The Journal of Neuroscience, September 1995, 15(g): 5989-5998

Structural and Functional Evolution of the / Superfamily: Vasopressin-Related Conopressin Is the Only Member Present in Lymnaea, and Is Involved in the Control of Sexual Behavior

R. E. Van Kesteren,’ A. B. Smit,’ R. P. J. De Lange,’ K. S. Kits,’ F. A. Van Golen,’ R. C. Van Der Schors,’ N. D. De With,’ J. F. Burke,* and W. P. M. Geraertsl ‘Graduate School Neurosciences Amsterdam, Institute of Neurosciences, Faculty of Biology, Department of Experimental Zoology, Vrije Universiteit, Amsterdam, The Netherlands and *Sussex Centre for Neuroscience, School of Biological Sciences, University of Sussex, Brighton, United Kingdom

It has been suggested that the gene duplication that led to Vasopressin and oxytocin are nonapeptides that are present in the formation of the vasopressin/oxytocin two-gene family all placental mammals. Although similar in structure, they serve occurred early during vertebrate evolution. However, the different functions (Ramsay, 1983). Vasopressin has anti- existence of both vasopressin- and oxytocin-related pep- activities and is involved in hydromineral regulation, whereas tides in invertebrates suggests that this duplication may oxytocin has and milk-ejection activities and regulates have occurred much earlier, although there is no evidence aspects of reproductive behavior. Peptides related to both vaso- for the co-occurrence of vasopressin- and oxytocin-related pressin and oxytocin are present in all vertebrate classes, except peptides in the same invertebrate species. We report here the cyclostomes, which have only the vasopressin-related pep- that in Lymnaea only the vasopressin-related peptide Lys- tide (reviewed by Acher, 1993). Peptides of the va- conopressin, but not an oxytocin-related peptide, is pres- sopressin family have a basic amino acid residue at position 8 ent. Moreover, it is very likely that an oxytocin-like cDNA in common, whereas oxytocin and related peptides have a neu- or gene is absent. The conopressin gene is expressed in tral amino acid residue at this position. The chemical nature of neurons that control male sexual behavior, and its gene this amino acid residue is thought to be of critical importance products are present in the penis nerve and the vas defer- in the interactions of the peptides with their respective receptors. ens. Conopressin induces muscular contractions of the The structure of the precursors of the vasopressin/ vas deferens and inhibits central neurons that control fe- oxytocin superfamily are very similar, with a signal peptide do- male reproductive behavior. Thus, although structurally re- main followed by the nonapeptide, neurophysin, and copeptin lated to vasopressin, conopressin has functional and be- domains (reviewed by Acher, 1993), although the copeptin do- havioral characteristics typical for oxytocin. Physiological main is lacking in the oxytocin and mesotocin precursors. The and receptor binding data suggest that conopressin and precursor are encoded by different genes that have an [Ile*]-conopressin, a synthetic oxytocin-like analog of cono- identical exon-intron organization in mammals (Schmale et al., pressin, are functionally equivalent in Lymnaea, and that 1983; Ivell and Richter, 1984; Ruppert et al., 1984; Sausville et the chemical nature of the amino acid residue at position al., 1985; Hara et al., 1990), bony fish (Morley et al., I990), and 8 does not result in a functional difference. Therefore, we cyclostomes (Heierhorst et al., 1992). Together with the phylo- suggest that invertebrates contain only a single member of genetic distribution of the peptides, these data suggest that the the vasopressin/oxytocin gene family and that the amino superfamily evolved from an ancestral gene by the mechanism acid change that distinguishes vasopressin from oxytocin of gene duplication. Because lampreys and hagfish have only vasotocin (Lane et al., 1988; Heierhorst et al., 1992), it has been is functionally neutral in invertebrates. hypothesized that the vasotocin gene is the present-day repre- [Key words: conopressin, vasopressin/oxytocin gene sentative of the ancestral gene, and that the gene duplication family, evolution, male sexual behavior, peptide sequenc- occurred between the radiation of cyclostomes and cartilaginous ing, gene cloning, receptor interaction, Lymnaea stagnalis] fish, about 450 million years ago (Acher, 1980). Vasopressin-related peptides are not restricted to the verte- brates, but have been identified in several invertebrate phyla as Recetved Oct. 3, 1994; revised Apr. 20, 1995; accepted Apr. 24, 1995. well, including insects (Proux et al., 1987), molluscs (Cruz et This work was supported by the Foundation for Life Sciences (SLW: Grant SOS-26.203), which is subsidized by the Netherlands Organization for Scien- al., 1987; McMaster et al., 1992) and annelids (Salzet et al., tific Research (NWO). Support was also received from the EC (BIO2CT- 1993). Oxytocin-related peptides have similarly been found in CT930169). We thank Dr. P. Van Veelen for the mass determination, Dr. G. Reich for performing the anti-oxytocin RIA, Mrs. E. R. Van Kesteren and Mr. molluscs (Reich, 1992) and annelids (Oumi et al., 1994). The J. C. Lodder for technical assistance, and Mrs. T. Laan for secretarial assistance. presence of both vasopressin- and oxytocin-related peptides in Correspondence should be addressed to Dr. R. E. Van Kesteren, Department invertebrates leads to an alternative hypothesis stating that du- of Zoology, Faculty of Biology, Vrije Universiteit, De Boelelaan 1087, 1081 HV Amsterdam. plication of the ancestral gene occurred prior to the divergence Copyright 0 1995 Society for Neuroscience 0270.6474/95/155989-10$05.00/O of vertebrates and invertebrates, over 600 million years ago (Van 5990 Van Kesteren et al. - Conopressin Controls Sexual Behavior in Lymnaea

Kesteren et al., 1992). However, this hypothesis has not been Li et al. (1989). In short, 1% of each fraction was dot blotted on nitro- verified by experimental data showing co-occurrence of vaso- cellulose paper, heat fixed at 106°C for 1.5 min, and then stained with antisera W 1E and 121. Corresponding immunoreactive fractions were pressin- and oxytocin-related peptides in a single invertebrate pooled and applied to a Nucleosil 120 C-l 8 column (3 urn, 4.6 X 150 species, or by demonstrating functional divergence of the two mm) equilibrated with solvent A. Elution was achieved by a linear types of peptide in invertebrates. Thus, species variation instead gradient- of 20-35% solvent B in 60 min. Fractions were Ivoohilized.d I of gene duplication may explain the appearance of either vaso- redissolved in bidistilled water, and screened in the DIA. The immu- noreactive fraction was applied to the same C-18 column equilibrated pressin- or oxytocin-related peptides in invertebrates. with 25 mM NH,Ac for the CNS material. Elution was achieved by a In this article, we attempt to distinguish between these pos- linear gradient of IO-40% of 25 mM NH,Ac in 60% acetonitrile in 60 sibilities by studying vasopressin- and oxytocin-related peptides min. The vas deferens and penis complex material was applied to the in the mollusc Lymnuea stugnalis. Previous cDNA cloning stud- same column equilibrated with 0.05% HCl, and elution was achieved ies (Van Kesteren et al., 1992) have identified a vasopressin-like by applying 0.05% HCl/60% acetonitrile in a linear gradient of O-5% in 5 min. and 5-35% in 60 min. FmCtiOnS were lvoohilized. redissolved precursor, pro-Lys-conopressin, in Lymnaea. To demonstrate that in bidistilled water, and screened using the DIA: L conopressin has the typical tertiary structure of vasopressin-re- Sequencing, co-elution, and muss determination. The immunoreac- lated peptides and is transported to peripheral target tissues, we tive fraction of the final HPLC step was subjected to automated Edman have now identified Lys-conopressin from both the CNS and the degradation using a model 473A pulse liquid protein sequencer (Ap- male copulatory organs. At the same time, we show evidence plied Biosystems). To verify the sequence and complete the structural characterization, synthetic Lys-conopressin was used in co-elution ex- for the absence of an oxytocin-related peptide or cDNA. More- periments employing the same HPLC system, solvents, and gradients over, genomic analysis strongly suggests that the conopressin as in the first and the third rpHPLC step described above. In addition, gene is a single copy gene in Lymnuea, and that related genes the purification procedure was repeated as described above, and the are absent. Although structurally related to vasopressin, cono- immunoreactive fraction of the final HPLC step was subjected to mass pressin has functions in the control of sexual behavior that in spectrometry using a Bio-ion 20/30 plasma desorption mass spectro- meter (Applied Biosystems). mammals usually are associated with oxytocin. The implications Anti-oxytocin radioimmunoussuy. CNS of 200 snails were homoge- of these findings for theories of the molecular and functional nized, prepurified, and subjected to HPGPC as described above. All evolution of the vasopressin/oxytocin superfamily are discussed. fractions were lyophilized, redissolved in bidistilled water, and tested in an anti-oxytocin radioimmunoassay as described (Reich, 1992). Materials and Methods Polymerase chain reaction (PCR) umplijicution of putative oxytocin encoding cDNAs. Two degenerate oligonucleotides, OTI and OT2, were Animals, peptides, and antibodies. Adult specimens of L. stugnalis (shell height 28-34 mm), bred in the laboratory under standard condi- synthesized, based on the putative oxytocin-like sequences Cys-(Phe/ tions, were used. Synthetic Lys-conopressin G was obtained from Bach- Tyr)-(Phe/Ile)-Arg-Asn-Cys-Pro-(Leu/Ile/Val)-Gly and the following em Feinchemikalien AG (Budendorf, Switzerland). [Ilex]-conopressin amidation and processing sequence Gly-Lys-Arg [OTl : 5’.GGAAGC- (Cys-Phe-Be-Arg-Asn-Cys-Pro-Be-Gly-amide) was synthesized using a TTG(TC)T(AT)(TC)(AT)T(ATC)(AC)G(GATC)AA(TC)TG(TC)CC- peptide synthesizer from Applied Biosystems (Foster City, CA). After (GATC)(ATC)T-3’; OT2: 5’.GGAAGCTTG(TC)CC(GATC)(ATC)T- formation of the intramolecular disulphide bridge, the final structure and (GATC)GG(GATC)GG(GATC)AA(GA)(AC)G-3’1. Both oligonucleo- purity of the peptide was confirmed by mass spectrometry and amino tides were provided with Hind III restriction site extensions on the 5’ acid sequencing. Antiserum WlE was a gift of Dr. J. P H. Burbach ends and used as primers in a PCR as described (Van Kesteren et al., (Rudolf Magnus Institute, Utrecht, The Netherlands) and recognizes the 1992). In short, cDNA was isolated from a XgtlO cDNA library of the carboxy-terminal sequence of Arg-vasopressin (Liu and Burbach, 1987), CNS (Vreugdenhil et al., 1988) and used as a template. As reverse and antiserum 121 was a gift of Dr. E W. Van Leeuwen (Netherlands primer, a h primer was used, which is complementary to a sequence in Institute for Brain Research, Amsterdam, The Netherlands) and recog- the left arm of XgtlO (X primer: S’-AGCAAGTTCAGCCTGGTT- nizes the carboxy-terminal sequence of Lys-vasopressin (Swaab and AAGTCC-3’). In the first PCR, cDNA was amplified between OTl and Pool, 1975). An anti-CDCH antiserum was used as described (Van Min- the X primer, using 40 cycles: 94°C for 1 min, 52°C for 1 min, and nen et al., 1989). A synthetic peptide fragment corresponding to the 11 72°C for 2 min. After amplification, 5 pl of the PCR mixture were amino-terminal amino acids (SLDTGMVTSRE) of Lymnaeu neurophy- reamplified under the same conditions, now using OT2 and the h primer. sin was a gift of Dr. R. H. M. Ebberink (Applied Biosystems, Maarssen, Amplified cDNA was digested with Eco RI and Hind III, cloned into The Netherlands). An antiserum to the fragment was raised in mice as Ml3mp18 and sequenced using the dideoxy chain termination method described (Van Minnen and Schallig, 1990). (Sanger et al., 1977). Purijicution of conopressin. Relevant tissues (CNS, or vas deferens Low stringency screening of Lymnaea cDNA and genomic libraries and penis complex together) of 1000 snails were homogenized in 0.1 for conopressin-related sequences. From an amplified cDNA library of M acetic acid and centrifuged. The supernatant was applied to Supel- Lymnaea CNS in hgtl0 (for details, see Vreugdenhil et al., 1988), as clean C-8 cartridges (Supelco Inc., Bellefonte, USA) previously acti- well as from an amplified Lyymnaea genomic library in EMBL3 (for vated with methanol and solvent A (7.5 mM trifluoroacetic acid). After details, see Bogerd et al., 1993), recombinant phages (3.0 X IO5 and loading, cartridges were washed with 4 ml solvent A, and peptides were 1.2 X IO’, respectively) were plated on 245 X 245 mm dishes (0.6 X eluted with 2 ml solvent B (7.0 mM trifluoroacetic acid in 60% aceto- lo5 phages/dish) and absorbed to replica Hybond N’ filters (Amersham nitrile). The effluents were lyophilized and redissolved in bidistilled Int., Amersham, UK). After prehybridization for 16 hr, the filters were water. High-performance gel permeation chromatography (HPGPC) and hybridized at 55°C for 60 hr to a i2P-labeled probe that was generated reverse-phase high-performance liquid chromatography (rpHPLC) were by primer extension on a single stranded Ml3 clone containing the full performed using an M-720 system controller, an M-730 data module, length conopressin cDNA. The hybridization solution contained 6X an M-600 solvent delivery system, and an M-441 UV/VB absorbance SSC (1X SSC = 150 mM NaCl, 15 mM sodium citrate; pH 7.4), 5X detector (Waters Assoc., Bedford, MA). The flow rate was 1 mllmin, Denhardt’s (according to Maniatis et al., 1982), 0.1% SDS and 10 kg/ and fractions of 1 ml were collected. ml salmon sperm DNA. The specific activity of the probe was > 3X The peptide extract was applied to tandemly arranged Protein-Pak lo8 dpm/pg. Filters were washed in 2X SSC/O.l% SDS at 55°C for 30 I-125 and I-300 gel permeation columns (Waters Assoc.) in several min and autoradiographed. Positive clones were replated at lower successive runs. The running solvent consisted of 30% acetonitrile in plaque density and screened with the cDNA probe. Genomic clones 7.5 mM TFA. Fractions containing the small peptides were pooled, ly- were also screened with synthetic oligonucleotides specific for either ophilized, redissolved in bidistilled water, and applied to a Nucleosil the leader sequence or the neurophysin domain of the conopressin gene. 300 C-18 column (5 pm, 4.6 X 250 mm), equilibrated with solvent A, Inserts from positive cDNA clones were subcloned and sequenced. Of in several successive runs. Elution was achieved by a linear gradient of three positive genomic clones, XDNA was mapped for restriction en- O-10% solvent B in 5 min, lo-35% solvent B in 35 min, and 35-100% donuclease sites for the enzymes Bgl II, Eco RI, Hind III, Pst I, Sal I, solvent B in 25 min. Fractions were lyophilized, redissolved in bidis- and Xba I by partial digestion and hybridization to an oligonucleotide tilled water, and tested in a dotting immunoassay (DIA) according to directed to the right arm of EMBL3. The clones were directly sequenced The Journal of Neuroscience, September 1995, 15(9) 5991

Figure 1. Purification of conopressin from the CNS of Lymnaeu. A, Chro- matogram of the first rpHPLC step, representing one-third of the total CNS material (330 animal equivalents). Syn- thetic Lys-conopressin co-eluted with fraction 39. B, Chromatogram of the fi- nal rpHPLC step, representing the tofdl CNS material (1000 animal equiva- lents). Synthetic Lys-conopressin co- eluted with fraction 48, which was also I I 1 I I I I used for molecular characterization. IO 20 30 40 SO 60 0 I;, 2b 30 40 50 60 Immunoreactive fractions are indicated t (min) t (min) by black squares at the time axis.

by linear I out amplification of the XDNA using 5’.end labeled conopressin tion) were carried as described (De Lange et al., 1994). Hybridiza- specific primers and dideoxy chain termination mixtures (Lasham and tion was performed in the presence of 30% formamide, with a xsS- Darlison, 1993). labeled oligonucleotide complementary to the neurophysin encoding Genomic Southern anulysis. Genomic DNA was isolated from 100 part of the conopressin cDNA (5’.CGGTCCGACACACTGTCCGGT- CNS using the G-nome DNA Kit (Bio 101, La Jolla, CA), and digested GCCACCTGGCCC-3’) in a final concentration of 10 fmol/p,l. with 200 units of either of the restriction enzymes Bgl II, Eco RI, Hind Eflects of Lys-conoprrssin und [Ile8]-conopressin on the vas deferens. III, Pst I, and Xha I (8 yg DNA/digestion). The digested DNA was The anterior part of the vas deferens was dissected, attached to a dis- electrophoresed on a 0.8% agarose gel and transblotted to Hybond N+ placement transducer (Li et al., 1992) in a 1 ml chamber tilled with filter. After prehybridization, the filter was hybridized for 60 hr to a ??P- snail Ringer (Geraerts et al., 1984). Peptides were added at different labeled PCR fragment containing the entire coding sequence of the concentrations under continuous superfusion with Ringer at a flow rate conopressin cDNA (specific activity: > 5X lox dpm/p,g). Hybridization of 1 ml/min and contractions were displayed on a recorder. Dose-re- conditions were as described above. The filter was washed in I X SSC sponse curves were constructed by plotting the contraction frequencies at 55°C for 30 min and autoradiographed for 20 hr. Thereafter, the filter (mean of four experiments) semilogarithmically versus the concentra- was washed in 0.2X SSC at 65°C for 30 min and autoradiographed for tion of the peptide. 7 d. Membrane potential ejj%cts of conopressin on caudodorsal cells. Immunoc)ltnchemistry and in situ hybridization. CNS and vas defer- Electrophysiological recordings were performed on Caudodorsal cells ens were dissected and fixed in a mixture of 1% paraformaldehyde and (CDC) in situ in the isolated CNS (Kits et al., 199 I). Membrane poten- 1% acetic acid or in Bouin’s fixative. Tissues were dehydrated, embed- tial recordings were made using microelectrodes (Clark, Reading, UK) ded in paraffin, cut to 7 pm sections, and alternately mounted on 0.5% filled with 0.1 M KCI and with resistances of 15-30 Ma connected to gelatine/0.5% chrome-alum coated microscope slides. For immunocy- high input resistance amplifiers built in the laboratory. Recordings were tochemistry, dewaxed slides were incubated at 4°C for 16-18 hr with plotted on paper with a Gould 2200 pen recorder. Bath saline had the primary antibodies diluted in Tris-buffered saline (TBS; 50 mM Tris, following composition: 30 mM NaCI, IO mM NaCH,SO,, 5 mM Na- 150 mM NaC1; pH 7.4) containing 0.25% gelatine and 5% Triton X- 100 HCO,, 1.7 mM KCl, 4 mM CaCI,, 1.5 mM MgCl,, 10 mM Hepes; pH (TBS.GT). Anti-neurophysin was diluted 1:200 and anti-Caudodorsal 7.8. Cell (anti-CDCH) 1:500. After two IO min rinses in TBS-GT, sections were incubated at 20°C for 1 hr with a secondary antibody coupled to horseradish peroxidase, diluted I : 100 in TBS. After rinsing Results in TBS-GT, horseradish peroxidase was visualized by incubation in Purijcation and structural identi$cation qf Lymnaea 0.05% 3,3 diaminobenLidine tetrahydrochloride and 0.01% H,O, in conopressin TBS. In situ hybridization procedures (pretreatment, labeling, hybridiza- After the first rpHPLC step, two separate immunoreactive frac- tions (fractions 28 and 39; Fig. IA) could be detected in CNS material with the antisera WlE and 121. Fraction 39 was further Table 1. Yield of PTH-amino acids in nine cycles of automated purified, and after a third rpHPLC step, two immunoreactive Edman degradation of purified Lys-conopressin fractions (47 and 48; Fig. 1B) could be detected. The peptide material in fraction 48 was sequenced without modification of Edman PTH-amino Yield cysteine residues (Table 1). In the first and the sixth cycle of the cycle acid (pm00 Edman degradation, no PTH-amino acids could be detected. As- 1 cys -C’ suming that the peptide contains cysteine residues at the corre- 2 Phe 26.80 sponding positions, its amino acid sequence [(Cys)-Phe-Ile-Arg- 3 Ile 28.31 Asn-(Cys)-Pro-Lys-Gly] is identical to the primary sequence of 4 Arg 2.06 Lys-conopressin (Cruz et al., 1987). Plasma desorption mass 5 Asn 13.17 spectrometry indicated that the peptide has an MH’ of 1036.7 6 cys -ci Da. HPLC analysis demonstrated co-elution of synthetic Lys- 7 Pro 15.40 conopressin with the native peptide in two rpHPLC systems (7.5 8 Lys 12.47 mM TFA and 25 mM NH,Ac). 9 GUY 15.47 Further purification and characterization of the immunoreac-

” PTH-cysteines are unstable during Edman degradation and, therefore, unde- tive substance of fraction 28 (Fig. 1A) revealed a novel peptide tectable. with the amino acid sequence Leu-Ala-Ser-Leu-Gly-Gly-Lys- 5992 Van Kesteren et al. - Conopressin Controls Sexual Behavior in Lymnaea

Ser-Ser-Ser-Arg-Pro-Ala-Pro-Arg-Gly-amide. The structure of the vasopressin and oxytocin genes of vertebrates (Fig. 2B). this peptide was confirmed by mass spectrometry and co-elution Exon I and the 5’ part of intron 1 are located on ACP56 and the with a synthetic form of the peptide. The carboxy-terminal se- 3’ part of intron 1 and the rest of the gene are located on both quence Pro-Arg-Gly-NH, explains its immunoreactivity to the ACP21 and ACP31 (Fig. 2). The exon sequences of ACP56 and anti-vasopressin antisera. However, because it lacks the struc- ACP21 are identical to the conopressin cDNA sequence (Van tural characteristics of the vasopressin/oxytocin superfamily Kesteren et al., 1992). Clone ACP31, however, has two conser- (e.g., the presence of cysteine residues that can form a disulphide vative third-base substitutions in the neurophysin encoding part bridge), it is not further considered here. of the gene (one in exon II and one in exon III) and numerous Lys-conopressin could also be purified from the vas deferens substitutions in the 3’ noncoding sequence and in the introns, as and the penis complex together (data not shown). Its identity well as an 18 base pair insertion in intron 2 (Fig. 2B). was established by amino acid sequencing and co-elution with In exon III, at 14 nucleotides after the first putative poly7 synthetic Lys-conopressin. adenylation signal (at nucleotide 1245 of the 2.5 kb cDNA; Van Kesteren et al., 1992), sequence identity between gene and Anti-oxytocin radioimmunoassay cDNA ends. Since many additional cDNA clones have been A peptide extract of 200 Lymnaea CNS was size fractionated sequenced, all containing poly(A)-tails at position 1245 (data not on HPGPC and all fractions were tested in a radioimmunoassay shown), we suggest that the rest of the cDNA is not a part of with 1251-oxytocin as the tracer (Reich, 1992). No immunoreac- the conopressin gene and became unspecifically ligated to the tivity significantly above background could be detected in these conopressin cDNA during construction of the library. fractions (data not shown). Genomic Southern analysis PCR amplijcation qf putative oxytocin encoding cDNAs Southern blots containing genomic DNA digested with five dif- Amplification of AgtlO cDNA using primer OT I in combination ferent restriction enzymes did not show different hybridization with the A primer yielded many DNA fragments when analized patterns when hybridized to a PCR fragment of the coding re- by electrophoresis on an agarose gel. Reamplification of the gion of the conopressin cDNA at low (1 X SSC, 55°C) or at high PCR mixture using primer OT2 in combination with the A prim- (0.2X SSC, 65°C) stringency (Fig. 3). In each lane, three to six er yielded a single DNA fragment of 700 base pairs (data not bands hybridized at low stringency and could not be washed off shown). This fragment was cloned, and 30 clones were se- at high stringency. Only in the lanes containing Bgl II and Pst quenced and shown to contain the conopressin cDNA. To ex- I digested DNA, a few faint bands were present that disappeared clude the possibility that the amplification product consisted of after high-stringency washing. the conopressin cDNA and a less abundant fragment represent- ing a related cDNA species, -300 additional clones were hy- Localization of conopressin and conopressin transcripts bridized to an oligonucleotide specific for the 3’ trailer sequence of the conopressin cDNA. No clones were found that did not To determine the site of synthesis of conopressin, anti-neuro- hybridize, demonstrating that the amplification product most physin antiserum and conopressin-specific oligonucleotides were likely consisted of only the conopressin cDNA. applied to alternate sections of the CNS. Both the antiserum and the oligonucleotide probe recognized one neuron at the caudal Low-stringency screening cf Lymnaea cDNA and genomic surface of each pedal ganglion (not shown) and neurons of the libraries ,fiw conopressin-related sequences anterior lobes of both cerebral ganglia (Fig. 4A,B). The anti- Low-stringency hybridization of a cDNA library of Lymnaea neurophysin antiserum stained also axon bundles in the penis CNS yielded 26 positive clones. After subcloning and sequence nerve (Fig. 4C) and along the vas deferens (Fig. 40). Positive analysis, two of these appeared to contain the conopressin cDNA staining was also observed in axons in the cerebral ganglia lo- sequence. All others contained sequences that were not related cated in close proximity of the CDC (Fig. 4E). When the alter- to conopressin, and were apparently isolated because of short nate sections were stained with anti-CDCH, to visualize the ax- stretches of DNA showing slight homology with the conopressin ons of the CDC, anti-neurophysin positive axons and anti-CDCH cDNA sequence (data not shown). positive axons could be observed within the cerebral ganglia in Low-stringency hybridization (i.e., washing conditions: 2X close proximity of each other (Fig. 4E,F). SSC/O.l% SDS at 55°C) of a Lymnaea genomic library yielded 17 clones that after two rounds of screening still hybridized to l?jfects of Lys-conopressin and [IleX]-conopressin on the vas the conopressin cDNA probe. Of these clones, 10 hybridized to deferens an oligonucleotide specific for the leader sequence of the cono- Application of Lys-conopressin or [Ile8]-conopressin to the dis- pressin cDNA, and 7 to an oligonucleotide specific for the neu- sected anterior vas deferens (the part of the vas deferens that rophysin domain. There was no overlap in these two groups of runs from the body wall to the penis; Barraud, 1957) elicited clones. One clone (hCP56) of the first group and two (ACP21 contractions that increased with frequency in a dose-dependent and XCP3 1) of the second were used for restriction enzyme map- manner. The contraction frequency during the first minute fol- ping and sequence analysis. The restriction map of hCP56 shows lowing application (mean of four experiments) was used to con- no overlap with those of XCP21 and ACP3 1, while the latter two struct dose-response curves (Fig. 5A). The calculated EC,,, val- have a large part in common. Within the overlapping part, how- ues were 4 ? 1.2 nM for Lys-conopressin and 1 + 0.5 nM for ever, ACP21 contains two Xba I sites that are absent in ACP3 1, [Ilex]-conopressin (mean -C SD). Desensitization of the cono- whereas ACP31 contains two Xba I sites and one Pst I site that pressin receptor by repeated applications of Lys-conopressin are absent in ACP21 (Fig. 2A). Sequencing of the three clones resulted in decreased responses to both Lys-conopressin and revealed that the conopressin gene is composed of three exons [Ile*]-conopressin, whereas the response to an unrelated -FF- interrupted by two introns at exactly the same locations as in amide peptide (Li et al., 1995) did not desensitize (Fig. 5B), The Journal of Neuroscience, September 1995, 75(9) 5993

A LCP 56 ____------_------

x x HEXXH H E E ?LcP21 1 1 I I II Ill

--> conopressin --> “europhysln hfP56 CTG CTG *cc GCC GCT GTC cl% *CA cm TCA CTG ACG GAC GCT TGT TTC ATC *Go MC TGT cc* lvvL GGT GGA AAG cG* TCG **a GAC ACG GGC ATG GTG *co Leu Leu Thr Ala Ala Val Leu Se= LBU Se= Leu Thr ASP Ala Cye Phe Ile A=g &sn Cys PEO lye Gly Gly Lys ?,=g Se= LB,, Asp Th= Gly jet "al Thr --,intron 1

hCP21 lyjC ATG AA0 TGT GGG CCA GGT GGC ACC DDA CA0 TOT OTC! GGA CC0 AGC ATC TGC TGT GGT CA0 GAC TTT GGC TGT CAT GTC GGG ACA GCG GAG GCG GCG GT,, hCP?Z ------~~------a ------. --- -.. . . -~. ~.. ..~ ~~~ ------~~~ ..~ . . . . '&S Met W'S Q'S Gly Pro Gly Gly Th= Gly Gln Cya Val Gly P=o Se= Ile WS CYS Gly Gin ASP Phe Gly Cys His Val Gly Thr Ala Glu Ala Ala val kPil1 WC CM CAD GAG AAC GAC AGC TCG ACC CC0 TOC CT0 OTC ZAG GGG GAG GCG TGT GGG TCA ADO OAT GCG GGC AAC TGT GTG GCA GAC GGC ATA TGC TGT GA* ,J&.p31 ~.. ------. .------...... ------...... _. .~~ --- --_ ..~ ~~~ ~~~ ~.. ..~ CYS Gln Gln Glu AS* ASP Se= Se= Th= P=O CYS bell Val LYS Gly Glu Ala Cys Gly Se= A=g ASD Ala Gly Asn Cys "al Ala Asp Gly 11e Cys cys ASP --> intron 2

kP21 ~TCTGGTHTCT~TT~TCTTCGATGTTTTTCATCTAG AA TCA TGT GCT On: MT GAC CGA TGC CGT GAT CTA GAT GGA AAC GCC CAG GCC MC CGO OOT GAC cw ATT ).cpj1 ------...... ------.. .._ ___ --- ..- .__ -__ ------~~~ . ~~~ ~.. lU Se= CY8 Ala Val Aan ASP A=g C,'S A=g ASP leu ASD Gly A8n Ala Gln Ala Asn A=g Gly ASI) Let, 11e hCF2l CA0 TTG ATA CAC aaa CTC CTG AAA GTG AGG GAT TaC GaC TM CCGGMG~MCTMDCCGCC~GC~TGCTTCTCAGCCATGC~G~AGTGGT~~CATG~G~~GTGGTCCT hCPjl ~~------G ------..G...... ~~~--~~----*------.-.~--~~....*...... ~~...... ~~ Gln Leu Ile His Q'S Leu Leu bys Val A=g AS,, Tyr Asp stop

F&o-e 2. Organization and nucleotide sequence of the Lynwea conopressin gene. A, Restriction maps of genomic clones hCP56, ACPZI, and ACP32. Black boxes indicate the positions of exons I, II, artd III. Recognition sites for the restriction endonucleases Bgl II (B), Eco RI (E), Hind III (H), Pst I (P), Sal I (S), and Xbn I (X) are indicated. Before exon II, the restriction maps of clones XCP2 1 and XCP3 I differ due to polymorphism (see Discussion). The size bar represents I kb. B, Nucleotide sequence of the conopressin gene. Relevant parts of clones XCP56, XCP2 I, and XCP3 I were sequenced. Exon sequences are shown in bold face, as well as the predicted amino acid sequence of the conopressin precursor. Arrows indicate the beginning of exon I, II, and III, intron I and 2, the signal peptide domain, the conopressin domain, and the neurophysin domain. The transcription initiation site (the beginning of exon I) was determined by size determination of reverse transcription products of the conopressin mRNA (data not shown). The polyadenylation signal is underlined. Parts of the introns that were not sequenced are represented by dots. Of clone XCP3 I, only the nucleotides that differ from clone XCPZI are shown, whereas the rest of the sequence is represented by dashes. suggestingthat Lys-conopressinand [Ile8]-conopressininteract of action potential frequency was observed over the sametime with the samereceptor. period (seealso Kits, 1980).

Mernbmne potentiul effects of conopressirl on CDC Discussion Electrical stimulationof CDC in situ in the isolatedCNS evoked A single copy conopressin gene is the 0111, member of the an afterdischargethat lasted -50 min. Bath applicationof cono- vusopressin/oxytocin gene funiil~ present in Lymnaea pressinresulted in a hyperpolarization of the membranepotential We usedseveral independentmethods to examine the presence and an inhibition of electrical activity (Fig. 6). Due to the hy- of vasopressin-and oxytocin-related peptidesand corresponding perpolarization, an increasein the spike amplitudeoccurred. To genes in Lymnnen: peptide purification and characterization, quantify the effect of conopressin,action potential frequencies PCR-amplification, and low-stringency hybridization of cDNA were measuredduring 40 set before and 40 set after application and genomic DNA. Two antisera(W IE and 121) that specifi- of the peptide. Before application, a frequency of 0.7 + 0.2 cally recognize the Pro-Lys/Arg-Gly-amide carboxy-terminal se- spikeslsecwas observed,whereas afterwards the frequency was quence typical of vasopressinand vasopressin-relatedpeptides 0.4 -C 0.1 spikes/set (mean ?ZSD; n = 4). Thus, conopressin (Swaaband Pool, 1975; Liu and Burbach, 1987) were employed significantly reduced the action potential frequency of the dis- to detect vasopressin-relatedpeptides in extracts of the brain. charge (p < 0.02) whereasin control discharges,no reduction One of the two immunoreactive peptides that were identified 5994 Van Kesteren et al. * Conopressin Controls Sexual Behavior in Lymnaea

- 21kb -

- 9.4 kb - - 6.6 kb - - 5.0 kb - - 4.3 kb - - 3.5 kb -

Figure 3. Genomic Southern blot analysisof the Lymnaea conopressin - 2.3 kb - gene.A blot containingLymnaea ge- nomicDNA digestedwith BgZII, Eco - 2.0 kb - RI, Hind III, Pst I, andXba I was hy- bridized at 55°C with a probe corre- - 1.6kb - spondingto the entirecoding region of - 1.3 kb - the conopressincDNA, andwashed at low stringency (1X SSC/O.l% SDS, - l.Okb - 55°C;left panel) andthereafter at high stringency (0.2X SSC/O.l% SDS, 65°C; right panel). No differences were observedbetween the hybridiza- tion patternsat thesetwo washingcon- ditions,except for somefaint bandsin lx SSC; 0.1% SDS 0.2x SSC; 0.1% SDS the Bgl II and the Pst I lanes,which disappeared after high-stringency washing.Size markers are indicated be- tweenthe two panels. 55 “C 65 “C from the brain has a primary structure identical to that of Lys- naea. In mammals,the vasopressinand oxytocin genesconsist conopressin,originally identified in C. geographus (Cruz et al., of three exons, with a remarkably high-sequenceidentity be- 1987), except that the cysteine residuesat positions 1 and 6 and tween the secondexon of each gene that is maintainedby the the carboxy-terminal amide could not be determined directly. mechanismof gene conversion (Ruppert et al., 1984). Conse- However, becausethe MH+ of the native peptide (1036.7 Da) quently, vasopressinand oxytocin genescross-hybridize, even correspondswith the expected MH+ of Lys-conopressin(1034.3 under high stringency conditions (seefor instance,Land et al., Da) and as it co-elutes with synthetic Lys-conopressinin two 1983). Members of multigene families are commonly identified rpHPLC systems,it must have a disulphide bridge involving by low-stringency hybridization using one of the genes as a cysteine residuesat positions 1 and 6, as well as an amidated probe (see for instance, Haeflinger et al., 1992; Garcia et al., carboxy-terminal glycine residue.Thus, Lymnaea conopressinis 1995). Therefore, we applied low-stringency hybridization with identical with Lys-conopressinand has the structural character- conopressincDNA to identify putative oxytocin-like cDNAs and istics typical for vasopressinand related peptides,including the genesin Lymnaea. In our experiments,hybridization conditions basic amino acid residueat position 8. were chosensuch that even sequenceswith only 50% sequence To investigate the presenceof oxytocin or oxytocin-related identity would have been detected(see Maniatis et al., 1982). A peptides in the Lymnaea CNS, we used a radioimmunoassay further lowering of the stringency resulted in a strong increase employing an antiserumthat was raisedagainst oxytocin. In dis- of backgroundhybridization on genomic Southernblots, where- placement studies,this antiserumrecognized the oxytocin-relat- as no additional bands appeared(data not shown). We first ed peptides isotocin, mesotocin, and cephalotocin, but not va- screeneda cDNA library of Lymnaea CNS under low-stringency sopressinor vasotocin (G. Reich, personalcommunication). No conditions in order to find conopressin-relatedcDNAs. In ad- immunoreactivepeptides were found in HPGPC fractions of the dition to two conopressinclones, 24 positive clones were iso- Lymnaea CNS. However, the antiserummay not recognize all lated; however, none of these clones encoded a conopressin- oxytocin-related peptides, and the absenceof a peptide related related peptide precursor. Moreover, amplification of putative to oxytocin in Lymnaea cannot be concluded on these grounds oxytocin-like sequencesfrom the same cDNA library using a alone. Therefore, we examined the possibility that multiple PCR approachthat was previously usedto isolate the conopres- membersof the vasopressin/oxytocin gene family exist in Lym- sin cDNA did not result in the isolation of oxytocin-like cDNAs. The Journal of Neuroscience, September 1995, 15(9) 5995

Figure 4. Localization of conopressin gene expression in anterior lobe neu- rons and of anti-neurophysin immuno- reactivity in the CNS, the penis nerve, and the vas deferens. A, In situ hybrid- ization of a section through the anterior lobe of the right cerebral ganglion with a proconopressin specific oligonucleo- tide probe. Arrows indicate anterior lobe neurons that express the conopres- sin gene. In the alternate section (B), the same neuronsare positively stained with the anti-neurophysin antiserum (arrows). C, Section through the origin of the penis nerve. Arrowheads indi- cate anti-neurophysin-positive axon bundles in the penis nerve. D, Cross- section through the vas deferens. Ar- rowheads indicate anti-neurophysin- positive axon bundles in the outer layer of the vas deferens. E, Section through the right cerebral ganglion. Arrow- heads indicate anti-neurophysin-posi- tive axons in the neuropile of the gan- glion. These axons occur in the same location as anti-CDCH nositive axons. as visualized in the alternate section (arrowheads in F), suggestingcono- pressinergicinnervation of the CDC system. Some of the CDC somata are visible at the bottom right-hand corner in Figure F (Scale bars: 50 p.m.)

Because both the screening and the PCR approach are restricted do not encodedifferent peptide precursorsand the differences to genes that are expressed in the CNS, we also performed a are almost exclusively restricted to the noncoding parts of the screening of a genomic DNA library under low-stringency con- gene. To verify this, we performed genomic Southern blot hy- ditions. This resultedin the isolation of 17 clonesthat hybridized bridizations (Fig. 3). The conopressincDNA probe identified to the conopressincDNA. All clonescontain parts of the cono- multiple bandsin each restriction enzyme digestionof genomic pressingene, as was verified by hybridization under high-strin- DNA when hybridized under low-stringency conditions (1 X gency conditions to oligonucleotidesspecific for the leader se- SSC at 55°C). After washingthe blot at high stringency (0.2X quence and the neurophysin domain of the conopressincDNA, SSC at 65”(Z),all bandsremained equally intens,except for some and by sequencingof the part encoding conopressinof all 10 faint bandsin the Bgl II and the Pst I lanes,which disappeared. clones that contain exon I. Together, these data show that an Since this was not observed in any of the other digestions,these oxytocin-related peptide or gene can not be identified in Lym- bandsare most likely due to aspecific hybridization under low naea using various well-establishedpeptide chemical and mo- stringency conditions. The Xba I lane (Fig. 3) showedthat the lecular biological methods,and strongly suggestthat conopres- sin is the only memberof the vasopressin/oxytocin superfamily two fragments(1.2 kb and 4.5 kb) containing exon II in clones in Lymnaea. XCP21 and XCP31 hybridized with different intensities, sug- Restriction enzyme mapping and nucleotide sequencingre- gestingthat they have an unequal distribution in the snail pop- vealed that the conopressin gene is split by two introns at exactly ulation and representpolymorphisms. Hybridization of the blot the sameposition as in the vasopressinand oxytocin genes(Fig. with exon-specific probes (not shown) revealed similar restric- 2). Thus, the exon-intron organization of the membersof vaso- tion site polymorphismsfor the other enzymes used,except for pressinloxytocin gene family has been conservedduring at least Bgl II. In conclusion,these data show that all hybridizing frag- 600 million years of evolution. Differences in the restriction mentsare derived from the conopressingene, and that there is maps and nucleotide sequences of clones XCP21 and ACP31 no evidence for an additional oxytocin-related genein the Lym- (Fig. 2) are very likely due to polymorphism, as the two clones naea genome. 5996 Van Kesteren et al. * Conopressin Controls Sexual Behavior in Lymnaea

Lys-conopressin [He 8 ]-conopressin GLTPNMNSLFF-amide

I I I I I I r -12 -1 1 -10 -9 -8 -7 -6 t t log [ligand] (M) Lys-conopressin [Ile 8 ]-conopressin GLTPNMNSLFF-amide

Figure 5. Excitatory effects of Lys-conopressin and [hex]-conopressin on the anterior vas deferens of Lymnaea. A, Dose-response curves of the effects of Lys-conopressin (b&k circles) and [hex]-conopressin (open squares) on the contraction frequency. The contraction frequency during the first minute after application of the peptide is expressed relative to the maximum frequency, and plotted semilogarithmically versus the peptide concentration. Means and SEM (error bars) are shown of three independent experiments. B, Responses to Lys-conopressin (3X lo-” M), [Ibex]- conopressin (3X 10mxM), and the -FF-amide peptide (GLTPNMNSLFF-amide; 1O-5 M) before (upper panel) and after (lower panel) desensitization of the conopressin receptor by repeated applications of Lys-conopressin. The response to the unrelated peptide does not desensitize.

Conopressin is involved in the control qf reproductive roendocrine cells that release egg-laying peptides and control behavior egg mass production and associated female reproductive behav- The conopressin gene is expressed in neurons in the anterior ior (Geraerts et al., 1991). Conopressin hyperpolarizes the CDC lobes of both cerebral ganglia (Fig. 4A,B). These neurons were membrane potential and terminates an electrically triggered previously shown to project into the penis nerve to innervate CDC discharge (Fig. 6), suggesting that conopressin may be able the penis complex and the vas deferens (Smit et al., 1992; A. to inhibit peptide release by the CDC system and can prevent Ter Maat, A. W. Pieneman, and Y.A. Van Duivenboden, unpub- egg laying. These actions are meaningful in the context of the lished data). In accordance, conopressin immunopositive axon simultaneous hermaphroditic mode of reproduction of the ani- bundles were observed in the penis nerve and in the vas deferens mal, which necessitates a temporal patterning of the otherwise (Fig. 4C,D), suggesting a possible role for conopressin in the conflicting mating and egg-laying behaviors. control of the activities of the male copulation structures. This was further supported by the isolation and structural identifica- The vasopressin and oxytocin gene ,fhmily: evolution of genes tion of Lys-conopressin from the penis complex and the vas and ,functions deferens together. Our results support the view that an oxytocin-like peptide or Conopressin potently induced muscle contractions in the vas gene is absent in Lymnaea. First, using immunological methods deferens at nM concentrations. The frequency of these contrac- and well-characterized antisera, only the vasopressin-related tions was dose dependent within a physiological range of cono- peptide Lys-conopressin, but no oxytocin-related peptide could pressin concentrations, with an EC,, of 4.3 nM (Fig. 5A). Thus, be identified in the CNS. Second, no conopressin-related cDNAs conopressin may be involved in the control of ejaculation of or genes could be identified using low-stringent hybridization as semen during intromission in Lymnaea, which nicely corre- well as PCR amplification of oxytocin-like sequences. Third, the sponds to the proposed role of oxytocin in the regulation of vas function of conopressin, i.e., control of reproductive behavior, is deferens activity and ejaculation of semen in mammals (Todd similar to that of oxytocin in mammals, suggesting that cono- and Lightman, 1986; Murphy et al., 1987). In addition to effects pressin might represent an ancestral peptide of the vasopressin/ on peripheral tissue, conopressin also affects central neurons in- oxytocin superfamily, endowed with structural characteristics of volved in the regulation of reproductive behavior. Effects of Lys- vasopressin and functional characteristics of oxytocin. There- vasopressin on identified molluscan neurons have been noticed fore, we hypothesize that according to the current model the before (Barker and Gainer, 1974). In the present study, we ob- gene duplication that led to a vasopressinloxytocin two-gene served that anti-neurophysin-positive axons occur in close prox- family occurred early during vertebrate evolution, and that in imity of the axons of the CDC (Fig. 4E,F), suggesting cono- invertebrates, only one gene is present. pressinergic innervation of the CDC system. The CDCs are neu- Thus, the presence in invertebrates of both vasopressin- and The Journal of Neuroscience, September 1995, 15(9) 5997

from a basic to an aliphatic residue occurred independently in the evolution of the various invertebrate phyla, without altering the functional properties of the peptide. In vertebrates, on the other hand, duplication of the gene followed by mutation of residue 8 in one of the genes, resulted in separate evolutionary lineages in which different receptors co-evolved with each pep- tide, yielding functionally distinct peptide-receptor combina- 20 mV tions. References Acher R (1980) Molecular evolution of biologically active polypep.. tides. Proc R Sot Lond [Biol] 210:21113. Acher R (1993) Neurohypophysial peptide systems: processing ma- chinery, hydroosmotic regulation, adaptation and evolution. Regul Pept 45:1-13. Barker JL, Gainer H (I 974) Peptide regulation of bursting pacemaker activity in a molluscan neurosecretory cell. Science 184: 137 I - 1372. Barraud EM (1957) The copulatory behaviour of the freshwater snail (Lymnaea stugrzalis L.). Br J Anim Behav 5:55-59. Bogerd J, Van Kesteren RE, Van Heerikhuizen H, Geraerts WPM, Veen- stra J, Smit AB, Joosse J (1923) Alternative splicing generates di- versity of VDI/RPD2 alpha peptides in the central nervous system of Lymnnea stugnulis. Cell Mol Neurobiol 13: 123-l 36. Garcia CK, Brown MS, Pathak RK, Goldstein JL (1995) cDNA cloning of MCT2, a second monocarboxylate transporter expressed in differ- ent cells than MCTl, J Biol Chem 270:1843-1849. Cruz LJ, De Santos V, Zafaralla GC, Ramilo CA, Zeikus R, Gray WR, Olivera BM (1987) Invertebrate vasopressinloxytocin homologs: characterization of peptides from Corzus geographus and Con~ls srriu- 0 10 20 30 40 50 tusvenoms. J Biol Chem 262:15821-15824. De Lange RPJ, Van Minnen J, Boer HH (1994) Expression and trans- time (min) lation of the egg-laying neuropeptide hormone genes during post- Figure 6. Membrane potential effects of conopressin on CDC. A, embrionic development of the pond snail Lymnueu stugnali~. Cell Membrane potential recording of a CDC in situ in the isolated CNS. Tissue Res 275:369-375. An afterdischarge was evoked by electrical stimulation of the CDC and Geraerts WPM, De With ND, Vreugdenhil E, Van Hartingsveldt W, Lys-conopressin (IO-” M) was applied at 2.5 min after the onset of the Hogenes TM (1984) Studies on the physiological role of a partially discharge (arrow). Conopressin rapidly hyperpolarized the CDC mem- purified small cardioactive neuropeptide of Lymnueu .stugnulix J brane potential and reduced the firing frequency. As a result of the Comp Physiol [B] 154:29-34. hyperpolarization, the spike amplitude increased. The spiking activity Geraerts WPM, Smit AB, Li KW, Vreugdenhil E, Van Heerikhuizen H from 20 set before until -5 min after application of conopressin is (1991) Neuropeptide gene families that control reproductive behav- shown. B, Frequency plots of a control CDC dicharge (black clots) and iour and growth in molluscs. In: Current aspects of the neurosciences a CDC discharge in the presence of conopressin (open dots). The con- (Osborne NN, ed), pp 255-304. London: MacMillan. trol discharges lasted for 48 min, whereas the discharges in the presence Haeflinger JA, Bruzzone R, Jenkins NA, Gilbert DJ, Copeland NG, Paul of conopressin lasted for 13 min (single representative of four indepen- DL (1992) Four novel members of the connexins family of gap dent experiments). junction proteins: molecular cloning expression and chromosomal mapping..J Biol Chem 267:2057-2064. . Hara Y, Battev J, Gainer H (1990) Structure of mouse vasooressinI and oxytocin g&es. Mol Brain Res’8:3 19-324. oxytocin-like peptides (i.e., peptides with a basic and peptides Heierhorst J, Lederis K, Richter D (1992) Presence of a member of with an aliphatic amino acid residue at position 8, respectively) the Tel-like transposon family from nematodes and Llrosophilu with- results from the occasional mutation of a single gene rather than in the vasotocin gene of a primitive vertebrate, the pacific hagfish from gene duplication, and the mutation of residue 8 from a Eptutretus stouti. Proc Nat1 Acad Sci USA 89:6798-6802. Ivell R, Richter D (1984) Structure and comparison of the oxytocin basic to an aliphatic residue may therefore be neutral with re- and vasopressin genes from rat. Proc Nat1 Acad Sci USA 81:2006- spect to peptide-receptor interactions and peptide functions in 2010. invertebrates. To test this, we synthesized [Ile8]-conopressin (i.e., Kits KS (1980) States of excitability in ovulation hormone producing Lys-conopressin with an isoleucine residue instead of a lysine neuroendocrine cells of Lymnueu stugnulis (Gastropoda) and their residue at position S), and compared its myoactive effect with relation to the egg-laying cycle. J Neurobiol 11:397410. Kits KS, Crest M, Bobeldijk RC, Lodder JC (1991) Glucose-induced that of Lys-conopressin. Isoleucine was substituted because oxy- excitation in molluscan central neurons producing insulin-related pep- tocin-related cephalotocin from Octopus, which is the closest tides. Pflugers Arch 417:597-604. related species to Lymnaea in which such a peptide has been Land H, Grez M, Ruppert S, Schmale H, Rehbein M, Richter D, Schlitz identified, contains isoleucine at position 8. Both peptides have G (1983) Deduced amino acid sequence from the bovine oxytocin- similar potencies on the Lymnaea vas deferens (Fig. 5A) and precursor cDNA. Nature 302:342-344. Lane TF, Sower SA, Kawauchi H (1988) Arginine vasotocin from the interact with the same receptor, since desensitization of the re- of the lamprey (Petromyzon mar&us): isolation and ceptor by Lys-conopressin resulted in a desensitized response to amino acid sequence. Gen Comp Endocrinol 70: 152- 157. [Ilex]-conopressin, but not to an unrelated -FF-amide peptide that Lasham A, Darlison MG (I 993) Direct sequencing of lambda-DNA also acts as an excitatory transmitter on the vas deferens (Li et from crude lysates using an improved linear amplification technique. Mol Cell Probes 7:67-73. al., 1995; Fig. 5B). Thus, the amino acid change that distinguish- Li KW, Geraerts WPM, Van Elk R, Joosse J (1989) Dot immunobind- es vasopressin from oxytocin in mammals, is functionally neu- ing assay of high performance liquid chromatographic fractions on tral in Lymnuea. This may explain why mutation of residue 8 poly(vinylidene difluoride) membrane. J Chromatogr 472:445-446. 5998 Van Kesteren et al. * Conopressin Controls Sexual Behavior in Lymnaea

Li KW, Smit AB, Geraerts WPM (1992) Structural and functional char- volving bovine vasopressin and oxvtocin precursor genes suggested acterization of neuropeptides involved in the control of male mating by nuseotide sequence. Nature 308:554-557. - -- behavior of Lymnae; s&g&is. Peptides 13:633-638. Salzet M. Bulet P. Van Dorsselaer A. Malecha J (1993) Isolation. struc- Li KW. El Filali Z. Van Golen FA. Geraerts WPM (1995) Identification tural characterization and biological function of a lysine-conopressin of a novel amide peptide, GLTPNMNSLFF-NHz,‘involved in the con- in the central nervous system of the Pharyngobdellid leech &pob- trol of vas deferens motility in Lymnaea sragnalis. Eur J Biochem dellu octoculutu. J Biochem 217:897-903. 229110-72. Sanger E Nicklen S, Coulson AR (1977) DNA sequencing with chain- Liu B, Burbach JPH (1987) Characterization of vasopressin and oxy- terminating inhibitors. Proc Nat1 Acad Sci USA 745463-5467. tocin immunoreactivity in the sheep and rat pineal gland: absence of Sausville E. Carnev D, Battev J (1985) The human vasooressin gene vasotocin and detection of a vasopressin-like peptide. Peptides 8:7- is linked to the oxytocin gene and is’ selectively expressed in acul- 11. tured lung cancer cell line. J Biol Chem 260:10236-10241. Maniatis T, Fritsch EG, Sambrook J (1982) Molecular cloning: a lab- Schmale H,“Heinsohn S, Richter D (1983) Structural organization of oratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Lab- the rat gene for the arginine vasopressin-neurophysm precursor. oratory. EMBO J 21763-767. McMaster D, Kobayashi Y, Lederis K (1992) A vasotocin-like peptide Smit AB, Jimtnez CR, Dirks RW, Croll RP, Geraerts WPM (1992) in Anlvsia kurodui ganglia: HPLC and RIA evidence for its identitv Characterization of a cDNA clone encoding multiple copies of the with’ Lys-conopressm G. Peptides 13:4 13-421. neuropeptide APGWamide in the mollusc Lymnuea sfugnulis. J Neu- Morley SD, Schonrock C, Heierhorst J, Figueroa J, Lederis K, Richter rosci 12:1709-1715. D (1990) Vasotocin genes of the teleost fish Catastomus commev- Swaab DE Pool CW (1975) Specificity of oxytocin and vasopressin soni: gene structure, exon-intron boundary, and hormone precursor immunofluorescence. J Endocrinol 66:263-272. organization. Biochemistry 29:2506-251 1. Todd K, Lightman SL (1986) Oxytocin release during coitus in male Murphy MR, Seckl JR, Burton S, Checkley SA, Lightman SL (1987) and female rabbits: effect of opiate receptor blockade with nalaxone. Changes in oxytocin and vasopressin secretion during sexual activity Psychoneuroendocrinology 11:367-371. in men. J Clin Endocrinol Metab 65:738-741. Van Kesteren RE, Smit AB, Dirks RW, De With ND, Geraerts WPM, Joosse J (1992) Evolution of the vasopressin/oxytocin superfamily: Oumi T, Ukena K, Matsushima 0, Ikeda T, Fujita T, Minakata H, Nom- characterization of a cDNA encoding a vasopressin-related precursor, oto K (1994) Annetocin: an oxytocin-related peptide isolated from preproconopressin, from the mollusc, Lymnueu stugnu1i.s. Proc Nat1 the earthworm, Eiseniafoetidu. Biochem Biophys Res Commun 198: Acad Sci USA 89:4593-4597. 393-399. Van Minnen J, Schallig H (1990) Demonstration of insulin-related sub- Proux JP, Miller CA, Li JP, Carney PL, Girardie A, Delaage M, School- stances in the central nervous system of pulmonates and Aplysiu cul- ey DA (1987) Identification of an arginine vasopressin-like diuretic ifornica. Cell Tissue Res 260:381-386. hormone from Locustu migrutoria. Biochem Biophys Res Commun Van Minnen J, Dirks RW, Vreugdenhil E, Van Diepen J (1989) Ex- 149:180-186. pression of the egg-laying hormone genes in peripheral neurons and Ramsay DJ (1983) Posterior pituitary gland. In: Basic and clinical endocrine cells in the reproductive tract of the mollusc Lymnueu stag- endocrinology (Greenspan FS, ed), pp 177-187. Englewood Cliffs, nulis. Neuroscience 33:35-46. NJ: Prentice-Hall. Vreugdenhil E, Jackson JF, Bouwmeester T, Smit AB, Van Minnen J, Reich G (1992) A new peptide of the oxytocin/vasopressin family iso- Van Heerikhuizen H, Klootwijk J, Joosse J (1988) Isolation, char- lated from nerves of the cephalopod Octopus vulgaris. Neurosci Lett acterization and evolutionary aspects of a cDNA clone encoding mul- 134:191-194. tiple neuropeptides involved in the stereotyped egg-laying behavior Ruppert S, Scherer G, Schutz G (1984) Recent gene conversion in- of the freshwater snail Lymnueu stugnulis. J Neurosci 8:4184-4191.