopressin-T and its L7P analogue provide new templates for the design of novel pharmacological agents with enhanced activity at the V 1a receptor.

EXPERIMENTAL PROCEDURES
Materials-t-Butoxycarbonyl-protected amino acids and reagents used during chain assembly and HPLC purification (dimethylformamide, dichloromethane, acetonitrile, and trifluoroacetic acid) were peptide synthesis grade purchased from Auspep (Melbourne, Australia) and Novabiochem (San Diego, CA). 4-methylbenzhydrylamine-NH 2  Isolation and Sequencing-Specimens of Conus tulipa were collected from the Great Barrier Reef, Australia. Venom ducts were dissected, and crude venom prepared as previously described (15). Initial fractionation of the venom was carried out by RP-HPLC, and fractions were stored at 4°C until further use.
Conopressin-T was purified from fraction 30 using a 1% linear gradient of 0 -60% solvent B in A over 60 min (A ϭ 0.05% trifluoroacetic acid (aqueous); B ϭ 0.045% trifluoroacetic acid, 90% acetonitrile) on an analytical C 18 Phenomenex column. The flow rate was 1 ml/min, and the absorbance was monitored at 214 nm. N-terminal sequencing was carried out on an Applied Biosystems Procise HT Protein Sequencer (BRF, Newcastle, Australia) using ϳ20 pmol of purified peptide.
Peptide Synthesis-Conopressin-T and analogues were synthesized using the method described in Ref. 13, with minor modifications. Briefly, conopressin-T and L7P-conopressin-T were synthesized manually using t-butoxycarbonyl chemistry with in situ neutralization protocols (16) on a 0.5-mmol scale for residues 9 and 8. The peptide was then nitrogen-dried and split into two syntheses of 0.25 mmol for the remaining 7 residues of conopressin-T and L7P-conopressin-T. After HF cleavage, the crude peptides were purified by semipreparative RP-HPLC using a linear gradient of 0 -60% B at 3 ml/min while monitoring UV absorbance at 214 nm. Air oxidation was carried out by dissolving 10 mg of the crude peptides in 45 ml of 0.1 M NH 4 HCO 3 (pH 8.25) with vigorous stirring at room temperature for 1 h. Whereas L7P-conopressin-T dissolved readily at concentrations up to 0.5 mg/ml, conopressin-T was found to aggregate instantly. To overcome this, 1 mg of reduced conopressin-T was first dissolved in a 2:1 ratio of solution A and B and then slowly added to the oxidation buffer. Prior to purification, the solution was acidified to pH 3 with neat trifluoroacetic acid and analyzed by analytical C 18 HPLC and electrospray-MS. Oxidized conopressin-T and L7P-conopressin-T were then purified by semipreparative RP-HPLC using the same chromatographic conditions as above. Pure fractions, as determined by analytical RP-HPLC and mass spectrometry, were pooled and lyophilized. Stock solutions of conopressin-T and L7P-conopressin-T were made in MilliQ water and quantified by analytical HPLC using a commercial AVP standard.
Cloning of Human OT Receptor-A 1.3-kb fragment of OT cDNA, containing the full coding region of the human OT receptor, was isolated from human placental cDNA constructed from total placental RNA using SuperScript TM III reverse transcriptase. The following forward (5Ј-GGTAGAG-GATTCCCGCTCATTTG-3Ј) and reverse (5Ј-GGGGAGG-GATACAAACTGATAGG-3Ј) primers were used to isolate the DNA fragment. This fragment was subcloned into the mammalian expression vector pcDNA3.1/V5-his-TOPO. Following sequence analysis, 8 nucleotide changes were found over the published OT sequence (NM_000916) encoding 7 amino acid changes with one change being conserved. The nucleotide changes were T47C, A281G, A256C, G532C, T646C, and A1121G, which correspond to amino acid changes of V16A, H73R, M86L, G178R, F216L, and N374S. These changes were mutated to the published sequence using a QuikChange TM XL site-directed mutagenesis kit.
OT Receptor Expression and Membrane Preparation-DNA was prepared and used for transfection in CHO-K1 cells. CHO-K1 cells were propagated in 150-mm plates with F-12 medium containing 10% Serum Supreme at 37°C in a humidified atmosphere with 5% CO 2 . Transient transfection was performed with an appropriate plasmid using the TransIT-CHO transfection reagent method. Briefly, the method involved combining TransIT-CHO reagent (72 l) in serum-free F-12 medium incubated for 5 min at room temperature. Following the addition of receptor plasmid (24 g), this mixture was incubated for 10 min at room temperature. CHO Mojo reagent (16 l) was added to the mixture and incubated for 15 min at room temperature prior to the addition to the cells.
The transfected cells were harvested within 24 h by the following method. Confluent 150-mm plates were washed with phosphate-buffered saline (10 mM phosphate buffer, pH 7.4, 2.7 mM KCl, 137 mM NaCl), harvested by scraping into ice-cold harvest buffer (50 mM Tris-HCl, pH 7.4, 5 mM MgCl 2 ), followed by homogenization with a Polytron homogenizer and centrifugation at 100 ϫ g for 10 min at 4°C. Supernatants were recovered and centrifuged at 22,000 ϫ g for 1 h at 4°C. Membrane pellets were resuspended in 0.5 ml of ice-cold assay buffer A (50 mM Tris-HCl, pH 7.4, 5 mM MgCl 2 , 0.1% bovine serum albumin, containing 10% glycerol and aliquots stored at Ϫ80°C until use.
Receptor Binding Studies-Receptor binding assays were performed using FlashBlue TM GPCR scintillation beads. Flash-Blue TM GPCR beads are 3-m polystyrene scintillating beads with wheat germ agglutinin covalently attached on the surface. These beads allow development of a homogeneous GPCR radioligand binding assay using cellular membranes. Briefly, FlashBlue TM GPCR SPA binding assays were performed in 96-well white polystyrene plates with clear flat bottoms ) was added to each membrane preparation, followed by the addition of various concentrations of competing compounds (1 pM to 10 M) in a total volume of 80 l containing assay buffer A or B (50 mM Tris-HCl, pH 7.4, 0.1% bovine serum albumin, and either 5 mM MgCl 2 (A) or 10 mM MgCl 2 (B)) (OT and V 2 assay, buffer A; V 1a and V 1b assay, buffer B). The final reaction volume per well comprised 20 l of compound/buffer, 20 l of FlashBlue TM GPCR beads (OTR, V 1a , and V 2 , 100 g; V 1b , 200 g), and 20 l of membrane, and the assay was initiated by the addition of 20 l of radioligand. The plate was then sealed with TopSeal-A sealing film and incubated with shaking for 1 h at room temperature. Radioligand binding was then assessed for 30 s/well on a 1450 Microbeta scintillation counter (Wallac).
All binding data were analyzed using GraphPad PRISM (GraphPAD Software, Inc., San Diego, CA). Each data point was performed in triplicate and derived from at least three separate experiments. The inhibitory dissociation constant (K i ) was calculated using the following formula: is the concentration of radioligand present, and K d is the dissociation constant of radioligand (17).
Inositol Phosphate Assays-Inositol phosphate accumulation was determined as previously described (18). Briefly, CHO cells stably transfected with AVP/OT receptors were plated at 100,000 cells/well. Cells were grown for 24 h in their respective culture medium (see above) and further incubated for another 24-h period in a serum-and inositol-free medium supplemented with 1 Ci⅐ml Ϫ1 myo-[2-3 H]inositol. Cells were then washed twice with Hanks' buffered saline medium, incubated for 15 min in this medium supplemented with 20 mM LiCl, and further stimulated for 15 min with increasing concentrations of analogues to be tested. The reaction was stopped by adding perchloric acid (5%, v/v). Total inositol phosphates (IPs) accumulated were extracted and purified on a Dowex AGI-X8 anion exchange chromatography column and counted.
NMR Spectroscopy-Samples for 1 H NMR measurements contained ϳ1 mM peptide in 95% H 2 O, 5% D 2 O (v/v) at pH ϳ3. Spectra were recorded at 290 K on a Bruker AVANCE-600 spectrometer. Two-dimensional NMR spectra were recorded in phase-sensitive mode using time-proportional phase incrementation for quadrature detection in the t1 dimension (19,20). The two-dimensional experiments consisted of a TOCSY using a MLEV-17 spin lock sequence with a mixing time of 80 ms, DQF-COSY, and NOESY with mixing times of 200 -300 ms. Solvent suppression was achieved using a modified WATERGATE sequence. Spectra were acquired over 6024 Hz with 4096 complex data points in F2 and 512 increments in the F1 dimension. 3 J HN-H␣ coupling constants were measured from a one-dimensional spectrum or from the DQF-COSY spectrum. Spectra were processed on a Silicon Graphics Indigo work station using XWINNMR (Bruker) software. The t1 dimension was zero-filled to 1024 real data points, and 90°p hase-shifted sine bell window functions were applied prior to Fourier transformation.
Structure Calculations-Preliminary structures were calculated using a torsion angle simulated annealing protocol within the program DYANA (21). Final structures were calculated using simulated annealing and energy minimization protocols within CNS version 1.1 (22). The starting structures were generated using random (,) dihedral angles and energy-minimized to produce structures with the correct local geometry. A set of 50 structures was generated by a torsion angle simulated annealing protocol (19,20). This protocol involved a high temperature phase comprising 4000 steps of 0.015 ps of torsion angle dynamics, a cooling phase with 4000 steps of 0.015 ps of torsion angle dynamics during which the temperature was lowered to 0 K, and finally an energy minimization phase comprising 500 steps of Powell minimization. Structures consistent with restraints were subjected to further molecular dynamics and energy minimization in a water shell, as described by Linge and Nilges (23). The refinement in explicit water involved the following steps. First, heating to 500 K via steps of 100 K, each comprising 50 steps of 0.005 ps of Cartesian dynamics. Second, 2500 steps of 0.005 ps of Cartesian dynamics at 500 K before a cooling phase where the temperature was lowered in steps of 100 K, each comprising 2500 steps of 0.005 ps of Cartesian dynamics. Finally, the structures were minimized with 2000 steps of Powell minimization. Structures were analyzed using PROMOTIF (24) and PROCHECK-NMR (25).

RESULTS
Isolation and Sequencing of Conopressin-T-A peptide with a monoisotopic mass of 1107.54 Da was purified from C. tulipa venom (Fig. 1, A-C). The Tris(2-carboxyethyl)phosphine-reduced peptide displayed a mass of 1109.5 Da, suggesting the presence of one disulfide bond (data not shown). N-terminal sequencing produced the sequence CYIQNCLRV with a calculated mass of 1110.53 Da, consistent with one disulfide bond (Ϫ2 Da) and C-terminal amidation (Ϫ1 Da). A BLAST search revealed that this sequence belonged to the vasopressin peptide family characterized by a disulfide-containing ring (residues 1-6) and a short exocyclic C-terminal tripeptide (residues 7-9). Conopressin-T had 7 of 9 residues identical to vasotocin (AVT), including the first six residues also found in oxytocin and mesotocin, and Arg 8 known to be essential for the pressor activity in vasopressin. However, two highly conserved residues, Pro 7 and Gly 9 , found in other vasopressin-like peptides as well as conopressin-G and conopressin-S are modified to Leu 7 and Val 9 in conopressin-T. De novo mass spectrometry sequencing (data not shown) unambiguously confirmed the sequence of Con-T as CYIQNCLRV-NH 2 .
Synthesis of Conopressin-T and L7P-Conopressin-T-Con-T and its analogues were synthesized on 4-methylbenzhydrylamine resin using solid-phase peptide synthesis t-butoxycarbonyl chemistry to further validate the primary sequence and allow characterization of its biological activity. Lyophilized crude peptides from HF cleavage were dissolved in 0.5% trifluoroacetic acid and analyzed on analytical C 18 Phenomenex RP-HPLC column. The chromatograms and electrospray-MS revealed that the syntheses contained Ͼ90% of the expected products; consequently, the oxidation was carried out directly on the crude peptides following the conditions described under "Experimental Procedures." During synthesis, we noticed that the replacement of Leu 7 by Pro 7 had a marked effect on solubility. L7P-Con-T dissolved readily in oxidation buffer to 0.5 mg/ml, whereas Con-T aggregated instantly in this solvent. This difference suggested that a Leu or a Pro at position 7 significantly modified the physical properties of the peptides. Oxidized synthetic and native Con-T were indistinguishable when co-injected in a C 18 Phenomenex analytical RP-HPLC column (Fig. 1D).

Conopressin-T and L7P-Conopressin-T Are Selective V 1a
Receptor Antagonists-Con-T and L7P-Con-T were tested for their ability to displace radioligand from human AVP and OT receptors expressed in HEK and CHO cells ( Fig. 2A and Table 1). Con-T showed the highest affinity for human OTR (K i ϭ 108 nM) and V 1a R (K i ϭ 319 nM) and no detectable activity at V 1b and V 2 receptors at up to 10 M peptide. In contrast, L7P-Con-T was 8-fold more potent than Con-T at the V 1a R (K i ϭ 37 nM) but had similar affinity for OTR (K i ϭ 132 nM). L7P-Con-T was also found to have a weak effect at V 2 R (K i ϭ 1.8 M), whereas no displacement of radioligand was detected at the V 1b R at up to 10 M peptide. For comparison, we also tested the activity of conopressin-S (Con-S) across these receptors. Con-S does not bind to V 2 R (up to 10 M peptide), has similar affinity for OTR (K i ϭ 175 nM) but is less potent at V 1a (K i ϭ 827 nM). In contrast to Con-T and L7P-Con-T, Con-S binds with high affinity to V 1b R (K i ϭ 8.3 nM), although the displacement of specific [ 3 H]AVP binding was incomplete (a 30 -40% resistant component remained; see Fig. 2A). Therefore, conopressin-T, -S, and -L7P have distinct pharmacological profiles on human receptors, with the lack of effect of Con-T on radioligand binding to both V 1b and V 2 receptors being unique (supplemental Fig. 1).
The functional properties of Con-T and L7P-Con-T were investigated on CHO cells expressing human receptors ( Fig. 2B and Table 2). Con-T did not stimulate phospholipase C activity in cells expressing V 1a R at Յ10 M peptide. However, Con-T at 10 M showed partial agonist activity at the V 1b R and OTR, producing 9 and 22% of AVP and OT maximal activity, respectively. L7P-Con-T displayed a similar profile, with no agonist activity at the V 1a R at Յ10 M and partial agonist activity at the V 1b R and OTR (4 and 28% of AVP and OT maximal activity, respectively) at 10 M peptide. In contrast, both Con-T and L7P-Con-T induced a potent and concentration-dependent inhibition of AVP-stimulated IP production in CHO cells expressing V 1a R, with K inact values of 329 and 90 nM, respectively (Table 2). These values correlated well with the binding affinities. Thus, both peptides are full antagonists at the human V 1a R subtype and weak partial agonists for the V 1b R and OTR (Fig. 2B). Due to the absence of significant binding to V 2 R, the functional activity of Con-T and L7P-Con-T was not investigated on this subtype.
Gly 9 3 Val Replacement in OT and AVP Acts as an Antagonist Switch at the V 1a Receptor-L7P-conopressin-T and vasotocin differs only at position 9, yet L7P-conopressin-T acts as a full antagonist at the V 1a receptor (Fig. 2B), whereas the closely related vasotocin is known to function as an agonist (29). To investigate if the Val 9 modification in conopressin-T could switch agonist to antagonist activity in related peptides, we tested this modification in OT and AVP. First, the binding properties of these analogues were assessed on human receptors ( Fig. 3A and Table 1). G9V-AVP was equipotent at V 1a R and V 2 R (ϳ25 nM) and 5-7-fold less potent at the OTR and V 1b R (114 and 170 nM, respectively). Compared with AVP, G9V-AVP affinity ranged from 5-fold less at V 2 R to 2000-fold less at V 1b R. In contrast, G9V-OT was equipotent at the V 1a and OT receptors (ϳ465 nM), whereas it was unable to fully displace radioligands at the V 1b and V 2 subtypes at concentrations up to 10 M. Compared with OT, G9V-OT affinity ranged from 13-fold less at V 1a R to 309-fold less at OTR.
The functional properties of G9V-AVP and G9V-OT were investigated on CHO cells expressing human receptors (Fig. 3B  and Table 2). Both peptides failed to stimulate phospholipase C activity in cells expressing V 1a R at Յ10 M peptide. However, G9V-AVP and G9V-OT both acted as partial agonists at the OTR, eliciting ϳ5-10% and ϳ25-30% of OT maximal activity, respectively. In contrast, both peptides induced a potent and concentration-dependent inhibition of AVP-stimulated IP production in CHO cells expressing V 1a R. The affinities (K inact ) of G9V-AVP (40 nM) and G9V-OT (391 nM) correlate well with the binding data at the V 1a receptor (25 and 466 nM, respectively). Clearly, the G9V replacement shifted the activity of both AVP and OT from full agonist to full antagonist at the V 1a R, supporting the hypothesis that position 9 can act as a functional switch at the V 1a receptor.
Solution Structure of Con-T and L7P-Con-T-NMR spectral assignments of both peptides were made using established techniques (26). The chemical shifts in the amide region are well dispersed, and restraints derived from the NOESY spectra allowed determination of well defined structures. Analysis of the nuclear Overhauser effects indicated that the proline residue in L7P-Con-T is predominantly in a trans conformation based on the strong nuclear Overhauser effects between the H␣ of Cys 6 and the H␦s of Pro 7 . In addition, there is evidence of a minor conformation in the L7P-Con-T spectra, including a second proline spin system. However, this minor conformation could not be assigned because of its low intensity and missing peaks. It is not uncommon for small peptides to have prolines adopting mixed cis/ trans conformations, which probably explains the second spin system observed.
The three-dimensional structures of conopressin-T and its L7P analogue were calculated using a simulated annealing protocol in CNS. The resulting families of structures had good structural and energetic statistics, as shown in Table 3. An ensemble of the 20 lowest energy structures for each peptide is shown in Fig. 4A. Analysis of the structures of each peptide with PROMOTIF (24) indicated that the major structural feature is a ␤-turn between residues 4 and 7. The structures of conopressin-T and L7P overlaid well over the defined regions of the structure (root mean square deviation for the backbone atoms of residues 1-6 is 0.52 Å). Although the C-terminal residues are disordered, the directions of the C-terminal tails clearly differ between the two structures as shown in Fig. 4, B and C.

DISCUSSION
Con-T isolated from C. tulipa venom is a novel vasopressinlike peptide. Investigation of its biological activity on human receptors revealed that Con-T acted selectively at V 1a and OT receptors. Functional studies demonstrated that Con-T was a full antagonist at V 1a R and a partial agonist at OTR. The pharmacological profile of an L7P analogue was similar but for increased affinity for the V 1a R. The partial agonist activity of both peptides at OTR indicates incomplete signaling through  the associated G protein (G q/11 , G i/0 ) perhaps because they bind and trap the OT receptor in a conformation that is unable to signal fully. Similarly, OT and AVP act as partial agonists on vasopressin and oxytocin receptors, respectively (27). The sequence of Con-T differs significantly from the previously reported conopressin-G and Con-S. First, the additional positively charged residue at position 4, which was thought to be characteristic of conopressins, is absent in Con-T. Interestingly, and probably ecologically and evolutionary relevant, Con-T has 7 of 9 residues (including the disulfide containing ring) identical to vasotocin, the teleost fish equivalent. In light of our pharmacological results, and since C. tulipa preys on fish, Con-T might specifically target teleost AVT receptors. So far, the only cloned receptor in teleost fish appears to be pharmacologically comparable with the mammalian V 1a and oxytocin receptors (28). This receptor controls the arterialvenous flows and is particularly important in gill hemodynamics and consequently respiration. However, conopressin-G and Con-S were found to be only moderately active on fish vasotocin receptor (29). Animal toxins usually shut down vital functions by blocking membrane receptors rather than by activating them, and although Con-T is yet to be tested on this fish receptor, it may act as an antagonist as demonstrated here for the human V 1a receptor. Alternatively, Con-T may also act synergistically with other lethal peptides present in the venom and/or modify the behavior of the fish to reduce the risk of injury to the cone snail. Interestingly, specific localization of staining for vasotocin in the brain of zebrafish was found to correlate with aggressive behavior (dominant versus subordinate social status) (30), and an antagonist might be expected to reduce such behavior. Other conopeptides are known to act centrally in fish, which have a leaky blood-brain barrier (31).
Recently, an unusual ␥-carboxyglutamate at position 8 in the sequence of conopressin-Vil from the venom of Conus villepinii  has been reported, but its biological activity was not determined (32). The discovery of multiple conopressin sequences supports the theory that the scaffold of the original vasopressin-like endogenous peptide in Conus has evolved for more specialized use in the venom (14). Finally, the exocyclic tripeptide of Con-T has a Leu in position 7 and a Val in position 9. To our knowledge, Con-T is the first naturally occurring vasopressin-like peptide with substitutions at these positions (Table 4). Although substitutions within the tocin ring and at position 8 have been largely investigated over the past 50 years using natural and nonnatural amino acids, there has been limited study on positions 7 and 9, probably because of their absolute conservation in this peptide family. AVT acts as an agonist on AVP receptors (33), and since L7P-Con-T differs from AVT only at position 9 and acts as a V 1a antagonist, it appears that modification at this position alone can switch peptide activity from agonist to antagonist. In support of this hypothesis, the modification G9V in OT and AVP indeed switched their activity to a full antagonist at the V 1a receptor. Thus, all endogenous peptides in the AVP-like family have a Gly 9 (see Table 4) and act as agonists on their respective receptors, whereas  Con-T, L7P-Con-T, G9V-OT, G9V-AVP, and a synthetic OT analogue with a Tyr 9 (34) act as antagonists. In contrast, all Val 9 -containing peptides investigated here acted as partial agonists at the OT receptor.

TABLE 2 Antagonist and agonist actions of Con-T, L7P-Con-T, G9V-AVP, and G9V-OT on IP signaling through human vasopressin and oxytocin receptors
The three-dimensional structures of conopressin-T and its L7P analogue were determined using NMR spectroscopy. Although the peptides are only 9 residues in length, well defined structures were determined for residues 1-6, presumably as a result of the constraints imposed by the disulfide bond between Cys 1 and Cys 6 . The C-terminal regions are disordered but are oriented differently between the two peptides, indicating that Leu 7 has a significant influence on the overall structure of conopressin-T. This structural shift probably contributes to the relatively poor solubility of Con-T. Given that the orientation of the exocyclic residues is influenced by Pro 7 , it appears that this relatively short stretch of peptide acts independently to determine peptide activity, since the pharmacological profiles of Con-T and L7P-Con-T are somewhat similar.
Comparison of the Con-T and L7P-Con-T structures with oxytocin and vasopressin highlights the similarity between all of the structures (supplemental Fig. 2). In particular, the loop between the two cysteine residues is structurally similar, with root mean square deviations of less than 1.5 Å for all structures. The C-terminal tails of Con-T and L7P-Con-T, although somewhat disordered, do not overlay well with the tail region of vasopressin or oxytocin. The proline residue in L7P-Con-T is in a trans conformation, but there is evidence for a minor conformation that may originate from a cis proline isomer. The proline geometry in OT and AVP has been extensively studied, with small percentages (ϳ10%) of cis isomers in the solution conformations of these molecules (35). This finding is consistent with the current study as the population of the minor conformation of L7P-Con-T is ϳ7% based on relative peak intensities. Previous studies have shown that a cis/trans conformational change plays a role in OT receptor binding and activation (36), but significant differences are not observed for the affinity of Con-T and L7P-Con-T for the OTR, suggesting that cis/trans isomerization is less important for the affinity of these peptides.
The V 1a receptor was recently reported to play a major role in spatial memory (37), heart hypertrophy (38), and hyperfiltration (39). Selective inhibition of this receptor subtype may therefore have important therapeutic applications. The antagonist action of conopressin-T, as well as the determination of its three-dimensional structure, offer novel insights into the structure-activity relationship within the vasopressin peptide family that may guide the synthesis of novel analogues modified at positions 7 and 9 toward the design of highly selective V 1a antagonists. The double underline represents conserved residues. The single underline represents conserved residues in all vasopressin-like peptides except in Con-T or Con-Vil.