HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by SPIJKER, S. Articles by GERAERTS, W. P. M. Search for Related Content PubMed PubMed Citation Articles by SPIJKER, S. Articles by GERAERTS, W. P. M.

A molluscan peptide -amidating enzyme precursor that generates five distinct enzymes

^a Department of Molecular and Cellular Neurobiology, Graduate School Neurosciences Amsterdam, Research Institute Neurosciences Vrije Universiteit, 1081 HV Amsterdam, The Netherlands; and

^b Department of Neuroscience, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mechanisms underlying the specificity and efficiency of enzymes,which modify peptide messengers, especially with the variable requirements of synthesis in the neuronal secretory pathway, are poorly understood. Here, we examine the process of peptide -amidation in individually identifiable Lymnaea neurons that synthesize multiple proproteins, yielding complex mixtures of structurally diverse peptide substrates. The -amidation of these peptide substrates is efficiently controlled by a multifunctional Lymnaea peptidyl glycine -amidating monooxygenase (LPAM), which contains four different copies of the rate-limiting Lymnaea peptidyl glycine -hydroxylating monooxygenase (LPHM) and a single Lymnaea peptidyl -hydroxyglycine -amidating lyase. Endogenously, this zymogen is converted to yield a mixture of monofunctional isoenzymes. In vitro, each LPHM displays a unique combination of substrate affinity and reaction velocity, depending on the penultimate residue of the substrate. This suggests that the different isoenzymes are generated in order to efficiently amidate the many peptide substrates that are present in molluscan neurons. The cellular expression of the LPAM gene is restricted to neurons that synthesize amidated peptides, which underscores the critical importance of regulation of peptide -amidation.—Spijker, S., Smit, A. B., Eipper, B. A., Malik, A., Mains, R. E., Geraerts, W. P M. A molluscan peptide -amidating enzyme precursor that generates five distinct enzymes.

Key Words: posttranslational modification • neuropeptide -amidation • PAM • mono-oxygenase • mollusk Lymnaea stagnalis

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
NEURONS INTEGRATE THE INFORMATION they receive, and many translate this into the release of multiple peptides and other transmitters for signal transmission to other cells ⁽¹⁾ . The released set of messengers can be viewed as a code to which a message-specific informational content is assigned by the presence of diverse molecules that occur in a specific ratio and are released in a specific spatio-temporal pattern ⁽²⁾ . In particular, the diversity of bioactive peptides contributes to the specific informational content of the message. It has been well established that processes such as regulated gene expression, alternative RNA splicing, as well as differential processing and targeting of peptides are important in generating neuron-specific sets of peptides.

In this study, we examine the process of peptide -amidation, a posttranslational modification that is critical for generating bioactive peptides. Peptide -amidation protects peptides from degradative enzymes and is often required for high-affinity interaction with cognate receptors ⁽³⁾ . It is a final step in a series of enzymatic events in which neuropeptides mature from proproteins within the endoplasmic reticulum, trans-Golgi network (TGN),¹ and secretory granules. It involves the sequential conversion of a peptidylglycine substrate into peptidyl--hydroxyglycine and conversion of the latter into an amidated peptide (⁴ , ⁵ ). Of the two reactions, the first, catalyzed by peptidylglycine -hydroxylating monooxygenase (PHM), is rate limiting and dependent on the penultimate position of the substrate (Xxx-Gly) (³ , ⁶ , ⁷ ). The second reaction is catalyzed by peptidyl--hydroxyglycine -amidating lyase (PAL). In vertebrates, PHM and PAL are synthesized together as the bifunctional protein peptidylglycine -amidating monooxygenase (PAM) ^8-10) , whereas in Drosophila ⁽¹¹⁾ and Calliactis ⁽¹²⁾ PHM is encoded by a distinct gene.

As a model in our study, we chose the central nervous system (CNS) of the mollusk Lymnaea stagnalis, because molluscan neurons often coexpress multiple proproteins yielding different amidated peptides that are colocalized in a single subset of large dense-core vesicles ⁽¹³⁾ . Moreover, a molluscan proprotein often yields peptides having different carboxy-terminal amidated residues (e.g., Ile-NH₂, Val-NH₂, Leu-NH₂/Met-NH₂) (A. B. Smit, unpublished observations; 14). We reasoned that the cocktails consisting of many different peptide substrates produced by molluscan neurons may place high demands on the performance of the peptide -amidating enzyme in the molluscan brain.

Here, we examined the specific structural characteristics of the molluscan -amidating enzyme from the Lymnaea CNS (LPAM) and performed a functional analysis of the enzyme(s) toward different Xxx-Gly substrates. Using the advantage offered by the Lymnaea CNS—namely, of large neurons that are identifiable according to shape, function, and chemical characteristics—we examined the cellular coexpression of LPAM and neuropeptides by single-cell reverse transcriptase polymerase chain reaction (RT-PCR) on cDNA generated from individual neurons of the Lymnaea CNS.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Animals and peptides
Adult specimens of L. stagnalis (shell height 28–34 mm), bred in the laboratory under standard conditions, were used. -N-Ac-Tyr-Gly-Gly was obtained from the Dutch Cancer Institute (NKI, The Netherlands). -N-Ac-Tyr-Phe-Gly and -N-Ac-Tyr-Val-Gly were synthesized as described by Husten and Eipper ⁽¹⁵⁾ . The peptidyl--hydroxyglycine substrate was prepared as described by Eipper et al. ⁽¹⁶⁾ .

Protein extraction
Total brain extracts of 100 snails were made in 1 ml TM [20 mM NaTES, 10 mM mannitol, and the inhibitor mix (final concentrations): 5 µg/ml aprotinin, 50 µg/ml lima bean trypsin inhibitor, 2 µg/ml pepstatin, 16 µg/ml benzamide-HCl, 2 µg/ml leupeptin, 0.3 mg/ml phenyl methyl sulfonyl fluoride, and 0.6 mg/ml Pefabloc]. Debris was removed by a 5 min centrifugation at 1000 x g. The soluble fraction was frozen, thawed, and separated from the particulate fraction by 15 min centrifugation at 435,000 x g. The membrane pellet was resuspended in TMT (TM with 1% Triton X-100). All solutions contained protease inhibitors as above. Linear gradients were 5–20% sucrose (1.8 ml) in TMT containing 150 mM NaCl. Aggregated material in the samples (0.117 ml soluble mixed with 0.133 TMT or 0.25 ml particulate) was pelleted at 8500 x g for 5 min. Supernatant was loaded on the gradient and centrifuged for 3 h at 200,000 x g in a swinging bucket rotor at 4°C. In total, 13 0.15 ml fractions ⁽¹⁷⁾ were collected from the top. Protein standards were analyzed simultaneously. Aliquots of these fractions were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis to locate the marker proteins.

Cloning of LPAM
Poly(A)⁺ RNA was isolated from dissected CNS of L. stagnalis using magnetic beads (Dynal A.S., Oslo, Norway) and reverse transcribed into hexanucleotide primed cDNA. Two degenerate oligonucleotides (Isogen Bioscience, Maarssen, The Netherlands) directed to sequences that are conserved among the vertebrate PHM enzymes, namely, Ala-Trp-Ala-Arg-Asn-Ala-Pro-Pro (sense, OL1: 5'-CTCAAGCTTCITG(T/C)GCI(A/C)GNAA(C/T)GCNCCNCC-3') and Pro-Gln-Ala-Phe-Tyr-Pro-Val-Glu (antisense, OL2: 5'-CGGGAATTC(T/C)TCIACIGG(A/G)TA(A/G)AANGC (T/C)TGNGG-3'), were used to amplify PHM-like sequences. PCR was performed on one animal equivalent of CNS cDNA using 150 pmol of OL1 and OL2 under standard conditions for 45 cycles (94°C, 20 s; 53°C, 30 s; 72°C, 1 min). Amplified cDNA of the expected size was cloned and sequenced. The cloned LPAM PCR product was used as a random primed [-³²P]dATP-labeled probe (specific activity >10⁹ cpm/µg) to screen approximately 200,000 clones of an amplified ZAP II cDNA library, as described previously ⁽¹⁸⁾ . To obtain the partial cDNA sequence of LPAM-2, total RNA isolated from the CNS was treated with DNase-I to remove traces of genomic DNA and reverse transcribed into hexanucleotide primed cDNA. Amplification of cDNA was performed in a 100 µl PCR in triplicate using a primer combination (sense, PAM S8: 5'-CGCGGATCCACTACCCCCCAACCCTG-3'; antisense, LPAL AS1: 5'-CGCGGATCCACGCCCCAA TCTGAC-3') that covers the noncatalytic exon A domain. Three ~140 bp products obtained were cloned into M13mp19 and sequenced.

Subcloning and nucleotide sequence analysis
The insert of the pBluescript II PAM cDNA was excised by digestion with Bam HI and further digested with either Mbo I, Rsa I, Hpa II, or Taq I. The fragments obtained were randomly cloned into a Bam HI-, Sma I-, or Acc I-cut M13mp18, respectively. LPAM PCR products, LPAM cDNA fragments subcloned in M13mp18, pBluescript II LPAM cDNA, and LPAM constructs for expression (see below) were sequenced in both orientations according to the dideoxy chain termination method ⁽¹⁹⁾ , by using T7 DNA polymerase, universal primers, and internal primers continued by primer walking. Sequence alignments were performed using the program Clustal W1.5.

5'-Rapid amplification of cDNA ends (5'-RACE)
Poly(A)⁺ RNA was isolated from the CNS using magnetic beads. 5'-RACE was carried out as described by Frohman et al. ⁽²⁰⁾ , with the following modifications: 0.5 µg mRNA, 400 pmol hexanucleotide primers in 30.5 µl water were heated at 90°C for 1 min, quenched on ice, 10 U of RNAsin (Promega, Madison, Wis.) was added, and reverse transcription was performed using Superscript II and the manufacturers protocol (Gibco BRL, Grand Island, N.Y.). After incubation, the cDNA was size-separated on a Pasteur pipette Sephadex G-50 column to remove nucleotides and hexanucleotide primers. Fractions of 150 µl were collected and fractions 3 to 7 were pooled, precipitated and resuspended in 5 µl water. For dATP tailing, the cDNA was heated at 90°C for 2 min and snap-cooled, then terminal deoxynucleotidyl-transferase components were added as described by the manufacturer (Boehringer Mannheim, Mannheim, Germany). The mixture was incubated at 37°C for 30 min and heat-inactivated at 65°C for 2 min, precipitated, and dissolved in 10 µl water. Aliquots (3 µl) were used for a first round of amplification with Super Taq polymerase (Boehringer Mannheim, Germany), by using 50 pmol PAM-specific antisense primer in combination with 50 pmol RoRi-(dT₁₇) [5'-ATCGATGGTCGACGCATGCGGATCCAAAGCTTGAATTCGAGCTC(T)₁₇-3'] or with ADA-(dT₂₀) [5'-GACTCGAGTCGACATCGA(T)₂₀-3'] for 10 cycles (94°C, 30 s; 34°C (RoRi) or 40°C (ADA), 45 s; 72°C, 2 min) in a volume of 50 µl. Then, 2% of this PCR reaction mixture was reamplified by using 50 pmol of the first set of adaptor primers Ro (5'-ATCGATGGTCGACGCATGCGGATCC-3') or ADA-I (5'-CGCTCTAGAGACTCGAGTCGACATCGA-3') in combination with 50 pmol PAM-specific antisense primer in a volume of 100 µl. We used Ro for 40 cycles (94°C, 30 s; 63°C, 45 s; 72°C, 2 min) or ADA-I for 5 cycles (94°C, 30 s; 56°C, 45 s; 72°C, 2 min), followed by 35 cycles (94°C, 30 s; 63°C, 45 s; 72°C, 2 min). To increase specificity, a second round of amplification was carried out on 1% of the previous PCR reaction mixture by using 50 pmol of a nested PAM-specific antisense primer in combination with 50 pmol nested adaptor primers Ri (5'-GGATCCAAAGCTTGAATTCGAGCTCT-3') or ADA-II (5'-CGCGAGCTCGAGTCGACATCGATTT-3') under conditions identical to those described above. Three sets of nested antisense PAM primers were used, OL4 (5'-CTTTTTATGAGTTGTATGGATCATG-3') and OL5 (5'-AGAGGATCCGATCTGGAGCTTCTGAAATCTTC-3'); OL6(5'-CACTGACTGGATGTGTCATAGAA-3') and OL7 (5'-GTGGGATCCGTCGATCGAGCAGCCAGTTTG-3'); OL8 (5'-GCACAAGAGTTTTAATGGTGGTC-3') and OL9 (5'-GCGAAGCTTCACCTTCTGGGAGAACTGTG-3'). Amplified cDNA was digested with BamHI and SacI or with Bam HI and HindIII, and separated on an agarose gel. In each cycle of 5'-RACE, both the RoRi set as well as the ADA-set were used in duplicate; at least three fragments were cloned from each individual PCR, both in M13mp18 and mp19, and sequenced in order to avoid the introduction of sequence errors arising from individual PCRs.

Size determination of LPAM mRNA
Total RNA from the CNS was isolated ⁽²¹⁾ , and two CNS equivalent to RNA (~20 µg) were fractionated on a 0.8% agarose-formamide gel and transferred to charged nylon membranes. The inserts of the LPHM-1, -2, -3, -4 and LPAL pCI.neo-plasmids (see below) and a subclone containing the exon A domain (bp 4221-4531) of the LPAM cDNA were used to make individual random primed [-³²P]dATP-labeled probes (specific activity >10⁹ cpm/µg). Only the exon A domain probe or a mixture of equal activities of each LPHM and the LPAL probe were used for hybridization (conditions as described above).

Single-cell RT-PCR
Individual neurons were isolated using a fine glass pipette (tip diameter, ~100 µm), washed in culture medium, and transferred to a glass plate. After removal of superfluous medium, 11 µl water containing 200 pmol hexanucleotides was added, mixed, and transferred to a tube; the sample was heated at 90°C for 1 min, quenched on ice, 10 U of RNAsin was added, and reverse transcription was performed using M-MLV reverse transcriptase and the manufacturers protocol (Promega) in a total volume of 25 µl. Amplification of cDNA (1 µl) was performed in a 100 µl PCR using LPAM- or neuropeptide-specific primers. LPAM primers (sense, OL10: 5'-TGTCTGTGTGCACCCAGCC-3'; antisense, PALC2: 5'-GCG-GATCCTGAAGGTGTGAAATCACTGC-3') or primersspecific for the presence (sense, Ex-ASI: 5'-GGAAACATCACAGAGAGAATC-3') or absence (sense, LPHM4/PALSI: 5'-GGGATGAAAGCAGCATCCA-3') of the exon A domain were used in combination with the LPAL primer (antisense, LPALAS1) as well as primers specific for neuropeptide transcripts for MIP-I, LYCP, ELH-I (CDCH-I), SCP, CP, and APGWa (data not shown). After 35 cycles of amplification, 10 µl of each PCR reaction was separated on an agarose gel and Southern blotted. After hybridization to individual random primed [-³²P]dATP-labeled probes (specific activity >10⁹ cpm/µg), the filters were washed and autoradiographed. Autoradiograms were scanned by using a flatbed scanner.

Phylogenetic analysis
The various PHM, rat dopamine ß-monooxygenase and Drosophila tyramine ß-hydroxylase amino acid sequences, analogous to rat PHM Ile⁶⁶ to Pro³⁴⁷, i.e., from the first identical amino acid residue between LPHM and rat PHM to the end of the PHM catalytic core, were aligned by using the program Clustal W1.5. Bootstrap analysis was then performed to generate 100 resampled data sets (with a 10-fold jumble of input order) using Seqboot. Protdist was used to compute distance between the protein sequences using the Dayhoff PAM001 matrix, followed by Fitch to calculate a distance matrix. The Consense program generated a consensus tree. Confidence limits were calculated by using the data obtained from bootstrapping. Programs used were made and distributed by Felsenstein ⁽²²⁾ .

Plasmid construction for functional expression
pCI.neo-plasmids encoding the separate domains of LPHM-1, -2, -3, -4 and LPAL were created using amplification of cDNA by PCR. To ensure correct synthesis, routing, and secretion, each enzyme domain was preceded by the consensus Kozak sequence ⁽²³⁾ and rat PAM signal and prosequence, followed at the carboxyl terminus by a rhodopsin epitope tag (GSGATETSQVAPA; 24) to visualize expression, and a stop codon. Constructs contained the following sequences: LPHM-1, Thr²²-Pro³⁴⁴; LPHM-2, Ile³⁵³-Ser⁶⁷⁵; LPHM-3, Thr⁶⁹⁰-Gly¹⁰¹³; LPHM-4, Gly¹⁰¹⁹-His¹³³³ LPAL, Phe¹⁴⁹⁶-Ser¹⁸⁶⁵ (numbers refer to LPAM-1). Specific primers for each catalytic domain were used in PCR amplification. PCR products were cloned in-frame into a pBluescript vector containing the rhodopsin epitope tag (pBS.rhodopsin). The Kozak/signal sequence/pro sequence was amplified from pBS.rPHMs ⁽²³⁾ using a specific primer, Koz-proA1 (antisense, 5'-GCGGAGCTCAAATGATCTGGTAGTTTCTTTAAAC-3', containing a SacI site at the 5'-end) in combination with a universal pBS primer and cloned into pBluescript II KS (Stratagene) to yield pBS.rKoz/ss/pro. cDNA regions derived from PCR were verified by sequencing. Finally, fragments of pBS.rhodopsin-LPHM/LPAL cDNA and the fragment of pBS.rKoz/ss/pro were ligated together in the pCI.neo-expression vector (Promega).

Cell culture and transfection
hEK-293 cell lines were maintained as described ⁽²⁵⁾ . pCI.neo-plasmids containing LPHM-1, -2, -3, and -4 and LPAL were transfected into hEK-293 cells using lipofectin and G-418 selection ⁽²⁵⁾ and screened by PHM or PAL assay ⁽¹⁰⁾ , Western blot, and immunostaining with a mouse anti-rhodopsin antibody ⁽²⁴⁾ , as described ⁽²³⁾ . Clones secreting the highest amount of PHM or PAL activity were selected for collection of the medium.

PHM and PAL assays
Aliquots of fractions collected from the sucrose gradients (see section on protein extraction) and aliquots of spent medium were assayed as described ⁽¹⁰⁾ by using 0.5 µM substrate and mono-¹²⁵I--N-acetyl-Tyr-Val-Gly (PHM) or mono-¹²⁵I--N-acetyl-Tyr-Val--hydroxyglycine. In addition, mono-¹²⁵I--N-acetyl-Tyr-Phe-Gly and mono-¹²⁵I--N-acetyl-Tyr-Gly-Gly (PHM) were used in kinetic studies. Assays were performed in duplicate or triplicate.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Endogenous PHM and PAL activities are present as single enzyme entities in Lymnaea central neurons
As a first step toward structural and functional identification of LPAM, we investigated the presence of endogenous PHM and PAL enzyme activities in the Lymnaea CNS. We first prepared crude soluble and particulate fractions from homogenates of Lymnaea CNS. Assays on the soluble fraction and proteins solubilized from the crude particulate fraction indicated that both PHM and PAL activity were present in each fraction. The soluble and solubilized crude particulate fractions were then separated on a sucrose gradient to estimate the molecular weight of both enzymes. Fractions were collected and assayed for PHM and PAL activities (Fig. 1 ). Both in soluble and particulate fractions, PHM and PAL activities were present as single enzyme entities (~45 kDa). The PHM activity in the particulate fraction is probably due to noncovalent attachment to the secretory granule membranes, a phenomenon that is also observed for soluble rat PHM ⁽²⁶⁾ and for many soluble intravesicular proteins (²⁷ , ²⁸ ). Since a precursor molecule containing PHM and PAL could not be identified, both enzymes may be generated from separate precursors, as in Drosophila ⁽¹¹⁾ and Calliactis ⁽¹²⁾ , or from a single precursor by endoproteolysis.

View larger version (36K): [in this window] [in a new window]   Figure 1. Endogenous LPHM and LPAL activities in the Lymnaea CNS. Fractions of a sucrose gradient of the soluble (A) and solubilized crude particulate (B) fractions of a CNS protein homogenate were tested for PHM and PAL activities, as described. The conversion rate is depicted as pmol/µl/h (left Y-axis). Protein molecular size standards (as indicated) were analyzed on a similar gradient and are depicted as relative density (right Y-axis).

The LPAM-1 cDNA sequence predicts a bifunctional peptide -amidating enzyme with four PHM domains and a single PAL domain
To identify the structural organization of the LPAM precursor, PCR using degenerate primers directed to conserved regions of vertebrate PHM ⁽⁵⁾ was performed on cDNA from the Lymnaea brain. A PCR product of the expected size (360 bp) (Fig. 2A ) encodes a protein fragment with 38% amino acid sequence identity to rat PHM. To identify the full sequence of the LPAM cDNA, we used the PCR fragment to screen 200,000 clones of a cDNA library from the cerebral ganglia of the Lymnaea brain ⁽¹⁸⁾ , yielding one positive clone of ~5 kb. Sequencing of the 5'-end of this clone, named LPAM clone 2.11, revealed that the longest open reading frame contains a partial PHM amino acid sequence, but no start codon. It is flanked by a 603 bp 3' untranslated region. A short poly(A) stretch is present at the 3' end of the cDNA clone, but a consensus sequence for polyadenylation is absent. To obtain the 5' part of the cDNA encoding the start codon, three cycles of 5' rapid amplification of cDNA ends (5'-RACE) were required ⁽²⁰⁾ . The assembled sequences (Fig. 2A ) gave rise to a cDNA of 6615 nucleotides (LPAM-1). The open reading frame consists of 5853 nucleotides and encodes a protein of 1951 amino acids with a predicted mass of 215.6 kDa. The coding region is flanked by a 156 bp 5' untranslated sequence containing several in-frame stop codons.

View larger version (38K): [in this window] [in a new window]   Figure 2. The sequence of LPAM-1 cDNA predicts a novel bifunctional peptide -amidating enzyme. A) Strategy used to obtain the sequence of the LPAM-1 cDNA. In addition to cloning and sequencing LPAM clone 2.11, isolated from the cerebral ganglia-specific cDNA library, three cycles of 5'-RACE yielded several overlapping clones from independent PCRs. Independent sequence analyses are depicted by horizontal arrows. The open reading frame (ORF) and the 5' and 3' untranslated regions (5' UTR and 3' UTR, respectively) are indicated. The location of oligonucleotides OL1 and OL2, used in the initial PCR, are indicated. B) Comparison of the predicted LPAM-1 and rat PAM-1 proteins. In rat PAM-1, the signal sequence (ss; black) is followed by a proregion (pro; dotted), which is absent in LPAM-1. The percentage sequence identity of the various domains in LPAM-1 with counterparts in rat PAM-1 are indicated. The scissors represent paired basic amino acid sequences, which are potential sites for endoproteolysis. Potential N-linked glycosylation sites are indicated by ellipses; Asn26, Asn67, Asn220, Asn396, Asn512, Asn551, Asn887, Asn1063, Asn1217, Asn1503, Asn1625, Asn1802, Asn1848). The location of oligonucleotide primers used in the RT-PCR are indicated by arrows above the figure (D). Transmembrane domain, TM; cytoplasmic domain, CD. C) Northern blot analysis. Various LPAM probes (see Materials and Methods) were hybridized to total RNA (20 µg) from the Lymnaea CNS. Two transcripts (7.0 kb: LPAM-1, and 6.5 kb: LPAM-2) were detected. Size markers (kb) are indicated. D) RT-PCR on random primed cDNA generated from DNase-treated total RNA, isolated from Lymnaea CNS, using primers spanning the exon A domain (indicated in panel B; see text). Cloning and sequencing of the 140 bp product revealed that it is identical to the LPAM-1 sequence, lacking a 465 bp sequence (LPAM-1 bp 4169-4633, underlined). The junctions are indicated by arrows. E) Northern blot analysis as in panel C, now using the exon A domain as a probe, detects only the 7.0 kb (LPAM-1) transcript. Size markers (kb) are indicated.

The predicted LPAM-1 protein is a putative multifunctional peptide -amidating enzyme. Surprisingly, it is organized differently from vertebrate PAM (Fig. 2B ). LPAM-1 contains a signal peptide, which is predicted to be cleaved after residue Ala^{21 (29)} . An adjacent propeptide, present in rat PAM-1, is absent in LPAM-1. In contrast to rat PAM-1, the LPAM-1 protein contains four distinct putative monooxygenase domains, LPHM-1 to -4, each of which exhibits 37–40% sequence identity to vertebrate PHM. In addition, a domain with 36% sequence identity to vertebrate PAL is present. This putative Lymnaea PAL domain (LPAL) is followed by a predicted transmembrane domain and a cytoplasmic domain with 39% sequence identity to the cytoplasmic domain of rat PAM-1. The LPAL domain is separated from the four LPHM domains by a putative exon A domain, which shows no significant homology to the corresponding region of rat PAM-1. A single consensus site for N-linked glycosylation is present in rat PAM-1, whereas one to four consensus sites for N-linked glycosylation are present in each LPHM domain and in the LPAL domain (Fig. 2B ).

Rat PAM-2 and PAM-3 arise via alternative splicing in the pre-PAM RNA of small exons, representing either the exon A domain or the transmembrane domain and/or part of the cytoplasmic domain (⁵ , ³⁰ , ³¹ ). Rat PAM-4 arises via alternative use of an alternative polyadenylation addition site, generating a soluble monofunctional PHM protein. To explore the possibility that the LPAM gene gives rise to different splice variants, we first performed Northern blot analysis of RNA from the CNS (Fig. 2C ). LPAM transcripts of ~7.0 kb and ~6.5 kb were detected. Taking into account the length of the LPAM-1 cDNA and the absence of a polyadenylation site, we concluded that the ~7.0 kb transcript represents the LPAM-1 cDNA. To identify the putatively alternatively spliced region in the 6.5 kb transcript (LPAM-2, see below), RT-PCR was performed on CNS cDNA by using the primer combinations indicated in Fig. 2B . Only the primer combination that covers the noncatalytic exon A domain yielded two products, a prominent band of ~140 bp in addition to the 630 bp product predicted by the LPAM-1 transcript (Fig. 2D ). Cloning and sequencing of the 140 bp PCR product revealed that it contains the LPAM-1 cDNA sequence, but lacks the exon A domain (bp 4169-4633). The exon A domain is not a retained intron, since it does not disrupt the open reading frame and exon/intron consensus sequences are absent at the boundaries (Fig. 2D ). To determine whether the 6.5 kb LPAM-2 transcript lacks the sequence of the exon A domain, Northern blot analysis was performed with RNA prepared from Lymnaea CNS, by using the exon A domain as a probe. Only the 7.0 kb LPAM-1 transcript was detected (Fig. 2E ). Together, these results strongly suggest that LPAM-1 and LPAM-2 are generated by alternative splicing of exon A. Further transcript diversity is absent.

LPAM expression is restricted to neurons that express amidated peptides
To address the question whether LPAM is strictly coexpressed with amidated peptides, we performed single-cell RT-PCR on cDNA, generated from identifiable neurons of the Lymnaea CNS. Neurons were selected for which the complement of peptides synthesized is known from previous studies, by using mass spectrometric methods applied to single neurons ^32-34) . We selected three cell types that are representative of neurons expressing amidated peptides: 1) the caudo-dorsal cells ⁽³²⁾ ; 2) the visceral-dorsal neuron 1, ⁽³⁴⁾ ; and 3) neurons from the right anterior lobe cluster ^35-38) . As examples of neurons that exclusively produce no -amidated peptides, only two cell types could be selected because most identifiable neurons in the Lymnaea CNS produce amidated peptides. These cell types are light green cells ⁽³⁹⁾ and the light yellow cells ⁽³³⁾ .

For each cell type, at least three neurons were isolated and cDNA was synthesized from each individual cell. PCR reactions on this cDNA were performed for the neuropeptide transcripts and for LPAM. We found that LPAM expression was restricted to neurons that express amidated peptides (Fig. 3A ), indicating that peptide -amidation gene expression is indeed tightly regulated.

View larger version (71K): [in this window] [in a new window]   Figure 3. Neurons expressing genes that encode amidated peptides express the LPAM gene. A) RT-PCR using neuropeptide transcript-specific primers and LPAM-specific primers (which do not discriminate between LPAM-1 and LPAM-2 transcripts; see panel C) on cDNA from individual neurons of five different cell types. For each cell type, three to five neurons were individually isolated and cDNA was synthesized from each individual neuron; PCR reactions were performed in duplicate for the neuropeptide transcripts and in quadruplicate for the LPAM transcripts. Specific probes were used to monitor PCR products formed in each PCR reaction by Southern blot analysis. PCR products obtained from the cDNA synthesis of each neuron (numbered lanes show the PCR product formed by using an individual neuron as template) are shown. Notice that only neurons expressing genes that encode amidated peptides express the LPAM gene. Abbreviations for neuropeptide transcripts: MIP (molluscan insulin-related peptide), LYCP (light yellow cell peptide), ELH (egg-laying hormone), SCP (small cardio-active peptide), CP (conopressin). Abbreviations for neurons: LGC (light green cells), LYC (light yellow cells), CDC (caudo-dorsal cells), VD1 (visceral-dorsal neuron 1), RAL (right anterior lobe neuron). B) RT-PCR on cDNA from neurons, as used in panel A, expressing genes encoding amidated peptides, using LPAM-1 specific primers (C, upper panel), or LPAM-2 specific primers (C, lower panel). C) Schematic representations of LPAM-1 and -2 to indicate the primers used in panels A and B.

To examine the cellular localization of the LPAM-1 and LPAM-2 transcripts in more detail, RT-PCR was performed on the same cDNA from neurons that were shown to express the LPAM gene (see above). Primer sets specific for the presence of the exon A domain (LPAM-1) and specific for the absence of exon A (LPAM-2) were used. PCR products representing the LPAM-1 transcript were detected only in a subset of neurons of the right anterior lobe cluster (Fig. 3B , upper panel), whereas the PCR product representing the LPAM-2 transcript is present in all neurons analyzed (Fig. 3B , lower panel).

LPAM-1 and LPAM-2 proteins are structurally multifunctional PAM zymogens
LPHM and LPAL characteristics
The interspecies sequence identity between LPHM-1 to LPHM-4 and rat PHM varies between 37% and 40% (Fig. 2B and Fig. 4A ). Identical amino acid residues and conserved substitutions are restricted to a number of small regions, which form the core of the enzyme and are essential for catalytic activity ⁽⁴⁰⁾ . The eight conserved Cys residues, which form four intrachain disulfide bonds—two in the amino-terminal and two in the carboxy-terminal domain of the catalytic core of rat PHM ⁽⁴¹⁾ —are conserved in the other type 2 copper monooxygenase, dopamine ß-monooxygenase (EC 1.14.17.1; DßM), as well as in all LPHM domains. In addition, the five essential His residues (rat PHM residues 107, 108, 172, 242, and 244) and Met³¹⁴, which are involved in copper binding, are present in all four LPHM domains (²³ , ⁴⁰ , ⁴² ). Arg²⁴⁰, Asn³¹⁶, and Tyr³¹⁸, which are involved in substrate binding, and Tyr⁷⁹, whose mutation affects the K_m for the peptidylglycine substrate, are also conserved.

View larger version (77K): [in this window] [in a new window]   Figure 4. Sequence alignments of LPAM with rat PAM. The number of amino acid residues, starting at Met1, is indicated at the end of each line. Residues conserved in all sequences are boxed, whereas conserved substitutions are indicated with asterisks below the sequence. A) Alignment of rat PHM and the four LPHM domains. Conserved Cys residues are indicated by open triangles. Residues important for catalytic activity, substrate binding, or copper binding are indicated by large hatched arrows. In rat (thin open arrow) and in Lymnaea (filled arrow), the (demonstrated or predicted) site of signal sequence cleavage is indicated; the two cleavage sites (filled arrowheads) within the rat proregion are also indicated. The protease-accessible region (exon 15, Met360-Gln392) is overlined. B) Alignment of rat PAL and LPAL domains and their respective transmembrane (TM; overlined) and cytoplasmic (CD, horizontal arrow) domains. Rat amino acid residues Tyr936 (filled ellipse) and potentially phosphorylated Ser937 (P in ellipse) are indicated. Potential sites for endoproteolysis are overlined (Arg779-Lys780, thin; Lys821-Lys822, boldface). Conserved Cys (open triangle) and His (open arrow) residues are indicated. C) Alignment of the LPAM-1 and rat PAM-1 exon A domains. Potential sites for endoproteolysis are overlined. D) Phylogenetic relationships of members of the type 2 copper monooxygenase family: Lymnaea PHM, Drosophila (DPHM, Genbank accession number U77426), rat (rPHM, M25732), human (hPHM, M33721), bovine (bPHM, M18683) and Xenopus [AE-I (XPHM-I, M18134) and AE-II (XPHM-II, M19032)] PHM domains, rat DßM (rDßM, L12407), and Drosophila TßH (DTßH, Z70316) domains were determined. The tree, constructed from 100 bootstrapped data sets (see Materials and Methods), is shown. The length of the branches represents protein distances. The numbers at the forks are a measure of variation of the branch and are related to a confidence limit for the positions of the branches (100 is no variation, 100% confidence). This is an unrooted tree. E) Alignment of the four LPHM domains. The predicted site of signal sequence cleavage (filled arrow) is indicated.

The amino acid sequence identity in the LPAL domain (Fig. 4B ) is also restricted to short stretches of residues, and four Cys and six His residues are conserved. PAL is a metalloenzyme, requiring divalent ions for optimal activity ⁽¹⁶⁾ . As in PHM, the conserved His residues in PAL may be involved in metal ion binding and the conserved Cys residues may be important for protein folding (²³ , ⁴¹ , ⁴² ).

The cytoplasmic domains of LPAM and rat PAM exhibit 39% sequence identity, indicating the presence of functionally important residues in this domain. In the cytoplasmic domain of rat PAM, residues Ser⁹³⁷ and Tyr⁹³⁶ are crucial for intracellular routing of PAM and for internalization of integral PAM protein after exocytosis (²⁵ , ⁴³ ). In LPAM-1, Ser¹⁹¹⁵ corresponds to rat PAM Ser⁹³⁷, but Tyr⁹³⁶, critical to internalization, is replaced by an Asn (Fig. 4C ). The cytoplasmic domain of LPAM is shorter by 18 residues, indicating the lack of an exon equivalent to rat exon B_b (⁵ , ³⁰ , ³¹ ). As in rat PAM, LPHM has a high percentage of acidic and charged residues near its carboxyl terminus, including several putative casein kinase II phosphorylation sites. The Pro-Glu-Ser-Thr sequence at the carboxy-terminal of rat PAM-1 is only weakly represented in LPAM.

Phylogenetic relatedness of PHM domains
We determined the phylogenetic relatedness of the LPHM domains to other type 2 copper monooxygenases, including Drosophila and vertebrate PHM, rat DßM, and Drosophila tyramine ß-hydroxylase (TßH), using a protein distance matrix to compute the phylogenetic distance between the protein sequences (Fig. 4D ). It demonstrates that the LPHM domains and DPHM are related to the PHM proteins of vertebrates. Bootstrap analysis of the tree and the data set shows that the assignment of the LPHM domains to the vertebrate PHM class is supported with 100% confidence. The extensive branch lengths of the four LPHM domains indicate that after multiplication of the LPHM domains, considerable structural diversification occurred, as delineated in Fig. 4E . The shared sequence identity of the PHM domains in Lymnaea (55%) is even lower than that of the PHM domains of the various vertebrate species (63%; human, bovine, rat, and Xenopus; 5).

The four LPHM domains and the LPAL domain display enzyme activity
The high sequence divergence among the LPHM domains might reflect random mutations and consequent loss of function or, alternatively, mutations that allow diversification of the catalytic properties of each LPHM domain. Because the endogenous LPHM and LPAL forms are single enzyme entities, we could address this issue and evaluate the four LPHM domains and the LPAL domain for their individual activity.

We made expression constructs of each putative domain of LPAM and created five sets of hEK-293 cell lines, each set stably expressing one of the four LPHM domains or the LPAL domain. Spent medium from each cell line was assayed for protein production by Western blot analysis (data not shown) and for hydroxylating (PHM) or peptide -amidating lyase (PAL) activity by using the synthetic tripeptide substrates -N-Ac-Tyr-Val-Gly or -N-Ac-Tyr-Val--hydroxyglycine, respectively ⁽¹⁰⁾ . The pH dependence of the reactions as well as the copper and ascorbate dependence of the LPHM domains were determined. Spent medium from cells expressing recombinant rat PHM and PAL served as controls. All four LPHM proteins and LPAL were active (Fig. 5 ). Because all four LPHM domains behaved similarly with respect to pH, copper, and ascorbate, only the results obtained with LPHM-3 are shown. As with rat PHM ⁽⁴⁴⁾ , the LPHM domain showed maximal activity at acidic pH; activity at pH 5 was about threefold higher than at pH 7 (Fig. 5A ). The pH range of LPAL was even broader than that of rat PAL ⁽¹⁶⁾ . As rat PHM ⁽²³⁾ , each LPHM domain showed a similar dependence on ascorbate (Fig. 5B ), with optimal activity observed with 0.5–1 mM ascorbate. Also, each LPHM domain showed dependence on copper (Fig. 5C ), with optimal activity observed with 0.5–2 µM CuSO₄. As shown for rat PHM ⁽⁴⁵⁾ , addition of diethyl-dithiocarbamate, a copper chelator, abolished nearly all LPHM activity, showing that the LPHM proteins require copper for catalytic activity. Thus, despite sequence divergence, each LPHM domain displays catalytic activity and shows pH and cofactor requirements similar to rat PHM.

View larger version (87K): [in this window] [in a new window]   Figure 5. Comparison of LPHM and LPAL activity to rat PHM and rat PAL activity. Spent media (LPHM-3, rat PHM, LPAL and rat PAL) were assayed as described (see Materials and Methods). Results are plotted as percentage of maximal activity (100%). LPHM-1, -2, and -4 displayed activity similar to LPHM-3. A) Assay pH was varied using 100 mM Na+ MES buffer of the indicated pH. B) Ascorbate concentrations were varied in the presence of 1.0 µM CuSO4 and 0.1 mg/ml catalase in 100 mM Na MES buffer, pH 5.0. C) Copper concentration was varied by adding 0 or 1 µM CuSO4, or 3 µM diethyl-dithiocarbamate (DDC). Numbers are means for assays performed in duplicate.

The LPHM enzymes display distinct substrate specificities
Rat PHM hydroxylates peptidylglycine substrates with different amino acid residues at the penultimate position, with different affinities and at different rates (³ , ⁶ , ⁷ , ²³ ). Therefore, we examined the possibility that each LPHM domain might show a distinctive pattern of kinetics toward different synthetic tripeptide substrates. Three substrates were chosen, for which rat PHM displays a high, a medium, and a very low affinity, respectively. For each LPHM domain and for each substrate, the data could be fit to Michaelis-Menten kinetics. The data obtained with each LPHM domain using the ga-N-Ac-Tyr-Val-Gly and ga-N-Ac-Tyr-Phe-Gly substrates are shown as Eadie-Hofstee plots (Fig. 6 A). The values obtained with ga-N-Ac-Tyr-Gly-Gly are presented together with the substrate affinity (K_m) (Fig. 6B ) and V_max (Fig. 6C ) for the other substrates, showing that each LPHM domain yielded a distinct set of values for K_m and V_max (Table 1 ).

View larger version (42K): [in this window] [in a new window]   Figure 6. Comparison of enzyme kinetics using substrates with different penultimate amino acid residues. A) Eadie-Hofstee plots for LPHM-1, -2, -3, and -4 using -N-Ac-Tyr-Val-Gly (Ac-YVG, square) or -N-Ac-Tyr-Phe-Gly (Ac-YFG, triangle) as the substrate. For each LPHM enzyme, the apparent Vmax using -N-Ac-Tyr-Val-Gly was defined as 10 units. The Vmax using -N-Ac-Tyr-Phe-Gly or -N-Ac-Tyr-Gly-Gly (see below) was evaluated using the same preparation of enzyme. The set of Vmax values can then be compared for the different LPHM domains. The best fit lines were determined using the program KaleidaGraph. B) Assays were performed as in panel A. Bar graphs for LPHM-1, -2, -3, and -4 and rat PHM show the affinity (Km) for each substrate. Kinetic parameters were determined in quadruplicate; the standard error of the mean is indicated. Km values of rat PHM for Ac-YVG and Ac-YFG were taken from Eipper et al. (23). C) Bar graphs show the relative Vmax for LPHM-1, -2, -3, and -4. For each LPHM domain, the Vmax using -N-Ac-Tyr-Val-Gly was defined as 10 units. The standard error of the mean is indicated.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
LPAM is a multifunctional PAM with a unique tandem array of divergent PHM domains
Although the molluscan brain uses a large array of amidated peptides as messenger molecules, virtually nothing is known about the specific features of peptide -amidation in molluscs. In this paper, we set out to identify the peptide -amidating enzymes from the snail, L. stagnalis.

Using a PCR amplification and cDNA cloning strategy, we identified a PAM sequence from the Lymnaea CNS. Many structural features uniquely identify LPAM as a true PAM homologue, having both PHM and PAL catalytic domains. LPAM and vertebrate PAMs share this type of organization, in contrast to Drosophila and Calliactis (¹¹ , ¹² ). In vertebrates, PHM and PAL are synthesized as a bifunctional protein on a single gene. In the fruit fly Drosophila, both PHM and PAL activities were demonstrated. The PHM and PAL enzymes do not form a bifunctional enzyme and are encoded by distinct genes (P. H. Taghert, personal communication, 11). LPAM is unique in that it has four strongly structurally divergent (Fig. 4) yet functional (Fig. 5) PHM domains. All four LPHM domains and the LPAL domain show typical sequence characteristics, in which all residues essential for catalytic activity and for copper and substrate binding have been conserved (Fig. 4) . Apart from these residues, sequence conservation is low.

During evolution, proteins have evolved frequently by duplication of entire genes and further mutation, giving rise to families of structurally related proteins. A clear example is represented by the prohormone convertases [reviewed by Seidah and Chrétien ⁽⁴⁶⁾ ], which form a family of structurally and functionally related serine proteases. In contrast, examples of intragenic duplication in genes encoding enzymes are rare. However, there is an example where intragenic duplication is thought to have given rise to two isoenzymes, i.e., endothelial angiotensin I-converting enzyme (⁴⁷ , ⁴⁸ ). The two enzymes that have evolved may, in some instances, convert different substrates (⁴⁹ , ⁵⁰ ). The LPAM precursor is a unique example of intragenic duplication and subsequent mutation that gave rise to the specific kinetic features of four (PHM) isoenzymes.

LPAM expression in the Lymnaea brain
Many neurons of the Lymnaea CNS use complex peptide cocktails as messages for communication with follower cells [reviewed by Nässel ⁽⁵¹⁾ ]. Therefore, diversity and ratio of the peptides in a cocktail must be strictly controlled in order to assign a message-specific informational content to the peptide cocktail. The -amidation of peptides is often essential for their bioactivity and, as such, to the specificity of peptide messages in the Lymnaea CNS. Clearly, in addition to the molecular mechanisms underlying the generation of different LPHM enzymes, a strictly regulated cellular expression of LPAM is an important mechanism assuring complete amidation of peptide substrates. It has been observed that in the rat CNS, PAM is expressed in regions where many amidated peptides are produced, such as the hippocampus and hypothalamus (⁵² , ⁵³ ). On the other hand, PAM immunoreactivity is also observed in neurons and other tissues not known to produce -amidated peptides (⁵² , ⁵⁴ ). However, in the rat brain it is not possible to examine cellular coexpression by applying single-cell analysis. Because of the large size of Lymnaea neurons, this issue can be examined by applying the very sensitive method of RT-PCR on individual identifiable neurons. Indeed, using this approach demonstrated a tightly regulated expression of LPAM exclusively in neurons that coexpress amidated peptides (Fig. 3) .

The LPAM gene, like the PAM gene in the rat, generates two major transcripts, LPAM-1 and LPAM-2, probably by alternative splicing of the exon A domain in the pre-mRNA (Fig. 2) . The exon A domain containing LPAM-1 transcript is coexpressed with LPAM-2 in a few neurons that express the peptide conopressin and APGWamide (Fig. 3) , whereas the LPAM-2 transcript is found in all other neurons that synthesize amidated peptides. Because the role of exon A in the LPAM-1 precursor is not understood (see below), the functional reason for the difference in cellular expression pattern of the LPAM-1 and LPAM-2 transcripts is not clear.

Processing of the LPAM precursor into five distinct enzymes
The LPAM-1 and 2 zymogens are endoproteolytically processed to five individual catalytic domains (Fig. 1) . Endoproteolytic cleavage has to occur at sites by a hitherto unknown protease. Furthermore, based on LPAL activity in the soluble fraction, LPAL is generally cleaved from the transmembrane domain (Fig. 1) . Because dibasic or multiple basic processing sites are absent carboxy-terminal to LPAL, the endoproteolytic cleavage that generates soluble LPAL must occur at a different site. Thus, the LPAM zymogens are cleaved at nonconventional sites, illustrating that enzymes different from the Kex2-related prohormone convertases ⁽⁴⁶⁾ are active in the secretory pathway. That the region carboxy-terminally from PHM is accessible for cleavage is suggested by the limited tryptic digestion of rat PAM in vitro, which results in endoproteolysis at the carboxyl terminus of rat PHM in a protease accessible region (exon 15: rat PHM Met³⁶⁰ to Gln³⁹²; 23) that is not required for catalytic activity. This cleavage increases the turnover number and affinity of PHM for its substrates in rat (⁴⁴ , ⁵⁵ ). The rat PHM protease accessible region has virtually no sequence identity with its counterpart in any of the four LPHM domains (Fig. 4A ).

In vivo, endoproteolytic processing of rat PAM-1 can occur at a cleavage site consisting of paired basic amino residues present in the noncatalytic linker (exon A or exon 16; 5), which separates the PHM domain from the PAL domain. When this cleavage site is used, monofunctional PHM is generated from integral membrane PAL (¹⁶ , ⁵⁶ ). The region in LPAM-1 corresponding to the exon A domain is present, but much longer (157 vs. 105 amino acid residues) and without significant sequence identity (Fig. 2B and Fig. 4C ). However, it contains the residues Arg¹³⁷⁸-His¹³⁷⁹-Lys¹³⁸⁰-Arg¹³⁸¹, which may serve both as a potential dibasic cleavage site and a consensus sequence for furin catalyzed cleavage ⁽⁵⁷⁾ . Conservation of cleavage sites without conservation of the overall sequence would be consistent with a role for exon A in the generation of soluble LPHM(1-4) domains. The functional relevance of a cleavage in exon A remains obscure because all the LPHM enzyme domains are solubilized anyway, very likely by endoproteolysis at the protease sensitive regions.

Each LPHM displays a different combination of substrate affinity and reaction velocity
The precursor of carboxypeptidase D (gp180; 58) encodes three (different) carboxypeptidase domains. Because only two of these domains show catalytic activity, the question rises whether, in analogy to this, all four LPHM domains and the LPAL domain maintained their activity. Clearly, single amino acid changes can obliterate all enzyme activity ⁽⁴¹⁾ . Individual LPHM domains expressed and secreted by hEK293 cells are capable of converting synthetic -N-Ac-Tyr-Xxx-Gly tripeptide substrates (Fig. 5 and Fig. 6 ). This demonstrated that, although the LPHM domains have strongly divergent sequences, they are all catalytically active. Next, the maximal conversion rates and affinity values of each domain toward the three substrates (i.e., -N-Ac-Tyr-Val-Gly, -N-Ac-Tyr-Phe-Gly and -N-Ac-Tyr-Gly-Gly) were tested (Table 1) . When the substrate affinities for the different substrates were compared, each LPHM domain displayed a unique set of affinity values. Whereas LPHM-1 and -2 showed high affinities (low K_m) toward the three substrates tested, LPHM-3 and -4 showed lower affinities. The affinity of rat PHM toward -N-Ac-Tyr-Gly-Gly was very low, with a K_m of 680 ±230 µM (see also ^{ref 7} ). In contrast, the LPHM domains displayed about 10-fold higher affinity for this substrate. In short, the affinity of the LPHM domains toward the three substrates were in most cases higher than the affinity of rat PHM toward the same substrates. Subsequently, the relative V_max values of the LPHM domains for the three different substrates were compared. Based on the relative V_max values for the -N-Ac-Tyr-Gly-Gly substrate, LPHM-2 and LPHM-4 form a class distinct from LPHM-1 and LPHM-3. Yet LPHM-2 and LPHM-3 show a relative low V_max for the -N-Ac-Tyr-Phe-Gly substrate, which is in contrast to LPHM-1 and LPHM-4.

Together, the results indicate that sequence divergence of the LPHM domains resulted in a unique combination of K_m and relative V_max values for each LPHM and for the three substrates studied. The physiological relevance of this finding could be the involvement of one or more LPHMs in brain fatty acid amide formation, as postulated for mammalian PHM (⁵⁹ , ⁶⁰ ). In addition, the four LPHM enzymes together may allow for maximal reaction velocities under conditions of various distinct substrate and varying substrate concentrations, a situation that occurs in secretory granules of different Lymnaea neurons.

	ACKNOWLEDGMENTS

 
We thank Ellen van Kesteren and Carla Berard for technical assistance. This work was supported by travel grant SIR-15-1606 from the Dutch Scientific Organization (N.W.O.) (S.S.) and by National Institutes of Health Service Grant DK-32949 (B.A.E., R.E.M.). Genbank accession numbers for Lymnaea PAM-1 and PAM-2 are AF109919 and AF109920, respectively.

	FOOTNOTES

 
^* Correspondence: Department of Molecular and Cellular Neurobiology, Graduate School Neurosciences Amsterdam, Research Institute Neurosciences Vrije Universiteit, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands. E-mail: absmit{at}bio.vu.nl

¹ Abbreviations: CNS, central nervous system; DTßH, Drosophila TßH; LPAL, Lymnaea PAL domain; LPAM, Lymnaea peptidyl glycine -amidating monooxygenase; PAL, peptidyl--hydroxyglycine -amidating lyase; PAM, peptidylglycine -amidating monooxygenase; PHM, peptidylglycine -hydroxylating monooxygenase; 5'-RACE, 5'-rapid amplification of cDNA ends; RT-PCR, reverse transcriptase polymerase chain reaction; TßH, Drosophila tyramine ß-hydroxylase; TGN, trans-Golgi network; TMT, TM with 1% Triton X-100.

Received for publication September 15, 1998. Revision received December 3, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Hökfelt, T. (1991). Neuropeptides in perspective: the last ten years. Neuron 7,867-879[Medline]
2. Brezina, V., Orekhova, I. V., Weiss, K. R. (1996). Functional uncoupling of linked neurotransmitter effects by combinatorial convergence. Science 273,806-810[Abstract]
3. Merkler, D. J. (1994). C-terminal amidated peptides: production by the in vitro enzymatic amidation of glycine-extended peptides and the importance of the amide to bioactivity. Enzyme Microb. Technol. 16,450-456[Medline]
4. Bradbury, A. F., Finnie, M. D. A., Smyth, D. G. (1982). Mechanism of C-terminal amide formation by pituitary enzymes. Nature (London) 298,686-688[Medline]
5. Eipper, B. A., Stoffers, D. A., Mains, R. E. (1992). The biosynthesis of neuropeptides: Peptide -amidation. Annu. Rev. Neurosci. 15,57-85[Medline]
6. Bradbury, A. F., Smyth, D. G. (1983). Substrate specificity of an amidating enzyme in porcine pituitary. Biochem. Biophys. Res. Commun. 112,372-377[Medline]
7. Tamburini, P. P., Young, S. D., Jones, B. N., Palmesino, R. A., Consalvo, A. P. (1990). Peptide substrate specificity of the -amidating enzyme isolated from rat medullary thyroid CA-77 cells. Int. J. Peptide Protein Res. 35,153-156[Medline]
8. Kato, I., Yonekura, H., Tajima, M., Yanagi, M., Yamamoto, H., Okamoto, H. (1990). Two enzymes concerned in peptide hormone -amidation are synthesized from a single mRNA. Biochem. Biophys. Res. Commun. 172,197-203[Medline]
9. Katopodis, A.G., Ping, D., May, S. W. (1990). A novel enzyme from bovine neurointermediate pituitary catalyzes dealkylation of -hydroxyglycine derivatives, thereby functioning sequentially with peptidylglycine -amidating monooxygenase in peptide amidation. Biochemistry 29,6115-6120[Medline]
10. Perkins, S. N., Husten, E. J., Eipper, B. A. (1990). The 108-kDa peptidylglycine -amidating monooxygenase precursor contains two separable enzymatic activities involved in peptide amidation. Biochem. Biophys. Res. Commun. 171,926-932[Medline]
11. Kolhekar, A. S., Roberts, M. S., Jiang, N., Johnson, R. C., Mains, R. E., Eipper, B. A., Taghert, P. H. (1997). Neuropeptide amidation in Drosophila: Separate genes encode the two enzymes catalyzing amidation. J. Neurosci. 17,1363-1376[Abstract/Free Full Text]
12. Hauser, F., Williamson, M., Grimmelikhuijzen, C. J. (1997). Molecular cloning of a peptidylglycine -hydroxylating monooxygenase from sea anemones. Biochem. Biophys. Res. Commun. 241,509-512[Medline]
13. Vilim, F. S., Price, D. A., Lesser, W., Kupfermann, I., Weiss, K. R. (1996). Costorage and corelease of modulatory peptide cotransmitters with partially antagonistic actions on the accessory radula closer muscle of Aplysia californica. J. Neurosci. 16,8092-8104[Abstract/Free Full Text]
14. Miller, M. W., Beushausen, S., Cropper, E. C., Eisinger, K., Stamm, S., Vilim, F. S., Vitek, A., Zajc, A., Kupfermann, I., Brosius, J., Weiss, K. R. (1993). The buccalin-related neuropeptides: isolation and characterization of an Aplysia cDNA clone encoding a family of peptide cotransmitters. J. Neurosci. 13,3346-3357[Abstract]
15. Husten, E. J., Eipper, B. A. (1991). The membrane-bound bifunctional peptidylglycine -amidating mono-oxygenase protein. Explorations of its domain structure through limited proteolysis. J. Biol. Chem. 266,17004-17010[Abstract/Free Full Text]
16. Eipper, B. A., Perkins, S. N., Husten, E. J., Johnson, R. C., Keutman, H. T., Mains, R. E. (1991). Peptidyl--hydroxyglycine -amidating lyase. J. Biol. Chem. 266,7827-7833[Abstract/Free Full Text]
17. Bennett, M. K., Calakos, N., Kreiner, T., Scheller, R. H. (1992). Synaptic vesicle membrane proteins interact to form a multimeric complex. J. Cell Biol. 116,761-775[Abstract/Free Full Text]
18. Smit, A. B., Spijker, S., Geraerts, W. P. M. (1992). Molluscan putative prohormone convertases: structural diversity in the central nervous system of Lymnaea stagnalis. FEBS Lett 312,213-218[Medline]
19. Sanger, F., Nicklen, S., Coulson, A. R. (1977). DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74,5463-5467[Abstract/Free Full Text]
20. Frohman, M. A., Dush, M. K., Martin, G. R. (1988). Rapid production of full length cDNA from rare transcripts: amplification using a single gene-specific oligonucleotide primer (PCR). Proc. Natl. Acad. Sci. USA 85,8998-9002[Abstract/Free Full Text]
21. Chomczynski, P., Sacchi, N. (1987). Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162,156-159[Medline]
22. Felsenstein, J. (1995) PHYLYP (Phylogeny Inference Package) version 3.57c, University of Washington, Seattle
23. Eipper, B. A., Quon, A. S. W., Mains, R. E., Boswell, J. S., Blackburn, N. J. (1995). The catalytic core of peptidylglycine -hydroxylating monooxygenase: Investigation by site-directed mutagenesis, Cu X-ray absorption spectroscopy, and electron paramagnetic resonance. Biochemistry 34,2857-2865[Medline]
24. Hodges, R. S., Heaton, R. J., Parker, J. M. R., Molday, L., Molday, R. S. (1988). Antigen–antibody interaction. J. Biol. Chem. 263,11768-11775[Abstract/Free Full Text]
25. Yun, H.-Y., Milgram, S. L., Keutmann, H. T., Eipper, B. A. (1995). Phosphorylation of the cytosolic domain of peptidylglycine -amidating monooxygenase. J. Biol. Chem. 270,30075-30083[Abstract/Free Full Text]
26. Oyarce, A. M., Eipper, B. A. (1993). Neurosecretory vesicles contain soluble and membrane-associated monofunctional and bifunctional peptidylglycine -amidating monooxygenase proteins. J. Neurochem. 60,1105-1114[Medline]
27. Hook, V. Y. H., Liston, D. (1987). Distribution of enkephalin-containing peptides within bovine chromaffin granules. Neuropeptides 9,263-267[Medline]
28. Yoo, S. H. (1993). pH-dependent binding of chromogranin B and secretory vesicle matrix proteins to the vesicle membrane. Biochem. Biophys. Acta 1179,239-246[Medline]
29. Von Heijne, G. (1983). A new method for predicting signal sequence cleavage sites. Nucleic Acids Res 14,4683-4690[Abstract/Free Full Text]
30. Kato, I., Yonekura, H., Yamamoto, H., Okamoto, H. (1990). Isolation and functional expression of pituitary peptidylglycine -amidating enzyme mRNA. FEBS Lett 269,319-323[Medline]
31. Eipper, B. A., Green, C. B.-R., Campbell, T. A., Stoffers, D. A., Keutman, H. T., Mains, R. E., Ouafik, L. (1992). Alternative splicing and endoproteolytic processing generate tissue-specific forms of pituitary peptidyl glycine -amidating monooxygenase (PAM). J. Biol. Chem. 267,4008-4015[Abstract/Free Full Text]
32. Li, K. W., Hoek, R. M., Smith, F., Jiménez, C. R., Van der Schors, R. C., Van Veelen, P. A., Chen, S., Van der Greef, J., Parish, D. C., Benjamin, P. R., Geraerts, W. P. M. (1994). Direct peptide profiling by mass spectrometry of single identified neurons reveals complex neuropeptide-processing pattern. J. Biol. Chem. 269,30288-30292[Abstract/Free Full Text]
33. Li, K. W., Jiménez, C. R., Van Veelen, P. A., Geraerts, W. P. M. (1994). Processing and targeting of a molluscan egg-laying peptide prohormone as revealed by mass spectrometric peptide fingerprinting and peptide sequencing. Endocrinology 34,1812-1819
34. Jiménez, C. R., Li, K. W., Dreisewerd, K., Spijker, S., Kingston, R., Bateman, R. H., Burlingame, A. L., Smit, A. B., Van Minnen, J., Geraerts, W. P. M. (1998). Direct mass spectrometric peptide profiling and sequencing of single neurons reveals differential peptide patterns in a small neuronal network. Biochemistry 37,2070-2076[Medline]
35. Van Kesteren, R. E., Smit, A. B., Dirks, R. W., De With, N. D., Geraerts, W. P. M., Joosse, J. (1992). Evolution of the vasopressin-related precursor, preproconopressin, from the mollusc, Lymnaea stagnalis. Proc. Natl. Acad. Sci. USA 89,4593-4597[Abstract/Free Full Text]
36. Li, K. W., Smit, A. B., Geraerts, W. P. M. (1992). Structural and functional characterization of neuropeptides involved in the control of male mating behavior of Lymnaea stagnalis. Peptides 13,633-638[Medline]
37. Bogerd, J., Li, K. W., Jiménez, C. R., Van der Schors, R. C., Ebberink, R. H. M., Gerearts, W. P. M. (1994). Processing, axonal transport and cardioregulatory functions of peptides derived from two related prohormones generated by alternative splicing of a single gene in identified neurons VD1 and RPD2 of Lymnaea. Mol. Brain Res. 23,66-72[Medline]
38. De Lange, R. P. J., Van Golen, F. A., Van Minnen, J. (1997). Diversity in cell specific co-expression of four neuropeptide genes involved in control of male copulation behaviour in Lymnaea stagnalis. Neuroscience 78,289-299[Medline]
39. Smit, A. B., Van Kesteren, R. E., Li, K. W., Van Minnen, J., Spijker, S., Van Heerikhuizen, H., Geraerts, W. P. M. (1998). Towards understanding the role of insulin in the brain: lessons from insulin-related signaling systems in the invertebrate brain. Prog. Neurobiol. 54,35-54[Medline]
40. Prigge, S. T., Kohlekar, A. S., Eipper, B. A., Mains, R. E., Amzel, L. M. (1997). Amidation of bioactive peptides—the structure of peptidylglycine -hydroxylating monooxygenase. Science 278,1300-1305[Abstract/Free Full Text]
41. Kolhekar, A. S., Keutman, H. T., Mains, R. E., Quon, A. S., Eipper, B. A. (1997). Peptidylglycine -hydroxylating monooxygenase: disulfide linkages define amino- and carboxy-terminal domains of the catalytic core. Biochemistry 36,10901-10910[Medline]
42. Yonekura, H., Anzai, T., Kato, I., Furuya, Y., Shizuta, S., Takasawa, S., Okamoto, H. (1996). Identification of five essential histidine residues for peptidylglycine monooxygenase. Biochem. Biophys. Res. Commun. 218,495-499[Medline]
43. Milgram, S. L., Mains, R. E., Eipper, B. A. (1996). Identification of routing determinants in the cytosolic domain of a secretory granule-associated integral membrane protein. J. Biol. Chem. 271,17526-17535[Abstract/Free Full Text]
44. Husten, E. J., Eipper, B. A. (1994). Purification and characterization of PAM-1, an integral membrane protein involved in peptide processing. Arch. Biochem. Biophys. 312,487-492[Medline]
45. Hilsted, L. (1990). Alpha-amidation of gastrin is impaired by diethyl-dithiocarbamate. Regul. Pept. 29,179-187[Medline]
46. Seidah, N. G., Chrétien, M. (1997). Eukaryotic protein processing: endoproteolysis of precursor proteins. Curr. Opin. Biotechnol. 8,602-607[Medline]
47. Soubrier, F., Alhenc-Gelas, F., Hubert, C., Allegrini, J., John, M., Tregear, G., Corvol, P. (1988). Two putative active centers in human angiotensin I-converting enzyme revealed by molecular cloning. Proc. Natl. Acad. Sci. USA 85,9386-9390[Abstract/Free Full Text]
48. Hubert, C., Houot, A. M., Corvol, P., Soubrier, F. (1991). Structure of the angiotensin I-converting enzyme gene. Two alternate promotors correspond to evolutionary steps of a duplicate gene. J. Biol. Chem. 266,15377-15383[Abstract/Free Full Text]
49. Jaspard, E., Wei, L., Alhenc-Gelas, F. (1993). Differences in the properties and enzymatic specificities of the two active sites of angiotensin I-converting enzyme (kinase II). Studies with bradykinin and other natural peptides. J. Biol. Chem. 268,9496-9503[Abstract/Free Full Text]
50. Rousseau, A., Michaud, A., Chauvet, M. T., Lenfant, M., Corvol, P. (1995). The hemoregulatory peptide N-acetyl-Ser-Asp-Lys-Pro is a natural and specific substrate of the N-terminal active site of human angiotensin-converting enzyme. J. Biol. Chem. 270,3656-3661[Abstract/Free Full Text]
51. Nässel, D. R. (1996). Peptidergic neurohormonal control systems in invertebrates. Curr. Opin. Neurobiol. 6,842-850[Medline]
52. Rhodes, C. H., Xu, R. Y., Angeletti, R. H. (1990). Peptidylglycine alpha-amidating monooxygenase (PAM) in Schwann cells and glia as well as neurons. J. Histochem. Cytochem. 38,1301-1311[Abstract]
53. Schäfer, M. K., Stoffers, D. A., Eipper, B. A., Watson, S. J. (1992). Expression of peptidylglycine -amidating mono-oxygenase (EC 1.14.17.3) in the rat central nervous system. J. Neurosci. 12,222-234[Abstract]
54. Ouafik, L. H., May, V., Keutmann, H. T., Eipper, B. A. (1989). Developmental regulation of peptidyl-glycine -amidating monooxygenase (PAM) in rat heart and atrium and ventricle. J. Biol. Chem. 264,5839-5845[Abstract/Free Full Text]
55. Husten, E. J., Tausk, F. A., Keutmann, H. T., Eipper, B. A. (1993). Use of endoproteases to identify catalytic domains, linker regions, and functional interactions in soluble peptidylglycine -amidating monooxygenase. J. Biol. Chem.