Advertisement

Conserved biophysical features of the CaV2 presynaptic Ca2+ channel homologue from the early-diverging animal Trichoplax adhaerens

Open AccessPublished:October 23, 2020DOI:https://doi.org/10.1074/jbc.RA120.015725
      The dominant role of CaV2 voltage-gated calcium channels for driving neurotransmitter release is broadly conserved. Given the overlapping functional properties of CaV2 and CaV1 channels, and less so CaV3 channels, it is unclear why there have not been major shifts toward dependence on other CaV channels for synaptic transmission. Here, we provide a structural and functional profile of the CaV2 channel cloned from the early-diverging animal Trichoplax adhaerens, which lacks a nervous system but possesses single gene homologues for CaV1–CaV3 channels. Remarkably, the highly divergent channel possesses similar features as human CaV2.1 and other CaV2 channels, including high voltage–activated currents that are larger in external Ba2+ than in Ca2+; voltage-dependent kinetics of activation, inactivation, and deactivation; and bimodal recovery from inactivation. Altogether, the functional profile of Trichoplax CaV2 suggests that the core features of presynaptic CaV2 channels were established early during animal evolution, after CaV1 and CaV2 channels emerged via proposed gene duplication from an ancestral CaV1/2 type channel. The Trichoplax channel was relatively insensitive to mammalian CaV2 channel blockers ω-agatoxin-IVA and ω-conotoxin-GVIA and to metal cation blockers Cd2+ and Ni2+. Also absent was the capacity for voltage-dependent G-protein inhibition by co-expressed Trichoplax Gβγ subunits, which nevertheless inhibited the human CaV2.1 channel, suggesting that this modulatory capacity evolved via changes in channel sequence/structure, and not G proteins. Last, the Trichoplax channel was immunolocalized in cells that express an endomorphin-like peptide implicated in cell signaling and locomotive behavior and other likely secretory cells, suggesting contributions to regulated exocytosis.
      Voltage-gated Ca2+ (CaV) channels serve essential functions in excitable cells, imparted by their capacity to translate electrical signals carried by Na+ and K+ channels, into cytoplasmic Ca2+ signals (
      • Clapham D.E.
      Calcium signaling.
      ). For example, CaV channels couple membrane excitation with fusion of presynaptic vesicles, muscle contraction, alterations in nuclear gene expression, and regulation of ciliary beating (
      • Catterall W.A.
      Voltage-gated calcium channels.
      ,
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ). CaV channels belong to a large family of pore-loop (P-loop) channels that includes voltage-gated Na+ (NaV) channels and K+ (KV) channels (
      • Pozdnyakov I.
      • Matantseva O.
      • Skarlato S.
      Diversity and evolution of four-domain voltage-gated cation channels of eukaryotes and their ancestral functional determinants.
      ), named after their four extracellular loop structures that come together in the pore to form the ion selectivity filter, a motif uniquely configured in different channels for selecting Ca2+, Na+, or K+ ions (
      • MacKinnon R.
      Pore loops: an emerging theme in ion channel structure.
      ). Humans and related animals possess three types of CaV channels, broadly classified as high and low voltage–activated, the former requiring strong depolarization for activation (i.e. CaV1 or L-type channels and CaV2 or N-, P-/Q-, and R-type channels) and the latter requiring only mild, sub-threshold depolarization (i.e. CaV3 or T-type channels) (
      ). Phylogenomic studies have established that most animals possess single gene copies of CaV1–CaV3 channels, whereas gene duplications in vertebrates gave rise to four CaV1 channels (CaV1.1–CaV1.4), three CaV2 channels (CaV2.1–CaV2.3), and three CaV3 channels (CaV3.1–CaV3.3) (
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ,
      • Pozdnyakov I.
      • Matantseva O.
      • Skarlato S.
      Diversity and evolution of four-domain voltage-gated cation channels of eukaryotes and their ancestral functional determinants.
      ,
      • Catterall W.A.
      • Perez-Reyes E.
      • Snutch T.P.
      • Striessnig J.
      International Union of Pharmacology. XLVIII. Nomenclature and structure-function relationships of voltage-gated calcium channels.
      ,
      • Moran Y.
      • Zakon H.H.
      The evolution of the four subunits of voltage-gated calcium channels: ancient roots, increasing complexity, and multiple losses.
      ,
      • Jeziorski M.C.
      • Greenberg R.M.
      • Anderson P.
      The molecular biology of invertebrate voltage-gated Ca2+ channels.
      ,
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ,
      • Spafford J.D.
      • Dunn T.
      • Smit A.B.
      • Syed N.I.
      • Zamponi G.W.
      In vitro characterization of L-type calcium channels and their contribution to firing behavior in invertebrate respiratory neurons.
      ). Teleosts have had a further duplication of CaV channel genes, with species like Danio rerio having seven CaV1, six CaV2, and five CaV3 genes (
      • Haverinen J.
      • Hassinen M.
      • Dash S.N.
      • Vornanen M.
      Expression of calcium channel transcripts in the zebrafish heart: dominance of T-type channels.
      ). Independently, the cnidarians (e.g. jellyfish) duplicated CaV2 and CaV3 channel genes, resulting in a repertoire of a single CaV1 channel, three CaV2 channels, and two CaV3 channels. The earliest diverging animal lineages possess only CaV2 channels (ctenophores), CaV1 channels (sponges), or an evolutionary precursor of CaV1 and CaV2 channels, dubbed CaV1/2 channels (sponges) (
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ,
      • Moran Y.
      • Zakon H.H.
      The evolution of the four subunits of voltage-gated calcium channels: ancient roots, increasing complexity, and multiple losses.
      ,
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ). The most early-diverging animals to possess all three CaV channel types (i.e. CaV1–CaV3) are the placozoans (
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ,
      • Moran Y.
      • Zakon H.H.
      The evolution of the four subunits of voltage-gated calcium channels: ancient roots, increasing complexity, and multiple losses.
      ,
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ), a phylum of simple seawater animals that includes the species Trichoplax adhaerens and Hoilungia hongkongensis (
      • Eitel M.
      • Francis W.R.
      • Varoqueaux F.
      • Daraspe J.
      • Osigus H.-J.
      • Krebs S.
      • Vargas S.
      • Blum H.
      • Williams G.A.
      • Schierwater B.
      • Wörheide G.
      Comparative genomics and the nature of placozoan species.
      ,
      • Srivastava M.
      • Begovic E.
      • Chapman J.
      • Putnam N.H.
      • Hellsten U.
      • Kawashima T.
      • Kuo A.
      • Mitros T.
      • Salamov A.
      • Carpenter M.L.
      • Signorovitch A.Y.
      • Moreno M.A.
      • Kamm K.
      • Grimwood J.
      • Schmutz J.
      • et al.
      The Trichoplax genome and the nature of placozoans.
      ). A unique feature of placozoans is that they lack neurons, synapses, and muscle (
      • Smith C.L.
      • Varoqueaux F.
      • Kittelmann M.
      • Azzam R.N.
      • Cooper B.
      • Winters C.A.
      • Eitel M.
      • Fasshauer D.
      • Reese T.S.
      Novel cell types, neurosecretory cells, and body plan of the early-diverging metazoan Trichoplax adhaerens.
      ,
      • Schierwater B.
      My favorite animal, Trichoplax adhaerens.
      ) and yet bear distinct cell types whose activity is coordinated for the purpose of motile behaviors such as feeding (
      • Smith C.L.
      • Pivovarova N.
      • Reese T.S.
      Coordinated feeding behavior in Trichoplax, an animal without synapses.
      ,
      • Senatore A.
      • Reese T.S.
      • Smith C.L.
      Neuropeptidergic integration of behavior in Trichoplax adhaerens, an animal without synapses.
      ), chemotaxis (
      • Smith C.L.
      • Reese T.S.
      • Govezensky T.
      • Barrio R.A.
      Coherent directed movement toward food modeled in Trichoplax, a ciliated animal lacking a nervous system.
      ,
      • Heyland A.
      • Croll R.
      • Goodall S.
      • Kranyak J.
      • Wyeth R.
      Trichoplax adhaerens, an enigmatic basal metazoan with potential.
      ,
      • Romanova D.Y.
      • Heyland A.
      • Sohn D.
      • Kohn A.B.
      • Fasshauer D.
      • Varoqueaux F.
      • Moroz L.L.
      Glycine as a signaling molecule and chemoattractant in Trichoplax (Placozoa): insights into the early evolution of neurotransmitters.
      ), phototaxis (
      • Heyland A.
      • Croll R.
      • Goodall S.
      • Kranyak J.
      • Wyeth R.
      Trichoplax adhaerens, an enigmatic basal metazoan with potential.
      ), and gravitaxis (
      • Mayorova T.D.
      • Smith C.L.
      • Hammar K.
      • Winters C.A.
      • Pivovarova N.B.
      • Aronova M.A.
      • Leapman R.D.
      • Reese T.S.
      Cells containing aragonite crystals mediate responses to gravity in Trichoplax adhaerens (Placozoa), an animal lacking neurons and synapses.
      ). Notably, despite lacking synapses, increasing evidence suggests that cellular communication in placozoans likely occurs in a protosynaptic manner, where regulated secretion of signaling molecules, such as neuropeptides and small-molecule transmitters, targets membrane receptors on other cells to exert an effect (
      • Senatore A.
      • Reese T.S.
      • Smith C.L.
      Neuropeptidergic integration of behavior in Trichoplax adhaerens, an animal without synapses.
      ,
      • Romanova D.Y.
      • Heyland A.
      • Sohn D.
      • Kohn A.B.
      • Fasshauer D.
      • Varoqueaux F.
      • Moroz L.L.
      Glycine as a signaling molecule and chemoattractant in Trichoplax (Placozoa): insights into the early evolution of neurotransmitters.
      ,
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ,
      • Varoqueaux F.
      • Williams E.A.
      • Grandemange S.
      • Truscello L.
      • Kamm K.
      • Schierwater B.
      • Jékely G.
      • Fasshauer D.
      High cell diversity and complex peptidergic signaling underlie placozoan behavior.
      ).
      In addition to their distinct voltages of activation, CaV channels are distinguished by their differential association with accessory CaVβ and CaVα2δ subunits, which are essential for the proper membrane expression and function of CaV1 and CaV2, but not CaV3 channels (
      • Catterall W.A.
      Voltage-gated calcium channels.
      ,
      ). Furthermore, although their cellular functions overlap in certain contexts, there are several functions for which the different channels have specialized, observed nearly ubiquitously in animals ranging from humans to fruit flies to nematode worms (
      • Catterall W.A.
      Voltage-gated calcium channels.
      ,
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ,
      • Tyson J.R.
      • Snutch T.P.
      Molecular nature of voltage-gated calcium channels: structure and species comparison.
      ). For example, endowed by their broadly conserved low activation voltages, CaV3 channels tend to regulate membrane excitability in neurons and muscle, often in the context of rhythmic excitation, or to boost sub-threshold excitation as occurs in neuron dendrites (
      • Perez-Reyes E.
      Molecular physiology of low-voltage-activated t-type calcium channels.
      ,
      • Senatore A.
      • Spafford J.D.
      Transient and big are key features of an invertebrate T-type channel (LCav3) from the central nervous system of Lymnaea stagnalis.
      ,
      • Smith C.L.
      • Abdallah S.
      • Wong Y.Y.
      • Le P.
      • Harracksingh A.N.
      • Artinian L.
      • Tamvacakis A.N.
      • Rehder V.
      • Reese T.S.
      • Senatore A.
      Evolutionary insights into T-type Ca2+ channel structure, function, and ion selectivity from the Trichoplax adhaerens homologue.
      ,
      • Jeong K.
      • Lee S.
      • Seo H.
      • Oh Y.
      • Jang D.
      • Choe J.
      • Kim D.
      • Lee J.-H.
      • Jones W.D.
      Ca-α1T, a fly T-type Ca2+ channel, negatively modulates sleep.
      ,
      • Steger K.A.
      • Shtonda B.B.
      • Thacker C.
      • Snutch T.P.
      • Avery L.
      The C. elegans T-type calcium channel CCA-1 boosts neuromuscular transmission.
      ,
      • McKay B.E.
      • McRory J.E.
      • Molineux M.L.
      • Hamid J.
      • Snutch T.P.
      • Zamponi G.W.
      • Turner R.W.
      CaV3 T-type calcium channel isoforms differentially distribute to somatic and dendritic compartments in rat central neurons.
      ,
      • Ivanov A.I.
      • Calabrese R.L.
      Intracellular Ca2+ dynamics during spontaneous and evoked activity of leech heart interneurons: low-threshold Ca currents and graded synaptic transmission.
      ,
      • Senatore A.
      • Guan W.
      • Boone A.N.
      • Spafford J.D.
      T-type channels become highly permeable to sodium ions using an alternative extracellular turret region (S5-P) outside the selectivity filter.
      ,
      • Shtonda B.
      • Avery L.
      CCA-1, EGL-19 and EXP-2 currents shape action potentials in the Caenorhabditis elegans pharynx.
      ). Instead, stronger depolarizing events, such as the action potential, activate CaV2 channels, which are the major drivers of fast, synchronous membrane fusion of synaptic vesicles at the nerve terminal (
      • Stanley E.F.
      The nanophysiology of fast transmitter release.
      ,
      • Südhof T.C.
      The presynaptic active zone.
      ,
      • Spafford J.D.
      • Chen L.
      • Feng Z.-P.
      • Smit A.B.
      • Zamponi G.W.
      Expression and modulation of an invertebrate presynaptic calcium channel α1 subunit homolog.
      ,
      • Spafford J.D.
      • Munno D.W.
      • van Nierop P.
      • Feng Z.-P.
      • Jarvis S.E.
      • Gallin W.J.
      • Smit A.B.
      • Zamponi G.W.
      • Syed N.I.
      Calcium channel structural determinants of synaptic transmission between identified invertebrate neurons.
      ,
      • Edmonds B.
      • Klein M.
      • Dale N.
      • Kandel E.R.
      Contributions of two types of calcium channels to synaptic transmission and plasticity.
      ,
      • Saheki Y.
      • Bargmann C.I.
      Presynaptic CaV2 calcium channel traffic requires CALF-1 and the α2δ subunit UNC-36.
      ,
      • Kawasaki F.
      • Zou B.
      • Xu X.
      • Ordway R.W.
      Active zone localization of presynaptic calcium channels encoded by the cacophony locus of Drosophila.
      ). Similarly, high voltage activation of post-synaptic CaV1 channels in muscles and neurons drives contraction and changes in nuclear gene expression, respectively (
      • Catterall W.A.
      Voltage-gated calcium channels.
      ,
      • Spafford J.D.
      • Dunn T.
      • Smit A.B.
      • Syed N.I.
      • Zamponi G.W.
      In vitro characterization of L-type calcium channels and their contribution to firing behavior in invertebrate respiratory neurons.
      ,
      • Senatore A.
      • Guan W.
      • Boone A.N.
      • Spafford J.D.
      T-type channels become highly permeable to sodium ions using an alternative extracellular turret region (S5-P) outside the selectivity filter.
      ,
      • Hara Y.
      • Koganezawa M.
      • Yamamoto D.
      The Dmca1D channel mediates Ca2+ inward currents in Drosophila embryonic muscles.
      ,
      • Collet C.
      Excitation-contraction coupling in skeletal muscle fibers from adult domestic honeybee.
      ,
      • Eberl D.F.
      • Ren D.
      • Feng G.
      • Lorenz L.J.
      • Van Vactor D.
      • Hall L.M.
      Genetic and developmental characterization of Dmca1D, a calcium channel α1 subunit gene in Drosophila melanogaster.
      ,
      • Erxleben C.
      • Rathmayer W.
      A dihydropyridine-sensitive voltage-dependent calcium channel in the sarcolemmal membrane of crustacean muscle.
      ,
      • Jospin M.
      • Jacquemond V.
      • Mariol M.-C.
      • Ségalat L.
      • Allard B.
      The L-type voltage-dependent Ca2+ channel EGL-19 controls body wall muscle function in Caenorhabditis elegans.
      ,
      • Di Biase V.
      • Franzini-Armstrong C.
      Evolution of skeletal type e–c coupling: a novel means of controlling calcium delivery.
      ,
      • Chan J.D.
      • Zhang D.
      • Liu X.
      • Zarowiecki M.
      • Berriman M.
      • Marchant J.S.
      Utilizing the planarian voltage-gated ion channel transcriptome to resolve a role for a Ca2+ channel in neuromuscular function and regeneration.
      ). Indeed, given the considerable overlap in biophysical, ion-conducting properties of CaV1 and CaV2 channels, it is unclear why they have generally persisted in their unique respective post- and presynaptic functions.
      Previously, we documented that the CaV2 channel from the placozoan T. adhaerens lacks an acidic C-terminal amino acid motif proposed to be critical for interactions with presynaptic scaffolding proteins, such as Mint and RIM, and broadly conserved in animals with synapses, such as chordates, arthropods, nematodes, and cnidarians (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ). CaV1 channels also bear deeply conserved C-terminal motifs for interactions with post-synaptic proteins like Shank and Erbin (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ). This suggests that a key evolutionary adaptation toward the specialization of CaV1 and CaV2 channels for distinct post- and presynaptic functions might have involved differential incorporation into protein complexes that would control trafficking and subcellular localization. Following the proposed CaV1/CaV2 split (
      • Moran Y.
      • Zakon H.H.
      The evolution of the four subunits of voltage-gated calcium channels: ancient roots, increasing complexity, and multiple losses.
      ,
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ), the two channel types might have also evolved biophysical features that distinguished them from each other. In the context of fast presynaptic exocytosis, ancestral CaV2 channels might thus have borne unique biophysical features that made them particularly well-suited for this role. Given that placozoans lack synapses but are the most early-diverging animals to possess both CaV1 and CaV2 channels, they present an opportunity to address this question. Here, we sought to explore whether the CaV2 channel from T. adhaerens exhibits biophysical features consistent with those of the major presynaptic CaV2 channel isotype from humans, CaV2.1. Cloning and in vitro expression of the Trichoplax CaV2 channel, coupled with whole-cell patch-clamp electrophysiology, allowed us to compare its ion-conducting properties with those of human CaV2.1 (
      • Hirano M.
      • Takada Y.
      • Wong C.F.
      • Yamaguchi K.
      • Kotani H.
      • Kurokawa T.
      • Mori M.X.
      • Snutch T.P.
      • Ronjat M.
      • De Waard M.
      • Mori Y.
      C-terminal splice variants of P/Q-type Ca2+ channel CaV2.1 α1 subunits are differentially regulated by Rab3-interacting molecule proteins.
      ). Remarkably, despite roughly 600 million years of divergence, the Trichoplax channel exhibited functional features similar to those of the human channel, and its biophysical properties differed from those of the previously cloned Trichoplax low voltage–activated CaV3 channel (
      • Smith C.L.
      • Abdallah S.
      • Wong Y.Y.
      • Le P.
      • Harracksingh A.N.
      • Artinian L.
      • Tamvacakis A.N.
      • Rehder V.
      • Reese T.S.
      • Senatore A.
      Evolutionary insights into T-type Ca2+ channel structure, function, and ion selectivity from the Trichoplax adhaerens homologue.
      ). Altogether, the work provides some important insights into the core features of synaptic CaV2 channels, contributing to our understanding of the evolution of CaV channel function in animals.

      Results

      Cloning of a CaV2 calcium channel homologue from T. adhaerens

      Previously, we identified a single putative Trichoplax CaV2 (TCaV2) channel transcript, bearing a complete protein-coding sequence, in a whole-animal mRNA transcriptome (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ,
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ). The TCaV2 channel ORF was verified in triplicate via cloning of the corresponding cDNA from whole-animal total RNA, producing a consensus sequence encoding a 2,093-amino acid protein with a predicted mass of ∼240 kDa (GenBankTM accession number MT506972). A Kyte-Doolittle hydrophobicity plot of the protein sequence showed hydrophobic peaks consistent with four repeat domains (DI–DIV), each with six transmembrane α helices (also known as segments 1–4 or S1–S4), and a long cytoplasmic C terminus (Fig. 1A). A maximum likelihood protein phylogeny of various CaV2 channels exhibited complete lineage sorting with respect to the leading metazoan phylogeny (
      • Whelan N.V.
      • Kocot K.M.
      • Moroz T.P.
      • Mukherjee K.
      • Williams P.
      • Paulay G.
      • Moroz L.L.
      • Halanych K.M.
      Ctenophore relationships and their placement as the sister group to all other animals.
      ), with TCaV2 and the CaV2 channel homologue from fellow placozoan H. hongkongensis forming a sister relationship with cnidarian and bilaterian CaV2 channels (Fig. 1B) and homologues from ctenophores forming the most distant clade of CaV2 channels. As reported previously, placozoans are the most early-diverging animals to possess all three types of metazoan voltage-gated calcium channels (CaV1–CaV3), unlike ctenophores and sponges that lack CaV3 (both phyla) and either CaV1 (ctenophores and most sponges) or CaV2 (sponges) channel homologues (
      • Moran Y.
      • Zakon H.H.
      The evolution of the four subunits of voltage-gated calcium channels: ancient roots, increasing complexity, and multiple losses.
      ,
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ,
      • Liebeskind B.J.
      • Hillis D.M.
      • Zakon H.H.
      Evolution of sodium channels predates the origin of nervous systems in animals.
      ).
      Figure thumbnail gr1
      Figure 1Conserved structural features of the Trichoplax CaV2 channel. A, Kyte–Doolittle plot of the TCaV2 protein sequence, revealing conserved hydrophobic peaks consistent with S1–S6 transmembrane segments arranged in four repeat domains (DI–DIV), separated by hydrophilic cytoplasmic linkers and N and C termini. The illustration above the plot denotes conserved features of CaV2 channels, including the positively charged S4 helices that make up the voltage sensors (red), the pore loops that make up the selectivity filter (purple), the AID required for interactions with the CaVβ accessory subunit (cyan), and the IQ region required for interactions with the Ca2+ sensor protein calmodulin (green). B, maximum likelihood phylogenetic tree of various CaV2 channel proteins, revealing the sister relationship of placozoan homologues with those from cnidarians and bilaterians. Consistent with the expected species phylogeny, the CaV2 channel homologues from ctenophores are the most early-diverging CaV2 channels. Bootstrap values for 1,000 ultrafast replicates are indicated at nodes, and branch lengths correspond to the bar on the bottom left indicating the unit of 0.5 substitutions per site. C, protein alignment of DI-IV pore-loop regions of various CaV2 channels, revealing complete conservation of the four-glutamate (EEEE) selectivity filter motif. D, protein alignment of the AID, revealing that the TCaV2 AID bears a conserved glycine-tyrosine-X-X-tryptophan-isoleucine (GY-WI) amino acid motif essential for the CaVβ interaction (blue asterisks). E, alignment of the DI-IV S4 helices bearing repeating positively charged lysine and arginine residues critical for voltage sensing.
      The TCaV2 channel bears a highly conserved Ca2+ ion selectivity filter motif of high voltage–activated CaV1 and CaV2 channels, made up of four negatively charged glutamate residues (i.e. EEEE motif) located within corresponding P-loop structures from each domain (Fig. 1, A and C). Also conserved is the α-interacting domain (AID), located within the cytosolic DI-DII linker; an α-helical structure unique to CaV1 and CaV2 channels that projects into the cytoplasm from the DI S6 helix and interacts with the cytoplasmic CaVβ subunit (Fig. 1D). It is notable that placozoan and ctenophore CaV2 AID sequences retain conserved glycine-tyrosine-X-X-tryptophan-isoleucine (GY-WI) amino acid motifs that are essential for the CaVβ interaction (
      • Opatowsky Y.
      • Chen C.-C.
      • Campbell K.P.
      • Hirsch J.A.
      Structural analysis of the voltage-dependent calcium channel β subunit functional core and its complex with the α1 interaction domain.
      ), where mutations lead to disrupted modulation by CaVβ (
      • De Waard M.
      • Scott V.E.
      • Pragnell M.
      • Campbell K.P.
      Identification of critical amino acids involved in α1-β interaction in voltage-dependent Ca2+ channels.
      ,
      • Berrou L.
      • Dodier Y.
      • Raybaud A.
      • Tousignant A.
      • Dafi O.
      • Pelletier J.N.
      • Parent L.
      The C-terminal residues in the α-interacting domain (AID) helix anchor CaVβ subunit interaction and modulation of CaV2.3 channels.
      ). However, placozoan and ctenophore AIDs lack the signature glutamine triplet (QQQ) motif found in most other CaV2 and CaV1 channels. Accordingly, the Trichoplax genome encodes a single CaVβ gene, as well as three CaVα2δ genes (
      • Srivastava M.
      • Begovic E.
      • Chapman J.
      • Putnam N.H.
      • Hellsten U.
      • Kawashima T.
      • Kuo A.
      • Mitros T.
      • Salamov A.
      • Carpenter M.L.
      • Signorovitch A.Y.
      • Moreno M.A.
      • Kamm K.
      • Grimwood J.
      • Schmutz J.
      • et al.
      The Trichoplax genome and the nature of placozoans.
      ,
      • Kamm K.
      • Osigus H.-J.
      • Stadler P.F.
      • DeSalle R.
      • Schierwater B.
      Trichoplax genomes reveal profound admixture and suggest stable wild populations without bisexual reproduction.
      ), and these are known to be expressed at the RNA level (
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ).
      Like other CaV2 channels, the S4 α helices (also known as S4 segments) of TCaV2 bear repeating positively charged lysine and/or arginine residues critical for voltage sensing (Fig. 1E) (
      • Catterall W.A.
      Ion channel voltage sensors: structure, function, and pathophysiology.
      ). Among the channels analyzed, S4 segments in DI and DIV generally show the strongest conservation, with the exception of ctenophore CaV2 channels that have shifted cationic charges in DIV toward the extracellular end of the S4 helix. Instead, S4 segments from DII and DIII are more variable for TCaV2 and other early-diverging homologues compared with bilaterian channels, notable because, at least for CaV1.2 channels, these particular segments contribute disproportionately toward voltage sensing and channel activation (
      • Pantazis A.
      • Savalli N.
      • Sigg D.
      • Neely A.
      • Olcese R.
      Functional heterogeneity of the four voltage sensors of a human L-type calcium channel.
      ). In DII, Trichoplax CaV2 also differs from other CaV2 channels, including H. hongkongensis CaV2, with one less cationic charge due to an arginine to proline substitution. Despite some differences, the Trichoplax CaV2 channel bears the core amino acid signatures required for gating the channel pore in response to changes in membrane voltage. This includes the highly conserved glutamate and aspartate residues located in S2 and S3 helices that counterbalance the positively charged arginine/lysine residues of S4 helices within each domain at rest and during channel activation, when S4 helices slide outward from the cell membrane upon depolarization (
      • Catterall W.A.
      Ion channel voltage sensors: structure, function, and pathophysiology.
      ) (Fig. S1).
      Last, we explored the conservation of motifs for association with the Ca2+ sensor protein calmodulin (CaM), which dynamically interacts with CaV1 and CaV2 channels to modulate their activity in response to changes in cytoplasmic Ca2+ concentration and for the purpose of intracellular Ca2+ signaling (
      • Dolmetsch R.E.
      • Pajvani U.
      • Fife K.
      • Spotts J.M.
      • Greenberg M.E.
      Signaling to the nucleus by an L-type calcium channel-calmodulin complex through the MAP kinase pathway.
      ). Ca2+-dependent regulation of CaV channels is likely ancient, observed in extant paramecia (
      • Brehm P.
      • Eckert R.
      Calcium entry leads to inactivation of calcium channel in Paramecium.
      ). Furthermore, the core C-terminal binding sites for CaM, known as pre-IQ and IQ motifs, are thought to have been present in the primordial ancestor of four-domain P-loop channels that gave rise to metazoan CaV1/2, CaV1, CaV2, CaV3, NaV, and NALCN channels (
      • Pozdnyakov I.
      • Matantseva O.
      • Skarlato S.
      Diversity and evolution of four-domain voltage-gated cation channels of eukaryotes and their ancestral functional determinants.
      ). A protein alignment of various CaV2 channels reveals considerable conservation of amino acid sequence within the pre-IQ and IQ domains (Fig. S2), which contrasts with other cytoplasmic regions that tend to be highly divergent among distant CaV homologues (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ). Trichoplax and Hoilungia CaV2 channel IQ domains bear key amino acids for interacting with CaM, including an invariable isoleucine comprising the namesake IQ motif with consensus sequence (I/L/V)QXXRXXXX(R/K) (
      • Jurado L.A.
      • Chockalingam P.S.
      • Jarrett H.W.
      Apocalmodulin.
      ). Also conserved is an isoleucine six residues upstream of the IQ moiety and a doublet of tyrosine (YY) residues just downstream. In crystal structures of CaV2.1 and CaV2.3, the isoleucine residue is found anchored within a hydrophobic pocket of the N-lobe (N terminus) of CaM, and the tyrosine residues embed within the C-lobe (C terminus; Fig. S2) (
      • Mori M.X.
      • Vander Kooi C.W.
      • Leahy D.J.
      • Yue D.T.
      Crystal structure of the CaV2 IQ domain in complex with Ca2+/calmodulin: high-resolution mechanistic implications for channel regulation by Ca2+.
      ). About 35 residues upstream of the pre-IQ motif, CaV1, and CaV2 channels possess putative EF-hand Ca2+-binding motifs that are structurally indispensable for Ca2+/CaM-dependent regulation, independent of their capacity to bind Ca2+ (
      • Zhou J.
      • Olcese R.
      • Qin N.
      • Noceti F.
      • Birnbaumer L.
      • Stefani E.
      Feedback inhibition of Ca2+ channels by Ca2+ depends on a short sequence of the C terminus that does not include the Ca2+-binding function of a motif with similarity to Ca2+-binding domains.
      ). We note here that placozoan as well as invertebrate CaV2 channels bear conserved EF-hand structures with amino acids capable of coordinating Ca2+ ions (Fig. S1). Conversely, the calmodulin-binding domain downstream of the IQ motif, reported for CaV2.1 channels (
      • Lee A.
      • Wong S.T.
      • Gallagher D.
      • Li B.
      • Storm D.R.
      • Scheuer T.
      • Catterall W.A.
      Ca2+/calmodulin binds to and modulates P/Q-type calcium channels.
      ), is not immediately evident in the sequences of invertebrate CaV2 channels (Fig. S1).

      TCaV2 expresses in vitro and is endogenously expressed in cells located around the periphery of the animal

      The cDNA of TCaV2 was cloned into the mammalian expression plasmids pIRES2-EGFP and pEGFP-C1, producing corresponding TCaV2 protein expression vectors pTCaV2-IR-EGFP and pEGFP-TCaV2 (Fig. 2A). Transfection of pTCaV2-IR-EGFP into HEK-293T cells permits bicistronic expression of the channel separately from enhanced GFP (EGFP), whereas pEGFP-TCaV2 expresses TCaV2 tagged with EGFP at its N terminus (Fig. 2A). Using a commercial monoclonal anti-GFP antibody, the full-length EGFP-TCaV2 fusion protein could be detected in protein lysates of HEK-293T cells transfected with pEGFP-TCaV2 as a band with an estimated molecular mass of ∼270 kDa (Fig. 2B). This corresponds to expected sum molecular weight of EGFP (28.9 kDa) plus TCaV2 (239.5 kDa). Like the CaV2 channel cloned from the snail Lymnaea stagnalis (
      • Spafford J.D.
      • Chen L.
      • Feng Z.-P.
      • Smit A.B.
      • Zamponi G.W.
      Expression and modulation of an invertebrate presynaptic calcium channel α1 subunit homolog.
      ), efficient in vitro expression of TCaV2 in HEK cells required co-transfection with vectors encoding mammalian CaV2 channel accessory subunits CaVβ (i.e. rat CaVβ1b) and CaVα2δ (i.e. rat CaVα2δ1), in lieu of Trichoplax subunits that were not part of this study (Fig. 2, B and C). This was also evident in fluorescence microscopy images of transfected cells, where fluorescence intensity of EGFP-TCaV2 was significantly higher when the HEK cells were co-transfected with CaVβ1b and CaVα2δ1 cDNAs (Fig. 2, D and E).
      Figure thumbnail gr2
      Figure 2TCaV2 is expressed as a full-length protein in vitro and in vivo. A, plasmid maps of pTCaV2-IR-EGFP and pEGFP-TCaV2 plasmid vectors for heterologous expression of the TCaV2 channel protein in mammalian cells. Whereas the pEGFP-TCaV2 vector expresses TCaV2 with an N-terminal EGFP fusion, pTCaV2-IR-EGFP permits expression of TCaV2 separately from EGFP. B, Western blotting of protein lysates from transfected HEK-293T cells with polyclonal anti-GFP antibodies reveals a band of about ∼270 kDa, consistent with the expected sum molecular weight of the TCaV2 plus the EGFP proteins. No such band is evident in untransfected cells, and its intensity is dramatically increased upon co-expression with the rat CaVβ1b and CaVα2δ1 accessory subunits. C, quantification of band intensity on triplicate Western blots, relative to corresponding total protein on lanes of Coomassie-stained gels. All average values ± S.D. (error bars) are expressed relative to the maximal value across all experiments. Lowercase letters denote significant differences using a Holm–Sidak test (p < 0.006) after one-way ANOVA (p < 0.001, F = 387.6; ). D, overlaid transmitted light and fluorescence images of HEK-293T cells depicting a dramatic increase in EGFP-TCaV2 fusion protein expression upon co-transfection with the rat CaVβ1b and CaVα2δ1 accessory subunits. Inset, confocal image of a positively transfected HEK-293T cell expressing the EGFP-TCaV2 protein, with EGFP fluorescence visible in regions outside of the nucleus (stained with DAPI, cyan), consistent with endomembrane and cell membrane localization. Scale bar, 100 μm (three larger panels) and 20 μm (inset)). E, quantification of average pTCaV2-IR-EGFP and pEGFP-TCaV2 fluorescence intensity ± S.D. in triplicate micrographs of separately transfected HEK-293T cells is consistent with Western blotting results, where co-expression of the CaVβ1b and CaVα2δ1 subunits dramatically increases protein expression of the EGFP-TCaV2 fusion protein, and EGFP expressed from the bicistronic vector.. Integrative density values were standardized to the maximal value across all experiments. Denoted p values for mean comparisons were generated using two-tailed tests. F, Western blotting of TCaV2 expressed in HEK-293T cells from the bicistronic pIRES2-EGFP vector, using rabbit polyclonal anti-TCaV2 antibodies directed against 142 amino acids in the II-III linker of the channel protein reveals strong ectopic expression of the Trichoplax calcium channel when co-expressed with rat CaVβ1b and CaVα2δ1. At least three bands are visible on the blots, with molecular masses of about 250, 220, and 85 kDa, that disappeared after preincubation of the anti-TCaV2 antibodies with a recombinant epitope peptide. G, Western blotting of Trichoplax whole-animal protein lysates using the anti-TCaV2 polyclonal antibodies revealed numerous bands with molecular masses of about 240, 200, 90, 75, 63, 60, and 50 kDa, all of which disappeared after preincubation with recombinant blocking peptide.
      We raised rabbit polyclonal antibodies against a recombinant peptide of 142 amino acids corresponding to the TCaV2 cytoplasmic II-III linker (Fig. S1), and tested their efficacy by Western blotting for untagged TCaV2 protein heterologously expressed in HEK-293T cells with the pTCaV2-IR-EGFP vector (co-transfected with rat CaVβ1b and CaVα2δ1; Fig. 2F). The antibodies labeled a band at around 250 kDa, which disappeared after preincubation of membranes with the corresponding recombinant epitope peptide. Bands of about 220 and 85 kDa were also present and disappeared with peptide preincubation. The same antibodies were then used to detect TCaV2 in protein lysates isolated from Trichoplax whole animals, labeling bands at ∼240, 200, 90, 75, 63, 60, and 50 kDa on a Western blot (Fig. 2G). Although it is possible that some of the smaller-molecular weight bands correspond to off-target proteins, we note that BLAST searching the epitope protein sequence against Trichoplax transcriptome (
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ) and genome (
      • Srivastava M.
      • Begovic E.
      • Chapman J.
      • Putnam N.H.
      • Hellsten U.
      • Kawashima T.
      • Kuo A.
      • Mitros T.
      • Salamov A.
      • Carpenter M.L.
      • Signorovitch A.Y.
      • Moreno M.A.
      • Kamm K.
      • Grimwood J.
      • Schmutz J.
      • et al.
      The Trichoplax genome and the nature of placozoans.
      ,
      • Kamm K.
      • Osigus H.-J.
      • Stadler P.F.
      • DeSalle R.
      • Schierwater B.
      Trichoplax genomes reveal profound admixture and suggest stable wild populations without bisexual reproduction.
      ) sequences fails to produce significant hits outside of the CaV2 channel. As observed with the HEK-293T lysates, all bands disappeared after peptide preincubation, suggesting that although present as a full-length protein, some of the TCaV2 channel protein within the lysates is fragmented. Altogether, it is apparent that TCaV2 is expressed as a full-length protein in HEK-293T cells when heterologously expressed in vitro and as a full-length endogenously expressed protein in Trichoplax.
      Applying the TCaV2 antibodies to whole-mount staining of fixed Trichoplax revealed expression around the periphery of the animal (Fig. 3A), in a region also labeled by fluorescent wheat germ agglutinin (WGA) that marks mucous-secreting type II gland cells, also referred to as mucocytes (
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ). At higher magnification, it is apparent that the TCaV2 protein is expressed within mucocytes, in small punctate regions adjacent to larger WGA-positive regions (Fig. 3A, inset). The latter likely represents clusters of mucous-containing vesicles labeled by WGA (
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ), demarking large cytoplasmic regions of separate mucocyte cells. Preincubation of the TCaV2 antibody with the blocking peptide, or staining in the absence of primary antibody, did not produce fluorescent signals above background levels, suggesting that the observed labeling was specific to TCaV2 (Fig. 3, B and C, respectively). Mucocytes, thought to constitutively secrete mucous for ciliary gliding/locomotion, also express the endomorphin-like peptide TaELP proposed to be subject to regulated secretion for the purpose of cell-cell signaling. Specifically, secreted TaELP is proposed to target receptors on ciliated ventral epithelial cells, pausing ciliary beating and hence causing locomotion to stop (
      • Senatore A.
      • Reese T.S.
      • Smith C.L.
      Neuropeptidergic integration of behavior in Trichoplax adhaerens, an animal without synapses.
      ,
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ). Three-dimensional rendering of the fluorescent images further reveals that although TCaV2 and WGA labeling overlap at the edge of the animal, TCaV2 is abundant along the dorsal epithelium, whereas mucocytes extend along the ventral epithelium (Fig. 3, D–F), as reported previously (
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ). Indeed, the dorsal staining observed for TCaV2 is consistent with staining patterns reported for two other Trichoplax regulatory peptides, SIFGamide and SITFamide, the former causing the animal to vigorously contract and “crinkle” and the latter slowing down ciliary locomotion (
      • Varoqueaux F.
      • Williams E.A.
      • Grandemange S.
      • Truscello L.
      • Kamm K.
      • Schierwater B.
      • Jékely G.
      • Fasshauer D.
      High cell diversity and complex peptidergic signaling underlie placozoan behavior.
      ). In some preparations, in addition to the labeling patterns noted above, the TCaV2 antibody labeled cells with branching filamentous structures consistent with fiber cells (Fig. 3, G and H), which are located between the dorsal and ventral epithelium and thought to be contractile in nature (
      • Smith C.L.
      • Varoqueaux F.
      • Kittelmann M.
      • Azzam R.N.
      • Cooper B.
      • Winters C.A.
      • Eitel M.
      • Fasshauer D.
      • Reese T.S.
      Novel cell types, neurosecretory cells, and body plan of the early-diverging metazoan Trichoplax adhaerens.
      ,
      • Rassat J.
      • Ruthmann A.
      Trichoplax adhaerens FE Schulze (Placozoa) in the scanning electron microscope.
      ,
      • Thiemann M.
      • Ruthmann A.
      Microfilaments and microtubules in isolated fiber cells of Trichoplax adhaerens (Placozoa).
      ).
      Figure thumbnail gr3
      Figure 3TCaV2 is expressed in cells concentrated along the dorsal periphery of Trichoplax. A, maximum projection fluorescence micrograph of a Trichoplax animal stained red for type II gland cells using fluorophore-conjugated wheat germ agglutinin (WGA), blue for nuclei with DAPI, and immunolabeled with the rabbit anti-TCaV2 polyclonal antibody and a fluorescent anti-rabbit secondary antibody (green). Inset, co-expression of TCaV2 along the periphery of cells labeled with WGA. B, control experiment with preincubation of the TCaV2 antibody with the blocking peptide. C, control experiment lacking primary antibody. D, three-dimensional reconstruction of the micrograph stack shown in A, oriented from the top down, reveals expression of TCaV2 within a cluster of cells located along the outer edge of the dorsal epithelium. E, side view of the three-dimensional reconstruction, with the dorsal epithelium oriented at the top. F, bottom view of three-dimensional reconstruction reveals sparse labeling along the ventral epithelium. G, some preparations revealed TCaV2 staining within cells located in the interior of the animal, with filamentous structures consistent with fiber cells. H, enlarged view of the outlined region shown in panel G. Scale bar, 20 μm (all panels).

      TCaV2 conducts high voltage–activated Ca2+ currents in vitro that are similar to the human P/Q-type channel

      Given that TCaV2 is expressed as a full-length protein in HEK-293T cells, we next sought to determine whether the recombinant channel could produce functional voltage-dependent Ca2+ currents in vitro using whole-cell patch-clamp electrophysiology. Voltage-clamp recordings of cells transfected with pTCaV2-IR-EGFP, along with expression vectors for rat CaVβ1b and CaVα2δ1, revealed large-amplitude Ca2+ currents in 3 mm external Ca2+ solution that could be elicited by depolarization from −100 mV to between −45 and +80 mV (Fig. 4A). The amplitudes of inward macroscopic Ca2+ currents were quite variable (Fig. S3) and, in some cells, were greater than 2,000 pA. Thus, to prevent voltage errors in our recordings that can be caused by large amplitude currents, we only used cells with peak inward currents near and below 1,000 pA and patch pipettes with minimal access resistance (see “Experimental procedures”). When only the rat CaVβ1b and CaVα2δ1 subunit cDNAs were transfected, no calcium currents could be recorded. In contrast to the low voltage–activated CaV3 channel cloned from Trichoplax (
      • Smith C.L.
      • Abdallah S.
      • Wong Y.Y.
      • Le P.
      • Harracksingh A.N.
      • Artinian L.
      • Tamvacakis A.N.
      • Rehder V.
      • Reese T.S.
      • Senatore A.
      Evolutionary insights into T-type Ca2+ channel structure, function, and ion selectivity from the Trichoplax adhaerens homologue.
      ), TCaV2 activation required strong depolarization, with inward Ca2+ currents first appearing at voltage steps to –35 mV and maximal peak macroscopic current occurring at −10 mV (Fig. 4B). To provide a comparative context to the TCaV2 electrophysiology experiments, the cloned human P/Q-type (CaV2.1) channel (
      • Hirano M.
      • Takada Y.
      • Wong C.F.
      • Yamaguchi K.
      • Kotani H.
      • Kurokawa T.
      • Mori M.X.
      • Snutch T.P.
      • Ronjat M.
      • De Waard M.
      • Mori Y.
      C-terminal splice variants of P/Q-type Ca2+ channel CaV2.1 α1 subunits are differentially regulated by Rab3-interacting molecule proteins.
      ) was also expressed in HEK-293T cells using the same conditions and recording solutions. Although TCaV2 is a high voltage–activated channel like other CaV2 channels, its voltage sensitivity is left-shifted compared with human CaV2.1 (hCaV2.1), the latter having a maximal inward current at +5 mV (Fig. 4B). Transformation of peak inward currents into conductance values, a process that removes the effect of driving force and enables visualization of macroscopic conductance as the channel population responds to depolarization, reveals that TCaV2 achieves half-maximal activation (V½) at −17.7 ± 2.2 mV, compared with −4.4 ± 2.8 mV for hCaV2.1 (Fig. 4D). Despite this roughly 13-mV difference, the rate of activation of TCaV2 and hCaV2.1 in response to depolarization is similar, with respective conductance curve slope (kactivation) values of 3.8 ± 0.7 and 3.8 ± 0.4 mV (not significantly different, two-tailed test p = 0.956). We also compared the inactivation properties of TCaV2 and hCaV2.1, which approximates the fraction of channels within a population available for activation at different values of resting membrane potential, albeit within a relatively transient time scale. Holding voltages ranging from −60 to +10 mV, held for 1 s, caused the amplitude of macroscopic Ca2+ currents elicited by a test pulse to 0 mV to gradually decline relative to a prepulse to 0 mV due to accumulating inactivation within the channel population (Fig. 4C). Plotting the ratio of maximal inactivated current amplitude versus test pulse current amplitude, as a function of the inactivating voltage, revealed an inactivation curve for TCaV2 with a V½ at −28.7 ± 1.8 mV and a slope (kinactivation) of 3.9 ± 0.5 mV (Fig. 4D). In contrast, hCaV2.1 exhibited a hyperpolarized shift in V½ of inactivation with a value of −34.4 ± 1.5 mV and also a slower rate of inactivation with a kinactivation of 6.7 ± 1.0 mV (two-tailed 3.94 test p < 0.001). Notable is that the relatively right-shifted inactivation curve of TCaV2, coupled with its left-shifted activation curve, produces a substantial window current voltage range between −35 and −10 mV (Fig. 4D, red fill), much more prominent than observed for hCaV2.1 (Fig. 4D, blue fill). This feature is more commonly attributed to CaV1 and CaV3 channels and represents a range of resting voltages through which a subset of channels would remain open to conduct constitutive Ca2+ currents into the cell (
      • Crunelli V.
      • Tóth T.I.
      • Cope D.W.
      • Blethyn K.
      • Hughes S.W.
      The “window” T-type calcium current in brain dynamics of different behavioural states.
      ,
      • Hirano Y.
      • Moscucci A.
      • January C.T.
      Direct measurement of L-type Ca2+ window current in heart cells.
      ).
      Figure thumbnail gr4
      Figure 4TCaV2 produces robust voltage-gated Ca2+ currents in vitro. A, sample macroscopic current traces recorded via whole-cell patch voltage clamp of HEK-293T cells transfected with pTCaV2-IR-EGFP and rat CaVβ1b and CaVα2δ1 subunits (bottom). The corresponding voltage-clamp protocol, with depolarizing voltage steps from −100 mV to various voltages, is depicted above. B, standardized average peak macroscopic current ± S.D. (error bars) plotted against corresponding voltage steps for TCaV2 and hCaV2.1 reveals a left-shifted maximal inward current for the Trichoplax channel. C, illustration of the voltage-clamp protocol used to assess inactivation of TCaV2 and hCaV2.1 (top). Peak amplitude of inward currents elicited by a test pulse following a 1-s pulse at various inactivating voltages is compared with that of a prepulse voltage step to 0 mV. Sample currents are shown below. D, plots of average inactivation ± S.D. and transformation of current-voltage plots into conductance plots reveal that, relative to hCaV2.1 (grey), the TCaV2 channel (black) is less sensitive to inactivation and more readily activated by small voltage steps. This results in a large window current at voltages where a subset of channels are not inactivated, and some become activated (pink fill). By comparison, the human channel has a much smaller window current component within the overlap between the inactivation and activation curves (blue fill). Average values for half-maximal activation and inactivation (V½act and V½inact), plus kactivation and kinactivation slope factors are depicted and were generated by fitting the activation and inactivation data with a Boltzmann function. E, voltage-clamp protocol used to assess recovery from inactivation of peak inward current after a 10-s inactivating pulse (top) and corresponding sample current traces recorded for the TCaV2 channel (bottom). F, plots of average recovery from inactivation ± S.D. of TCaV2 and hCaV2.1 reveal a slower recovery from inactivation for the Trichoplax channel. Inset, recovery data for the first 5 s, with inflections in the curves indicative of bimodal recovery from inactivation for both channels. Biexponential curve fitting over the data produced larger τ values for TCaV2 than hCaV2.1 (τ1 and τ2), especially for the slower recovery component.
      A feature of CaV channels that determines their continued contribution to rises in intracellular Ca2+ during prolonged bouts of excitation (e.g. action potential burst firing) is their recovery from inactivation. Whereas a population of CaV channels with a fast recovery from inactivation can remain active throughout a train of action potentials, those with slow recovery tend to accumulate inactivation and hence contribute less Ca2+ influx toward the end of an action potential burst (
      • Adams P.J.
      • Garcia E.
      • David L.S.
      • Mulatz K.J.
      • Spacey S.D.
      • Snutch T.P.
      CaV2.1 P/Q-type calcium channel alternative splicing affects the functional impact of familial hemiplegic migraine mutations: implications for calcium channelopathies.
      ). Recovery from inactivation of TCaV2 and hCaV2.1 was assessed by determining the peak current that could be elicited by a step to 0 mV at different time intervals succeeding a 10-s inactivating pulse to 0 mV (Fig. 4E). Consistent with previous reports, the hCaV2.1 channel exhibited bimodal recovery from inactivation (
      • Adams P.J.
      • Garcia E.
      • David L.S.
      • Mulatz K.J.
      • Spacey S.D.
      • Snutch T.P.
      CaV2.1 P/Q-type calcium channel alternative splicing affects the functional impact of familial hemiplegic migraine mutations: implications for calcium channelopathies.
      ), with respective time constants for fast and slow components of the recovery process of 0.7 ± 0.3 s (τ1) and 7.5 ± 3.5 s (τ2) (Fig. 4F). Interestingly, the TCaV2 channel also exhibited bimodal recovery from inactivation, with a similar fast component (τ1 = 1.6 ± 0.8 s; p = 0.713) but a much slower slow component compared with hCaV2.1 (τ2 = 40.0 ± 6.1 s; p < 0.001 for Holm–Sidak test after two-way ANOVA with p < 0.001 and F ≥ 90.175 for all comparisons; Table S1). Altogether, the slower recovery from inactivation kinetics of TCaV2 is evident after 3 s of hyperpolarization following the inactivating pulse, where 68.4 ± 3.9% of hCaV2.1 channels had recovered, compared with only 36.9 ± 9.6% of TCaV2 channels (Fig. 4F, inset). Similarly, TCaV2 required much more time for full recovery from inactivation, at roughly 135 s (99.2 ± 3.8% recovery) compared with 40 s for hCaV2.1 (98 ± 2.7% recovery).

      The kinetic properties of TCaV2 macroscopic currents resemble those of hCaV2.1 in their voltage dependence

      We compared the kinetic properties of TCaV2 and hCaV2.1 activation and inactivation by fitting monoexponential curves over the rise and decay phases of macroscopic currents, producing corresponding time constants (τactivation and τinactivation). Both channels exhibited accelerating activation kinetics with increasing depolarization, with roughly 2-fold decreases in τactivation at +60 mV compared with 0 mV (Fig. 5A) (p < 0.001 and F ≥ 99.123 for one-way repeated measures ANOVAs for TCaV2 and hCaV2.1; Table S1). Nevertheless, activation of the Trichoplax channel was much slower than hCaV2.1, with respective τactivation values of 10.5 ± 1.5 and 2.6 ± 0.3 ms at 0 mV, decreasing to 3.6 ± 0.7 and 0.5 ± 0.1 ms at +60 mV (p < 0.001 for Holm–Sidak test after two-way ANOVA; p < 0.001 and F ≥ 18.201 for all comparisons; Table S1). Kinetics of inactivation for both channels were also voltage-dependent, but in contrast to activation kinetics, τinactivation showed a general deceleration with increasing depolarization (Fig. 5B; p < 0.001 and F ≥ 14.301 for one-way repeated measures ANOVAs for TCaV2 and hCaV2.1; Table S1). Within the voltage range tested, hCaV2.1 showed first acceleration in inactivation kinetics from −5 to 0 mV, followed by deceleration toward +40 mV. This is in contrast to TCaV2, which exhibited faster inactivation kinetics at −5 mV and a continual voltage-dependent deceleration until +40 mV. Altogether, the trajectories of respective τinactivation curves over the tested voltage range were similar, with TCaV2 exhibiting slower inactivation kinetics than hCaV2.1 at the voltages of 5 and 10 mV (p < 0.05 for Holm–Sidak test after two-way ANOVA; p = 0.029 and F = 4.824 for variation due to channels, p < 0.001 and F = 12.942 for variation due to voltage, p = 0.063 and F = 2.036 for variation due to the interaction of channel and voltage; Table S1).
      Figure thumbnail gr5
      Figure 5Although generally slower, the kinetics of TCaV2 macroscopic currents resemble those of hCaV2.1 in their voltage dependence. A, plot of average τ values ± S.D. obtained by monoexponential curve fitting over the activation phase of macroscopic currents elicited by different depolarizing voltages. Changing τactivation values for both TCaV2 and hCaV2.1 indicates accelerating activation with stronger depolarization. Nevertheless, activation of TCaV2 was significantly slower at all depolarizing voltages. B, plot of average τ values ± S.D. (error bars) obtained by monoexponential curve fitting over the inactivation phase of macroscopic currents elicited by different depolarizing voltages. Both TCaV2 and hCaV2.1 exhibit increasing τinactivation values with stronger depolarization. C, protocol used to assess deactivation kinetics, with hyperpolarizing pulses to varying voltages following a brief step to 0 mV (top). Sample tail current traces recorded for the TCaV2 channel are shown below. D, plot of average τdeactivation values ± S.D. for TCaV2 and hCaV2.1 revealing accelerating deactivation upon stronger hyperpolarization for both channels. Across all voltages, the Trichoplax channel exhibits slower deactivation kinetics than hCaV2.1. Letters above the data points indicate statistically significant differences resulting from paired comparisons using a Holm–Sidak test after respective one-way repeated measures ANOVA for TCaV2 and hCaV2.1 τ values (i.e. where p < 0.01). Thus, plotted values bearing the same letter are not statistically different from each other. Single and double cyan asterisks denote respective post hoc Holm–Sidak p values of <0.05 and <0.001 for paired comparisons of τ values for TCaV2 and hCaV2.1 currents at different voltages, after two-way ANOVA ().
      Another important factor that determines the amount of Ca2+ influx through CaV channels is their deactivation kinetics, which reflects how quickly open channels transition to a closed, activatable state upon membrane hyperpolarization. During action potential repolarization and hyperpolarization, CaV channels with slow deactivation remain open longer to conduct a surge of Ca2+ into the cytoplasm (facilitated by the increased driving force for inward Ca2+ flow at negative potentials), compared with fast deactivating channels that would quickly close (
      • Bahamonde M.I.
      • Serra S.A.
      • Drechsel O.
      • Rahman R.
      • Marcé-Grau A.
      • Prieto M.
      • Ossowski S.
      • Macaya A.
      • Fernández-Fernández J.M.
      A single amino acid deletion (ΔF1502) in the S6 segment of CaV2. 1 domain III associated with congenital ataxia increases channel activity and promotes Ca2+ influx.
      ). To compare the deactivation kinetics of TCaV2 and hCaV2.1, monoexponential curves were fitted over decaying macroscopic currents elicited through open channels upon hyperpolarization to between −120 and –30 mV (Fig. 5C). Similar to τinactivation, τdeactivation for both channels exhibited voltage-dependent deceleration, most striking for TCaV2 with τdeactivation increasing from 0.5 ± 0.2 ms at −120 mV to 8.7 ± 2.5 ms at −30 mV, compared with 0.2 ± 0.1 ms at −120 mV to 0.5 ± 0.1 ms at –30 mV for hCaV2.1 (Fig. 5D). At the more depolarized voltages of −40 and −30 mV, there is a theoretical but perhaps small possibility that channel inactivation might be contributing to the current decay, especially for TCaV2, which undergoes marginal activation and considerable inactivation at these voltages (Fig. 4D). However, we note that the decaying currents were best fit with monoexponential functions, indicative of a single component process. At all voltages, TCaV2 deactivation was slower than hCaV2.1 deactivation (p < 0.001 for Holm–Sidak test after two-way ANOVA; p < 0.001 and F ≥ 6.842 for all comparisons; Table S1).

      TCaV2 resembles hCaV2.1 and other high voltage–activated calcium channels by conducting larger Ba2+ than Ca2+ currents

      The nonphysiological cation Ba2+ is used as a surrogate for Ca2+ in electrophysiological recordings of CaV channels, likely for its tendency to produce large currents compared to other divalent cations, plus its ability to potently block K+ channel currents (
      • Hille B.
      Ionic Channels of Excitable Membranes.
      ). Generally, high voltage–activated CaV channels such as L-type (CaV1), N-type, and P/Q-type (CaV2) conduct larger macroscopic Ba2+ than Ca2+ currents (
      • Hille B.
      Ionic Channels of Excitable Membranes.
      ,
      • Feng Z.-P.
      • Hamid J.
      • Doering C.
      • Jarvis S.E.
      • Bosey G.M.
      • Bourinet E.
      • Snutch T.P.
      • Zamponi G.W.
      Amino acid residues outside of the pore region contribute to N-type calcium channel permeation.
      ,
      • Bourinet E.
      • Zamponi G.W.
      • Stea A.
      • Soong T.W.
      • Lewis B.A.
      • Jones L.P.
      • Yue D.T.
      • Snutch T.P.
      The α1E calcium channel exhibits permeation properties similar to low-voltage-activated calcium channels.
      ), whereas T-type (CaV3) channels tend to vary in this respect between paralogous subtypes (i.e. CaV3.1–CaV3.3) and species (
      • Perez-Reyes E.
      Molecular physiology of low-voltage-activated t-type calcium channels.
      ,
      • Senatore A.
      • Spafford J.D.
      Transient and big are key features of an invertebrate T-type channel (LCav3) from the central nervous system of Lymnaea stagnalis.
      ,
      • Smith C.L.
      • Abdallah S.
      • Wong Y.Y.
      • Le P.
      • Harracksingh A.N.
      • Artinian L.
      • Tamvacakis A.N.
      • Rehder V.
      • Reese T.S.
      • Senatore A.
      Evolutionary insights into T-type Ca2+ channel structure, function, and ion selectivity from the Trichoplax adhaerens homologue.
      ,
      • Jeong K.
      • Lee S.
      • Seo H.
      • Oh Y.
      • Jang D.
      • Choe J.
      • Kim D.
      • Lee J.-H.
      • Jones W.D.
      Ca-α1T, a fly T-type Ca2+ channel, negatively modulates sleep.
      ,
      • Senatore A.
      • Guan W.
      • Boone A.N.
      • Spafford J.D.
      T-type channels become highly permeable to sodium ions using an alternative extracellular turret region (S5-P) outside the selectivity filter.
      ,
      • McRory J.E.
      • Santi C.M.
      • Hamming K.S.
      • Mezeyova J.
      • Sutton K.G.
      • Baillie D.L.
      • Stea A.
      • Snutch T.P.
      Molecular and functional characterization of a family of rat brain T-type calcium channels.
      ). We sought to compare the permeation properties of TCaV2 and hCaV2.1 for Ca2+ and Ba2+ ions. Voltage steps from –100 to +10 mV at 30-s intervals (Fig. 6A) produced stable TCaV2 macroscopic currents while perfusing 20 mm external Ca2+. Currents increased in amplitude ∼4.3-fold upon switching to 20 mm external Ba2+ (Fig. 6B). This is in contrast to hCaV2.1, which only showed a ∼1.6-fold increase when Ca2+ was replaced with Ba2+. Notably, the peak Ba2+ currents for TCaV2 decayed in amplitude with sequential voltage pulses, a rare phenomenon also reported for CaV channels recorded in the somatic membrane of mollusc neurons (
      • Kostyuk P.
      • Doroshenko P.
      • Martynyuk A.
      Fast decrease of the peak current carried by barium ions through calcium channels in the somatic membrane of mollusc neurons.
      ). This use-dependent decay prevented us from accurately recording current-voltage data for the channel for comparing Ba2+ versus Ca2+ permeation properties across a range of voltage steps. However, we noticed that the decay process could be stopped by preceding the depolarizing voltage steps with 50-ms hyperpolarizing prepulses to −200 mV, similar to what was reported for the decaying CaV channel Ba2+ currents recorded in snail neurons (
      • Kostyuk P.
      • Doroshenko P.
      • Martynyuk A.
      Fast decrease of the peak current carried by barium ions through calcium channels in the somatic membrane of mollusc neurons.
      ). With the prepulse, the increase in peak current for TCaV2 when switching from Ca2+ to Ba2+ was reduced to resemble that of the human channel, with only a ∼1.5-fold increase (Fig. 6B). We note that although the hyperpolarization step had an effect on TCaV2 channel Ba2+ currents, it did not affect its Ca2+ currents or the hCaV2.1 Ca2+ or Ba2+ currents, verified across a range of voltages (−50 to +80 mV, not shown). To further understand the decaying Ba2+ currents for TCaV2, we sought to rule out the anomalous mole fraction effect that can occur when perfusing solutions containing Ca2+ and Ba2+ over cells (
      • Almers W.
      • McCleskey E.
      Non-selective conductance in calcium channels of frog muscle: calcium selectivity in a single-file pore.
      ). Cells expressing TCaV2 channels were therefore placed into a bath containing only 20 mm Ba2+ solution. However, the decay process was still evident (Fig. 6C), ruling out the anomalous mole fraction effect. Furthermore, to rule out the possibility that Ba2+ enhances accumulation of inactivated channels, intersweep intervals were increased to 60 s. Under these conditions, peak Ba2+ currents once again exhibited a decay in amplitude in the absence of a hyperpolarizing prepulse, which was not observed when a hyperpolarizing prepulse was applied (Fig. 6C).
      Figure thumbnail gr6
      Figure 6TCaV2 resembles hCaV2.1 by conducting larger macroscopic Ba2+ than Ca2+ currents. A, voltage step protocols to record inward currents at 10 mV without (left) and with (right) a 200-mV, 50-ms prepulse. B, peak currents for TCaV2 and hCaV2.1 elicited by the protocols depicted in A, while perfusing 20 mm external Ba2+ or Ca2+ solutions, normalized against the peak amplitude of the first traces recorded in Ca2+. Without a prepulse, the Ba2+ currents through the TCaV2 are markedly larger and decay upon subsequent voltage steps. C, representative plot of relative peak TCaV2 inward currents elicited by the protocols in A, but with a longer interpulse interval of 60 s, indicates that the decaying currents in the presence of Ba2+ are not likely to be due to accumulated inactivation. D, illustration of the voltage-clamp protocol used to compare relative current-voltage properties of TCaV2 and hCaV2.1 while perfusing 20 mm external Ca2+ or Ba2+ (left). Sample Ca2+ and Ba2+ current traces recorded from a single cell expressing TCaV2 are shown on the right. E, current-voltage plots of TCaV2 and hCaV2.1 average peak inward current ± S.D. (error bars), relative to maximal Ba2+ current, reveal similar increases in current amplitude and leftward shifts in maximal inward current for both channels when Ba2+ is used as a charge carrier. F, transformation of the current-voltage plots in E reveals left-shifted activation curves in the presence of external Ba2+, particularly for TCaV2. Average values ± S.D. for V½act, V½inact, kactivation, and kinactivation slope factors were generated by fitting the activation data with a Boltzmann function. The reported p values in the plot are for Holm–Sidak comparisons of mean V½ values for current activation curves for TCaV2 and hCaV2.1 in Ca2+ and Ba2+ after two-way ANOVA ().
      In light of these observations, we introduced a similar −200-mV prepulse to generate TCaV2 and hCaV2.1 current-voltage data while perfusing 20 mm Ca2+ or Ba2+ (Fig. 6D). We also selected cells with peak Ca2+ current amplitudes of roughly 100 pA, such that voltage errors did not arise when switching to Ba2+ due to larger current amplitudes. For both channels, plots of peak current at different voltages revealed similar increases in maximal inward current of roughly 1.6-fold in Ba2+ compared with Ca2+, with both channels reaching maximal current at +20 mV in the presence of Ca2+, shifting leftward to 0 and +10 mV, respectively, for TCaV2 and hCaV2.1 in the presence of Ba2+ (Fig. 6E). Notably, the overlapping IV data for TCaV2 and hCaV2.1 in 20 mm Ca2+, a condition that resembles the physiological Ca2+ concentration in seawater, contrast with our previous observations in 3 mm Ca2+, where peak TCaV2 currents were left-shifted relative to hCaV2.1 (Fig. 4B). Transformation of the current-voltage data to relative conductance plots revealed similar voltages of half-maximal activation in 20 mm Ca2+ for the two channels (TCaV2 V½ = 5.3 ± 2.0 mV; hCaV2.1 V½ = 8.4 ± 2.6 mV; Fig. 6F), although statistically significantly different from each other (p < 0.05 for Holm–Sidak test after two-way ANOVA; p < 0.001 and F ≥ 29.602 for variation due to ions and channels, and p = 0.079 and F = 3.435 for variation due to the interaction of ions and channels; Table S1). Activation curves were left-shifted in the presence of Ba2+ (TCaV2 V½ = −10.0 ± 1.1 mV; hCaV2.1 V½ = −3.7 ± 1.8 mV), more so for TCaV2 (p < 0.001 for Holm–Sidak test after two-way ANOVA; Table S1). Furthermore, kactivation slope values in Ba2+ were statistically significantly different between the two channels (p < 0.05 for Holm–Sidak test after two-way ANOVA; p = 0.310 and F = 1.086 for variation due to ion type, p < 0.05 and F ≥ 4.716 for variation due to channels as well as ion and channel interactions; Table S1), although they were not in Ca2+.

      TCaV2 exhibits low sensitivity to general CaV channel blockers Cd2+ and Ni2+ and the selective peptide blockers ω-conotoxin-GVIA and ω-agatoxin-IVA

      CaV1 and CaV2 channels are known for their low micromolar sensitivity to block by the metal cation Cd2+ and relative insensitivity to Ni2+ compared with CaV3 channels (
      • Perez-Reyes E.
      Molecular physiology of low-voltage-activated t-type calcium channels.
      ,
      • Simms B.A.
      • Zamponi G.W.
      Neuronal voltage-gated calcium channels: structure, function, and dysfunction.
      ). Accordingly, perfusion of external Cd2+ ions at increasing concentrations blocked hCaV2.1 Ca2+ currents with an IC50 of 1.0 ± 0.2 μm (Fig. 7A), whereas Ni2+ required a much stronger concentration, with an IC50 of 448.7 ± 38.4 μm (Fig. 7B). TCaV2 was roughly 1.4-fold less sensitive to Ni2+ than hCaV2.1 with an IC50 of 648.3 ± 105.7 μm (p < 1E−2 using a Mann–Whitney U test) and 1.9-fold less sensitive than the Trichoplax CaV3 channel with a previously reported IC50 of 335.0 ± 6.5 μm (mean ± S.E.) (
      • Smith C.L.
      • Abdallah S.
      • Wong Y.Y.
      • Le P.
      • Harracksingh A.N.
      • Artinian L.
      • Tamvacakis A.N.
      • Rehder V.
      • Reese T.S.
      • Senatore A.
      Evolutionary insights into T-type Ca2+ channel structure, function, and ion selectivity from the Trichoplax adhaerens homologue.
      ). Compared with hCaV2.1, TCaV2 was markedly less sensitive to Cd2+, with an IC50 of 20.6 ± 2.8 μm (p < 1E−11, two-tailed test), and requiring 100–300 μm for complete block compared with only 10–30 μm (Fig. 7B). Thus, TCaV2 differs from vertebrate high voltage–activated channels in its relative insensitivity to Cd2+, making it more similar to the expressed CaV1 and CaV2 channels cloned from the snail L. stagnalis (
      • Spafford J.D.
      • Dunn T.
      • Smit A.B.
      • Syed N.I.
      • Zamponi G.W.
      In vitro characterization of L-type calcium channels and their contribution to firing behavior in invertebrate respiratory neurons.
      ).
      Figure thumbnail gr7
      Figure 7TCaV2 is relatively insensitive to block by divalent metal cations Cd2+ and Ni2+ and CaV2 isotype–specific peptide toxins ω-conotoxin GVIA and ω-agatoxin IVA. A, dose-response curve for block of peak macroscopic currents of TCaV2 and hCaV2.1 with increasing concentrations of perfused external Cd2+ ions. TCaV2 is significantly less sensitive to Cd2+ block than hCaV2.1, with a roughly 19-fold higher IC50 (p < 1E−11 using a two-tailed test). B, dose-response curves for peak current block of TCaV2 and hCaV2.1 with increasing concentrations of perfused extracellular Ni2+ ions. TCaV2 is slightly less sensitive to Ni2+ block than hCaV2.1 (p < 1E−2 using Mann–Whitney comparison of mean IC50 values). C, plot of average % block ± S.D. of peak macroscopic currents at 0 mV for TCaV2 and hCaV2.1 by external application of 2 μm ω-conotoxin GVIA, a CaV2.2-specific blocker. D, plot of average % block ± S.D. (error bars) of peak currents for TCaV2 and hCaV2.1 by 1 μm ω-agatoxin IVA, revealing complete (98%) block of the human channel compared with only 15% block for TCaV2 (p < 0.001 using a two-tailed test). E, protein alignment of the DIII pore-loop region of various CaV2 channels reveals presence/absence of key residues important for high-affinity block of CaV2.2 channels by ω-conotoxin GVIA. Orange back-colored residues reflect positions determined important for block of mammalian channels (
      • Feng Z.-P.
      • Doering C.J.
      • Winkfein R.J.
      • Beedle A.M.
      • Spafford J.D.
      • Zamponi G.W.
      Determinants of inhibition of transiently expressed voltage-gated calcium channels by ω-conotoxins GVIA and MVIIA.
      ), whereas cyan back-colored residues are either conserved or share similar chemical properties. F, protein alignment of the DIV S3-S4 linker containing a key glutamate residue associated with affinity block of CaV2.1 channels in mammals (
      • Winterfield J.R.
      • Swartz K.J.
      A hot spot for the interaction of gating modifier toxins with voltage-dependent ion channels.
      ) (orange back color). Notably, these residues are also conserved in channels that are not highly sensitive to ω-agatoxin-IVA.
      Next, we tested the effects of the cone snail and spider peptide toxins, ω-conotoxin-GVIA and ω-agatoxin-IVA, respectively, on TCaV2. For mammalian CaV2 channels, ω-conotoxin-GVIA selectively blocks N-type (CaV2.2) channels (
      • Olivera B.M.
      • McIntosh J.M.
      • Cruz L.J.
      • Luque F.A.
      • Gray W.R.
      Purification and sequence of a presynaptic peptide toxin from Conus geographus venom.
      ,
      • McCleskey E.
      • Fox A.
      • Feldman D.
      • Cruz L.
      • Olivera B.
      • Tsien R.
      • Yoshikami D.
      ω-conotoxin: direct and persistent blockade of specific types of calcium channels in neurons but not muscle.
      ,
      • Aosaki T.
      • Kasai H.
      Characterization of two kinds of high-voltage-activated Ca-channel currents in chick sensory neurons.
      ), whereas ω-agatoxin-IVA selectively blocks P/Q-type (CaV2.1) channels (
      • Mintz I.M.
      • Venema V.J.
      • Swiderek K.M.
      • Lee T.D.
      • Bean B.P.
      • Adams M.E.
      P-type calcium channels blocked by the spider toxin ω-Aga-IVA.
      ,
      • Venema V.
      • Swiderek K.
      • Lee T.
      • Hathaway G.
      • Adams M.
      Antagonism of synaptosomal calcium channels by subtypes of ω-agatoxins.
      ). Neither TCaV2 nor hCaV2.1 were sensitive to 2 μm ω-conotoxin-GVIA, only being blocked by 13.2 ± 1.5% and 16.4 ± 2.9%, respectively (Fig. 7C; p = 0.1 for two-tailed test). Interestingly, work on zebrafish calcium channels involved in synaptic transmission revealed that both CaV2.2 and CaV2.1 channels were highly sensitive to ω-conotoxin-GVIA and relatively insensitive to ω-agatoxin-IVA, indicating that the specificity of these compounds to select CaV2 channel isotypes is not conserved among vertebrate homologues (
      • Wen H.
      • Linhoff M.W.
      • Hubbard J.M.
      • Nelson N.R.
      • Stensland D.
      • Dallman J.
      • Mandel G.
      • Brehm P.
      Zebrafish calls for reinterpretation for the roles of P/Q calcium channels in neuromuscular transmission.
      ). A key locus for ω-conotoxin-GVIA sensitivity has been identified in the DIII pore-loop of mammalian CaV2.2 channels, which contains an EF-hand structure previously implicated in Ca2+ permeation (
      • Feng Z.-P.
      • Hamid J.
      • Doering C.
      • Jarvis S.E.
      • Bosey G.M.
      • Bourinet E.
      • Snutch T.P.
      • Zamponi G.W.
      Amino acid residues outside of the pore region contribute to N-type calcium channel permeation.
      ). Here, directed mutations were found to severely disrupt blocking affinity to the toxin (
      • Feng Z.-P.
      • Doering C.J.
      • Winkfein R.J.
      • Beedle A.M.
      • Spafford J.D.
      • Zamponi G.W.
      Determinants of inhibition of transiently expressed voltage-gated calcium channels by ω-conotoxins GVIA and MVIIA.
      ). Alignment of this region revealed that both hCaV2.1 and TCaV2, which are poorly blocked by ω-conotoxin-GVIA, lack the key residues associated with high sensitivity identified in rat CaV2.2 and conserved in the human orthologue (Fig. 7E). In contrast, the CaV2.1 and CaV2.2 channel isotypes from zebrafish, which are highly sensitive, resemble human CaV2.2 in this region with tandem negatively charged glutamate and aspartate residues and a C-terminal glutamine residue.
      As expected, 1 μm ω-agatoxin-IVA potently blocked the human CaV2.1 channel (98.5 ± 0.6% block), but this was not the case for TCaV2 (15.0 ± 5.2% block) (Fig. 7D; p < 0.001 for two-tailed test). TCaV2 lacks a key glutamate residue in the DIV S3-S4 linker associated with affinity block of CaV2.1 channels in mammals (
      • Winterfield J.R.
      • Swartz K.J.
      A hot spot for the interaction of gating modifier toxins with voltage-dependent ion channels.
      ). However, the presence of this residue in channels that are not highly sensitive to ω-agatoxin-IVA, including zebrafish CaV2.1 channels, CaV2.2 channels, and protostome invertebrate CaV2 channels, suggests that additional sites are important (Fig. 7F).

      TCaV2 does not exhibit voltage-dependent G-protein inhibition in vitro

      A compelling feature of synaptic CaV2 channels is their voltage-dependent inhibition by G-protein β and γ heterodimers, which enables direct neuromodulatory input mediated by G protein–coupled receptors (GPCRs) on presynaptic Ca2+ influx and synaptic transmission (
      • Tedford H.W.
      • Zamponi G.W.
      Direct G protein modulation of Cav2 calcium channels.
      ,
      • Dunlap K.
      • Fischbach G.
      Neurotransmitters decrease the calcium conductance activated by depolarization of embryonic chick sensory neurones.
      ). Indeed, even simple neural circuits exhibit complex neuromodulation, which permits reconfiguring of fast electrochemical machinery for altered input/output properties of neural circuits (
      • Marder E.
      Neuromodulation of neuronal circuits: back to the future.
      ). Our understanding of how and when fast electrochemical signaling machinery and slow neuromodulatory signaling became integrated is limited, and this represents a fundamental question about nervous system evolution (
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ,
      • Katz P.S.
      • Lillvis J.L.
      Reconciling the deep homology of neuromodulation with the evolution of behavior.
      ). With respect to CaV2 channels, neurotransmitter/ligand activation of select GPCRs activates associated Gβγ dimers, which, in a membrane-delimited manner, bind and inhibit CaV2 channels via a direct interaction with regions of the N terminus, C terminus, and I-II linker (
      • Tedford H.W.
      • Zamponi G.W.
      Direct G protein modulation of Cav2 calcium channels.
      ). G-protein inhibition can be temporarily alleviated by bouts of elevated excitation (
      • Bean B.P.
      Neurotransmitter inhibition of neuronal calcium currents by changes in channel voltage dependence.
      ), permitting voltage-dependent facilitation of presynaptic Ca2+ influx and exocytosis. Previously, the CaV2 channel cloned from the snail L. stagnalis was reported to lack voltage-dependent G-protein inhibition (
      • Huang X.
      • Senatore A.
      • Dawson T.F.
      • Quan Q.
      • Spafford J.D.
      G-proteins modulate invertebrate synaptic calcium channel (LCav2) differently from the classical voltage-dependent regulation of mammalian Cav2. 1 and Cav2.2 channels.
      ), suggesting that this important neuromodulatory process is unique to vertebrates and closely related animals. However, subsequent work in cultured Lymnaea neurons revealed that endogenous presynaptic CaV channel currents could be inhibited by activation of dopamine D2 GPCRs (
      • Dong N.
      • Lee D.W.
      • Sun H.-S.
      • Feng Z.-P.
      Dopamine-mediated calcium channel regulation in synaptic suppression in L. stagnalis interneurons.
      ) and that G-protein inhibition of calcium currents occurs and can be alleviated by strong depolarizing prepulses (
      • Dong N.
      • Lee D.W.
      • Sun H.-S.
      • Feng Z.-P.
      Dopamine-mediated calcium channel regulation in synaptic suppression in L. stagnalis interneurons.
      ,
      • de Hoog E.
      • Lukewich M.K.
      • Spencer G.E.
      Retinoid receptor-based signaling plays a role in voltage-dependent inhibition of invertebrate voltage-gated Ca2+ channels.
      ). This suggests that GPCR signaling and specifically Gβγ modulation of presynaptic CaV2 channels was present in the common ancestor of animals with bilateral symmetry (i.e. protostomes and deuterostomes).
      Here, we sought to explore whether G-protein inhibition of CaV2 channels might be conserved in placozoans. We note from an ongoing transcriptome study that Trichoplax expresses over 656 GPCRs, as well as all core intracellular GPCR signaling machinery, including putative G-protein αi/o subunits associated with G-protein inhibition of CaV2 channels (
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ,
      • Tedford H.W.
      • Zamponi G.W.
      Direct G protein modulation of Cav2 calcium channels.
      ,
      • Holz G.G.
      • Rane S.G.
      • Dunlap K.
      GTP-binding proteins mediate transmitter inhibition of voltage-dependent calcium channels.
      ). We reasoned that the absence of G-protein inhibition observed for the Lymnaea CaV2 channel in vitro might have been due to divergence between Gβγ subunits in the human cell line used for the electrophysiology experiments and endogenous Gβγ subunits in isolated Lymnaea neurons, as well as the binding sites for Gβγ along intracellular regions of the CaV2 channels. Hence, we searched for G-protein β and γ subunits in the Trichoplax transcriptome, identifying four Gγ subunits (Gγ1–4), and two Gβ subunits (Gβ1–2). Protein alignment and secondary/tertiary structure prediction of the Trichoplax Gβ subunits and representative homologues from other animals revealed conserved N-terminal amphipathic helices important for interactions with the Gγ subunit and an array of β strands partitioned into seven tryptophan-aspartate (WD) domain repeats, predicted to fold into a 7-bladed β-propeller configuration (Fig. S4) (
      • Wall M.A.
      • Coleman D.E.
      • Lee E.
      • Iñiguez-Lluhi J.A.
      • Posner B.A.
      • Gilman A.G.
      • Sprang S.R.
      The structure of the G protein heterotrimer Giα1β1γ2.
      ,
      • Lambright D.G.
      • Sondek J.
      • Bohm A.
      • Skiba N.P.
      • Hamm H.E.
      • Sigler P.B.
      The 2.0 Å crystal structure of a heterotrimeric G protein.
      ). Also evident is that the Trichoplax2 subunit is more divergent from other Gβ subunits compared with Gβ1. A similar analysis of the Trichoplax Gγ subunits revealed a conserved set of tandem N-terminal α helices, which interact with helices of the Gβ subunit (
      • Sondek J.
      • Bohm A.
      • Lambright D.G.
      • Hamm H.E.
      • Sigler P.B.
      Crystal structure of a G-protein βγ dimer at 2.1 Å resolution.
      ), and C-terminal cysteine residues that become isoprenylated for integration of the Gβγ heterodimer into the plasma membrane (Fig. S4). Notably, the Trichoplax Gβγ proteins differ at some key amino acid positions important for effector function, as reported in yeast (
      • Grishin A.V.
      • Weiner J.L.
      • Blumer K.J.
      Biochemical and genetic analysis of dominant-negative mutations affecting a yeast G-protein γ subunit.
      ), but bear determinant amino acids in the Gβ subunits that are required for interactions with mammalian CaV2.2 channels (
      • Doering C.J.
      • Kisilevsky A.E.
      • Feng Z.-P.
      • Arnot M.I.
      • Peloquin J.
      • Hamid J.
      • Barr W.
      • Nirdosh A.
      • Simms B.
      • Winkfein R.J.
      • Zamponi G.W.
      A single Gβ subunit locus controls cross-talk between protein kinase C and G protein regulation of N-type calcium channels.
      ,
      • Tedford H.W.
      • Kisilevsky A.E.
      • Peloquin J.B.
      • Zamponi G.W.
      Scanning mutagenesis reveals a role for serine 189 of the heterotrimeric G-protein β1 subunit in the inhibition of N-type calcium channels.
      ) (i.e. Tyr111, Asp153, and Ser189; Fig. S4).
      We sought to clone the Trichoplax Gβγ subunits for in vitro co-expression with the Trichoplax CaV2 channel. PCR amplification from a whole-animal poly(A) cDNA library was successful for three of the four Gγ subunits (Gγ1–3), and one of the Gβ subunits (Gβ1). Despite repeated attempts, we were unable to amplify the Gγ4 and Gβ2 subunits, perhaps due to low-level mRNA expression. Thus, the Gγ1–3 and Gβ1 were cloned in triplicate into the bicistronic mammalian expression vector pIRES2-DsRed2 that, in addition to the cloned G-protein subunit, expresses the DsRed2 red fluorescent protein that permits co-detection of TCaV2 and G proteins via green and red fluorescence, respectively. The consensus sequences for the Trichoplax G proteins were submitted to NCBI with accession numbers AZJ50981.1 (Gγ1), AZJ50982.1 (Gγ2), AZJ50983.1 (Gγ3), and AZJ50980.1 (Gβ1). As a positive control for electrophysiological experiments, we synthesized and cloned the cDNAs for human Gγ2 (NM_053064.5) and Gβ1 (NM_002074.5) into the pIRES2-DsRed2 vector for co-expression with the human CaV2.1 channel. To assess the occurrence of voltage-dependent G-protein inhibition, we used a protocol comprising a 50-ms test pulse to 0 mV (for TCaV2) or +10 mV (hCaV2.1), followed by a ±150-mV prepulse for 50 ms, and a subsequent test pulse to capture the changes in current amplitude and/or channel kinetics resulting from the prepulse (Fig. 8A). Under these conditions, the human channel exhibited voltage-dependent facilitation of peak current amplitude, with amplitudes increasing roughly 32% after the prepulse (Fig. 8 (A and B); +PP/−PP ratio = 0.88 ± 0.06 without Gβ1γ2 and 1.20 ± 0.11 with Gβ1γ2; p < 0.001 for Holm–Sidak test after one-way ANOVA with p < 0.001 and F = 22.959; Table S1). Human CaV2.1 activation kinetics were also accelerated by a prepulse in the presence of Gβ1γ2, with τactivation values of 10.55 ± 6.14 ms without a prepulse, compared with 3.26 ± 1.74 ms with a prepulse (p = 0.002 for Holm–Sidak test after two-way ANOVA, p < 0.001 and F = 21.868 for variation due to G proteins, p = 0.001 and F = 12.646 for variation due to prepulse, p = 0.08 and F = 2.714 for variation due to G protein × prepulse; Table S1). Instead, in the absence of co-transfected Gβ1γ2 subunits, the human channel did not exhibit prepulse-dependent acceleration of activation kinetics (τactivation values of 1.20 ± 0.36 and 1.16 ± 0.36 with and without prepulse; Fig. 8C), but kinetics were faster compared with all conditions where G proteins were co-expressed (p < 0.05 for Holm–Sidak test after two-way ANOVA; Table S1). Hence, as expected, the 150-mV prepulse temporarily relieved G-protein inhibition of the hCaV2.1 channel, resulting in larger-amplitude currents and faster activation kinetics (
      • Tedford H.W.
      • Zamponi G.W.
      Direct G protein modulation of Cav2 calcium channels.
      ). In contrast, we did not observe voltage-dependent G-protein inhibition of the TCaV2 channel co-expressed with the Trichoplax1γ1–3 subunits (Fig. 8, A–C). Plus/minus prepulse ratios of peak current were statistically indistinguishable, with mean values of 0.77 ± 0.05 (no Gβγ), 0.77 ± 0.07 (Gβ1γ1), 0.79 ± 0.05 (Gβ1γ2), and 0.80 ± 0.07 (Gβ1γ3). Activation kinetics were also unaffected, with τactivation values of 15.52 ± 3.26 ms (−PP, no Gβγ), 16.03 ± 2.20 ms (+PP, no Gβγ), 13.34 ± 1.67 ms (−PP, Gβ1γ1), 14.34 ± 1.59 ms (+PP, Gβ1γ1), 15.34 ± 1.90 ms (−PP, Gβ1γ2), 15.35 ± 2.27 ms (+PP, Gβ1γ2), 13.94 ± 1.04 ms (−PP, Gβ1γ3), and 14.90 ± 0.83 ms (+PP, Gβ1γ3).
      Figure thumbnail gr8
      Figure 8Absence of voltage-dependent Gβγ inhibition for the TCaV2 channel in vitro. A, illustration of voltage-clamp protocol used to assess G-protein inhibition of TCaV2 and hCaV2.1 channels in vitro (top). Sample current traces (normalized to maximal current) elicited by voltage steps to either 0 mV (for TCaV2) or +10 mV (for hCaV2.1), before and after a strong depolarizing 50-ms prepulse to +150 mV, are show below. TCaV2 currents did not exhibit facilitation in the presence/absence of co-expressed Trichoplax1γ1–3 heterodimers. Conversely, in the presence of co-expressed human Gβ1γ2 heterodimers, hCaV2.1 exhibited voltage-dependent facilitation, evident as larger-amplitude currents after the prepulse with faster activation kinetics. Furthermore, the Trichoplax1γ1–3 subunits caused voltage-dependent facilitation of hCaV2.1 activation kinetics, but not current amplitude. B, average (AVG) facilitation of peak macroscopic current amplitude after the prepulse ± S.D. (error bars) (i.e. amplitude after prepulse/amplitude before prepulse). Mean values were not statistically significantly different for TCaV2 (denoted as n.s.), in contrast to hCaV2.1 with a significantly higher amplitude ratio in the presence of human Gβ1γ2 subunits (letters above plotted values denote p < 0.001 for paired comparisons using post hoc Holm–Sidak tests after a one-way ANOVA; ). C, average τactivation values ± S.D. for monoexponential curves fit over the activation phase of macroscopic currents before and after the 150-mV prepulse. Mean τactivation values were not statistically different for TCaV2 (n.s.), whereas the prepulse caused acceleration of hCaV2.1 activation kinetics in the presence of both human Gβ1γ2 and Trichoplax1γ1-3. In the absence of co-transfected G proteins, hCaV2.1 activation kinetics were faster than all conditions where G proteins were present and did not significantly change after a prepulse. Letters above plotted values denote p < 0.001 for paired post hoc Holm–Sidak tests, where lowercase letters denote comparisons between the three G-protein conditions without a prepulse, and capital letters denote comparisons between the three G-protein conditions with a prepulse (after a two-way ANOVA; ). The reported p values in the plot are for Holm–Sidak comparisons of τactivation values within G-protein conditions before and after the prepulse.
      Strikingly, co-transfection of the human CaV2.1 channel with the Trichoplax G proteins caused slower current activation compared with control conditions (τactivation = 8.96 ± 6.51 versus 1.20 ± 0.36 in the absence of co-transfected G proteins, p < 0.001 for Holm–Sidak test after two-way ANOVA; Table S1; Fig. 8A). This inhibition was voltage-dependent and could be relieved with a depolarizing prepulse, as observed with human G proteins, decreasing from 8.96 ± 6.51 to 3.44 ± 2.40 (Fig. 8C; p = 0.008 for Holm–Sidak test after two-way ANOVA; Table S1). However, unlike the human G proteins, the Trichoplax homologues did not elicit voltage-dependent inhibition/facilitation of peak current amplitude (+PP/−PP ratio of 0.90 ± 0.09 versus 0.88 ± 0.06 in the absence of G proteins; Fig. 8, A and B). Altogether, it appears as though the Trichoplax CaV2 channel does not exhibit direct Gβγ inhibition, at least under our in vitro conditions. Nevertheless, the Trichoplax G proteins were able to interact with the human channel to produce voltage-dependent inhibition. This suggests that the adaptive changes required to render CaV2 channels sensitive to G-protein modulation occurred primarily via emergent changes in channel sequence/structure, and perhaps the CaVβ subunit (
      • Tedford H.W.
      • Zamponi G.W.
      Direct G protein modulation of Cav2 calcium channels.
      ), that permitted interactions with the Gβγ subunits. Accordingly, TCaV2 exhibits considerable sequence divergence from vertebrate CaV2.1 and CaV2.2 channels at cytoplasmic N-terminal and I-II linker regions that are required for interactions with Gβγ (
      • Tedford H.W.
      • Zamponi G.W.
      Direct G protein modulation of Cav2 calcium channels.
      ,
      • Agler H.L.
      • Evans J.
      • Tay L.H.
      • Anderson M.J.
      • Colecraft H.M.
      • Yue D.T.
      G protein-gated inhibitory module of N-type (Cav2.2) Ca2+ channels.
      ) (Fig. S1).

      Discussion

      Insights into Trichoplax biology inferred from the TCaV2 channel

      Placozoans provide a unique opportunity for exploring the evolution of CaV channel properties and cellular functions, in part because they are the most early-diverging animals to possess CaV1–CaV3 channels (
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ,
      • Moran Y.
      • Zakon H.H.
      The evolution of the four subunits of voltage-gated calcium channels: ancient roots, increasing complexity, and multiple losses.
      ,
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ) and also because of their morphological simplicity, bearing only six cell types distinguishable by ultrastructure (
      • Smith C.L.
      • Varoqueaux F.
      • Kittelmann M.
      • Azzam R.N.
      • Cooper B.
      • Winters C.A.
      • Eitel M.
      • Fasshauer D.
      • Reese T.S.
      Novel cell types, neurosecretory cells, and body plan of the early-diverging metazoan Trichoplax adhaerens.
      ,
      • Rassat J.
      • Ruthmann A.
      Trichoplax adhaerens FE Schulze (Placozoa) in the scanning electron microscope.
      ), and absence of true tissues. Our work here characterizing the functional properties of the CaV2 channel from T. adhaerens revealed that despite upwards of 600 million years of divergence, TCaV2 conducts high voltage–activated Ca2+ currents with similar profiles to those of human CaV2.1 and other cloned CaV2 channels (
      • Catterall W.A.
      Voltage-gated calcium channels.
      ,
      ), such as the homologues from the snail L. stagnalis (
      • Spafford J.D.
      • Chen L.
      • Feng Z.-P.
      • Smit A.B.
      • Zamponi G.W.
      Expression and modulation of an invertebrate presynaptic calcium channel α1 subunit homolog.
      ) and the honeybee Apis mellifera (
      • Cens T.
      • Rousset M.
      • Collet C.
      • Charreton M.
      • Garnery L.
      • Le Conte Y.
      • Chahine M.
      • Sandoz J.-C.
      • Charnet P.
      Molecular characterization and functional expression of the Apis mellifera voltage-dependent Ca2+ channels.
      ). Previously, we showed that the Trichoplax CaV3 channel conducts low voltage–activated Ca2+ currents similar to orthologues from other animals (
      • Smith C.L.
      • Abdallah S.
      • Wong Y.Y.
      • Le P.
      • Harracksingh A.N.
      • Artinian L.
      • Tamvacakis A.N.
      • Rehder V.
      • Reese T.S.
      • Senatore A.
      Evolutionary insights into T-type Ca2+ channel structure, function, and ion selectivity from the Trichoplax adhaerens homologue.
      ). Thus, it appears as though the core biophysical features of CaV2 channels that distinguish them from at least CaV3 channels were established very early on during evolution. Given that CaV3 channels predate animals and that CaV1 and CaV2 channels likely evolved from a premetazoan CaV1/2-like channel (
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ,
      • Moran Y.
      • Zakon H.H.
      The evolution of the four subunits of voltage-gated calcium channels: ancient roots, increasing complexity, and multiple losses.
      ,
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ), it is perhaps not surprising that extant Trichoplax CaV2 and CaV3 channels retain distinct functional profiles. This is also apparent in phylogenetic and sequence/structural analyses, where TCaV2 and TCaV3 are more similar to their counterparts in other animals than to each other, retaining all differentiating structures. Specifically, TCaV2 bears a conserved AID in the I-II cytoplasmic linker (required for interactions with CaVβ), C-terminal pre-IQ and IQ motifs (for interactions with calmodulin), and an EEEE Ca2+ selectivity filter motif. TCaV3, on the other hand, bears a conserved helix-loop-helix gating brake structure in the I-II linker (in lieu of the AID) and an EEDD selectivity filter motif (
      • Smith C.L.
      • Abdallah S.
      • Wong Y.Y.
      • Le P.
      • Harracksingh A.N.
      • Artinian L.
      • Tamvacakis A.N.
      • Rehder V.
      • Reese T.S.
      • Senatore A.
      Evolutionary insights into T-type Ca2+ channel structure, function, and ion selectivity from the Trichoplax adhaerens homologue.
      ). Less clear are the differences between CaV2 and CaV1 channels, in that they exhibit overlapping biophysical properties and share similar structural features. Perhaps an exception is a deeply conserved α-helical structure in the C terminus of CaV1 channels, involved in interactions with cAMP-dependent protein kinase-anchoring protein 15 (AKAP15), which is required for enhancement of macroscopic calcium current by β-adrenergic receptor (GPCR) signaling (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ,
      • Hulme J.T.
      • Lin T.W.-C.
      • Westenbroek R.E.
      • Scheuer T.
      • Catterall W.A.
      β-Adrenergic regulation requires direct anchoring of PKA to cardiac CaV1.2 channels via a leucine zipper interaction with A kinase-anchoring protein 15.
      ). Currently, we are conducting a functional characterization of the Trichoplax CaV1 channel, which will complete the characterization of the placozoan CaV channel repertoire. A key comparison here will be of Ca2+-dependent inactivation and/or facilitation of the Trichoplax CaV1 and CaV2 channels, which is possibly one of the key functional differences between these two channel types. These feedback processes are mediated by Ca2+ influx through open channels binding to calmodulin proteins preassociated with C-terminal pre-IQ and IQ motifs, which triggers alterations in channel gating (
      • Ben-Johny M.
      • Yue D.T.
      Calmodulin regulation (calmodulation) of voltage-gated calcium channels.
      ). In other words, CaV1 channels tend to exhibit pronounced Ca2+-dependent inactivation, whereas CaV2 channels show no to moderate inactivation and, in some cases, Ca2+-dependent facilitation (
      • Ben-Johny M.
      • Yue D.T.
      Calmodulin regulation (calmodulation) of voltage-gated calcium channels.
      ,
      • Taiakina V.
      • Boone A.N.
      • Fux J.
      • Senatore A.
      • Weber-Adrian D.
      • Guillemette J.G.
      • Spafford J.D.
      The calmodulin-binding, short linear motif, NSCaTE is conserved in L-type channel ancestors of vertebrate Cav1.2 and Cav1.3 channels.
      ). Conversely, CaV2 channels are generally more readily inactivated by voltage than CaV1 channels (
      • Catterall W.A.
      Voltage-gated calcium channels.
      ). Interestingly, recent work has shown that vertebrate and invertebrate CaV3 channels are also regulated by Ca2+/calmodulin, but through structural determinants that are different from those of CaV1 and CaV2 channels (
      • Cazade M.
      • Bidaud I.
      • Lory P.
      • Chemin J.
      Activity-dependent regulation of T-type calcium channels by submembrane calcium ions.
      ,
      • Chemin J.
      • Taiakina V.
      • Monteil A.
      • Piazza M.
      • Guan W.
      • Stephens R.F.
      • Kitmitto A.
      • Pang Z.P.
      • Dolphin A.C.
      • Perez-Reyes E.
      • Dieckmann T.
      • Guillemette J.G.
      • Spafford J.D.
      Calmodulin regulates Cav3 T-type channels at their gating brake.
      ). Physiologically, the differences observed between CaV1 and CaV2 in this regard become apparent during prolonged bouts of excitation. Here, CaV2 channel activity is more susceptible to membrane voltage, where repeated and strong depolarization causes accumulated inactivation and channel silencing, whereas CaV1 channels are less susceptible to inactivation by voltage and, rather, respond to rising levels of cytoplasmic Ca2+ (
      • Liu Z.
      • Ren J.
      • Murphy T.H.
      Decoding of synaptic voltage waveforms by specific classes of recombinant high-threshold Ca2+ channels.
      ). If this key difference was established early on, and perhaps conserved in Trichoplax, this could in part explain why the two channels have retained several nonoverlapping cellular functions broadly within animals.
      A notable feature of Trichoplax and placozoans in general is that, despite our knowledge that they express most genes required for fast neural electrochemical signaling (
      • Eitel M.
      • Francis W.R.
      • Varoqueaux F.
      • Daraspe J.
      • Osigus H.-J.
      • Krebs S.
      • Vargas S.
      • Blum H.
      • Williams G.A.
      • Schierwater B.
      • Wörheide G.
      Comparative genomics and the nature of placozoan species.
      ,
      • Srivastava M.
      • Begovic E.
      • Chapman J.
      • Putnam N.H.
      • Hellsten U.
      • Kawashima T.
      • Kuo A.
      • Mitros T.
      • Salamov A.
      • Carpenter M.L.
      • Signorovitch A.Y.
      • Moreno M.A.
      • Kamm K.
      • Grimwood J.
      • Schmutz J.
      • et al.
      The Trichoplax genome and the nature of placozoans.
      ,
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ,
      • Kamm K.
      • Osigus H.-J.
      • Stadler P.F.
      • DeSalle R.
      • Schierwater B.
      Trichoplax genomes reveal profound admixture and suggest stable wild populations without bisexual reproduction.
      ), including CaV channels and voltage-gated Na+ and K+ channels, we know very little about the presence and function of endogenous electrical activity in these animals. This is in contrast to other early-diverging lineages, such as sponges, ctenophores, and cnidarians, for which extensive electrophysiological data have been acquired (
      • Senatore A.
      • Raiss H.
      • Le P.
      Physiology and evolution of voltage-gated calcium channels in early diverging animal phyla: cnidaria, placozoa, porifera and ctenophora.
      ). A challenge in this respect is that dissociated Trichoplax cells are difficult to distinguish, are quite small (roughly 1 μm in diameter), and have apparent extracellular matrices that make patch-clamp and sharp electrode recording difficult. Very recently, a first report of endogenous electrical activity of Trichoplax and H. hongkongensis, recorded from immobilized whole animals using extracellular electrodes, revealed the presence of action potentials that could be elicited by injection of a depolarizing current (
      • Romanova D.Y.
      • Smirnov I.V.
      • Nikitin M.A.
      • Kohn A.B.
      • Borman A.I.
      • Malyshev A.Y.
      • Balaban P.M.
      • Moroz L.L.
      Action potentials and Na+ voltage-gated ion channels in Placozoa.
      ). Furthermore, extracellular recording of isolated crystal cells, involved in Trichoplax gravitactic behavior (
      • Mayorova T.D.
      • Smith C.L.
      • Hammar K.
      • Winters C.A.
      • Pivovarova N.B.
      • Aronova M.A.
      • Leapman R.D.
      • Reese T.S.
      Cells containing aragonite crystals mediate responses to gravity in Trichoplax adhaerens (Placozoa), an animal lacking neurons and synapses.
      ), also revealed busts of action potentials upon stimulation. This study has therefore confirmed that electrogenic genes are indeed active in placozoans and that electrical signaling is likely important for Trichoplax cell biology and physiology. Key questions that emerge include: how are electrogenic genes differentially deployed in placozoan cell types, and what is the nature and purpose of electrical activity in these cells? Our work here on TCaV2 and previously on the Trichoplax CaV3 channel reveal functional properties that only make sense in the context of fast oscillations in membrane voltage (e.g. graded and action potentials), consistent with the recent description of action potentials. For example, both channels have voltage properties that would render them inactivated and hence nonfunctional at depolarized membrane voltages, suggesting that cells expressing them must retain negative voltages through membrane shuffling of Na+, K+, and Cl ions by pumps and exchangers. The distinct and conserved activation properties of TCaV3 and TCaV2, the former being low voltage–activated and the latter high voltage–activated, indicate a conserved duality in CaV channel function in Trichoplax. Specifically, TCaV3 channels, endowed by their low activation voltages, likely contribute toward regulating membrane excitability and action potential generation, whereas TCaV2 channels respond to stronger depolarizing events to elicit Ca2+ influx and any downstream consequences. Other voltage properties of TCaV2 (e.g. the observed window current that represents a constant trickle of cytoplasmic Ca2+ influx within a discrete range of membrane voltages) can serve functions in regulating membrane voltage and/or Ca2+ signaling (
      • Perez-Reyes E.
      Molecular physiology of low-voltage-activated t-type calcium channels.
      ,
      • Crunelli V.
      • Tóth T.I.
      • Cope D.W.
      • Blethyn K.
      • Hughes S.W.
      The “window” T-type calcium current in brain dynamics of different behavioural states.
      ,
      • Hirano Y.
      • Moscucci A.
      • January C.T.
      Direct measurement of L-type Ca2+ window current in heart cells.
      ).
      We note that, compared with the hCaV2.1 channel, TCaV2 is somewhat hyperexcitable, at least out under in vitro conditions, in the sense that it is less susceptible to inactivation and more readily activated by depolarization. Of course, observations in vivo could be dramatically different, because hCaV2.1 is active at temperatures near 37 °C, whereas TCaV2 is active at temperatures closer to 24–28 °C (
      • Pearse V.B.
      • Voigt O.
      Field biology of placozoans (Trichoplax): distribution, diversity, biotic interactions.
      ). Nevertheless, it is apparent that TCaV2 does not require a very hyperpolarized resting membrane potential to remain active, showing moderate to minimal inactivation at membrane voltages between −30 and −40 mV compared with hCaV2.1. This is in stark contrast to the recently characterized CaV2a channel from the cnidarian Nematostella vectensis (Fig. 1B), one of three CaV2 channel paralogues that appears to have specialized for stinging cell (cnidocyte) discharge. Expressed in HEK-293T cells, the recombinant channel produced high voltage–activated currents and a very left-shifted inactivation curve, rendering it susceptible to inactivation even at hyperpolarized potentials (
      • Weir K.
      • Dupre C.
      • van Giesen L.
      • Lee A.S.
      • Bellono N.W.
      A molecular filter for the cnidarian stinging response.
      ). Like hCaV2.1, TCaV2 exhibited biphasic recovery from inactivation, with a fast component similar to the human channel but a much slower secondary component. Thus, TCaV2 would be more susceptible to accumulated inactivation during bouts of prolonged excitation, resulting in a more substantial decline in Ca2+ influx over time. Last, we note that the kinetic properties for activation, inactivation, and deactivation are generally slower for TCaV2 compared with hCaV2.1, differences that are likely amplified when considering the acceleration of kinetics of hCaV2.1 at warm physiological temperatures and the slowing down of kinetics of TCaV2 at cooler seawater temperatures. An additional consideration that might further differentiate TCaV2 and hCaV2.1 in vivo is that the Trichoplax channel is surrounded by different salt compositions in seawater, including a roughly 5-fold higher external Ca2+ concentration. Nevertheless, despite some differences, we note that TCaV2 exhibits the core functional features of other CaV2 channels involved in synaptic transmission. This includes a dependence on the accessory subunits CaVβ and CaVα2δ, where efficient in vitro expression required co-expression with the rat subunits CaVβ1b and CaVα2δ1. This is similar to what was observed for the CaV2 channel cloned from the snail (
      • Spafford J.D.
      • Chen L.
      • Feng Z.-P.
      • Smit A.B.
      • Zamponi G.W.
      Expression and modulation of an invertebrate presynaptic calcium channel α1 subunit homolog.
      ), suggesting that the molecular determinants for interacting with these subunits (i.e. the AID for CaVβ and extracellular regions for CaVα2δ) are strongly conserved. Here, we did not clone and co-express the Trichoplax CaVβ or CaVα2δ subunit cDNAs; however, we note from our transcriptome work that the animal expresses one CaVβ subunit and three CaVα2δ subunit genes (
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ). Future studies will be needed to explore the molecular and functional properties of these divergent CaV channel accessory subunits.

      Specialization of CaV2 channels for fast synchronous exocytosis

      We were unable to identify a high-affinity pharmacological compound to block the TCaV2 channel in vitro that would facilitate exploration of its contributions to Trichoplax cellular physiology and behavior (
      • Dolphin A.C.
      A short history of voltage-gated calcium channels.
      ). A pertinent question is whether TCaV2 and Ca2+ influx play a role in regulated secretion, given that the animal expresses all of the necessary machinery, including the SNARE complex and associated genes, the exocytosis Ca2+ sensors synaptotagmin and complexin, and an array of “neuropeptides” that actively modulate Trichoplax motile behavior (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ,
      • Eitel M.
      • Francis W.R.
      • Varoqueaux F.
      • Daraspe J.
      • Osigus H.-J.
      • Krebs S.
      • Vargas S.
      • Blum H.
      • Williams G.A.
      • Schierwater B.
      • Wörheide G.
      Comparative genomics and the nature of placozoan species.
      ,
      • Srivastava M.
      • Begovic E.
      • Chapman J.
      • Putnam N.H.
      • Hellsten U.
      • Kawashima T.
      • Kuo A.
      • Mitros T.
      • Salamov A.
      • Carpenter M.L.
      • Signorovitch A.Y.
      • Moreno M.A.
      • Kamm K.
      • Grimwood J.
      • Schmutz J.
      • et al.
      The Trichoplax genome and the nature of placozoans.
      ,
      • Smith C.L.
      • Varoqueaux F.
      • Kittelmann M.
      • Azzam R.N.
      • Cooper B.
      • Winters C.A.
      • Eitel M.
      • Fasshauer D.
      • Reese T.S.
      Novel cell types, neurosecretory cells, and body plan of the early-diverging metazoan Trichoplax adhaerens.
      ,
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ,
      • Kamm K.
      • Osigus H.-J.
      • Stadler P.F.
      • DeSalle R.
      • Schierwater B.
      Trichoplax genomes reveal profound admixture and suggest stable wild populations without bisexual reproduction.
      ). Based on ultrastructural studies, Trichoplax cells contain both dense core and pale vesicles (
      • Smith C.L.
      • Varoqueaux F.
      • Kittelmann M.
      • Azzam R.N.
      • Cooper B.
      • Winters C.A.
      • Eitel M.
      • Fasshauer D.
      • Reese T.S.
      Novel cell types, neurosecretory cells, and body plan of the early-diverging metazoan Trichoplax adhaerens.
      ,
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ), suggesting that like other animals, they can secrete both peptide and small-molecule transmitters, respectively. However, the absence of highly clustered vesicles along the cell membrane, as occurs in the synapse active zone, suggests that Trichoplax cells do not carry out robust, synchronous secretion akin to that at the nerve terminal (
      • Smith C.L.
      • Varoqueaux F.
      • Kittelmann M.
      • Azzam R.N.
      • Cooper B.
      • Winters C.A.
      • Eitel M.
      • Fasshauer D.
      • Reese T.S.
      Novel cell types, neurosecretory cells, and body plan of the early-diverging metazoan Trichoplax adhaerens.
      ,
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ). Instead, Ca2+-dependent secretion in Trichoplax might be more similar to asynchronous, neuroendocrine secretion. In this regard, if co-expressed, there could be complementary contributions from CaV1 and/or CaV3 as occurs in neuroendocrine cells, which, in the case of CaV3, would permit graded subthreshold exocytosis (
      • Weiss N.
      • Hameed S.
      • Fernández-Fernández J.M.
      • Fablet K.
      • Karmazinova M.
      • Poillot C.
      • Proft J.
      • Chen L.
      • Bidaud I.
      • Monteil A.
      • Huc-Brandt S.
      • Lacinova L.
      • Lory P.
      • Zamponi G.W.
      • De Waard M.
      A Cav3.2/syntaxin-1A signaling complex controls T-type channel activity and low-threshold exocytosis.
      ).
      A key consideration regarding the role of CaV2 and other CaV channels in driving exocytosis in Trichoplax is the proximity of the channels to the exocytotic machinery. This is because the presynaptic calcium sensors that trigger vesicle fusion require substantial increases in cytoplasmic Ca2+ concentration (
      • Schneggenburger R.
      • Neher E.
      Intracellular calcium dependence of transmitter release rates at a fast central synapse.
      ,
      • Bertram R.
      • Sherman A.
      • Stanley E.F.
      Single-domain/bound calcium hypothesis of transmitter release and facilitation.
      ), which, on a global scale, would lead to cellular toxicity (
      • Clapham D.E.
      Calcium signaling.
      ). Instead, cytoplasmic Ca2+ plumes from open CaV channels are spatially restricted by rapid removal via Ca2+ pumps, exchangers, and chelation agents, resulting in regions near the channel pore of only 20–100 nanometers where Ca2+ concentrations reach appreciable levels (i.e. Ca2+ nanodomains) (
      • Stanley E.F.
      The nanophysiology of fast transmitter release.
      ). In synapses where CaV2 nanodomains are positioned very close to vesicles (i.e. “nanodomain coupling”), excitation-secretion coupling is thought to be more efficient and to require less total Ca2+ than synapses where the channels are further away (
      • Stanley E.F.
      The nanophysiology of fast transmitter release.
      ,
      • Südhof T.C.
      The presynaptic active zone.
      ,
      • Stanley E.F.
      Single calcium channels on a cholinergic presynaptic nerve terminal.
      ). When channels are positioned further, the probability of release and fidelity declines (
      • Stanley E.F.
      The nanophysiology of fast transmitter release.
      ,
      • Schneggenburger R.
      • Neher E.
      Intracellular calcium dependence of transmitter release rates at a fast central synapse.
      ) toward a configuration referred to as microdomain coupling. At microdomain synapses, plumes of Ca2+ from separate open channels are thought to sum into larger plumes, where they collectively saturate vesicular calcium sensors of fusion-ready vesicles located in the vicinity (
      • Eggermann E.
      • Bucurenciu I.
      • Goswami S.P.
      • Jonas P.
      Nanodomain coupling between Ca2+ channels and sensors of exocytosis at fast mammalian synapses.
      ). An advantage of microdomain synapses is that they are capable of activity-dependent facilitation, where repetitive bouts of excitation, such as trains of action potentials, lead to incremental rises in cytoplasmic Ca2+ and a nonlinear increase in the probability of release (
      • Stanley E.F.
      The nanophysiology of fast transmitter release.
      ).
      Given its ubiquity (
      • Clapham D.E.
      Calcium signaling.
      ), it is likely that cytoplasmic Ca2+ sequestration is also active in Trichoplax. Hence, should CaV channels indeed be driving exocytosis, they must be positioned somewhat close to vesicles, perhaps comprising functional modules held together by specific protein-protein interactions. This would be consistent with the proposal that physical coupling between CaV2 channels and one or more vesicles creates a functional module that can be aggregated at synapses but also deployed more sparsely for nonsynaptic exocytosis (
      • Stanley E.F.
      The calcium channel and the organization of the presynaptic transmitter release face.
      ), as is likely to be the case in Trichoplax. Conceivably then, evolution of the presynaptic terminal involved a proteomic aggregation of CaV2 channel-vesicle functional modules, permitting fast, synchronous secretion. Worth noting is that immunolocalization of TCaV2 in WGA-positive cells, which co-express the secretory endomorphin-like peptide (
      • Senatore A.
      • Reese T.S.
      • Smith C.L.
      Neuropeptidergic integration of behavior in Trichoplax adhaerens, an animal without synapses.
      ,
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ), revealed clustered expression along the outward-facing edge of cells (Fig. 3A, inset), perhaps representing regions for vesicle fusion. However, the nature of the required apposition between CaV channels and vesicles would be unclear, because different values of proximity are functional and known to exist (i.e. nanodomain versus microdomain) (
      • Stanley E.F.
      The nanophysiology of fast transmitter release.
      ,
      • Eggermann E.
      • Bucurenciu I.
      • Goswami S.P.
      • Jonas P.
      Nanodomain coupling between Ca2+ channels and sensors of exocytosis at fast mammalian synapses.
      ). Something that confounds this matter further is that the molecular underpinnings that differentiate nanodomain versus microdomain arrangements are not entirely clear, and in many synapses, there appears to be a developmental shift from a microdomain configuration to nanodomain (
      • Stanley E.F.
      The nanophysiology of fast transmitter release.
      ,
      • Eggermann E.
      • Bucurenciu I.
      • Goswami S.P.
      • Jonas P.
      Nanodomain coupling between Ca2+ channels and sensors of exocytosis at fast mammalian synapses.
      ,
      • Fedchyshyn M.J.
      • Wang L.-Y.
      Developmental transformation of the release modality at the calyx of Held synapse.
      ). We also have a limited understanding of how, and along which animal lineages, these various presynaptic arrangements evolved.
      Previously, we noted that the TCaV2 channel lacks an acidic amino acid motif at its extreme C terminus with a consensus sequence of (D/E)(D/E/H)WC-COOH, which is conserved in cnidarian and bilaterian CaV2 channels and essential for interactions with the PDZ domains of presynaptic scaffolding proteins Mint (
      • Spafford J.D.
      • Munno D.W.
      • van Nierop P.
      • Feng Z.-P.
      • Jarvis S.E.
      • Gallin W.J.
      • Smit A.B.
      • Zamponi G.W.
      • Syed N.I.
      Calcium channel structural determinants of synaptic transmission between identified invertebrate neurons.
      ,
      • Maximov A.
      • Südhof T.C.
      • Bezprozvanny I.
      Association of neuronal calcium channels with modular adaptor proteins.
      ,
      • Gardezi S.R.
      • Li Q.
      • Stanley E.F.
      Inter-channel scaffolding of presynaptic CaV2.2 via the C terminal PDZ ligand domain.
      ) and RIM (
      • Hirano M.
      • Takada Y.
      • Wong C.F.
      • Yamaguchi K.
      • Kotani H.
      • Kurokawa T.
      • Mori M.X.
      • Snutch T.P.
      • Ronjat M.
      • De Waard M.
      • Mori Y.
      C-terminal splice variants of P/Q-type Ca2+ channel CaV2.1 α1 subunits are differentially regulated by Rab3-interacting molecule proteins.
      ,
      • Kaeser P.S.
      • Deng L.
      • Wang Y.
      • Dulubova I.
      • Liu X.
      • Rizo J.
      • Südhof T.C.
      RIM proteins tether Ca2+ channels to presynaptic active zones via a direct PDZ-domain interaction.
      ,
      • Graf E.R.
      • Valakh V.
      • Wright C.M.
      • Wu C.
      • Liu Z.
      • Zhang Y.Q.
      • DiAntonio A.
      RIM promotes calcium channel accumulation at active zones of the Drosophila neuromuscular junction.
      ). TCaV2 also lacks additional C-terminal motifs, upstream of the extreme C terminus, associated with CaV2 channel presynaptic localization and/or function (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ,
      • Gardezi S.R.
      • Nath A.R.
      • Li Q.
      • Stanley E.F.
      Characterization of a synaptic vesicle binding motif on the distal CaV2.2 channel C-terminal.
      ,
      • Lübbert M.
      • Goral R.O.
      • Satterfield R.
      • Putzke T.
      • van den Maagdenberg A.M.
      • Kamasawa N.
      • Young Jr., S.M.
      A novel region in the CaV2.1 α1 subunit C-terminus regulates fast synaptic vesicle fusion and vesicle docking at the mammalian presynaptic active zone.
      ,
      • Snidal C.A.
      • Li Q.
      • Elliott B.B.
      • Mah H.K.-H.
      • Chen R.H.
      • Gardezi S.R.
      • Stanley E.F.
      Molecular characterization of an SV capture site in the mid-region of the presynaptic CaV2.1 calcium channel C-terminal.
      ). With respect to the Ca2+ nanodomain arrangement, RIM has received considerable attention, because it has the capacity to directly interact with CaV2 channels and the vesicular protein Rab3 (
      • Wang Y.
      • Okamoto M.
      • Schmitz F.
      • Hofmann K.
      • Südhof T.C.
      Rim is a putative Rab3 effector in regulating synaptic-vesicle fusion.
      ), and its genetic deletion in both vertebrates and invertebrates causes reduced localization of CaV2 channels at the synapse active zone and disrupted synaptic transmission (
      • Südhof T.C.
      The presynaptic active zone.
      ,
      • Kaeser P.S.
      • Deng L.
      • Wang Y.
      • Dulubova I.
      • Liu X.
      • Rizo J.
      • Südhof T.C.
      RIM proteins tether Ca2+ channels to presynaptic active zones via a direct PDZ-domain interaction.
      ,
      • Graf E.R.
      • Valakh V.
      • Wright C.M.
      • Wu C.
      • Liu Z.
      • Zhang Y.Q.
      • DiAntonio A.
      RIM promotes calcium channel accumulation at active zones of the Drosophila neuromuscular junction.
      ,
      • Kushibiki Y.
      • Suzuki T.
      • Jin Y.
      • Taru H.
      RIMB-1/RIM-binding protein and UNC-10/RIM redundantly regulate presynaptic localization of the voltage-gated calcium channel in Caenorhabditis elegans.
      ). Although Trichoplax possesses a RIM homologue, the gene lacks a PDZ domain (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ), and in conjunction with the absence of a CaV2 channel (D/E)(D/E/H)WC-COOH motif, it is unlikely that TCaV2 is incorporated into homologous RIM-associated proteomic complexes, as reported in animals with synapses. However, this does not preclude the possibility that other redundant presynaptic interactions are present and conserved, where, for example, RIM can interact with the calcium channel CaVβ subunit (
      • Kiyonaka S.
      • Wakamori M.
      • Miki T.
      • Uriu Y.
      • Nonaka M.
      • Bito H.
      • Beedle A.M.
      • Mori E.
      • Hara Y.
      • De Waard M.
      • Kanagawa M.
      • Itakura M.
      • Takahashi M.
      • Campbell K.P.
      • Mori Y.
      RIM1 confers sustained activity and neurotransmitter vesicle anchoring to presynaptic Ca2+ channels.
      ) and another CaV2 channel-binding protein, RIM-BP (
      • Südhof T.C.
      The presynaptic active zone.
      ). Furthermore, additional interactions that operate independently of RIM (
      • Gardezi S.R.
      • Nath A.R.
      • Li Q.
      • Stanley E.F.
      Characterization of a synaptic vesicle binding motif on the distal CaV2.2 channel C-terminal.
      ,
      • Lübbert M.
      • Goral R.O.
      • Satterfield R.
      • Putzke T.
      • van den Maagdenberg A.M.
      • Kamasawa N.
      • Young Jr., S.M.
      A novel region in the CaV2.1 α1 subunit C-terminus regulates fast synaptic vesicle fusion and vesicle docking at the mammalian presynaptic active zone.
      ,
      • Snidal C.A.
      • Li Q.
      • Elliott B.B.
      • Mah H.K.-H.
      • Chen R.H.
      • Gardezi S.R.
      • Stanley E.F.
      Molecular characterization of an SV capture site in the mid-region of the presynaptic CaV2.1 calcium channel C-terminal.
      ) could also be conserved, at motifs that are not immediately detectable in protein alignments due to rapid divergence in ligand specificity, as reported for ligands of Src Homology 3 domains (
      • Li S.S.-C.
      Specificity and versatility of SH3 and other proline-recognition domains: structural basis and implications for cellular signal transduction.
      ).
      Interestingly, we recently discovered that Trichoplax possesses a second class of RIM homologues (dubbed RIM-II), which does bear a PDZ domain but with differences in key regions that suggests different ligand specificity compared with the canonical RIM (i.e. RIM-I) (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ). RIM-II is broadly conserved in animals, present even in chordates, but was lost multiple times independently, including in vertebrates. Notably, ctenophores, proposed to have independently evolved the synapse (
      • Moroz L.L.
      • Kohn A.B.
      Independent origins of neurons and synapses: insights from ctenophores.
      ), have RIM-II and lack RIM-I, making them the only animals with synapses to not have a RIM-I homologue. Whether RIM-II functions at the synapse is not known; however it is expressed in neurons and neuroendocrine cells in the snail, consistent with a role in secretion (
      • Piekut T.
      • Wong Y.Y.
      • Walker S.E.
      • Smith C.L.
      • Gauberg J.
      • Harracksingh A.N.
      • Lowden C.
      • Novogradac B.B.
      • Cheng H.-Y.M.
      • Spencer G.E.
      • Senatore A.
      Early metazoan origin and multiple losses of a novel clade of RIM pre-synaptic calcium channel scaffolding protein homologues.
      ). Future work exploring the proteomic interactions and subcellular localization of the Trichoplax CaV2 channel will help clarify its positioning relative to the exocytotic machinery and the homology of protein complexes for its localization.
      Indirect evidence that Trichoplax is capable of regulated secretion comes from studies on neuropeptide homologues and the small-molecule transmitter glycine that, when applied ectopically, elicit behaviors that emulate those observed naturally. For example, Trichoplax expresses endomorphin-like peptide in secretory cells that line the edge of the flat, disc-shaped animal, and ectopic application of this peptide causes Trichoplax to stop moving via cessation of ciliary beating on its ventral epithelium (
      • Senatore A.
      • Reese T.S.
      • Smith C.L.
      Neuropeptidergic integration of behavior in Trichoplax adhaerens, an animal without synapses.
      ). Other compounds, also expressed in secretory cells at various anatomical locations, can similarly alter Trichoplax behavior, including increased rotation, flattening, or crinkling/writhing into a ball (
      • Varoqueaux F.
      • Williams E.A.
      • Grandemange S.
      • Truscello L.
      • Kamm K.
      • Schierwater B.
      • Jékely G.
      • Fasshauer D.
      High cell diversity and complex peptidergic signaling underlie placozoan behavior.
      ,
      • Sebé-Pedrós A.
      • Chomsky E.
      • Pang K.
      • Lara-Astiaso D.
      • Gaiti F.
      • Mukamel Z.
      • Amit I.
      • Hejnol A.
      • Degnan B.M.
      • Tanay A.
      Early metazoan cell type diversity and the evolution of multicellular gene regulation.
      ). More recently, ectopic application of the small-molecule transmitter glycine was found to elicit concentration-dependent effects on Trichoplax behavior, with increased frequency of ciliary beating occurring at low (micromolar) concentrations and whole-body contractions at millimolar concentrations (
      • Romanova D.Y.
      • Heyland A.
      • Sohn D.
      • Kohn A.B.
      • Fasshauer D.
      • Varoqueaux F.
      • Moroz L.L.
      Glycine as a signaling molecule and chemoattractant in Trichoplax (Placozoa): insights into the early evolution of neurotransmitters.
      ). Altogether, these various observations suggest that these compounds are causative agents that underlie changes in Trichoplax behavior and, by extension, that their secretion must occur in a regulated fashion such that behaviors can be coordinated.
      Here, using a rigorously verified custom antibody, we show that the CaV2 channel is expressed in type-II gland cells also known to express the endomorphin-like peptide and mucous-bearing vesicles that stain with WGA (
      • Mayorova T.D.
      • Hammar K.
      • Winters C.A.
      • Reese T.S.
      • Smith C.L.
      The ventral epithelium of Trichoplax adhaerens deploys in distinct patterns cells that secrete digestive enzymes, mucus or diverse neuropeptides.
      ). TCaV2 was also expressed in other cells along the outer edge of the dorsal epithelium, in areas consistent with other peptide-expressing cells (
      • Varoqueaux F.
      • Williams E.A.
      • Grandemange S.
      • Truscello L.
      • Kamm K.
      • Schierwater B.
      • Jékely G.
      • Fasshauer D.
      High cell diversity and complex peptidergic signaling underlie placozoan behavior.
      ). Future work will involve determining the cellular co-expression of the three Trichoplax CaV channels, to provide a framework for appreciating the complementary and differential contributions of the different channels to cellular physiology. Previously, we documented that the TCaV3 channel is also expressed in cells along the periphery of the animal (
      • Smith C.L.
      • Abdallah S.
      • Wong Y.Y.
      • Le P.
      • Harracksingh A.N.
      • Artinian L.
      • Tamvacakis A.N.
      • Rehder V.
      • Reese T.S.
      • Senatore A.
      Evolutionary insights into T-type Ca2+ channel structure, function, and ion selectivity from the Trichoplax adhaerens homologue.
      ). However, because both antibodies were generated in rabbits, co-localization of the TCaV2 and TCaV3 has not been performed; nor has co-localization with TCaV1. We hope that ongoing generation of custom polyclonal antibodies in rats will permit effective co-localization experiments.
      Downstream of the secretion process, questions also remain about the receptors that make cellular communication possible in Trichoplax. For both neuropeptides and glycine, the most likely receptors are GPCRs and peptide- or glycine-gated ion channels. Based on genomic work by ourselves and others, Trichoplax was found to express over 656 GPCRs, many of which are homologous to known neuropeptide receptors. Additionally, Trichoplax expresses an array of intracellular signaling components, including the G-protein βγ subunits sequenced and cloned in this study (
      • Srivastava M.
      • Begovic E.
      • Chapman J.
      • Putnam N.H.
      • Hellsten U.
      • Kawashima T.
      • Kuo A.
      • Mitros T.
      • Salamov A.
      • Carpenter M.L.
      • Signorovitch A.Y.
      • Moreno M.A.
      • Kamm K.
      • Grimwood J.
      • Schmutz J.
      • et al.
      The Trichoplax genome and the nature of placozoans.
      ,
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ,
      • Kamm K.
      • Osigus H.-J.
      • Stadler P.F.
      • DeSalle R.
      • Schierwater B.
      Trichoplax genomes reveal profound admixture and suggest stable wild populations without bisexual reproduction.
      ). Inferred from studies done in other early-diverging animals (
      • Williams E.A.
      • Verasztó C.
      • Jasek S.
      • Conzelmann M.
      • Shahidi R.
      • Bauknecht P.
      • Mirabeau O.
      • Jékely G.
      Synaptic and peptidergic connectome of a neurosecretory center in the annelid brain.
      ,
      • Jékely G.
      Global view of the evolution and diversity of metazoan neuropeptide signaling.
      ), it is likely that some of the regulation of Trichoplax behavior by secreted substances occurs through GPCR signaling, in particular processes that are slower and long-lasting, akin to neuromodulation in the nervous system (
      • Marder E.
      Neuromodulation of neuronal circuits: back to the future.
      ). Trichoplax also expresses genes for degenerin/ENaC sodium channels that, in molluscs, vertebrates, and cnidarians, can be gated by neuropeptides (
      • Osmakov D.I.
      • Koshelev S.G.
      • Ivanov I.A.
      • Andreev Y.A.
      • Kozlov S.A.
      Endogenous neuropeptide nocistatin is a direct agonist of acid-sensing ion channels (ASIC1, ASIC2 and ASIC3).
      ,
      • Lingueglia E.
      • Deval E.
      • Lazdunski M.
      FMRFamide-gated sodium channel and ASIC channels: a new class of ionotropic receptors for FMRFamide and related peptides.
      ,
      • Golubovic A.
      • Kuhn A.
      • Williamson M.
      • Kalbacher H.
      • Holstein T.W.
      • Grimmelikhuijzen C.J.
      • Gründer S.
      A peptide-gated ion channel from the freshwater polyp Hydra.
      ) and are proposed to mediate synaptic transmission in hydra (
      • Gründer S.
      • Assmann M.
      Peptide-gated ion channels and the simple nervous system of Hydra.
      ). In our ongoing effort to identify peptide-gated channels in Trichoplax, we recently reported that one of the 11 known degenerin/ENaC homologues functions as a Na+ leak channel sensitive to block by external Ca2+ and H+ ions (
      • Elkhatib W.
      • Smith C.L.
      • Senatore A.
      A Na+ leak channel cloned from Trichoplax adhaerens extends extracellular pH and Ca2+ sensing for the DEG/ENaC family close to the base of Metazoa.
      ), whereas others are gated by protons similar to acid-activated channels from vertebrates and other deuterostomes.
      W. Elkhatib and A. Senatore, unpublished observations.
      Whether some of the Trichoplax degenerin/ENaC channels can also be activated by peptides remains to be determined, a capability that would enable much faster and transient peptidergic signaling than GPCRs, playing out over milliseconds compared to seconds or longer. Last, we note that Trichoplax also expresses a considerable number of ionotropic glutamate receptors homologous to NMDA/AMPA/kainate receptors (
      • Romanova D.Y.
      • Heyland A.
      • Sohn D.
      • Kohn A.B.
      • Fasshauer D.
      • Varoqueaux F.
      • Moroz L.L.
      Glycine as a signaling molecule and chemoattractant in Trichoplax (Placozoa): insights into the early evolution of neurotransmitters.
      ) that, based on work done in ctenophores (
      • Alberstein R.
      • Grey R.
      • Zimmet A.
      • Simmons D.K.
      • Mayer M.L.
      Glycine activated ion channel subunits encoded by ctenophore glutamate receptor genes.
      ), are possibly more sensitive to glycine than they are to glutamate. Indeed, continued functional characterization of these various receptors will be of value toward our understanding of cellular communication in Trichoplax, in addition to understanding the capacity of CaV2 and other CaV channels for driving regulated secretion of signaling compounds that target these receptors.

      On the absence of Gβγ modulation of TCaV2 in vitro

      Arguably, understanding how the nervous system evolved requires a deep understanding not only of the emergence of fast electrical signaling through synapses and neural circuits, but also how slow neuromodulatory processes co-evolved to regulate the fast signaling machinery (
      • Wong Y.Y.
      • Le P.
      • Elkhatib W.
      • Piekut T.
      • Senatore A.
      Transcriptome profiling of Trichoplax adhaerens highlights its digestive epithelium and a rich set of genes for fast electrogenic and slow neuromodulatory cellular signaling.
      ,
      • Katz P.S.
      • Lillvis J.L.
      Reconciling the deep homology of neuromodulation with the evolution of behavior.
      ). Even simple neural circuits are subject to extensive and complex neuromodulation, which can alter ion channel properties and synaptic proteins differently in different neurons for changes in excitability, synaptic connectivity, neural circuit function, and, ultimately, behavior (
      • Marder E.
      Neuromodulation of neuronal circuits: back to the future.
      ). Such an integration occurs for presynaptic CaV2 channels, where various transmitters bind their cognate GPCRs to exert modulatory action on the channels via two distinct pathways: 1) a relatively fast pathway, mediated by direct binding of G-protein βγ heterodimers for voltage-dependent inhibition (
      • Stanley E.
      • Mirotznik R.
      Cleavage of syntaxin prevents G-protein regulation of presynaptic calcium channels.
      ) and 2) a slower pathway that involves downstream second messengers and effector enzymes, such as protein kinases A and C, which phosphorylate CaV channels and their associated proteins to alter their function (
      • Burke Jr., K.J.
      • Bender K.J.
      Modulation of ion channels in the axon: mechanisms and function.
      ,
      • Zamponi G.W.
      • Currie K.P.
      Regulation of CaV2 calcium channels by G protein coupled receptors.
      ). Generally, binding of Gβγ to CaV2 channels shifts their voltage dependence of activation to more depolarized potentials and causes a slowing down of activation kinetics, leading to reduced macroscopic current and Ca2+ influx (
      • Tedford H.W.
      • Zamponi G.W.
      Direct G protein modulation of Cav2 calcium channels.
      ,
      • Zamponi G.W.
      • Currie K.P.
      Regulation of CaV2 calcium channels by G protein coupled receptors.
      ). Strong depolarizations can alleviate Gβγ binding, permitting a temporary relief of neuromodulatory inhibition of presynaptic CaV2 channels, observed for example during bouts of heightened electrical activity, such as action potential bursts. This form of regulation appears conserved between vertebrates and invertebrates, present in neurons isolated from the snail central nervous system (
      • Dong N.
      • Lee D.W.
      • Sun H.-S.
      • Feng Z.-P.
      Dopamine-mediated calcium channel regulation in synaptic suppression in L. stagnalis interneurons.
      ,
      • de Hoog E.
      • Lukewich M.K.
      • Spencer G.E.
      Retinoid r