Evolutionarily Conserved Multiple C2 Domain Proteins withTwo Transmembrane Regions (MCTPs) and Unusual Ca2�

Binding Properties*

Received for publication, June 30, 2004, and in revised form, November 3, 2004Published, JBC Papers in Press, November 4, 2004, DOI 10.1074/jbc.M407305200

Ok-Ho Shin, Weiping Han, Yun Wang, and Thomas C. Sudhof‡

From the Center for Basic Neuroscience, Department of Molecular Genetics and Howard Hughes Medical Institute,University of Texas Southwestern Medical Center, Dallas, Texas 75390

C2 domains are primarily found in signal transductionproteins such as protein kinase C, which generally con-tain a single C2 domain, and in membrane traffickingproteins such as synaptotagmins, which generally con-tain multiple C2 domains. In both classes of proteins, C2domains usually regulate the respective protein’s func-tion by forming Ca2�-dependent or Ca2�-independentphospholipid complexes. We now describe MCTPs (mul-tiple C2 domain and transmembrane region proteins), anovel family of evolutionarily conserved C2 domain pro-teins with unusual Ca2�-dependent properties. MCTPsare composed of a variable N-terminal sequence, threeC2 domains, two transmembrane regions, and a shortC-terminal sequence. The invertebrate organisms Cae-norhabditis elegans and Drosophila melanogaster ex-press a single MCTP gene, whereas vertebrates expresstwo MCTP genes (MCTP1 and MCTP2) whose primarytranscripts are extensively alternatively spliced. Most ofthe MCTP sequences, in particular the C2 domains, arehighly conserved. All MCTP C2 domains except for the sec-ond C2 domain of MCTP2 include a perfect Ca2�/phospho-lipid-binding consensus sequence. To determine whetherthe C2 domains of MCTPs actually function as Ca2�/phos-pholipid-binding modules, we analyzed their Ca2� andphospholipid binding properties. Surprisingly, we foundthat none of the three MCTP1 C2 domains interacted withnegatively charged or neutral phospholipids in the pres-ence or absence of Ca2�. However, Ca2� titrations moni-tored via intrinsic tryptophan fluorescence revealed thatall three C2 domains bound Ca2� in the absence of phospho-lipids with a high apparent affinity (EC50 of �1.3–2.3 �M).Our data thus reveal that MCTPs are evolutionarily con-served C2 domain proteins that are unusual in that the C2domains are anchored in the membrane by two closelyspaced transmembrane regions and represent Ca2�-bind-ing but not phospholipid-binding modules.

The C2 domain is defined as a sequence motif in a compari-son of the primary structures of different protein kinase Cisoforms and named in an unbiased fashion as the “secondconstant sequence” of protein kinase C isozymes (1). Laterstudies revealed that a large number of proteins include C2

domains, with �200 such proteins in the human genome alone(2). We observed that in synaptotagmin 1, a synaptic vesicleprotein that binds Ca2� and contains two C2 domains (3), C2

domains are autonomously folded Ca2�-binding modules (4, 5).Subsequently, most C2 domain proteins were found to formCa2�-dependent phospholipid complexes, although some ap-pear to bind to phospholipids in the absence of Ca2� (e.g.PTEN) (6) and others constitute protein-interaction domainsinstead of binding to either Ca2� or phospholipids (e.g. theN-terminal C2 domain of Munc13-1 (7) or the C-terminalRIM1� C2 domain (8)). Although the Ca2�-binding properties ofmany C2 domain proteins remain to be examined, the largenumber of C2 domain proteins in the vertebrate genome makesit likely that this domain represents the second most commonCa2� binding motif after the EF-hand motif.

Most C2 domain proteins are either signal transduction en-zymes, such as protein kinase C, or membrane traffickingproteins, such as synaptotagmin 1. At least some isoforms of allmajor signal transduction enzymes, from ubiquitin ligases tokinases to various phospholipases, contain a C2 domain. With-out exception, these proteins are soluble cytosolic enzymes thatinclude a single C2 domain. In contrast, membrane traffickingproteins generally include at least two C2 domains, although afew proteins such as the �-RIM isoforms (9) and some splicevariants of piccolo/aczonin and intersectin (10–12) contain onlya single C2 domain. In membrane trafficking proteins the dif-ferent C2 domains often feature conserved sequence differ-ences, indicating that the C2 domains are functionally special-ized. For example, although in synaptotagmin 1 the C2A andC2B domains both bind Ca2� and phospholipids with similaraffinities (13), the C2B domain contains an additional “bottom”�-helix that is absent from the C2A domain but is conserved inall of the C2B domains of synaptotagmins (14). Membranetrafficking proteins with multiple C2 domains either have noTMR1 or a single TMR either at the N terminus (e.g. synapto-tagmins; see Fig. 1) (15) or the C terminus (e.g. ferlins) (16).

Work over the last few decades established Ca2� as themajor intracellular second messenger in eukaryotic cells, withspecificity achieved by the spatial segregation of Ca2� signals.In vertebrate genomes, proteins containing EF-hand Ca2�-binding sites are more common than C2 domain proteins (2).However, Ca2� binding in EF-hand proteins often does notserve a direct regulatory function but instead acts in Ca2�

buffering (e.g. parvalbumin and calbindin) or subserves a struc-tural role (reviewed in Refs. 17 and 18). Among the EF-handproteins with a regulatory Ca2�-binding site, one particular

* The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked“advertisement” in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

The nucleotide sequence(s) reported in this paper has been submittedto the GenBankTM/EBI Data Bank with accession number(s) AY656715,AY656716, and AY656717.

‡ To whom correspondence should be addressed. Tel.: 214-648-1876;Fax: 214-648-1879; E-mail: Thomas.Sudhof@UTSouthwestern.edu.

1 The abbreviations used are: TMR, transmembrane region; EYFP,enhanced yellow fluorescent protein; GST, glutathione S-transferase;MCTP, multiple C2 domain and transmembrane region protein.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 280, No. 2, Issue of January 14, pp. 1641–1651, 2005© 2005 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 1641

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

protein, calmodulin, appears to mediate more Ca2�-dependentregulatory actions than all other EF-hand proteins combinedand is in fact expressed as an identical sequence from multipleindependent genes (reviewed in Refs. 19 and 20).

This situation seems to be completely different for C2 domainproteins. All functionally characterized proteins containing aC2 domain that binds Ca2� also act as Ca2� sensors, and nosingle C2 domain protein dominates the Ca2�-dependent regu-lation of a cell. Thus, to fully understand the targets of Ca2�-signaling in cells it is essential to characterize all principal C2

domain proteins. Only a complete overview of the Ca2�-de-pendent properties of different C2 domain proteins will provideinsight into how Ca2� signaling works. In view of these consid-erations, we have searched for conserved C2 domain proteinsthat might function as widely distributed Ca2� sensors andhave focused on membrane-anchored C2 domain proteins be-cause these are most likely involved in membrane traffic.

MATERIALS AND METHODS

Sequence Analyses and Data Bank Searches—Various data banks ofthe National Center for Biotechnology Information (NCBI) and CeleraGenomics were searched for multiple C2 domain proteins, expressedsequence tags, splice variants in reported full-length sequences, and thegenes of MCTPs using programs available on the NCBI web site withdefault settings. The MCTP cDNA sequences were submitted to Gen-BankTM (accession numbers AY656715, AY656716, and AY656717).

Expression and Purification of Recombinant Proteins—The DNAsencoding the human MCTP1L-C2A (residues 243–391), MCTP1L-C2B(residues 451–598), MCTP1L-C2C (residue 608–755), and MCTP2-C2C(residues 491–638) domains were PCR amplified, subcloned intopGEX-KG vector, and verified by DNA sequencing. All recombinantproteins were produced as bacterial GST fusion proteins, purified es-sentially as described (21), additionally treated with Benzonase, andwashed with 20 mM CaCl2 and high salt buffers to remove the bacterialcontaminants that stick to C2 domains and alter their properties (22).

Construction and Expression of Vectors Encoding Various EYFP Fu-sion Proteins of MCTP2 Fragments to Test the Functionality of PutativeTransmembrane Regions—pCMV-EYFP, pCMV-EYFP-MCTP-TM1(MCTP2 residue 641–750; does not correspond to a physiological splicevariant), pCMV-EYFP-MCTP-TM2 (MCTP2 residue 641–878 with res-idue 696–736 deletion; junction sequence: STLRSTIAFA-deleted se-quence-ESTDIDDEEDE, precisely corresponding to a splice variantobserved in random cDNA sequences in GenBankTM; see Fig. 2B), andpCMV-EYFP-MCTP-TM1-TM2 (MCTP2 residue 641–878) were gener-ated by standard procedures and transfected into COS-7 cells usingDEAE-dextran. After 2 days, cells were washed three times usingphosphate-buffered saline, and 0.5 ml of buffer (50 mM HEPES-NaOH,pH 7.2, 100 mM NaCl, 4 mM sodium EGTA, 2 mM MgCl2, 1 mM dithio-threitol, 1 mM phenylmethylsulfonyl fluoride, and a protease inhibitormixture from Roche Diagnostics) was added to each 100-mm plate.Scraped cells were passed through the 27-gauge, 0.5-inch needle (BDBiosciences) five times, and the cell suspension was centrifuged at500 � g for 10 min to precipitate unbroken cells and nuclei. Thesupernatant was additionally centrifuged at 100,000 � g for 1 h, andtotal membrane and cytosol fractions were separated. All fractions wereadjusted to the same volume, and Western blotting using a polyclonalantibody against GFP (T3743) was performed.

Centrifugation phospholipid binding assays were carried out withpurified soluble GST fusion proteins in buffer A (50 mM HEPES-NaOH,pH 6.8, 0.1 M NaCl, and 4 mM sodium EGTA). The C2 domain GSTfusion proteins were incubated with liposomes of defined phospholipidcomposition in buffer A containing variable amounts of CaCl2, SrCl2, orBaCl2 to provide defined concentrations of free Ca2�, Sr2�, or Ba2�,respectively (calculated using EqCal for Windows software from Bio-soft, Ferguson, MO). For liposome generation, dried phospholipids (ob-tained from Avanti Co.) were resuspended in buffer A containing 0.5 M

sucrose and sonicated. The resulting ”heavy“ liposomes were then iso-lated by centrifugation. After the incubations of C2 domains with lipo-somes, liposomes with bound C2 domains were re-isolated by centrifu-gation essentially as described (14, 23), and bound proteins wereprecipitated, resuspended in 30 �l of 2� SDS sample buffer, and ana-lyzed by SDS-PAGE and Coomassie Blue staining.

Fluorescence Spectroscopy—Purified GST-C2 domain proteins (3 mgin 1 ml of buffer B (40 mM Tris-HCl, pH 8, 0.1 M NaCl, and 0.5 mM

sodium EGTA)) were cleaved with 10 units of thrombin (Amersham

Biosciences) and reconstituted in buffer B by rotation for 18 h at roomtemperature, and complete cleavage was confirmed by SDS-PAGE.GST-protein was removed by three sequential incubations with 100 �lof glutathione-Sepharose 4B (Amersham Biosciences) for 1 h at 4 °C,and the absence of GST was confirmed by SDS-PAGE. C2 domainproteins without GST were used without further purification for record-ings of fluorescence spectra after �1:1,000 dilution using buffer B.Fluorescence emission spectra were acquired in an LS55 luminescencespectrometer (PerkinElmer Life Sciences) with an excitation wave-length of 282 nm. For Ca2� titrations, aliquots of Ca2� were added tothe cuvette from a concentrated stock solution, and fluorescence wasmonitored at the stated emission maximum wavelengths. The concen-tration of free Ca2� was calculated with the EqCal program, and thedata were analyzed using GraphPad software. Reversibility of theCa2�-induced increases in tryptophan fluorescence was tested bythe addition of an excess of EGTA.

Antibody Production—Polyclonal antibodies to human MCTPs weregenerated using the recombinant proteins U5837 (MCTP1L residues243–391), U5987 (MCTP1L residues 608–755), U6380 (MCTP2 resi-dues 1–324), and U5985 (MCTP2 residues 491–638). All proteins wereinitially produced as GST fusion proteins, but the GST moiety wasremoved by thrombin cleavage before the proteins were employed asantigens for antibody production.

Cell Culture and Confocal Imaging—HEK293 cells were plated onpolylysine-coated (1 mg/ml in 0.1 M borate buffer; Sigma) 18-mm diam-eter, 1.5 coverglasses (VWR Scientific), and maintained in Dulbecco’smodified Eagle’s medium (Invitrogen) supplemented with 10% fetalbovine serum at 37 °C and 5% CO2. One day after plating, cells weretransfected with the EYFP-MCTP2 fusion vectors using FuGENE-6(Roche Applied Science). Two days after transfection, cover glasses withcells were incubated in phosphate-buffered saline containing 10 �M

FM5–95 (Molecular Probes) for 5 min on ice, and then mounted in animaging chamber (Warner Instruments) for observation. Confocal im-ages were acquired on a Leica TCS2 laser-scanning confocal microscopeusing a 100� oil objective lens (numerical aperture, 1.3).

Cracked PC12 cell secretion assays were carried out with freeze-thawpermeabilization of PC 12 cells (13, 23). 70% confluent PC12 cells wereloaded with [3H]norepinephrine for 24 h, washed with physiologicalsaline (145 mM NaCl, 5.6 mM KCl, 2.2 mM CaCl2, 0.5 mM MgCl2, 5.6 mM

glucose, and 15 mM HEPES-NaOH, pH 7.4), harvested by pipetting astream of Ca2�-free ice-cold buffer C (120 mM potassium glutamate, 20mM potassium acetate, 2 mM EGTA, and 20 mM HEPES-NaOH, pH 7.2),and washed twice with same buffer. Cell “ghosts” were prepared byfreezing cells overnight at �80 °C, thawing on ice for 2 h, and washingthe ghosts 3� with 6 ml of buffer C containing 1% bovine serumalbumin. Standard secretion reactions (�20 reactions per 100-mmplate; 0.1 ml of total volume in 1.5-ml tubes) contained washed cellghosts, 2 mM ATP, 2 mM MgCl2, 10 �l of rat brain cytosol (10 g/liter) inbuffer C, and 6 �M of recombinant C2 domain protein with variousconcentrations of Ca2� to obtain the indicated free concentrations (cal-culated by EqCal). Reactions were incubated for 30 min at 30 °C andterminated on ice, and the samples were centrifuged at 4 °C for 3 minat 20,800 � g. Supernatants and pellets solubilized in 1% Triton X-100were analyzed by liquid scintillation counting.

Miscellaneous Procedure—SDS-PAGE and immunoblotting wereperformed using standard procedures (24, 25). Immunoblots were de-veloped by enhanced chemiluminescence (Amersham Biosciences).

RESULTS

Characterization of MCTP Sequences—In searching for pro-teins containing C2 domains and transmembrane regions, weidentified four classes of evolutionarily conserved proteins asfollows: (i) synaptotagmins, which are expressed in at least 15isoforms and are defined by the presence of a single N-terminalTMR and two C-terminal C2 domains with characteristic se-quence motifs (reviewed in Ref. 15); (ii) ferlins, which contain3–6 C2 domains and a C-terminal TMR (reviewed in Ref. 16);and (iii) two novel protein families that comprise 3–6 C2 do-mains with either N- or C-terminal TMRs (see schematic draw-ing in Fig. 1). We refer to the proteins containing an N-terminalTMR and multiple cytoplasmic C2 domains as E-Syts (for “ex-tended synaptotagmins”), because the topology of these pro-teins resembles that of synaptotagmins. One member of theE-Syt family was described previously as an unnamed plasmamembrane protein of adipocytes (26); these proteins will be

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins1642

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

examined in a later study. The present report will focus on thesecond class of novel C2 domain proteins, namely the proteinsreferred to here as MCTPs (for multiple C2 domain and TMRproteins”), because at least some splice variants of these pro-teins contain multiple TMRs in addition to the C2 domains (Fig.1). We focused on MCTPs because RNA interference experi-ments in Caenorhabditis elegans revealed that the C. elegansMCTP homolog (1H206) is an essential gene whose ablationleads to early embryonic lethality (27), and we undertook amolecular characterization of MCTPs as a first step towardunderstanding their essential functions.

We assembled full-length sequences and cDNA clones forhuman MCTPs using commercially available expressed se-quence tag clones and PCR on human cDNA. Data banksearches revealed that all vertebrates contain two MCTP geneswith the same overall architecture and a high degree of se-quence identity, whereas invertebrate animals (C. elegans andDrosophila melanogaster) contain a single MCTP gene. Forcomparison of MCTPs, we aligned the human MCTP1 andMCTP2 and the Drosophila and C. elegans MCTP sequences(Fig. 2A). The Drosophila sequences were assembled from twoseparate predicted transcription units in the data bank(CG33148 and CG33146). Although Drosophila MCTP is pre-dicted from the genome sequence to represent two separategenes, the precise colinearity of the Drosophila sequence withthe human and C. elegans sequences (Fig. 2A) and the corre-sponding Anopheles gambiae sequence (data not shown) indi-cates that this represents a single transcription unit.

The sequence alignments show that the N termini of MCTPsare highly variable; furthermore, both human MCTP1 and C.elegans MCTP are expressed with two alternative 5�-sequences

(referred to as MCTP1L and MCTP1S and as MCTPL andMCTPS, respectively), possibly because the genes contain twoindependent promoters. Motif searches demonstrated that allMCTPs contain three C2 domains (referred to as C2A, C2B, andC2C domains; shown in red in Fig. 2A) followed by two trans-membrane regions (shown in gray). Only short linkers separatethe three C2 domains, whereas a longer region connects the C2

domains to the TMRs (presumptive cytoplasmic non-C2 domainsequences are shown in yellow in Fig. 2A). The sequence thatconnects the two TMRs is short and highly charged. All of theMCTP sequences except for the N-terminal region are highlyconserved, with the highest degree of conservation of MCTPsbeing in the C2C domain, the second TMR, and the shortC-terminal cytoplasmic sequence. The latter includes a 12-residue stretch that is identical in all species and isoforms(NNELLDFLSRVP; Fig. 2A).

MCTPs Contain Two Functional TMRs—The proposedtransmembrane topography for MCTPs (Fig. 1) is based on theabsence of an N-terminal signal peptide in the MCTP se-quences and the fact that all known C2 domains are cytoplas-mic. With this topography, MCTPs are the only C2 domainproteins that include more than one TMR. Interestingly, anal-ysis of the MCTP sequences from expressed sequence tagsreveals evidence of at least two sites of alternative splicing.First, the linker that separates the C2A and C2B domain variesin size dependent on alternative splicing. Second, we observedin multiple independent clones alternative splicing of the firstTMR (Fig. 2B). As a result of this alternative splicing, somemRNAs for both MCTP1 and MCTP2 only encode the secondTMR. If correct, this alternative splicing would convert theconserved C-terminal cytoplasmic sequence of MCTPs into anon-cytoplasmic sequence, i.e. turn the topography of the Cterminus around (Fig. 1).

Although the predicted TMRs of MCTPs have all the featuresof classical TMRs, including a high degree of hydrophobicity,prediction of TMRs is not always accurate. To test whethereach TMR is individually capable of anchoring MCTPs into amembrane, we generated constructs that encode EYFP fusionproteins of MCTP2 (Fig. 3). In these proteins, EYFP is fused tothe C-terminal part of MCTP2 containing either both TMRs oronly the first or second TMR. Of these proteins, the one carry-ing a deletion of the first TMR corresponded precisely to thesequence of the expressed sequence tag clones containing al-ternatively spliced MCTP variants that lack the first TMR (seeFig. 2B).

We first transfected the EYFP-MCTP2 fusion constructs intoCOS cells and examined whether the presence of either TMRwould render the protein particulate as expected for a TMRprotein (Fig. 3). Indeed, we found that EYFP itself was soluble,but when fused to MCTP2 fragments containing either one ofthe two TMRs it became particulate. We next examined thelocalization of the transfected EYFP-MCTP2 fusion proteins intransfected HEK293 cells. All three proteins were localized tointracellular vesicular structures, suggesting that each TMRby itself is competent to anchor the protein to membranes andindicating that MCTP2 may normally be a vesicular protein(Fig. 4).

Structure of the MCTP Genes—Using public and the CeleraGenomics databases, we determined the organization and chro-mosomal localizations of the human MCTP genes. MCTP1 isencoded by a large gene (�600 kb) on chromosome 5q15,whereas MCTP2 is encoded by a smaller gene (�200 kb) onchromosome 15q26 (Fig. 5). The C-terminal half of the MCTPs,which exhibits the most sequence similarity, contains the sameexon-intron organization (Fig. 2A). Small differences in thenumber and placement of introns interrupting homologous se-

FIG. 1. Structures of proteins containing multiple C2 domainsanchored to membranes by a transmembrane region. Fourclasses of proteins were identified in data bank searches, namely syn-aptotagmins (Syts), extended synaptotagmins (E-Syts; to be describedin the future), ferlins, and multiple C2 domain and TMR proteins orMCTPs. Proteins are depicted embedded in a membrane (green) bytransmembrane regions (labeled TM) with presumptive non-cytoplas-mic sequences colored blue and cytoplasmic sequences colored red (forthe C2 domains) and yellow (for all other cytoplasmic sequences). N andC termini are indicated. In the case of the MCTPs, two major C-terminal splice variants result in the same sequence (shown in purple),being either cytoplasmic (when two TMRs are present) or non-cytoplas-mic (when one of the two TMRs is spliced out). C2 domains wereassigned based on the conserved domain data base of the NCBI; someproteins, especially ferlins, may have additional unpredicted C2 do-mains that do not precisely fit the consensus sequence as well asalternative transcripts with fewer C2 domains.

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins 1643

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

FIG. 2. Structure of MCTPs. A, primary amino acid sequences of human MCTP1 and MCTP2 (hMCTP1 and hMCTP2), Drosophila MCTP(DmMCTP), and C. elegans MCTP (CeMCTP). Residues that are present at the same position in at least two of the four sequences are highlightedwith a color code as follows: yellow letters on a red background, cytoplasmic C2 domain sequences; black letters on a yellow background, cytoplasmicnon-C2 domain sequences; white letters on a gray background, TMRs; yellow letters on a blue background, extracytoplasmic loop between the TMRs;and yellow letters on a purple background, C-terminal sequence. Sequences are numbered on the right; note that the C. elegans sequence includes a66-residue insertion at amino acid number 535 that is missing from the other species. Alternatively spliced sequences in the linker between the C2Aand C2B domains are underlined. B, alternative splicing of the C-terminal TMRs of MCTPs. The full-length variant and two variants for MCTP1(GenBankTM accession number AK057694 and AK058012) and one variant for MCTP2 (GenBankTM accession number R98750) are aligned. Note thatthe splice variants truncate TMR1 but leave TMR2 intact, thereby causing an inversion of the topology of the C terminus (see Fig. 1).

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins1644

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

quences are observed in the N-terminal half of the proteins(Tables I and II). The different N-terminal sequences observedin MCTP1L and MCTP1S are encoded by distinct exons (re-ferred to as 1a and 1b) that are separated by 200 kb in thegenome (Fig. 5). Most introns disrupt the coding sequence ofthe MCTPs at the same position, most often the “0” positionthat lies precisely between individual codons. The gene orga-nization explains the alternative splicing that we observed. Thealternative splicing of sequences separating the C2A and C2Bdomains is due to the presence or absence of exon 7 in MCTP1and MCTP2, and that of the first TMR is due to the presence orabsence of exon 18 in MCTP1 and exon 17 in MCTP2.

Tissue Distribution of MCTPs—Data bank analysis revealedthat the expressed sequence tags encoding MCTP1 and 2 wereisolated from a large number of tissues such as brain, bonemarrow, pancreas, spleen, thymus, placenta, blood vessel, andkidney. To test if the protein encoded by the MCTP mRNAs isactually synthesized and determine which tissues contain thehighest steady-state levels of these proteins, we produced twoantibodies against recombinant proteins derived from MCTP1and MCTP2 (Fig. 6A). Immunoblotting of various rat tissues

revealed that MCTP1 is highly expressed in skeletal muscleand to a lesser degree in heart muscle, whereas MCTP2 wasprimarily detectable in heart muscle and testis (Fig. 6B). Theimmunoblotting results for MCTP1 are likely to be an accuratereflection of the protein levels because two different antibodies,raised to non-overlapping MCTP1 epitopes, gave the same re-sults, whereas only one of the two MCTP2 antibodies wasusable. MCTP1 in heart muscle appears to be slightly smallerthan in skeletal muscle, possibly because of alternative splic-ing. Because of the limited sensitivity of the antibodies, the

FIG. 2—continued

FIG. 3. Analysis of the first (TM1) and second (TM2) putativetransmembrane regions of MCTP2. Constructs encoding the pro-teins schematically diagrammed on the left were transfected intoHEK293 cells and analyzed by subcellular fractionation and immuno-blotting as shown on the right. In the constructs containing only asingle TMR of MCTP2, the junctions were designed to correspondprecisely to alternatively spliced variants that could arise by the inclu-sion or exclusion of in-frame exons (see gene structures described be-low). Such variants were detected in expressed sequence tag data banksfor MCTP1 and MCTP2 lacking TM1 (see Fig. 2B). Transfected cellswere lysed and separated into soluble and particulate fractions bycentrifugation and analyzed by immunoblotting for EYFPs as shown onthe right, using EYFP alone as a control.

FIG. 4. Localizations of EYFP-MCTP2 fusion proteins in trans-fected HEK293 cells. EYFP-MCTP2 fusion proteins encoded by theconstructs described in Fig. 3 were expressed in HEK293 cells, and thecells were labeled with FM5-95 to selectively stain the plasma mem-brane. Images shown were obtained in a confocal microscope to visual-ize the FM5-95 fluorescence (left column, red) and EYFP fluorescence(middle column, green). Two examples are shown for each construct.Scale bar at the bottom of the right column (Merged) (2 �m) applies toall sections.

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins 1645

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

results do not exclude the possibility of a low amount of expres-sion of MCTP1 in non-muscle tissues. Nevertheless, the resultsclearly demonstrate that by far the highest levels of MCTPs arepresent in excitable muscle cells and that MCTPs are notenriched in brain, another tissue with a large number of excit-able cells.

Structure of the MCTP C2 Domains—Because the threeclosely spaced C2 domains are the major feature of the cyto-plasmic sequences of MCTPs, these domains presumably de-termine the function of these proteins. RNA interference ex-periments in C. elegans revealed that the MCTP homolog(1H206) is an essential gene and that its ablation leads to early

FIG. 5. Structures of the humanMCTP1 and MCTP2 genes and loca-tion of introns in the domain struc-ture. Gene structures were reconstructedfrom the sequences of the human genomebased on the assembled cDNA sequence.For each gene, the nucleotide scale andchromosomal localizations are shown atthe top, and the distribution of exons inthe genomic DNA is shown in the middle.At the bottom the domain structures ofMCTPs are depicted with the position ofthe introns indicated by vertical arrows;the numbers above each arrow designatethe location in the codon where the introndisrupts the coding sequence (0, intron isexactly between codons; 1, intron is afterfirst nucleotide of a codon; 2, intron isafter the second nucleotide of a codon).Note that most exons in the MCTP genesare flanked by introns that disrupt thecodons in the same frame; these exonscan, in principle, be alternatively splicedwithout a loss of reading frame. Preciselocations of exons and sequences of exon/intron junctions are shown in Tables Iand II. aa, amino acids.

TABLE IGenomic organization of the human MCTP1 gene (5q15)

Data are based on the analysis of human genome sequences in the Celera and GenBankTM databases. Nucleotide numbers correspond to thoseof the assembled Celera genome. UTR, untranslated region. Initiator and stop codons are boldfaced and underlined.

Exon no. Nucleotide no. Size (bp) Intron/exon junction sequences

1a 2934151–2933432 720 5�-UTR �����ATG,GAG,CCC,CGG,—GGC,AGC,AGC,CAG,gtgagtgccggt1b 2731058–2731002 57 ATG,CTA,GAC,AGC,—ATT,TGT,AAT,AAG,gtaagtaacatt2 2667060–2666943 118 tgttttccagAAA,ATA,ATA,AAC,—CGA,GAT,CGA,GGA,Ggtaagagcata3 2602937–2602795 143 tatttaacagGG,ACG,AGT,GAT,—TTG,TAT,ATA,AAG,gtgagccatttat4 2592004–2591924 81 ggcatctcagGTA,TTT,GAC,TAT,—GAG,TTA,AAC,AGgtacttttgaactg5 2589771–2589660 112 tctatcacagG,CCC,ACA,GAT,GTG,—TCC,AGG,GAT,GTG,gtaagttccc6 2581568–2581520 49 ttatttacagACA,ATG,CTA,ATG,—AGA,TCA,AGT,AAG,gtaattcttagc7 2573598–2573539 60 ttgtttctagGAA,CTT,TCA,GAA,—CTC,TTT,TGG,AGG,gtaatgtctaaa8 2567550–2567473 78 ttccatccagACG,TGC,GGC,AGG,—AAA,AAT,GTA,CAA,gtaagttcaacc9 2562554–2562383 172 ttgccaaaagTTT,CAG,ACC,CAA,—TAC,AAG,AGC,AAGgtaaccttgtctc

10 2558958–2558828 131 ttttttccagATT,ATG,CCA,AAA,—GAT,TTC,ATT,GGC,AGgtttgtgcaa11 2544413–2544226 188 ttgtctctagG,TGC,CAG,GTC,GAC,—TTA,AAG,AGA,TAT,gtacgtatgt12 2538549–2538456 94 ggtctttcagAGC,CCA,TTG,AGG,—GCC,GAC,GTC,ACT,Ggtaagagatga13 2522819–2522716 104 tctcttgtagGA,AAA,AGT,GAC,CCA,—GTC,TTC,ACG,TTgtaagtaggca14 2520994–2520892 103 tattatgcagC,AAC,ATT,AAA,GAT,—CCA,TTG,CTG,TCT,aagttttatt15 2520550–2520449 102 tttcccgcagATT,CAA,AAT,GGT,—ATT,TTT,AAT,GCT,gtaagtcttaac16 2520068–2519994 75 tgatgtatagGTG,AAA,GCC,AGC,—CTC,TCT,AAA,CAG,gtaaaaagcaag17 2518029–2517910 120 ttccttacagCTG,CTA,CTA,AGA,—GCT,GCT,TTT,GTG,gtatgatcatac18 2448709–2448590 120 tctcctccagCTC,TTT,CTC,TTT,—CAA,CGT,GAT,ACA,gtaagtctctac19 2428740–2428687 54 ctctctgcagGTA,GTG,GAG,GAC,—AAA,GAT,GAC,AAG,gtaattatcttt20 2364464–2364354 111 tgcattttagGAC,AGT,GAA,AAA,—GAA,AGG,AT(T/A),AAG,AAgtgagt21 2360504–2360395 110 ttccattcagT,ACT,TTC,AAC,TGG,—GTC,CTT,GTC,TGG,Ggtaagtaaa22 2358186–2358089 98 tttgtttcagGC,ATC,AAT,AAA,—GAT,GTA,CAA,GTG,gtatgtaggtttc23 2357144–2357073 69 tctgtttaaacagGTG,CAA,TAC,CAA,—AAT,CTT,GGC,tag��� 3�-UTR

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins1646

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

embryonic lethality (27). Most, but not all, C2 domains functionas Ca2�-binding modules (reviewed in Ref. 28). As a first ap-proach toward understanding the function of MCTPs, we there-fore studied its C2 domains. Specifically, we investigated thetwo major activities associated with C2 domains, namely Ca2�

binding and interaction with phospholipid membranes.Alignment of the C2 domain sequences of MCTPs reveals

that all three C2 domains belong to class 2 C2 domains. Class 1and 2 C2 domains have similar structures composed of aneight-stranded �-sandwich but differ in strand topology. Inclass 2 C2 domains, the �-strand corresponding to the first�-strand of class 1 C2 domains is transplanted to the end of the C2

domain, making this the last �-strand (referred to as �1�; see Fig.5). As a result, the topology of class 2 C2 domains represents acircular permutation of the topology of class 1 C2 domains (28),with the N- and C termini being on the top of the domain in class1 C2 domains and at the bottom of the domain in class 2 C2

domains. Although all MCTP C2 domains belong to class 2, theyare otherwise not very similar (Fig. 7). The C2A, C2B, and C2Cdomains are well conserved between various MCTP isoforms, butthe degree of identity between C2 domains is low.

Most C2 domains bind Ca2� with a low intrinsic bindingaffinity in the absence of phospholipid membranes but with ahigh apparent affinity in the presence of phospholipid mem-branes. This behavior probably results from the fact that thecoordination spheres for bound Ca2� ions are incomplete inthese C2 domains in the absence of phospholipids but are com-pleted by the phospholipid head groups. The Ca2� bindingmode of C2 domains has been best characterized for the C2

domain from protein kinase C (29, 30), the C2A and C2B do-mains of synaptotagmin 1 (5, 14, 22), and the C2 domain fromphospholipase C� (31). In all of the C2 domains studied, Ca2� isbound exclusively by the loops emerging from the top of the�-sandwich. Ca2� is coordinated by five aspartate, glutamate,

TABLE IIGenomic organization of the human MCTP2 gene (15q26)

Data are based on the analysis of human genome sequences in the Celera and GenBank databases. Nucleotide numbers correspond to those ofthe assembled Celera genome. UTR, untranslated region. Initiator and stop codons are boldfaced and underlined.

Exon no. Nucleotide no. Size (bp) Intron/exon junction sequences

1 9807022–9807486 465 5�-UTR �����ATG,GAT,CTG,GAT,—GAA,GAG,CCA,GAG,gtgagaataggg2 9823038–9823100 63 ttctttgcagAAG,CTA,TGT,GGA,—GAA,GAG,CCA,GAG,gtgagtggcatt3 9824285–9824393 109 atctgtgcagGTA,CCG,GGG,GAA,—CGA,GAT,CGC,TGT,Ggtaagacctgg4 9848046–9848188 143 ccttttaaagGC,ACA,AGT,GAT,—CTA,CGT,GTG,AAG,gtaatcacagata5 9848955–9849031 77 ctctttgtagGTA,TAT,GAT,CGA,—GAG,CTT,AAC,AGgtaccgtattttta6 9849569–9849680 112 ttgcttgtagA,ACA,ACT,GAA,CAT,—TTC,AAG,AGA,CAC,gtaagtggga7 9853885–9853920 36 aaatctttagCGT,TGG,TCA,AAT,—AGT,GCC,AGC,AAG,gtaaatatactt8 9864893–9865057 165 tttgctttagTCC,TCT,TTG,ATA,—TAT,AAA,AGT,AAG,gtaagttccatc9 9867238–9867368 131 tgttttctagACA,CTG,TGT,AAG,—GAA,CGT,CTG,GGC,ACgtgagtcccc

10 9876361–9876547 187 acttttctag(A/G),TGT,AAA,GTG,—ACC,CAG,CGA,TATgtgagtgtttt11 9878843–9878936 94 caaccctcagTGC,TTA,CAG,AAC,—GCA,GAT,TTC,TCA,Ggtacaggacat12 9892778–9892880 103 atctttccagGG,AAG,AGT,GAC,—AAA,GTT,TTT,ACA,TTgtaagtgcttt13 9894179–9894281 103 tttgttacagT,CCC,ATT,AAA,GAT,—CCC,TTG,CTG,TCC,gtaagtttcc14 9907717–9907818 102 tttgtgacagATT,AGA,GAT,GGA,—ATA,TAT,AAT,CCG,gtaagtctagct15 9908677–9908751 75 tcttttctagGTG,AAA,GCA,AGT,—CTG,TCC,AAA,AAG,gtgggtcgctac16 9910656–9910775 120 ttaaaatcagATC,TTA,TCA,AGA,—ATA,GCA,TTC,GCG,gtaagcttcctt17 9948932–9949054 123 tttcaatcagGTA,TTT,TTG,ATC,—CAG,GAC,AGC,CAG,gtaagcaaggat18 9951675–9951716 42 ttcctttcagGAG,AGC,ACA,GAC,—GAA,GAT,GAC,AAG,gtgcgtatgttc19 9966893–9967002 110 tgttttctagGAA,TCT,GAG,AAA,—GAA,AGG,ATT,AAG,AAgtaagttcta20 9979089–9979198 110 ttgtctacagC,ACA,TTT,AAC,TGG,—TTA,ATC,TGG,Ggtaagtttggaat21 9985452–9985549 98 tatttttcagGC,ATA,AAT,AAA,—GAT,GTT,CAA,AAG,gtatgtaatgaat22 9987722–9987790 66 ttgttttcacagGTG,CAG,TAT,GCA,—AGC,GCT,CTC,tag��� 3�-UTR

FIG. 6. Antibodies to MCTPs andthe tissue distribution of MCTP1 andMCTP2 expression. A, location of anti-body epitopes in the domain structure ofMCTPs. Two different epitopes were usedfor raising antibodies to each MCTP asindicated. B, immunoblots of rat tissueswith the two independent MCTP1 anti-bodies (left), one of the two MCTP2 anti-bodies (top right), and valosin-containingprotein (VCP) antibodies (bottom right;used as a loading control). Positions ofmolecular mass markers are shown onthe left. Note that the two independentMCTP1 antibodies result in exactly thesame pattern of reactivity on the multi-tissue immunoblots despite being raisedto distinct epitopes, suggesting thatMCTP1 is primarily expressed in skeletalmuscle and, to a lesser extent, in heart.The second MCTP2 antibody was not suc-cessful (not shown).

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins 1647

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

or asparagine residues that are present in the top loops. All ofthese residues are conserved in the MCTP C2 domains, includ-ing the Drosophila and C. elegans MCTP C2 domains, except forthe C2B domain of MCTP2, which lacks two of the five Ca2�-binding residues (Fig. 7). In addition, the other conserved res-idues of the Ca2�-binding loops of C2 domains are also retainedin the MCTP domains (e.g. the typical GXSD sequence of thefirst top loop that is also present in all synaptotagmins) (28),indicating that these domains have the expected features ofCa2�/phospholipid binding domains. However, recent results(32) revealed that the Ca2� binding properties of C2 domainscannot be predicted from sequence analyses because the C2Bdomain of synaptotagmin 4, despite a perfect Ca2�-bindingconsensus sequence, exhibits no intrinsic or phospholipid-de-pendent Ca2�-binding. This raises the question of whetherMCTPs indeed function as Ca2�-binding molecules.

Phospholipid Binding Properties of the MCTP C2 Do-mains—We first tested whether the MCTP C2 domains bind tophospholipids similarly as other C2 domains. We produced allthree C2 domains of MCTP1 and the C2C domain of MCTP2 asrecombinant GST fusion proteins. We then tested the bindingof these domains to liposomes with five different phospholipidcompositions that included neutral phospholipids, phosphati-dylinositol phosphates, and negatively charged phospholipids(Fig. 8). Binding was examined in the presence of EGTA, Ca2�,Sr2�, and Ba2� using a sensitive centrifugation assay thatdetects Ca2�-dependent phospholipid binding by measuringthe amount of C2 domain protein that can be isolated withliposomes in the presence of Ca2� but not EGTA (23). Thisbinding assay is important in assessing the ability, or lackthereof, of C2 domains to interact with phospholipids, becausethe pull-down assay that we originally developed (4) and hasbeen widely adopted does not detect weaker interactions suchas the binding of the synaptotagmin C2B domain (14). Surpris-ingly, using the centrifugation assay we detected no significantCa2�-dependent binding of any MCTP C2 domain to any of thephospholipid membranes (Fig. 8). Some weak binding was ob-served, especially for the C2C domain of MCTP1, but thisbinding never reached the level observed with typical phospho-lipid-binding C2 domains. This result indicates that MCTPs arenot phospholipid-binding molecules for either charged or neu-tral phospholipids, raising the question of whether they are atall involved in Ca2�-binding (32).

Intrinsic Ca2� Binding to MCTP C2 Domains—To monitorCa2� binding to MCTP1 C2 domains, we purified the recombi-nant C2 domains as GST fusion proteins, cleaved them off theGST moiety with thrombin, and examined their intrinsic tryp-tophan fluorescence as a function of Ca2� (Fig. 9). This exper-iment was prompted by the observation that all MCTP C2

domains contain a tryptophan in the middle of �-strand 5 (seeFig. 7) and that the C2B domains additionally contain a tryp-tophan in top loop 3 and in the bottom N-terminal sequence.

The tryptophan fluorescence spectra of the C2 domains ex-hibited characteristic differences in the number of fluorescencemaxima and their emission wavelength. In every C2 domain,however, the addition of a saturating concentration of Ca2�

increased the intrinsic tryptophan fluorescence in a mannerthat was fully reversible upon the addition of excess EGTA(Fig. 9). The Ca2�-induced fluorescence increase was quitelarge (�10%) for the C2A and the C2B domains and smaller forthe MCTP1 C2C domain. For this reason we also studied theC2C domain from MCTP2, because the C2C domain of thisisoform contains an extra tryptophan (Fig. 7), and we includedthis domain in all other assays as well. Indeed, the MCTP2 C2Cdomain does exhibit a significantly larger Ca2�-dependent flu-orescence increase (Fig. 9). Ca2� induced no shift in emissionmaxima, and Mg2� had no effect on any tryptophan fluores-cence property of the C2 domains (data not shown).

We next exploited the tryptophan fluorescence changes totitrate Ca2� binding (Fig. 10). Saturable Ca2�-dependent in-creases in tryptophan fluorescence were observed for all C2

domains, with half-maximal changes between 1.3 and 2.3 �M

free Ca2�. The signal-to-noise ratio was robust for the C2A andC2B domains, but the changes in the C2C domains were rathersmall (Fig. 10). However, multiple independent experimentsprovided similar results, indicating that the changes observedcorrespond to a reliable Ca2� binding event.

Effect of Purified MCTP C2 Domains on Ca2�-triggered Se-cretion from Permeabilized PC12 Cells—The properties of theMCTP C2 domains, high affinity intrinsic Ca2� binding and nophospholipid binding, are unexpected. The apparent Ca2� af-finity of the C2 domains approximately corresponds to that ofsecretion from PC12 cells. In previous studies we have shownthat C2 domains with an apparent high Ca2� affinity in phos-pholipid complexes potently inhibit Ca2�-induced exocytosisfrom permeabilized PC12 cells (23). However, for traditional C2

FIG. 7. Comparison of the C2 domain sequences of MCTP1 and MCTP2. The C2 domain sequences of the human MCTPs are aligned.Residues that are shared among at least three of the six sequences are highlighted with a color code that reflects the predicted secondary structureof the C2 domains as follows: blue, �-strands (labeled on top; note that because all of the C2 domains are class 2 C2 domains, the numbering startswith �2 and ends with �1�); yellow, top loops; green, bottom loops. In addition, presumptive Ca2�-binding ligands are marked by a white typefaceon a black background, and tryptophans that contribute to the intrinsic fluorescence of the domains are shown on a red background.

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins1648

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

domains, altering phospholipid binding always leads to achange in apparent Ca2� affinity, because phospholipid bind-ing and Ca2� binding are interdependent (33). Because theMCTP C2 domains have the requisite Ca2� affinity to interferewith PC12 cell exocytosis but do not bind phospholipids, theymay still function in Ca2�-dependent exocytosis. To address thispossibility, we compared the effect of purified MCTP C2 domainswith those of synaptotagmin 7 C2 domains (which are potentinhibitors of exocytosis) on Ca2�-induced secretion in permeabi-lized PC12 cells (Fig. 11). We observed a strong inhibitory effect

of the synaptotagmin 7 C2 domains on secretion but no effect bythe MCTP C2 domains. These results indicate that consistentwith the tissue distribution of MCTPs (Fig. 6), MCTPs are notcomponents of the secretory machinery. Furthermore, these re-sults show that for inhibition in the permeabilized PC12 cellassay, a high Ca2� binding affinity is not sufficient for an effect.

DISCUSSION

In the present study we describe a novel family of C2 domainproteins that exhibit unusual properties indicative of a role in

FIG. 8. Lack of Ca2�-dependent phospholipid binding by MCTP C2 domains. Purified recombinant C2 domain GST fusion proteins weretested with liposomes of the composition described below each set of gel images (PE, phosphatidylethanolamine; PC, phosphatidylcholine; PIP2,phosphatidylinositol bisphosphate; PIP, phosphatidylinositol phosphate; PS, phosphatidylserine). C2 domains were incubated with the liposomesin the presence of the concentration of divalent cations indicated at the top. Afterward, liposomes were isolated by centrifugation through a sucrosecushion, and proteins bound to the liposomes were examined by SDS-PAGE and Coomassie Blue staining. Images shown are CoomassieBlue-stained gels from a representative experiment repeated three times. Note that even very sticky, highly charged liposomes (e.g. liposomescontaining 5% PIP2) were unable to bind MCTP C2 domains.

FIG. 9. Fluorescence spectra of puri-fied recombinant C2 domains fromMCTPs and Ca2�-dependent changes.Intrinsic fluorescence spectra were re-corded of 3 �M purified C2 domains lackingthe GST moiety in 40 mM Tris-HCl, pH 8,0.1 M NaCl, and 0.5 mM sodium EGTA atan excitation wavelength of 282 nm. Afterthe initial spectra were measured, a secondspectrum of the same sample was obtainedupon the addition of 100 �M free Ca2�, anda third spectrum was recorded upon thefurther addition of 1 mM sodium EGTA.Data shown are from a representative ex-periments independently repeated multi-ple times.

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins 1649

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

Ca2� signaling. These properties, which differentiate MCTPsfrom other previously studied C2 domain proteins, are de-scribed in the next two paragraphs.

First, the architecture of the MCTPs is unique in that theyare composed of a large, presumptively cytoplasmic sequenceprimarily composed of three class 2 C2 domains and two TMRs.With the MCTPs there are now four families of membrane-anchored multiple C2 domain proteins (Fig. 1). MCTPs are theonly membrane-bound C2 domain proteins that contain twofunctional TMRs (Figs. 3 and 4).

Second, the functional properties of MCTPs are unique inthat they bind Ca2� but not phospholipids. We only showed thisfor MCTP1 and for the C2C domain of MCTP2, but the C2Adomain of MCTP2 is highly homologous to that of MCTP1 andthus is likely to also bind Ca2�. The C2B domain of MCTP2,

however, lacks the consensus sequence for C2 domain Ca2�-binding sites (Fig. 7) and is unlikely to bind Ca2�. InvertebrateMCTPs from C. elegans and D. melanogaster also contain allCa2�-binding site sequences (Fig. 2A), consistent with the no-tion that Ca2� binding is evolutionarily conserved and that theC2B domain of MCTP2 is the only C2 domain of an MCTP thatdoes not bind Ca2�.

With these properties, MCTPs are veritable Ca2�-bindingmachines in which three Ca2�-binding C2 domains are at-tached to the membrane. The high affinity of all three MCTP1C2 domains for Ca2� without phospholipid binding was unex-pected. We recently showed that the C2B domain of synapto-tagmin 4 contains a perfect predicted Ca2�-binding site with ahigh degree of sequence identity to that of the C2B domain ofsynaptotagmin 1, but, nevertheless, it does not bind Ca2� (32);thus, the presence of a predicted Ca2�-binding site is not suf-ficient to deduce the Ca2� binding properties of a C2 domain. Inextension of this conclusion, the current data indicate that thepresence of a consensus Ca2�-binding site in a C2 domain alsodoes not allow the prediction of the mode of Ca2� binding. Thestandard mode for C2 domains is to bind Ca2� in a complexwith phospholipids in which the phospholipid head groups areessential to complete the coordination spheres of the boundCa2� ions (33). The present data show that this mode of Ca2�

binding is not universally true in that the MCTP Ca2�-bindingsites must have reasonably complete coordination spheres forCa2� in the absence of phospholipids. The sequences of theMCTP C2 domains, however, provide no clue as to how this mightwork. It is possible that in the top loops of the C2 domains,negatively charged residues, in addition to the canonical fiveaspartates, contact the Ca2� ions. However, no conserved, addi-tional negatively charged residues are present that could fulfillthis role. Alternatively, it is possible that Ca2� ions bind betweentwo C2 domains, with the coordination sphere formed by residuesfrom the top loops of two different C2 domains.

The Ca2� binding properties and structures of MCTPs dem-onstrate that these proteins function in Ca2� signaling at themembrane. Elucidating the nature of this function will requirea genetic approach that builds on the observation that inC. elegans MCTP is an essential gene whose ablation causesembryonic lethality (27). The essential nature of MCTP is con-sistent with an important role in Ca2� signaling that could, in

FIG. 10. Ca2� titration of the trypto-phan fluorescence of purified recom-binant C2 domains from MCTPs. In-trinsic fluorescence of the indicatedpurified C2 domains from MCTPs (3 �M)was monitored as function of the freeCa2� concentration (excitation, 282 nm;emission of the MCTP1 C2A domain, 332nm; emission of the MCTP1 C2B domain,344 nm; emission of the MCTP1 C2C do-main, 328 nm; emission of the MCTP2C2C domain, 328 nm). Ca2� concentra-tions were clamped with Ca2�-EGTAbuffers (see “Materials and Methods”).Data shown are means � S.D. (n � 3 forC2A and C2B domains of MCTP1; n � 7for C2C domains from MCTP1 andMCTP2; each experiment was performedin triplicate). The numbers indicated dis-play the Ca2� concentration that caused ahalf-maximal fluorescence change, as cal-culated by curve fitting using GraphPadPRISM version 3.02 software.

FIG. 11. Effect of MCTP C2 domains on Ca2�-induced exocyto-sis from permeabilized PC12 cells. PC12 cells were loaded with3H-labeled norepinephrine, cracked, and preincubated with 6 �M puri-fied C2 domain GST fusion proteins. Exocytosis was measured as theamounts of norepinephrine released during a 30-min incubation periodafter the addition of a buffer containing no Ca2� or 10 �M Ca2�. Datashown are means � S.D. (n � 3; each experiment was performedin duplicate).

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins1650

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

principle, consist of a Ca2�-controlled regulatory function or anactivity as a Ca2� buffer. Although a Ca2� buffer functioncannot be excluded, a Ca2� regulatory function appears morelikely. With the generally fast Ca2� binding, the relatively lowCa2� affinities, and the obligatory formation of phospholipidcomplexes by C2 domains (reviewed in Ref. 28), C2 domainproteins are more suited than EF-hand proteins for some Ca2�

regulatory functions and less suited for Ca2� buffering func-tions. In the case of MCTPs, however, this assumption may notapply because of their relatively high Ca2� affinity in theabsence of phospholipids. In addition, MCTPs are exceptionalamong C2 domain proteins because they contain two TMRs, oneof which (TMR1) may be spliced in and out in both isoforms(Fig. 2B). Alternative splicing that removes the first TMRwould retain the membrane-bound nature of MCTPs, becausethe second TMR is sufficient to anchor the cytoplasmic C2

domains to the membrane (see Figs. 3 and 4) but would convertthe normally cytoplasmic C-terminal sequence of MCTP into anextracellular sequence (see Fig. 1).

It is interesting that of the four families of proteins thatcontain both C2 domains and a TMR, the two families that havebeen functionally studied (synaptotagmins and ferlins) are in-volved in membrane fusion (reviewed in Refs. 15 and 16).Members of the third family, referred to here as extendedsynaptotagmins or E-Syts because of their structural similarityto synaptotagmins (Fig. 1), have not been studied beyond thecloning of one of their isoforms, and the fourth family consistsof the MCTPs examined here. The fact that synaptotagminsand ferlins are involved in membrane traffic strongly supportsthe notion that MCTPs might also act in membrane traffic.Furthermore, the presence of two evolutionarily conservedTMRs, when one would have been sufficient to anchor the C2

domains to the membrane, and the presence of three C2 do-mains with distinct conserved sequences argue against a Ca2�

buffer function. Again, genetic experiments will have to ad-dress this important issue.

Acknowledgments—We thank Ms. I. Kornblum, A. Roth, andE. Borowicz for technical assistance.

REFERENCES

1. Coussens, L., Parker, P. J., Rhee, L., Yang-Feng, T. L., Chen, E., Waterfield,M. D., Francke, U., and Ullrich, A. (1986) Science 233, 859–866

2. International Human Genome Sequence Consortium (2001) Nature 409,860–921

3. Perin, M. S., Fried, V. A., Mignery, G. A., Jahn, R., and Sudhof, T. C. (1990)Nature 345, 260–263

4. Davletov, B. A., and Sudhof, T. C. (1993) J. Biol. Chem. 268, 26386–263905. Sutton, R. B., Davletov, B. A., Berghuis, A. M., Sudhof, T. C., and Sprang, S. R.

(1995) Cell 80, 929–9386. Lee, J. O., Yang, H., Georgescu, M. M., Di Cristofano, A., Maehama, T., Shi, Y.,

Dixon, J. E., Pandolfi, P., and Pavletich, N. P. (1999) Cell 99, 323–3347. Betz, A., Thakur, P., Junge, H. J., Ashery, U., Rhee, J. S., Scheuss, V.,

Rosenmund, C., Rettig, J., and Brose, N. (2001) Neuron 30, 183–1968. Schoch, S., Castillo, P. E., Jo, T., Mukherjee, K., Geppert, M., Wang, Y.,

Schmitz, F., Malenka, R. C., and Sudhof, T. C. (2002) Nature 415, 321–3269. Wang, Y., and Sudhof, T.C. (2003) Genomics 81, 126–137

10. Wang, X., Kibschull, M., Laue, M. M., Lichte, B., Petrasch-Parwez, E., andKilimann, M. W. (1999) J. Cell Biol. 147, 151–162

11. Fenster, S. D., Chung, W. J., Zhai, R., Cases-Langhoff, C., Voss, B., Garner,A. M., Kaempf, U., Kindler, S., Gundelfinger, E. D., and Garner, C. C. (2000)Neuron 25, 203–214

12. Pucharcos, C., Casas, C., Nadal, M., Estivill, X., and de la Luna, S. (2001)Biochim. Biophys. Acta. 1521, 1–11

13. Sugita, S., Shin, O.-H., Han, W., Lao, Y., and Sudhof, T.C. (2002) EMBO J. 21,270–280

14. Fernandez, I., Arac, D., Ubach, J., Gerber, S. H., Shin, O.-H., Gao, Y., Ander-son, R. G. W., Sudhof, T. C., and Rizo, J. (2001) Neuron 32, 1057–1069

15. Sudhof, T. C. (2002) J. Biol. Chem. 277, 7629–763216. Bansal, D., and Campbell, K. P. (2004) Trends Cell Biol. 14, 206–21317. Lewit-Bentley, A., and Rety, S. (2000) Curr. Opin. Struct. Biol. 10, 637–64318. Nakayama, N., Kawasaki, H., and Kretsinger, R. (2000) Top. Biol. Inorg.

Chem. 3, 29–5819. Toutenhoofd, S. L., and Strehler, E. E. (2000) Cell Calcium 28, 83–9620. Chin, D., and Means, A. R. (2000) Trends Cell Biol. 10, 322–32821. Guan, K. L., and Dixon, J. E. (1991) Anal. Biochem. 192, 262–26722. Ubach, J., Lao, Y., Fernandez, I., Arac, D., Sudhof, T. C., and Rizo, J. (2001)

Biochemistry 40, 5854–586023. Shin, O.-H., Rizo, J., and Sudhof, T.C. (2002) Nat. Neurosci. 5, 649–65624. Laemmli, U. K. (1970) Nature 227, 680–68525. Johnston, P. A., Jahn, R., and Sudhof, T. C. (1989) J. Biol. Chem. 264,

1268–127326. Morris, N. J., Ross, S. A., Neveu, J. M., Lane, W. S., and Lienhard, G. E. (1999)

Biochim. Biophys. Acta 1431, 525–53027. Maeda, I., Kohara, Y., Yamamoto, M., and Sugimoto, A. (2001) Curr. Biol. 11,

171–17628. Rizo, J., and Sudhof, T. C. (1998) J. Biol. Chem. 273, 15879–1588229. Shao, X., Davletov, B. A., Sutton, R. B., Sudhof, T. C., and Rizo, J. (1996)

Science 273, 248–25130. Verdaguer, N., Corbalan-Garcia, S., Ochoa, W. F., Fita, I., and Gomez-Fernan-

dez, J. C. (1999) EMBO J. 18, 6329–633831. Essen, L. O., Perisic, O., Lynch, D. E., Katan, M., and Williams, R. L. (1997)

Biochemistry 36, 2753–276232. Dai, H., Shin, O.-H., Machius, M., Tomchick, D. R., Sudhof, T. C., and Rizo, J.

(2004) Nat. Struct. Mol. Biol. 11, 844–84933. Fernandez-Chacon, R., Konigstorfer, A., Gerber, S. H., Garcıa, J., Matos, M. F.,

Stevens, C. F., Brose, N., Rizo, J., Rosenmund, C., and Sudhof, T. C. (2001)Nature 410, 41–49

Ca2� Binding by Novel Membrane-bound C2 Domain Proteins 1651

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

Ok-Ho Shin, Weiping Han, Yun Wang and Thomas C. Südhof Binding Properties2+Regions (MCTPs) and Unusual Ca

Domain Proteins with Two Transmembrane2Evolutionarily Conserved Multiple C

doi: 10.1074/jbc.M407305200 originally published online November 4, 20042005, 280:1641-1651.J. Biol. Chem. 

  10.1074/jbc.M407305200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/280/2/1641.full.html#ref-list-1

This article cites 33 references, 9 of which can be accessed free at

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from