Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 18 February 2019

Invertebrate Genomics Provide Insights Into the Origin of Synaptic Transmission

Abstract and Keywords

During the evolution of synapses, existing molecules were exapted to serve in specific synaptic roles. Recent increased availability of assembled transcriptomes from organisms that evolved before and after the appearance of the earliest synapses provides the opportunity to trace molecular adaptations important for development of fast synaptic transmission. We discuss issues that affect transcriptome assembly and phylogenetic analysis, and which therefore impact this analysis. We use relatively recent transcriptomes of pre-bilaterians to examine the molecular evolution of three types of critical synapse-specific proteins: vesicular transporters, synaptotagmins and ionotropic glutamate receptors. The results emphasize the fundamental difficulties in defining the specific point at which a protein “assumes” a synaptic function. Nevertheless, the analysis informs our understanding of several major evolutionary topics, including the evolution of synaptic vesicles and the identity of the first neurotransmitter used for fast, synchronous transmission. This analysis is also relevant for the current discussion of whether neuronal and synaptic function evolved separately, once in ctenophores and once in cnidarians and the main bilaterian lineage.

Keywords: vesicular transporters, synapse, evolution, transcriptome, synaptotagmin, synaptic transmission, synaptic vesicles, ctenophores, cnidarians, glutamate receptors

As sequencing of transcriptomes and genomes becomes less expensive and can be accomplished in ever shorter times with advances in next-generation sequencing (NGS) techniques, there is the expectation that our molecular and genetic understanding of evolution will advance dramatically. In particular, it is anticipated that we should be able to resolve questions about the evolutionary origin of the nervous system and complex behavior. These are fundamental issues that have intrigued evolutionary biologists, ethologists, and neurobiologists for much of the past century. With new genomic and transcriptomic sequencing, novel insights about the evolution of neuronal proteins arise and lead to high-impact publications in leading journals. The high level of interest that these papers attract is not surprising because at the core of these questions, we are seeking to understand the human mind and how it came to evolve. Nevertheless, as with many of the most exciting and unexpected scientific conclusions, some skepticism and reexamination is appropriate. As this chapter will suggest, although NGS and modern genomic and transcriptomic techniques are extremely powerful, reaching an understanding of the origin of the nervous system is a complex process that is potentially fraught with serious problems.

This introductory section addresses some general problems in this evolutionary analysis. As the production of reads of genomic or transcriptomic sequences becomes faster and less expensive, researchers have available hundreds of millions or billions of reads. Because current technologies can sequence hundreds of millions of nucleotide base pairs from a single DNA or RNA sample, bioinformatic tools have been developed to process these immense data sets and efficiently assemble genomes or transcriptomes. Assembly of genomes faces the challenge of unambiguous assembly across repeated sequences that result from highly conserved paralogs, and short repeats, which are common in intronic sequences, and across recurring, conserved regulatory sequences. There are a number of strategies that have been employed to address some of these assembly challenges, including generating libraries with different fragment lengths or using multiple platforms (e.g., a combination of PacBio long reads and Illumina short reads) to generate better genome assemblies. Similarly, while transcriptome assemblies benefit from the shorter total length of transcript sequences and fewer repeated regions compared with the genome, they still face assembly challenges. Some of these challenges include the large differences in relative expression levels of transcripts (some are selectively expressed in rare cell types, whereas others are ubiquitous “housekeeping” genes) that can confound sequencing as well as the downstream assemblers. In addition to bioinformatic challenges, there are upstream issues with both library preparation and with sequencing; for example, GC content can affect both library preparation and sequencing. Another challenge is that some signaling or stress-induced transcripts are minimally expressed, except under specific circumstances. To enable efficient assembly of massive numbers of reads, modern assemblers use a kmer graph-based approach to assemble the reads. By examining subregions within reads, known as kmers, to build a graph and using this graph to assemble the reads, the assemblers reduce the computational burden of the massive quantity of reads. However, the assembly process is statistical and stochastic, and thus not all resulting assembled sequences are biologically correct. Some errors in assembly of both transcriptomes and genomes are inevitable. That the assembly process is neither automatic nor straightforward can be appreciated from the fact that more than four dozen different assembler programs have been developed in an attempt to make assembly more efficient and more accurate.

Gene Loss and Assembly Errors

When one observes that an expected gene is missing in a group of animals that diverged early, it is reasonable to assume that this gene evolved in a later lineage. For example, the evidence that Porifera lack neurons and neuron-associated genes has been argued as supporting the hypothesis that basal ctenophores evolved their nervous system independently of cnidarian and bilaterian animals (Moroz, 2009, 2015; Moroz et al., 2014). However, loss of genes is an important mechanism in evolution. Rather than being exceptional, large-scale gene loss is a frequent occurrence in evolutionary lineages. To consider one example, the family of WNT-A genes, which are secreted signaling molecules that control developmental decisions, is present in echinoderms, but not in chordates, and is also found in mollusks and most arthropods, but not in Drosophila (Albalat & Canestro, 2016). Within Cnidaria, the parasitic Myxozoa lost much of their genomes; for example, the microscopic Kudoa iwatai lost more than 70% of its protein coding genes compared with cnidarians from other families (Chang et al., 2015). The flatworm Schmidtea mediterranea has lost nearly half of the ancestral gene families. The tunicate Oikopleura has undergone extensive gene loss, even compared with other tunicate groups. The chicken lineage appears to have lost genes at an order of magnitude higher rate than the mouse lineage, and the ancestor of the pufferfish Dichotomyctere underwent particularly high rates of gene loss. Among arthropods, the crustacean lineage has had substantially less gene loss than insects. Among insects, the red flour beetle Tribolium ancestor lost substantially fewer genes than most insects, whereas the honeybee ancestor lost genes at approximately twice the rate of these beetles (Albalat & Canestro, 2016).

The absence of a gene in a genome or transcriptome may give rise to interesting interpretations. However, the quality of both genome and transcriptome assemblies varies dramatically, and the absence of a gene may simply reflect incompleteness of the assembly. How much weight can be given to the absence of a gene? How often are expected genes simply missed due to low sequencing coverage or problems with an assembly? Is this an occurrence that is so infrequent that it can be readily ignored? Here we highlight three examples from research on molluskan models. In our recent sequencing and assembly project that is designed to achieve a complete transcriptome for the important neurobiological model system Aplysia californica, we were struck by the minimal evidence for expression of two genes in the central nervous system (CNS): neurotrophin and cytoplasmic polyadenylation element binding protein (CPEB). CPEB has been suggested to play a central role in long-term synaptic facilitation by regulating protein expression at synapses (Si et al., 2003, 2010). Yet, despite deep coverage of Aplysia CNS transcripts in Illumina sequencing, with read sets exceeding 1 billion, neither the short nor long form of the CPEB transcript was successfully assembled by either the Trinity or Oases assemblers. One could easily reach the conclusion that CPEB is not expressed in Aplysia; however, we were using the published sequence for Aplysia CPEB as a reference. In contrast, assemblies of reads from several peripheral tissues provided complete coverage of the CPEB transcript (although in two long fragments). Aplysia neurotrophin has also been implicated in long-term synaptic plasticity (Kassabov et al., 2013). Remarkably, neurotrophin is expressed in foot muscle and ovotestis, yet the expression levels in CNS were too low for assembly of more than 25% of the transcript length (which would be too short a sequence for identification of the gene, had it not been previously cloned). Presumably, this important signaling molecule is upregulated in the context of memory formation and also during synaptic development. A second example of incomplete transcript coverage in an assembly comes from the sponge Spongilla lacustris (Riesgo et al., 2014). Its transcriptome includes three ionotropic glutamate receptors; however, these sequences are shorter than 10% of the length of a full vertebrate glutamate receptor. One might conclude sponges lack functional glutamate receptors, except that the more complete transcriptome of Oscarella demonstrates major expansion of these ion channels. A third example comes from the vertical lobe of the cephalopod octopus, where synaptic plasticity has been implicated in memory formation. Winters et al. (2015, 2016) reported the absence of both nitric oxide synthase (NOS) and choline acetyl transferase in a vertical lobe transcriptome, raising the suggestion that cephalopods do not use either acetylcholine (ACh) or nitric oxide (NO) in synapse modification during learning. These investigators also reported the absence of glutamic acid decarboxylase, which could lead to the conclusion that the vertical lobes do not employ the major inhibitory neurotransmitter, gamma-amino butyric acid (GABA). However, pharmacological analyses do suggest that both ACh and NO are signaling molecules in the vertical lobes (Shomrat et al., 2011, 2015; Turchetti-Maia et al., 2016), and Casini et al. (2012) found immunoreactive ChAT in vertical lobe fibers, as did Hochner and colleagues (personal communication).

The converse error can also occur in assemblies: the apparent presence of a gene that is not actually found in a particular species. On occasion, a contaminating sequence can lead to the erroneous conclusion that a missing gene is actually expressed (see discussion of synaptotagmins in Pleurobrachia later in this chapter). Finally, there can be errors in annotation, where a sequence is incorrectly assigned. Indeed, most annotations of genomes are automated and do not reflect real phylogenetic analysis. For an example, see our later discussion of annotations of Vgluts/Sialins.

Computational Phylogenetics

The generation of phylogenetic trees through computational phylogenetics is frequently used to analyze evolutionary relationships among related genes or prologs. Although the resulting trees provide clear representations of evolutionary relationships, it is not sufficiently appreciated that these trees represent computational inferences from assembled sequences. Each of the various methods for phylogenetic predictions, such as distance matrix, maximum parsimony, maximum likelihood, and Bayesian methods, is statistical and stochastic; these approaches are subject to a number of limitations and potential errors, which can result in the misplaced grouping of orthologs in phylogenetic trees. For example, long-branch attraction, where rapidly evolving genes tend to become artifactually linked, and the stochastic statistical nature of phylogenetic predictions can lead to erroneous evolutionary conclusions (Yang & Rannala, 2012). Groupings in trees may also be sensitive to the choice of species and the specific genes to include, and of course to the specific method used. (Joe Felsenstein lists ~400 phylogeny software packages on the PHYLIP website.) In our generation of trees discussed later using RAxML, we have observed that the selection of species included in each phylogenetic analysis somewhat affects the outcome. Moreover, when genes are exapted (or adapted) for a new function, they often diverge very quickly, making it problematic to determine the ancestral gene.

Evolution of Synaptic Transmission

What were the evolutionary forces that led to the need for a nervous system? Perhaps the earliest function for synaptic transmission, as suggested by Cavalier-Smith (2017), may have been in hydrozoan planula larvae, for signaling to nematocytes to initiate attachment to substrates via nematocyst release (Namikawa et al., 1993). Cnidarian nematocytes are actually postsynaptic effectors (Scappaticci & Kass-Simon, 2008), and they may have evolved to mediate substrate attachment. With the evolution of early animals more generally came the ability to capture prey through rapid strike behavior. This was accompanied by the competing selective pressure to successfully escape from predators, either through rapid movement or withdrawal into a protected niche. Thus, for both predators and prey, speed was essential. Therefore, a key evolutionary function for the nervous systems of early animal species was enabling rapid movement in response to external stimuli. For movements to be effective, they also need to be appropriately coordinated to achieve the correct movement trajectory, toward prey or away from potential predators. Movement by multicellular organisms depends on specialized tissues coordinating movements; sensory receptors transduce external signals, interneuronal networks integrate signals from the environment, and coordinated contraction of muscle fibers is essential. For signaling throughout an animal and for integration of signals, which is essential for appropriate directional movement, rapid synaptic transmission is essential, as is efficient propagation of action potentials. Fast synaptic transmission is also critical for efficient activation of muscle fibers to produce movements.

In summary, the evolution of fast synaptic transmission is one of the key components in the evolution of the nervous system. The use of modern genomic methods to understand the evolution of fast synaptic transmission is the general focus of this review chapter.

It is remarkable that many of the key synaptic genes actually evolved prior to metazoans. Thus, during their evolution, early nervous systems appropriated genes used in nonmetazoan species, using these genes for new functions, in a process known as exaptation or cooption (Gould & Vrba, 1982). This is particularly true for ionotropic glutamate receptors, as well as genes in the synaptotagmin family. Glutamate receptors resembling those utilized in neuronal synapses evolved in bacteria and have importance signaling functions in plants. Synaptotagmin-like proteins in plants have been suggested to mediate Ca2+-triggered exocytosis (Wang et al., 2015; Kim et al., 2016).

Cellular communication clearly predates the nervous system, and many of the features of synaptic transmission are shared with other nonneuronal forms of cellular communication. For example, peptidergic transmission is present in fungi and most of the molecular machinery for exocytosis (i.e., SNARES, synaptotagmin, Munc [Burkhardt et al., 2011], and complexins [Yang et al., 2015]) that is critical for synaptic transmission had already evolved before the synaptic vesicle, presumably to mediate calcium-dependent release of secretory granules containing peptides. Indeed, adaptations to speed may have been first achieved through by the evolution of ligand-gated peptide receptors, which appear to be much more common in primitive nervous systems (Grunder & Assmann, 2015; Jekely et al., 2015).

The Earliest Nervous Systems

Understanding the evolution of the nervous system requires careful examination of prebilaterians, in particular ctenophores and cnidarians. It is generally accepted that cnidarians, such as Nematostella, Acropora, and Hydra, have a nervous system, but the relationship to the nervous systems of bilaterians is uncertain; in particular, the neurotransmitters used by cnidarians are not well defined. Placazoa and Porifera (sponges) are two prebilaterian phyla that do not have a defined nervous system, although these animals clearly contain specialized secretory cells that play roles in behaviors that require coordination between cell types. For example, flask cells in sponges sense cues from the environment and trigger settlement and metamorphosis via a Ca2+-dependent process, which possibly involves a neurosecretory-like mechanism (Nakanishi et al., 2015). The placazoan Trichoplax, which is arguably the simplest of metazoa, senses algae and responds by halting movement and releasing granules locally from a specialized lipophil secretory cell in a process that lyses the algae (Smith et al., 2014, 2015). Comb jellies (Mnemiopsis and Pleurobrachia) have a primitive nervous system, though it has recently been suggested that the ctenophore nervous system evolved independently of the cnidarian and bilaterian nervous systems (Moroz et al., 2014). Given that comb jellies apparently diverged before sponges and placazoans, if cnidarian and ctenophore nervous systems were derived from a common ancestor, then sponges and placozoans must have both lost a number of nervous system specializations (Fig. 1); this has been argued to be unlikely by some, though not all, researchers (Ryan, 2014; Moroz & Kohn, 2016). Although the loss of neuronal genes may seem implausibly maladaptive, as mentioned earlier, this occurred in the evolution of the Myxozoa, microscopic parasitic cnidarians that underwent a dramatic reduction in genome size (Chang et al., 2015). Some investigators argue that the phylogenetic evidence that Ctenophora diverged earliest from the metazoan lineage is ambiguous (Jekely et al., 2015), and that more appropriate phylogenetic analyses place Ctenophora and Cnidaria as sister phyla (Philippe et al., 2009); this would eliminate the need to invoke independent origins of the nervous system or the loss of neuronal genes in sponges and placazoans. A more recent analysis by Feuda et al. (2017) who incorporated patterns of site specific amino acid replacement in their models concluded that Porifera is the sister lineage to other animals, placing Ctenophera closer to Cnidaria, with Placazoa interposed between these two phyla (Fig. 1). This conclusion is consistent with the proposal of Cavalier-Smith (2017) that sponges evolved directly from single-celled choanoflagellates, an event that represents the origin of multicellularity. Finally, to determine which synapse-related genes were present in the ancestors to all metazoans, we have also examined unicellular holozoans, the sister groups to metazoans, which branched off earlier (e.g., Capsaspora, Salpingoeca, Monosiga, and Creolimax). To compare these prebilaterian species to bilaterians, we have chosen a small group of bilaterians. The list of both bilaterian and prebilaterian species is given in Table 1.

Invertebrate Genomics Provide Insights Into the Origin of Synaptic TransmissionClick to view larger

Figure 1 Evolutionary relationships of holozoan groups analyzed in this chapter. Major divisions are indicated by red lines. Two alternative models place either Porifera or Ctenophora as a sister group, diverging early from the main animal lineage (indicated by broken or dashed lines, respectively) (see Feuda et al., 2017). Two alternative hypotheses to account for the lack of neurons and nervous systems in Placazoa are indicated by arrowheads and Xs. In one model, the primitive nervous system evolved once (green arrowhead), but genes specifying neurons and synapses were then lost in the poriferan and placazoan lineages (green Xs). (In the Porifera as sister lineage model, gene loss is required only in the placazoan lineage, hence the absence of an X on the broken line to Porifera.) In an alternate hypothesis, the nervous system evolved independently in the ctenophore lineage and in the cnidarian-bilaterian lineage (blue arrowheads); no loss of genes is required for this second hypothesis. Note, because a number of synaptic genes are present in Trichoplax, gene loss does not mean that all nervous system-related genes were lost, but rather select genes required for specifying development of neuronal cell types and synaptic connections.

Table 1 Phylogenetic Groups and Species Analyzed

Unicellular Holozoa

Common Name


Creolimax fragrantissima

Single-celled protist †


Capsaspora owczarzaki

Single-celled protist ‡


Monosiga brevicollis

Single-celled choanoflagellate §


Salpingoeca rosetta

Colonial choanoflagellate§


Class: †Ichthyosporea, ‡Filasterea,



Mnemiopsis leidyi

Comb jelly


Pleurobrachia bachei

Comb jelly



Spongilla lacustris

Freshwater sponge


Amphimedon queenslandica



Oscarella carmela

Slime sponge



Trichoplax adhaerens




Nematostella vectensis

Starlet sea anemone


Acropora digitifera

Stony coral


Hydra vulgaris

Freshwater hydroid



Schmidtea mediterranea




Capitella teleta

Polychaete worm



Crassostrea gigas



Aplysia californica

Sea hare


Octopus bimaculoides




Daphnia pulex

Water flea


Drosophila melanogaster



Apis mellifera




Saccoglossus kowalevskii

Acorn worm



Brachiostoma floridae




Danio rerio

Zebra fish


Homo sapiens



Key Proteins Important for the Evolution of Fast Synaptic Transmission

What proteins should be examined to elucidate the evolution of the nervous system? Fast synaptic transmission requires synaptic vesicles. The evolution of the synaptic vesicle required a number of innovations: the sorting of the molecular machinery for release into a new type of vesicle; the formation of an active zone for release; vesicular proteins that mediate rapid vesicle exocytosis; and mechanisms for endocytic recovery of the vesicle lipid and vesicle membrane proteins. First, we focus on a key and defining feature of synaptic vesicles, the presence of a vesicular transporter to concentrate the neurotransmitter into the vesicle. Critically, the use of transporters to refill the vesicle allowed for efficient and rapid reuse of synaptic vesicles, an important evolutionary advantage for fast and efficient neurotransmission. Next, we examine the calcium sensor that is optimized for fast coordinated exocytosis, synaptotagmin. Finally, we examine the glutamate ligand-gated receptors, a key element required for fast synaptic transmission.

Vesicular Transporters

While all vesicular transporters are members of the solute carrier (SLC) family of transporters, distinct groups of vesicular transporters have derived independently from different branches of this family. The glutamate vesicular transporters (VGAT) derived from SLC17, whereas the vesicular monoamine transporters (VMAT) and the vesicular acetylcholine transporter (VAChT) derived from SLC18. Although these two families of transporters are closely linked in the SLC family, they had diverged before the evolution of vesicular transporters (see later discussion). The vesicular GABA transporters (VGATs) derived from SLC32, a completely distinct family of transporters (Fredriksson et al., 2008). Thus, these distinct groups of transporters presumably were selected and adapted for their role as vesicular neurotransmitter transporters independently. We will therefore examine each of these families separately.

Transporters for Acetylcholine and Biogenic Amines and the SLC18 Family

The SLC18 family contains three distinct types of transporters implicated in vesicular storage: VMAT, which transports all the biogenic amines (serotonin, dopamine, norepinephrine, epinephrine, and histamine, and also tyramine and octopamine in invertebrates); VAChT, which transports acetylcholine; and the recently identified VPAT, which transports polyamines and has been implicated in gliotransmission and secretion of spermine and spermidine from mast cells (Hiasa et al., 2014; Takeuchi et al., 2017). Phylogenetic analysis of this family (Fig. 2) demonstrates that while the VMAT and VAChT families form a well-defined clade in bilaterians, there are no members of these clades in prebilaterians, such as ctenophores and cnidarians. In contrast, cnidarians express SLC18 genes in the VPAT clade. There were no VPAT clade members found in other prebilaterians, although an SLC18 member distinct from these three main clades was found in all prebilaterians including Trichoplax, Mnemiopsis, and Porifera. Members of the SLC18 family were also seen in the unicellular holozoans Creolimax and Capsaspora, which predated metazoans. Thus SLC18 had diverged from SLC17 well before the evolution of synaptic vesicles.

VMAT and VAChT are closer to each other than they are to VPAT, suggesting initial divergence of the family into the ancestor to VPAT and the ancestor to VMAT and VAChT, followed by duplication and divergence of VMAT and VAChT. This divergence does not appear to have occurred before the bilaterian ancestor. It was surprising to not find a VMAT and VAChT family member in prebilaterians such as the cnidarians Hydra and Nematostella, as acetylcholine and biogenic amines are proposed to be used as neurotransmitters in these early animals (Kass-Simon & Pierobon, 2007), and nicotinic receptors and G protein–coupled receptors related to biogenic amine receptors are conserved (Anctil, 2009). VPAT has the ability to transport some biogenic amines (Hiasa et al., 2014), so it is possible that cnidarian members of this family, or other SLC18 family members could be vesicular transporters for acetylcholine or biogenic amines in prebilaterians. However, a search for genes coding the enzymes used for synthesis of ACh and biogenic amines suggests that cnidarians may not use either of these types of neurotransmitter. Choline acetyltransferase is a close relative of carnitine acetyltransferase, but all prebilaterian members of this family segregate with the carnitine acetyltransferases (Anctil, 2009). Tyrosine hydroxylase and tryptophan hydroxylase are key enzymes required in the synthesis of dopamine and norepinephrine and of serotonin, respectively. The closest hydroxylase in Nematostella segregates with phenylalanine hydroxylases, not tyrosine or tryptophan hydroxylase (Anctil, 2009). Thus, if prebilaterians used the SLC18 ancestor or VPAT as a transporter for biogenic amines or some other type of neurotransmitter, it may be a distinct transmitter from those used in the bilateria.

Invertebrate Genomics Provide Insights Into the Origin of Synaptic TransmissionClick to view larger

Figure 2 RAxML analysis of peptide sequences of SLC18 transporters. The three major clades of vertebrate transporters, VAChT, VMAT, and VPAT, are indicated by blue bars. The VMAT and VAChT clades are closer to each other than to the VPAT clade and contain no prebilaterian members, whereas the VPAT clade contains several cnidarian transporters. Other noteworthy groupings are indicated by green bars. Trichoplax expresses four distinct SLC18 transporters (¶) that have no close relationship to each other or to other SLC18 family members. There are two Mnemiopsis SLC18 members (§) that show weak similarity to Capsaspora SLC18s (†) and another that shows similarity to a distinct ichthyosporean transporter from Creolimax (††). There are also two poriferan SLC18 members that are related to one another, but not strongly to any other SLC18 family (*). The outgroup used was human VGAT (not shown in tree).

Transporters for Glutamate, Aspartate, and Nucleotides and the SLC17 Family

The SLC17 family has three transporters linked to vesicular transmission: vesicular glutamate transporters (VGLUTs, transporting glutamate); vesicular excitatory amino acid transporters (VEATs, transporting glutamate and aspartate [also known as sialins, which transport sialic acid in endosomes and lysosomes]); and vesicular nucleotide transporters (VNUTs, transporting ATP and other nucleotides). There are other members of the SLC17 family that are phosphate transporters (SLC17A1-4), but we have not studied the phylogeny of this branch of the family (Fredriksson et al., 2008), as there is no evidence that this branch participates in vesicular storage of neurotransmitters. We performed phylogenetic analysis of all transporters related to these three families, as well as of uncharacterized SLC17 members more closely related to these families than to the phosphate transporters. VGLUTs and VNUTs had already separated into separate clades before the bilaterian ancestor (Fig. 3). In contrast, the VEAT or sialin clade evolved in the Deuterostome lineage, as these transcripts are first found in Branchiostoma. The most ancient of these families appears to be the VNUTs, with support for the presence of this family in the filasterian Capsaspora. VNUTs have recently been shown to mediate ATP release from dorsal horn neurons in a mechanism that contributes to neuropathic pain (Masuda et al., 2016), as well as being important for concentrating catecholamines in chromaffin granules (Estevez-Herrera et al., 2016). However, while it is clear that VNUT is present in secretory granules, it is still somewhat controversial whether it is present in synaptic vesicles (Larsson et al., 2012).

Invertebrate Genomics Provide Insights Into the Origin of Synaptic TransmissionClick to view larger

Figure 3 RAxML analysis peptide sequences of SLC17 transporters. The three groups of characterized vertebrate transporters, VNUT, VGLUT, and VEAT/Sialin, are indicated by blue bars. Other notable groupings are indicated by green bars. There are two subclades of VGLUT, the first of which includes vertebrate transporters and representatives of the two major classes of protostomes. There is a subclade of VGLUT that contains cnidarian and some bilaterian transporters (*), suggesting a duplication in the bilaterian ancestor; this first subclade retained more similarity to cnidarian VGLUTs, but it was lost in deuterostomes, whereas the second subclade includes the well-characterized vertebrate VGLUTs. A number of expansions or clades containing SLC17 members found only in a small subset of organisms, and that are not part of the VNUT, VGLUT, or VEAT/Sialin clades, are indicated. These are found in mollusks, cnidarians, and poriferins. An expansion in Aplysia includes one oyster member (‡). The most ancient of these clades is from choanoflagellates (¶). Surprisingly, one of these small clades contains both protostome and deuterostome members, suggesting an ancient origin (§). There are four Mnenopsis SLC17 members, two of which form a family of their own (††) and two of which are weakly related to a separate expansion of Cnidarian SLC17 members. The outgroup used was human VGAT (not shown in tree).

VGLUTs form a clade with reasonable bootstrap support, with some members in cnidarians and somewhat surprisingly also in Trichoplax, which lacks a nervous system or any other distinct tissue. No VGLUTs were found in either ctenophores or sponges. There appear to be two families of VGLUTs. One is present in Cnidaria and some invertebrates, and the other is only present in bilaterians. Some protostome invertebrates, such as Aplysia and Daphnia, contain members of both families. Thus, it appears that VGLUTs were duplicated in the bilaterian ancestor, and then the subclade that retained the closer similarity to the prebilaterian members of the VGLUT family was lost in deuterostomes.

There are many SLC17 family members in prebilaterians that are neither VGLUTs nor VNUTs; it is not appropriate to classify them as sialins/VEATs as they are no more similar to VEATs than they are to VGLUTs (Fig. 3). The annotation of most of these proteins as “sialin” or “sialin-like” results from automated annotation pipelines and is not based on any phylogenetic analysis. Thus, despite the annotation, these SLC17 members are unlikely to be orthologs of VEAT/Sialin and, thus, whether they are likely to transport glutamate or aspartate is unclear. The most ancient members are in choanoflagellates, which surprisingly form a well-supported clade with SLC17s from one of the sponges, Amphimedon (¶ in Fig. 3), but not with the SLC17 members in a distinct sponge, Oscarella. There are expansions of the SLC17 family in many prebilaterians, including Porifera, as well as in ctenophores and cnidarians. The expansion in Porifera weakens the argument that the expansion of SLC17 members in ctenophores is evidence of their exaptation as synaptic vesicle transporters (Moroz et al., 2014; Moroz & Kohn, 2016), although it certainly does not rule this out. There are also expansions of the SLC17 family in many bilaterian groups, including in mollusks, particularly in Octopus, and also C. elegans, which was not included in our analysis (Fredriksson et al., 2008). There is the general assumption that expansion of a gene family is a consequence of increased importance, particularly in neuronal functions; however, as suggested here, it is extremely difficult to assess from genomic or transcriptomic analyses the nature of the expanded functional roles.

It can be difficult to determine if a transporter participates in synaptic transmission. A good example of this is the example of VEAT and aspartate. Although VEATs clearly can transport aspartate and are present in vesicles from synaptosomal prepations (Miyaji et al., 2008), whether aspartate is released from synaptic vesicles in mammalian neurons and functions as a neurotransmitter is controversial. Very little aspartate transport occurs in highly purified synaptic vesicles (Maycox et al., 1988), and neither VEAT nor VNUT has been identified in sensitive proteomics of highly purified synaptic vesicles (Takamori et al., 2006). Moreover, despite the ability of VEAT to transport glutamate, with knockout of mouse VGLUT, no activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) or N-Methyl-D-aspartate (NMDA) receptors occurs, suggesting if VEAT contributes to synaptic transmission in mammalian neurons, it would be only in specialized cases (Herring et al., 2015). Given the known role of VEAT/sialin in the endosome/lysosomal system, it is possible that its main role is nonsynaptic (Sagne & Gasnier, 2008). It will thus be difficult to assign a synaptic vesicular transporter role to the many bilaterian and prebilaterian SLC17 members not closely related to VGLUTs or VNUTs.

The earliest vesicular transporter family to form a clade containing vertebrate transporters known to store neurotransmitter is the VNUTs. Interestingly, ligand-gated ATP receptors (P2X receptors) are ancient (being present in plants, algae, and animals) (Hou & Cao, 2016); thus, the first neurotransmitter could have been ATP.

Transporters for GABA and the SLC32 Family

GABA is an ancient signaling molecule and is an important intercellular signal in plants (Ramesh et al., 2017). Unlike ACh and the biogenic amines, where one can partially trace the use of the transmitter by examining biosynthetic enzymes, glutamic acid decarboxylases were present as far back as bacteria (Yunes et al., 2016). The SLC32 family, to which VGAT belongs, is also very different from the other transporter families; whereas SLC17 and 18 are members of the major facilitator superfamily of transporters (MFS), VGAT is related to vacuolar amino acid transporters in yeast, which are members of the amino acid/auxin permease family (AAAP). The MFS and AAAP families have been distinct since bacteria.

The VGAT clade includes transcripts from Cnidaria; in contrast, there are no members of the VGAT clade in ctenophores, placazoans, or sponges; weak hits in the transcriptome databases of these other prebilaterians fail the reverse-blast test, recovering members of the SLC32-like family or closely related SLC36 family. We confirmed this by including selected members of the family in the phylogeny. Indeed, this phylogeny suggests that SCL32-like, non-VGAT, members of this family should really be reclassified as part of the SLC36 family because it is closer to SLC36 than to VGATs (Fig. 4). Thus, the SLC32/VGAT family clearly evolved in the common ancestor to Cnidaria and bilaterians. The VGAT genes, after their separation from SLC36/SCL32-like genes, have undergone substantial expansions in cnidarians, with over 12 members found in Hydra. Some of these are members of the bilaterian VGAT clade and some mark Cnidarian specific expansions.

Invertebrate Genomics Provide Insights Into the Origin of Synaptic TransmissionClick to view larger

Figure 4 RAxML analysis of peptide sequences of SLC32 transporters. The VGAT (SLC32) transporters form a clade, which includes bilaterian and cnidarian transporters (blue bar). This VGAT family is clearly distinct from the two most closely related transporter families, the SLC32-like and SLC36 clades (blue bars), which are closer to each other than to the SLC32/VGATs. The SLC 36 family is conserved throughout animals and includes choanoflagellates, while the SLC32-like family is similarly ancient but was lost at some point in the deuterostome lineage. Green bars represent the three distinct subclades of the VGAT family in cnidarians. One of these (¶) is the closest to the bilaterian VGAT family; these transporters are present in all cnidarians, although there are expansions of this family in Hydra and Nematostella. The other two clades of cnidarian VGATs represent independent expansions in Hydra (‡) and Acropora and Nematostella (§). The expansion in Hydra contains eight distinct genes encoding VGAT/SLC32 orthologs. The outgroup is the Creolimax vesicular amino acid transporter (VAAT) from the SLC5 family.

What Do Transporters Tell Us About Evolution of the Synaptic Vesicle?

Like many “specialized” neuronal functions, the function of vesicular transporters was predated by a generalized function, quite similar to the neuronal one, making it challenging to identify the “first” organism that uses a specific transporter in a synaptic vesicle. Here, we have used the criterion that if a transporter gene from an organism is part of a clade containing known vesicular transporters (normally the vertebrate transporters), then we assume that this protein has a vesicular transporter function. The VNUT, VGLUT, and VGAT clades all contain prebilaterian genes, suggesting that these were the first vesicular transporters. In contrast, the VMAT and VAChT transporter families do not contain prebilaterian members, and thus they were exapted after the synaptic vesicle had already evolved. It has been argued that the lack of VGLUT in Mnemiopsis suggests the independent evolution of the ctenophore nervous system. Our analysis would agree with the lack of a VGLUT in Mnemiopsis, but this ctenophore does have members of the SLC17 and SLC18 families that have expanded. Thus, glutamate uptake into synaptic vesicles in ctenophores may be mediated by a transporter in the SLC17 or SLC18 family. However, if the nervous system and synaptic vesicles truly evolved independently in ctenophores, compared with other metazoa, it is just as likely that one of the many other transporter families was exapted in the ctenophore ancestor for use in an independently evolved synaptic vesicle.

The Synaptotagmin Family of Vesicular Ca2+ Sensors

At presynaptic terminals in neurons, synaptotagmins are Ca2+ binding proteins with dual C2 domains that mediate tightly synchronized, action potential-triggered fusion of synaptic vesicles with the presynaptic membrane. The “canonical” synaptotagmins at synapses, for example, mammalian Syt1 and Syt2, are transmembrane proteins that have their N terminus projecting into the synaptic vesicle lumen, a transmembrane domain, a linker, and tandem C2 domains situated in the cytoplasm. Each C2 domain binds Ca2+ ions, which are coordinated by a conserved series of five aspartate residues. Note there are synaptotagmin isoforms with C2 domains that do not bind Ca2+; most lack these conserved aspartate residues (Sudhof, 2012). In most neurons, synaptotagmins in the Syt1/2 family are essential for synchronous, rapid release of synaptic vesicles. This function is conserved throughout bilaterians. In chromaffin cells, Syt7 plays a key role in mediating Ca2+-dependent exocytosis of large dense-core vesicles (Sugita et al., 2001), along with Syt1 (Schonn et al., 2008), and is enriched in the large dense-core vesicles in these secretory cells (Matsuoka et al., 2011). A distinct Syt, Syt10, has been implicated in the release of large cytoplasmic vesicles containing IGF1 from olfactory bulb neurons (Cao et al., 2011).

In the brain, Syt7 is stably localized to the plasma membrane (Sugita et al., 2001) and does not contribute to extremely rapid, tightly synchronized Ca2+-induced exocytosis at synapses. However, when Syt1 is knocked out, a component of less well-synchronized Ca2+-evoked release is revealed, which depends on Syt7 (Bacaj et al., 2013; Luo et al., 2015). Thus, some synaptotagmin isoforms that are localized to the presynaptic plasma membrane can trigger Ca2+-dependent exocytosis, albeit with slower kinetics, presumably by interacting with SNARE complexes. Some synaptotagmins also play a role in synaptic plasticity (Kaeser-Woo et al., 2013; Liu et al., 2017); for example, Syt7 is required for the increased release of vesicles during paired-pulse facilitation (Jackman, Turecek, Belinsky, & Regehr, 2016).

Considering possible roles for synaptotagmins in prebilaterian animals, it seems plausible that either Syt1 or Syt7 orthologs could mediate Ca2+-dependent synaptic transmission. However, the precise functional roles of the various synaptotagmin isoforms are still being clarified, even in mammals; thus, other Ca2+ binding Syts could conceivably play a role in synaptic transmitter release in prebilaterians.

Sudhof (2012) suggests that synaptotagmins are absent from plants and unicellular eukaryotic species. One might infer that synaptotagmins evolved specifically to mediate rapid synaptic vesicle exocytosis at neuronal terminals. However, this concept of the cotemporaneous evolution of prototypical synaptotagmins and synapses is not based on analysis of recently assembled prebilaterian transcriptomes or genomes. Examination of several primitive species suggests key synaptotagmins utilized for synaptic vesicle release evolved earlier in basal species that lack synapses. Strikingly, as suggested by our analysis that follows, one group of prebilaterians, ctenophores, may have evolved neuronal synapses in the absence of any “canonical,” vesicle-associated synaptotagmins.

Very remotely related synaptotagmin-like proteins in plants have been suggested to mediate vesicle fusion (Wang et al., 2015), although this proposal is still fairly speculative. These plant synaptotagmins isoforms resemble extended synaptotagmins or E-Syts due to the presence of SMP domains in both. The E-Syts are evolutionarily primitive and are found throughout eukaryotes; for example, the ichthyosporean protist Creolimax has a synaptotagmin similar to E-Syts. However, the E-Syt family in bilaterians (and the homologous tricalbins in yeast) are ER proteins that mediate phospholipid transfer to the plasma membrane through their SMP domains (Saheki & De Camilli, 2017). In contrast to the E-Syts, which have a series of three to five C2 domains, plant Syts contain only two C2 domains and are substantially shorter than E-Syts. Thus, these plant synaptotagmins may be somewhat intermediate between an E-Syt and the canonical synaptotagmins at synapses. Because one plant synaptotagmin isoform, SytB, is localized to Golgi rather than to the ER (Wang et al., 2015), it seems possible that a member of this synaptotagmin family evolved into the vesicle-associated synaptotagmins that play critical roles at synapses. Intriguingly, SytA in Arabidopsis binds to the plant syntaxin PEN1 in a Ca2+-dependent manner (Kim et al., 2016), reminiscent of the interaction of the C2B domain of Syt1 with syntaxin and SNAP-25 (Brewer et al., 2015).

Craxton (2004, 2010) has done extensive phylogenetic analysis of synaptotagmin sequences, focusing primarily on bilaterians. We sought to identify synaptotagmin homologs in phylogenetically early prebilaterian animal groups for which transcriptomes or genomes were available. The combined C2A and C2B protein domains of either Syt1 or Syt7 from humans were used as query sequences for BLAST (Fig. 5). A wide range of prebilaterian species have C2-containing synaptotagmin-like genes. However, only a subset of these C2 domains actually bind Ca2+. For each of the candidate C2 domain-containing genes found, we identified the most similar mouse proteins, which allowed us to remove more distantly related sequences that have relatively little similarity to synaptotagmins. (Thus, some primitive species analyzed [Table 1] may not appear in the synaptotagmin evolutionary tree [Fig. 5].) We also eliminated most sequences lacking the canonical aspartate residues, except for those where glutamate is substituted for aspartate. In the case of a few species, such as Mnemiopsis, where we had difficulty identifying any synaptotagmin candidates with apparently functional, Ca2+-binding C2 domains, the most similar sequences were included in this analysis. In some of the simpler prebilaterian species, we included sequences where only one of the C2 domains had the full series of conserved aspartates for Ca2+ binding, again so as to include at least some synaptotagmin candidates. (Note that Ca2+ binding to only one of the C2 domains may be sufficient for Ca2+-evoked transmitter release [Shin et al., 2009; Yoshihara et al., 2010; Lee et al., 2013].)

Of the cnidarians analyzed, Nematostella and Acropora have synaptotagmins that are relatively similar to human Syt1/2 and human Syt7, with all the canonical C2 domain aspartates. Hydra has two Syt1/2-like isoforms, each of which lacks a single canonical aspartate in one C2 domain. As noted earlier in Drosophila and mouse, Ca2+ binding to the C2B domain is sufficient for Ca2+-evoked transmitter release (Shin et al., 2009; Yoshihara et al., 2010; Lee et al., 2013). Thus, it seems likely that cnidarians had evolved synchronous synaptic vesicle transmitter release with both vesicular transporters and a Syt1/2 ortholog for synchronous release. Hydra also has a synaptotagmin similar to Syt7 in which both the C2A and C2B domains contain all five aspartates. (It should be noted that Craxton [2010] reported a substantially larger expansion of synaptotagmins in Nematostella, possibly because our criteria for inclusion in this analysis were more selective.)

Invertebrate Genomics Provide Insights Into the Origin of Synaptic TransmissionClick to view larger

Figure 5 RAxML analysis of peptide sequences of synaptotagmin genes in prebilaterians. Spliced C2A and C2B domains from human Syt1 and Syt7 sequences were used for separate tblastn or blastp searches of ctenophores (Mnemiopsis and Pleurobrachia) and cnidarians (Acropora, Hydra, and Nematostella), as well as of unicellular holozoans (Capsaspora, Creolimax, Monosiga, and Salpingoeca), Trichoplax, poriferans (Amphimedon and Oscarella), Schmidtea, Drosophila, and the two mollusks for which there are the most extensive transcriptomes available, Crassostrea and Aplysia (Riviere et al., 2015; Shrestha et al., 2015). Danio and human synaptotagmins are included in the analysis. Human Doc2A served as the outgroup; Doc2B and a human extended synaptotagmin (HSyt1) were also included. Dark blue bars indicate some of the major groupings of human synaptotagmins. The light blue bar indicates the larger group that encompasses Syts 1, 2, and 8 (and also Syt5, not shown). Green bars indicate groupings of invertebrate synaptotagmin isoforms of interest. Of the cnidarian synaptotagmin-like transcripts (¶), one clade clusters with the Syt1/2/8 group, a second clade clusters with Syt7, whereas a third clade is not closely related to any mammalian synaptotagmins. Of three Trichoplax transcripts, one clusters with the Syt1/2/8 group and is 70% similar to human Syt1; a second clusters with the Syt7 group and is 76% similar to human Syt7; the last transcript, although synaptotagmin-like, is not associated with any specific mammalian synaptotagmin type. The Oscarella transcript, which is situated close to the branch point with the Syt1/2/8 family, is actually most similar (69%) to human Syt9. One Mnemiopsis transcript (Ml1776, §) has weak (45%) similarity to mammalian Syt10, whereas the second (Ml062236, ‡) is most similar to mammalian Doc2.

More interestingly, the placazoan Trichoplax, which lacks neurons and muscles, expresses a Syt7 ortholog and a distinct Syt that clusters with the larger Syt1/2/8 family (Fig. 5), as also described by Craxton (2010) (see also Srivastava et al., 2008). The slime sponge Oscarella (class Homoscleromorpha) has a synaptotagmin-like transcript in this same large Syt1/2/8 group of transcripts (Fig. 5). Both of these synaptotagmins from Trichoplax and Oscarella have two C2 domains with the full series of five canonical aspartates that mediate Ca2+ binding, and both diverged from the Syt1/2/8 clade before Syt8, which does not bind Ca2+ (Sudhof, 2012), differentiated from Syt1 and Syt2. Thus, it appears that during the evolution of Syt8, this synaptotagmin isoform lost three of the aspartates in each of its two C2 domains. A second sponge, Amphimedon (class Demospongiae) has a synaptotagmin (Aq86524) with a full set of aspartates residues in both C2 domains, but without a clear relationship to any mammalian synaptotagmin. The observation that Trichoplax has two synaptotagmins that are more similar to both cnidarian and bilaterian isoforms than are the closest synaptotagmin homologs in sponges is consistent with the recent conclusion that Porifera is a sister lineage to all animals other than Ctenophora, including the very primitive placazoans (Simakov et al., 2013; Liebeskind et al., 2015). Consideration of the cnidarian and placazoan synaptotagmins leads to the conclusion that a fully functional Syt1 family member was present in the prebilaterian ancestor. The finding that Trichoplax has a Syt 1/2 ortholog as well as a putative VGLUT would be consistent with the well-coordinated release of vesicles for cellular communication predating neurons and synapses. Deeper examination of the roles of these orthologs in Trichoplax could shed light on the early steps in nervous system evolution.

Surprisingly, there was no evidence for classical synaptotagmins in the two ctenophore species that we examined, despite the recent conclusion that Syt7 evolved before ctenophores diverged from other metazoans (Moroz & Kohn, 2015). Specifically, we observed no synaptotagmin sequence in either a Cufflinks or Trinity assembly of the Mnemiopsis transcriptome, nor in the genome of this ctenophore (J. F. Ryan et al., 2013). Two of the most synaptotagmin-like genes found in Mnemiopsis do not have a full set of aspartates for Ca2+ binding in either C2 domain, nor are they particularly closely related to the synaptotagmins involved in synaptic release. For a second ctenophore species, Pleurobrachia bachei, tblastn searches of the genome and several transcriptome assemblies from adults yielded no valid candidate synaptotagmin genes.(Predicted synaptotagmin sequences in a Pleurobrachia Illumina RNA-seq assembly [available on the Neurobase website, Moroz et al., 2014], appear to actually be squid synaptotagmin, possibly from Doryteuthis or a closely related species. A portion of this same sequence appears in a 454 assembly. This error in two ctenophore assemblies, both available on the Neurobase website, could have resulted from contaminating mRNA or cDNA.) A syt7-like sequence found in the Pleurobrachia 454 assembly lacks C2 domains. Thus, despite several brief statements in the literature that ctenophores express synaptotagmins (Moroz et al., 2014; Moroz, 2015), this is not supported by recent genome or transcriptome assemblies. These results suggest that the synaptic vesicle-associated synaptotagmins may have evolved in the basal lineage of placazoans, cnidarians, and bilaterians; however, as discussed earlier, the lack of any single gene in an assembly should be considered with some caution because this could be due to incompleteness of the assembly.

An important question is how good is the evidence that either Cnidaria or Ctenophora have vesicle-mediated synaptic transmission? Pharmacology alone is not definitive, as transmitters can be released by nonvesicular mechanisms. Westfall and her colleagues have done extensive electron micrographic studies of cnidarian synapses, providing good evidence that vesicles are aggregated in an orderly manner at neuron-to-neuron junctions in Hydra and also in several other cnidarians, Calliactis, Anthopleura, Chrysaora, and Metridium (Westfall & Kinnamon, 1978, 1984; Westfall, 1996). Synapses in a number of ctenophore species were characterized at the electron micrographic level by Hernandez-Nicaise (1973), providing clear evidence for small synaptic vesicles. Electrophysiological recordings from the comb plate cells (or polster cells) of ctenophores by Moss and Tamm (1986, 1987) demonstrated graded synaptic potentials that are tightly synchronized with stimulation and with rise times of <15 ms. Some of these synapses displayed frequency-dependent facilitation. Thus, these ctenophore synapses resemble many synapses in higher invertebrates and in vertebrates.

Assuming that the ctenophore vesicles do release neurotransmitters, how do they achieve Ca2+-dependent vesicle fusion if they lack true “canonical” synaptotagmins? Are Syt1 and Syt2 the only Ca2+ sensors for release of synaptic vesicles? When synaptotagmins reside in the plasma membrane, they can function to trigger Ca2+-dependent release, although less synchronously, as is the case with Syt7 in mammalian neurons. However, ctenophores also lack a true Syt7 ortholog. DOC2, a tandem C2 domain protein, lacking transmembrane domains, has been implicated in regulating spontaneous release. Despite the presence of C2 domains that contain all five canonical aspartates, DOC2 in mammals does not function as a Ca2+ sensor, nor does it participate in evoked release (Pang et al., 2011). Otoferlin, a member of the ferlin family of C2 domain-containing proteins, functions instead of synaptotagmin as the Ca2+ sensor for vesicle release in auditory and vestibular hair cells, which have specialized ribbon synapses (Roux et al., 2006; Dulon et al., 2009; Beurg et al., 2010). Both ctenophore species studied have both ferlins and transcripts in the DOC2 family. It seems possible that a distinct C2 domain protein, a member of the ferlin or the DOC2 family, mediates Ca2+-dependent exocytosis at ctenophore synapses, assuming that transmitters in this system are indeed released from the candidate synaptic vesicles seen in electron microscopy.

Ionotropic Glutamate Receptors

Ionotropic glutamate receptors (iGluRs) are large integral membrane glutamate-gated ion channels, composed of four subunits, each of which is larger than 100 kD (Traynelis et al., 2010). For the individual subunits, the amino terminal is located extracellularly, as is the dual-lobe ligand-binding domain. This ligand binding domain has a clamshell or Venus flytrap configuration, which is formed by the S1 and S2 segments. The S1 segment is located on the amino terminal side of transmembrane helix M1, whereas the S2 segment is located between transmembrane helices M3 and M4. One additional membrane segment M2 forms a membrane reentrant loop that contributes to the channel pore.

Vertebrate ionotropic glutamate receptors form three distinct classes: kainate-, AMPA-, and NMDA-type glutamate receptors. Functional AMPA and kainate receptors are assembled as both homo- and heterotetramers, whereas functional NMDA receptors require heteromeric tetramers. A fourth type of glutamate receptor, the delta receptor, does not appear to have functional ligand-gated ion channel activity normally, but it may function as a developmental signaling or scaffold protein and will not be discussed here. Glutamate receptor pores are permeable to monovalent cations, with Na+ nearly as permeable as K+. Some forms, including NMDA receptors, also have substantial Ca2+ permeability (Traynelis et al., 2010).

The Venus flytrap structure of vertebrate glutamate receptors is reminiscent of soluble periplasmic binding proteins found in gram-negative bacteria, which bind a variety of nutrients and couple to transporters or chemotaxis receptors in the inner membrane (Lampinen et al., 1998). The tetrameric structure of ionotropic glutamate receptors has similarity to the organization of voltage-gated K+ channels; moreover, there is modest sequence similarity between the pore lining M2 segment of glutamate receptors and the P segment of K+ channels, which also lines the channel pore (Kuner et al., 2003), particularly in the pattern of the pore lining residues. This similarity was first appreciated by Wood et al. (1995), who proposed the possibility of a common origin for K+ channels and glutamate receptors. This seemed highly speculative, given the limited sequence similarity. Remarkably, a glutamate-gated K+ selective channel with similarities to bilaterian glutamate receptors, called GluR0, was identified in the bacterium Synechocystis and proposed as the possible evolutionary missing link between K+ channels and eukaryotic glutamate receptors (Chen et al., 1999). GluR0 homologs can be found in other bacteria (Martinac et al., 2008). In three other bacterial species, including Nostoc punctiforme, glutamate receptors were identified that, unlike GluR0, actually have P segments more similar to the reentrant loops of eukaryotic glutamate receptors than of K+ channels (Kuner et al., 2003; Lee et al., 2008). The structure of these various bacterial glutamate receptors is remarkably similar to bilaterian glutamate receptors; except for the absence of an M4 domain, these receptors apparently all have the pore-forming, reentrant loop helix. Unlike the Synechocystis GluR0 channel, these eukaryote-like glutamate receptors have not been studied biophysically. Nevertheless, the possibility that a gene coding for the periplasmic binding protein was fused to a gene coding for a voltage-sensitive ion channel to give rise to a ligand-gated channel is intriguing. Although an evolutionary sequence has been suggested with the Synechocystis glutamate-gated K+ channel evolving into bacterial channels with bilaterian-like pore selectivities, such as the Nostoc receptor, the different groups of bacterial channels could have evolved independently. Thus, this type of hybrid protein could have evolved more than once in prokaryotes.

Plants, including the model system Arabidopsis, expresses a number of glutamate receptor-like genes. Indeed, there are 20 putative glutamate receptor genes in Arabidopsis, which have predicted structures similar to bilaterian glutamate receptors, with S1 and S2 segments, three transmembrane domains, and the intramembrane loop (Chiu et al., 2002; Gilliham et al., 2006). The most extensively studied of these channels mediates Ca2+ influx and membrane depolarization in response to several amino acids, though all with very low affinity (Meyerhoff et al., 2005; Qi et al., 2006; Ni et al., 2016). Some of these glutamate receptors participate in responses of plants to cold and injury and also in root development. Most remarkably, when an Arabidopsis leaf is damaged, a depolarizing signal propagates slowly, over many seconds, to neighboring leaves, a mechanism that involves glutamate receptors (Mousavi et al., 2013). Thus, these glutamate receptors may participate in signal transmission across several centimeters.

Two single-cell, nonmetazoan eukaryotes, Capsaspora (a filasterean) and Salpingoeca (a choanoflagellate), seem to entirely lack ionotropic glutamate receptors. One might conclude that glutamate receptors evolved later in the metazoan lineage, except for the presence of highly conserved, functional glutamate receptors in plants and also in bacteria. Thus, these primitive, single-cell eukaryotes are likely to have lost the genes for these receptors. Interestingly, Trichoplax, which lacks a nervous system or synapses, expresses seven ionotropic glutamate receptor transcripts (Fig. 6). These genes apparently diverged early from ancestral bilaterian glutamate receptors and cluster in two distinct groups (* in Fig. 6). Although these Trichoplax receptors are not closely related to bilaterian glutamate receptors, they appear unambiguously to be members of the glutamate receptor family. Blasting mouse proteins with the two most distinct Trichoplax sequences (based on the RAxML analysis) yielded top hits that are all orthologous to glutamate receptors, with ≥56% similarity. (The other five putative glutamate receptors from Trichoplax had comparable similarities.)

Invertebrate Genomics Provide Insights Into the Origin of Synaptic TransmissionClick to view larger

Figure 6 RAxML analysis of peptide sequences of ionotropic glutamate receptors. Transcriptome assemblies were searched by tblastn or blastp using the spliced S1, S2, TM1, TM2, and TM3 domains of each of three human iGluRs separately: GluA2, GluK5, and GluN2B. The nicotinic α7 ACh receptor was included as an outgroup (not shown in tree); the 5-HT3 ionotropic serotonin receptor was also included as a second member of the Cys-loop superfamily of ionotropic receptors, which is entirely distinct from the iGluRs. The three major groups of ionotropic glutamate receptors characterized in vertebrates, AMPA, kainate, and NMDA types are indicated by dark blue bars. A substantial number of invertebrate receptors are distinct from these three types, frequently clustering in their own groups, indicated by green bars. Clades of invertebrate receptors within the larger AMPA and kainate groups are indicated by light blue bars. Note the expansion of Trichoplax iGluRs; these seven receptors cluster into two clades (*). The sponge Oscarella, which, like Trichoplax, also lacks neurons or synapses, expresses a large group of iGluRs in their own clade (§). Mnemiopsis has a small group of iGluRs, which are distinct from those from other phyla (‡). Within the larger AMPA and kainate receptor groups, the vertebrate clades are entirely distinct from the clusters of lophotrochozoan and ecdysozoan receptors in these same groups. Note the three very large expansions of invertebrate kainate receptors, distinct from vertebrate kainate receptors, with large numbers of receptors from both mollusks and insects († and ¶, signifying arthropod and molluskan receptors, respectively). There is also a clade of invertebrate AMPA receptors (¶†), with a large expansion of genes in Aplysia, and single representatives from Drosophila and Daphnia; interestingly, there is no transcript in this subgroup of AMPA receptors from Crassostrea. Unlike kainate and AMPA receptors, NMDA receptors from mollusks and Drosophila are similar to subtypes of vertebrate NMDA receptors; thus, there is not a distinct invertebrate subgroup of NMDA receptors. There are also expansions in three distinct invertebrate clades of glutamate receptors, with two clades of three oyster receptors (¶∆), one of which includes Aplysia receptors, and a third clade primarily of arthropod receptors, with a single oyster transcript (†¶∆).

Do invertebrate GluRs, clearly align with the distinct NMDA, AMPA, and kainate groups of receptors in vertebrates? Not all of the invertebrate glutamate receptors are associated with these three classes. Nevertheless, for the bilaterian species analyzed, most have members within the NMDA receptor group and other receptors that are similar to AMPA and/or kainate receptors. For example, both Daphnia and Drosophila have several receptors clustered with the vertebrate AMPA receptors and also several isoforms more closely associated with the vertebrate kainate receptors. Daphnia and Drosophila also have receptors that cluster with two distinct subclasses of NMDA receptors. Similarly, the gastropod mollusk Aplysia has distinct members of the AMPA, kainate, and NMDA receptor families (Fig. 6). In contrast, the sponge Oscarella has 10 glutamate receptor transcripts that form a cluster entirely distinct from vertebrate glutamate receptors, which represents a major expansion (§ in Fig. 6). As discussed earlier, neither Trichoplax nor Oscarella has synapses. Interestingly, in ctenophores, which have effective synaptic transmission (Moss & Tamm, 1986, 1987), the iGluRs are also clustered as a unique group (‡ in Fig. 6).

Substantial expansions of glutamate receptors are evident in a number of other invertebrate clades. Aplysia expresses six related AMPA receptors in a clade that does not include any vertebrate AMPA receptors (¶† in Fig. 6); although there were no Crassostrea members found in this clade, there are single representatives from Drosophila and Daphnia, indicating that this ancient group antedated the divergence of the lophotrochozoan and ecdysozoan lineages. Similarly, a small group of Drosophila and Daphnia kainate-type iGluRs includes an oyster and an Aplysia member (†¶). There are also oyster and Drosophila expansions of kainate-type receptors that contain exclusively molluscan or arthropod receptors, respectively (¶ and † in Fig. 6). Three other groups of invertebrate iGluRs that include exclusively either molluskan receptors or arthropod and molluskan receptors are entirely distinct from the canonical groups of AMPA, kainate, and NMDA receptors of vertebrates (¶∆ and †¶∆ in Fig. 6).

Among both the AMPA and kainate receptors, the invertebrate family members do not cluster with specific vertebrate receptor subtypes but form their own clusters (¶, †, †¶, and ¶† in Fig. 6). Thus, the vertebrate subtypes of receptors within the larger AMPA and kainate groups apparently diverged after the divergence of the ancestral deuterostomes and protostomes. In contrast, the subgroups of NMDA receptors; that is, the GluN1, GluN2, and GluN3 subtypes all include arthropod and/or molluskan iGluR transcripts. Thus, these subtypes of NMDA recpetors appear to have diverged early, before the divergence of the protostome and deuterostome lineages. The single Acropora transcript found among the GluN1 receptors suggests that of all the groups of ionotropic glutamate receptors that are utilized by vertebrates, GluN1 may have evolved earliest in a prebilaterian ancestor. It is worth noting that no NMDA receptor family members were identified in Mnemiopsis.

In mammals there are four AMPA receptor (GluA) genes, five kainate receptor (GluK) genes, and seven NMDA receptor (GluN) genes. The proliferation of synaptic proteins within the mammalian lineage has been proposed to be the necessary substrate for complex forms of synaptic plasticity, learning, and cognition (Emes et al., 2008; Nithianantharajah et al., 2013; T. J. Ryan et al., 2013; Grant, 2016). Grant (2016) states “vertebrate behavioural complexity is a product of synapse proteome complexity.” One group of synaptic proteins to which these authors attributes substantial importance is the NMDA receptors (T. J. Ryan et al., 2013, p. 372). However, examination of ionotropic glutamate receptors across phylogeny does not support this perspective that diversity of isoforms determines behavioral complexity. For example, the oyster Crassostrea expresses 17 distinct ionotropic glutamate receptors. Based on the recent Aplysia transcriptome, this gastropod mollusk expresses more than 20 distinct ionotropic glutamate receptor isoforms in the CNS, some of which are analyzed in Figure 6. Thus, Aplysia glutamate receptors have diverged more widely than mammalian glutamate receptors. These Aplysia receptors include at least two isoforms of NMDA receptors, and possibly more. It is worth noting that there are a dozen Drosophila iGluRs; seven of these Drosophila iGluRs cluster with kainate receptors, which also represents a substantial expansion. T. J. Ryan et al. (2013) have demonstrated the importance of specific NMDA receptor cytoplasmic termini for different distinct forms of learning and behavior and for long-term potentiation. However, we have no similar analysis of impact of molluskan NMDA receptor structure on plasticity or learning; thus, there is no reason to believe the diversity of molluskan NMDARs does not have similarly important physiological functional consequences.

It is also striking how there are four distinct invertebrate-specific or phylum-specific groups of glutamate receptors have diverged from these three vertebrate groups. The sponge Oscarella expresses 10 closely related iGluRs (§ in Fig. 6). Although these receptors are clearly related to vertebrate iGluRs, they are not substantially more similar to any single vertebrate receptor iGluR group. The closest related GluRs in this RAxML analysis are a handful of Trichoplax receptors. The oyster Crassostrea expresses a group of three closely related iGluRs that are distinct from any others in this analysis. Another noteworthy group ( in Fig. 6) includes Aplysia, oyster, Drosophila, and Daphnia receptors; this group is also not similar to any vertebrate GluRs.

Mnemiopsis expresses three closely related iGluRs that are distinct from any bilaterian glutamate receptors. This is intriguing because this species lacks VGLUTs. This raises the question of which transporter might transport glutamate into vesicles at synapses, if Mnemiopsis uses glutamate as a transmitter.


Many of the key synaptic proteins that we analyzed evolved for related functions well before the advent of the most primitive nervous systems (Sakarya et al., 2007; Srivastava et al., 2010; Burkhardt et al., 2011). In most cases, it is difficult from phylogenetic analysis alone to determine when these proteins were exapted for their present roles as key synaptic proteins.

For example, prior to their incorporation into synaptic vesicles, related vesicular transporters functioned in other membrane-bound compartments. An expansion of a particular transporter family may be due to its assumption of a new role in neurotransmitter uptake, but this expansion could also reflect some extrasynaptic or nonneuronal function. One indication of the adoption of a transporter into a synaptic role is the formation of a tight clade of transporters that cluster with the known synaptic vesicle transporters of vertebrates, as seen in the VAChT and VMAT clades in bilaterians. However, this does not preclude that these or other closely related transmitters used a distinct transporter in synaptic vesicles in earlier evolving groups; note that a distinct vesicular transporter would represent an additional evolutionary event and therefore would be somewhat less likely. This is particularly relevant for the question of which transmitters were used in cnidarians, which have a much larger families of SLC17 and SLC32 transporters than most bilaterians.

What does our analysis say about the origins of synaptic vesicles? Synaptic vesicles are derived from endosomes; they share the vacuolar ATPase (V-ATPase) and other features with endosomes, and they are also formed by endocytosis, either from the plasma membrane or from a special class of early endosomes (Watanabe & Boucrot, 2017). Vacuoles in yeast have active uptake of amino acids dependent on the V-ATPase (Sekito et al., 2008), a mechanism similar to that in synaptic vesicles. Thus, the mechanism of transporter import of amino acids into a storage endosome-related structure clearly predates the use of this transporter function for synaptic vesicles. Similarly, exocytotic release from lysosomes (another endosome-derived vesicle) is also evolutionarily ancient (Charette & Cosson, 2007). From phylogenetic analysis of synaptic proteins, one would assume that Trichoplax uses glutamate as a transmitter released from synaptic vesicles. Its genome contains VGLUT, a Syt1/2 ortholog and bona fide ligand-gated glutamate receptors. However, Trichoplax has neither neurons nor synaptic vesicles, although there are specialized secretory cells that contain neuropeptides (Schuchert, 1993; Smith et al., 2014). An intriguing possibility is that VGLUT and VNUT were first used to transport glutamate and ATP into dense-core vesicles. The presence of both ligand-gated glutamate and ATP receptors and ligand-gated peptide receptors related to the ASIC family, in combination with dense-core vesicles, may have allowed fast cellular chemical transmission before the evolution of true synapses or synaptic vesicles. Indeed, even in vertebrates, many transporters that we have discussed (VPAT, VNUT, VMAT) are found in dense-core vesicles (Henry, Botton et al., 1994; Henry, Gasnier, et al., 1994; Sakamoto et al., 2014; Takeuchi et al., 2017).

In contrast, synaptic vesicle proteins are noticeably absent in ctenophores, despite both the presence of structures morphologically resembling synaptic vesicles and electrophysiological evidence for synaptic transmission (Hernandez-Nicaise, 1973; Moss & Tamm, 1986, 1987). Transporters clearly have evolved into synaptic vesicle transporters multiple times in evolution, and thus ctenophore utilization of a distinct transporter would not necessarily indicate independent evolution of synaptic vesicles. Nevertheless, the lack of a synaptic vesicle synaptotagmin (i.e., Syt1, 2, or 7) strongly suggests the independent evolution of fast neurotransmitter release in this organism. Although the iGluRs identified in Mnemiopsis do not cluster with any group of vertebrate glutamate receptors, one iGluR has a glycine binding site, much as do mammalian NMDARs. Interestingly, this iGluR also has an interdomain salt bridge that affects glycine binding (Yu et al., 2016). This suggests that glycine regulation of a Mnemiopsis glutamate receptor may be modulated in distinct ways, but that this glycine binding site is functionally important, as in vertebrate NMDA receptors. Clearly there are multiple ways that family members can be adapted for similar functional roles in the nervous system. J. F. Ryan et al. (2013) considered the evidence for the independent evolution of the nervous system in the ctenophore lineage. These authors found that more than 75% of the postsynaptic genes that they examined were present in Mnemiopsis, suggesting that a simple nervous system was already present in the metazoan ancestor. This interpretation is nevertheless consistent with the possibility suggested here that fast synaptic transmission evolved independently in ctenophore and the cnidarian/bilaterian lineages. Interestingly, in their conclusion, J. F. Ryan et al. (2013) are noncommittal as to whether the ctenophore nervous system had an independent origin.

The First Neurotransmitter

What is the earliest neurotransmitter stored in synaptic vesicles that is still utilized at bilaterian synapses? The oldest transporter is VNUT, and ligand-gated ATP channels are also ancient (Hou & Cao, 2016); however, even in vertebrates, it is still not clear whether ATP is released from synaptic vesicles, rather than via other mechanisms. Cnidarians have VGLUTs and VGATs, as well as ligand-gated receptors, for these families, so it seems likely that glutamate and GABA were the first small neurotransmitters stored in synaptic vesicles that have been conserved. The large expansion of related transporters in Cnidaria suggests that cnidarians also use many other transmitters that have not been characterized. This will be interesting to investigate in the future.

In vertebrates and in bilaterians generally, neuropeptides activate metabotropic receptors to initiate modulatory responses with relatively slow kinetics. Interestingly, peptides in the FMRFamide family, which is highly conserved across evolution, activate an ion channel in the DEG/ENaC family in gastropod mollusks, including Aplysia (Lingueglia et al., 1995; Cottrell, 1997), and also in Hydra (Golubovic et al., 2007). Some members of this ion channel family in vertebrates are gated by proton binding or sodium binding (Immke & McCleskey, 2003; Horisberger & Chraibi, 2004). Thus, it appears possible that a distinct form of ionotropic receptor gated by peptides mediated relatively rapid signaling predating the nervous system (i.e., in Trichoplax). As noted earlier, some of the vesicular transporters may have seen their first use in dense-core vesicles in an earlier stage of rapid signaling antedating synaptic vesicles or synapses.

Inhibitory Neurotransmission

It is noteworthy that GABA receptor-like sequences are not found in the Mnemiopsis or Pleurobrachia transcriptomes. The unicellular holozoans Creolimax and Capsaspora also do not have transcripts very similar to GABA receptors. GABA receptor–like genes are similarly absent in Trichoplax and sponges, such as Spongilla, Amphimedon, and Oscarella. Thus, it appears that GABA receptors first appeared in Cnidaria and are found throughout bilaterians, which matches the appearance of the VGAT clade of vesicular transporters. VGATs are present in cnidarians, but not in the earlier evolving prebilaterians, that is, sponges, Trichoplax, and ctenophores. It seems plausible that GABA-mediated inhibitory synapses evolved when nervous systems were sufficiently complex to require true integration of inputs from multiple presynaptic neurons. It may be the case that glutamate-gated chloride channels appeared earlier in evolution than GABA receptors, as Creolimax and Capsaspora have transcripts with weak similarity to these channels.

Gene Diversity and Evolution of Complex Cognitive Processes

It has been suggested that the expansion of multiple gene families involved in synaptic transmission is related to the complexity of the vertebrate nervous system and a “higher evolved” brain, as mentioned earlier. Grant and his colleagues have done several interesting analyses of expansions of postsynaptic genes in vertebrates (Grant, 2009, 2016; Emes & Grant, 2012; Nithianantharajah et al., 2013). Gene families that have been studied in this context include the membrane-associated guanylate kinases (MAGUKs) and NMDA receptors (Emes et al., 2008; T. J. Ryan et al., 2013). These researchers concluded that gene duplication and diversification are fundamental contributors to the diverse behavioral repertoires and complex cognitive processes of vertebrates. However, as discussed in this chapter, examination of a number of critical synaptic proteins reveals many expansions in invertebrate species with what are considered to be quite simple nervous systems. Indeed, there have been a number of expansions of transporter genes in invertebrates, particularly SLC17 family members, not represented in vertebrates. The extensive expansions of iGluR families in invertebrates are particularly noteworthy; a number of these groups of iGluRs are absent in the vertebrate lineage. Moreover, the oyster Crassostrea and the marine snail Aplysia have substantially more iGluR isoforms than do humans (Riviere et al., 2015; Shrestha et al., 2015). Note that although Crassostrea is substantially simpler behaviorally than Aplysia, the diversity of iGluRs is at least as great. Results indicating that some of these specific genes are critical for neuronal plasticity and cognitive function in the mouse are not actually a strong test of this overall complexity hypothesis that isoform diversity underlies cognitive sophistication and behavioral complexity. One could similarly knock out expression of key genes in an invertebrate species and obtain comparable deficits. For example, during learning, Aplysia expresses three isoforms of persistently active, truncated protein kinase C (PKC), known as PKMs. Selective expression of isoform-specific dominant negative constructs in individual neurons demonstrated that each form specifically participated in either associative or nonassociative long-term learning and specifically in either the pre- or postsynaptic compartment (Hu, Adler, et al., 2017; Hu, Ferguson, et al., 2017). These findings would be consistent with the conclusion that the diversity of PKMs in Aplysia contributes to behavioral complexity in this marine snail. It is noteworthy that currently in mammals, fewer PKM isoforms have been implicated in plasticity than in Aplysia. Demonstrating that members of a particular gene family in vertebrates are functionally important does not provide evidence that it is the expansion of this gene family that drove increased cognitive ability. Indeed, the functional impact of these expansions is extremely difficult to evaluate experimentally. In general, obtaining evidence that the complexity of a synaptic gene family contributes to a species’ cognitive capacity is problematic, as the logic is somewhat circular. If specific knockout of a member of a gene family resulted in a specific behavioral deficit, this would be interpreted as consistent with the complexity hypothesis. However, if no phenotype is observed in the knockout, this would lead to the conclusion that this diversity of isoforms permits compensation in case of a genetic deficit.

In contrast to the focus on expansions of synaptic and signaling molecules across phylogeny, which may well have little relationship to behavioral complexity, a stronger argument could be made that nervous systems with more complex wiring patterns should require larger arrays of guidance molecules to support axon path finding and synapse specificity. Thus, the dramatic expansion of the protocadherin family of cell adhesion molecules in octopus has been suggested to contribute to the assembly of local neural circuits and the specificity of synaptic connections in this highly evolved invertebrate with extensive behavioral complexity. Interestingly, octopus expresses nearly three-fold as many protocadherins as humans (Albertin et al., 2015). One might expect to be able to detect clear anatomical differences in the distribution of cell surface signaling molecules among distinct anatomical pathways. These differences would lead to precise predictions as to the developmental consequences of knocking out a specific cell surface axon guidance molecule. Development of additional invertebrate genetic models, with more complex central nervous systems than Drosophila and C. elegans, should enable experimental tests of the functional impact of diversity in cell adhesion molecules. The small sepioloid squid Euprymna scolopes, which can be reared in the lab and which completes its life cycle in 4 months, is one candidate species for lab culture and development of genetic strains (Lee et al., 2009; Zepeda et al., 2017), as is the dwarf cuttlefish Sepia bandensis. These species may enable new models for generating and rearing transgenic animals using improved technologies, such as CRISPR/Cas9.


For evolutionary analysis, sequences were chosen from various assemblies available, including the Aplysia transcriptome assembly (, assemblies provided by authors, or from the NCBI database (Table 2). Blastp or tblastn searches were conducted using Geneious R10. Reverse Blast searches (blastp or blastx), using the candidate sequences as queries, were done to ensure that sequences found were not more closely related to another protein family distinct from the one under investigation. For example, reverse Blast searches of the best match from the Trichoplax database with VGAT return SLC36 proton-coupled amino acid transporters, not other members of the SLC32 family.

Table 2 Transcriptome Assembly Resources


Source for Assembly

Acropora digitifera

Amphimedon queenslandica

Oleg Simakov (Srivastava et al, 2010)

Apis mellifera


Aplysia californica

Brachiostoma floridae

Capitella teleta

Capsaspora owczarzaki

Crassostrea gigas

Creolimax fragrantissima

Danio rerio

Daphnia pulex

Drosophila melanogaster

Homo sapiens


Hydra vulgaris

Mnemiopsis leidyi

Monosiga brevicollis

Nematostella vectensis

Octopus bimaculoides

Oscarella carmela

Pleurobrachia bachei

Saccoglossus kowalevskii

Salpingoeca rosetta

Schmidtea mediterranea

Spongilla lacustris

Trichoplax adhaerens

Dendrograms were generated by first producing sequence alignments with MUSCLE (, which were then analyzed with RAxML [Stamatakis, 2014] using raxmlGUI_v1.5b1; [Silvestro & Michalak, 2012]); RAxML GUI settings were ML + rapid bootstrap, and PROT CAT [I] and LG for amino acid substitution models, with 500 or 1,000 replicates. Trees were then generated with Figtree v1.4.3 and the nodes labeled with percentage of trees with each specific node in consensus tree.

In some RAxML analyses (e.g., the phylogeny of glutamate receptors), we used both the full predicted amino acid sequences and also the relatively conserved residues, identified using the Gblocks method (Castresana, 2000; Talavera & Castresana, 2007). Following a MUSCLE alignment, Gblocks ( with all three options for a less stringent selection was used to generate alignments of the relatively conserved amino acid positions, which were then used in RAxML. Although the resulting RAxML trees with and without Gblocks differed in some details, the overall conclusions reached were similar.

Published transcriptome assemblies (Table 2) were obtained from the following studies: Acropora (Shinzato et al., 2011), Amphimedon (Srivastava et al., 2010), Brachiostoma (Putnam et al., 2008), Capitella (Simakov et al., 2013), Capsaspora (Sebe-Pedros et al., 2013), Crassostrea (Riviere et al., 2015), Creolimax (de Mendoza et al, 2015), Daphnia (Colbourne et al., 2011), Mnemiopsis (J. F. Ryan et al., 2013), Monosiga (King et al., 2008), Nematostella (Putnam et al., 2007), Octopus (Albertin et al., 2015), Pleurobrachia (Moroz et al., 2014), Saccoglossus (Simakov et al., 2015), Spongilla (Riesgo et al., 2014), and Trichoplax (Srivastava et al., 2008).


Albalat, R., & Canestro, C. (2016). Evolution by gene loss. Nature Reviews Genetics, 17(7), 379–391. doi:10.1038/nrg.2016.39Find this resource:

Albertin, C. B., Simakov, O., Mitros, T., Wang, Z. Y., Pungor, J. R., Edsinger-Gonzales, E., … Rokhsar, D. S. (2015). The octopus genome and the evolution of cephalopod neural and morphological novelties. Nature, 524(7564), 220–224. doi:10.1038/nature14668Find this resource:

Anctil, M. (2009). Chemical transmission in the sea anemone Nematostella vectensis: A genomic perspective. Comparative Biochemistry and Physiology Part D: Genomics and Proteomics, 4(4), 268–289. doi:10.1016/j.cbd.2009.07.001Find this resource:

Bacaj, T., Wu, D., Yang, X., Morishita, W., Zhou, P., Xu, W., … Sudhof, T. C. (2013). Synaptotagmin-1 and synaptotagmin-7 trigger synchronous and asynchronous phases of neurotransmitter release. Neuron, 80(4), 947–959. doi:10.1016/j.neuron.2013.10.026Find this resource:

Beurg, M., Michalski, N., Safieddine, S., Bouleau, Y., Schneggenburger, R., Chapman, E. R., … Dulon, D. (2010). Control of exocytosis by synaptotagmins and otoferlin in auditory hair cells. Journal of Neuroscience, 30(40), 13281–13290. doi:10.1523/JNEUROSCI.2528-10.2010Find this resource:

Brewer, K. D., Bacaj, T., Cavalli, A., Camilloni, C., Swarbrick, J. D., Liu, J., … Rizo, J. (2015). Dynamic binding mode of a Synaptotagmin-1-SNARE complex in solution. Nature Structural and Molecular Biology, 22(7), 555–564. doi:10.1038/nsmb.3035Find this resource:

Burkhardt, P., Stegmann, C. M., Cooper, B., Kloepper, T. H., Imig, C., Varoqueaux, F., … Fasshauer, D. (2011). Primordial neurosecretory apparatus identified in the choanoflagellate Monosiga brevicollis. Proceedings of the National Academy of Sciences USA, 108(37), 15264–15269. doi:10.1073/pnas.1106189108Find this resource:

Cao, P., Maximov, A., & Sudhof, T. C. (2011). Activity-dependent IGF-1 exocytosis is controlled by the Ca(2+)-sensor synaptotagmin-10. Cell, 145(2), 300–311. doi:10.1016/j.cell.2011.03.034Find this resource:

Casini, A., Vaccaro, R., D'Este, L., Sakaue, Y., Bellier, J. P., Kimura, H., & Renda, T. G. (2012). Immunolocalization of choline acetyltransferase of common type in the central brain mass of Octopus vulgaris. European Journal of Histochemistry, 56(3), e34. doi:10.4081/ejh.2012.e34Find this resource:

Castresana, J. (2000). Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution, 17(4), 540–552.Find this resource:

Cavalier-Smith, T. (2017). Origin of animal multicellularity: Precursors, causes, consequences-the choanoflagellate/sponge transition, neurogenesis and the Cambrian explosion. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 372(1713). doi:10.1098/rstb.2015.0476Find this resource:

Chang, E. S., Neuhof, M., Rubinstein, N. D., Diamant, A., Philippe, H., Huchon, D., & Cartwright, P. (2015). Genomic insights into the evolutionary origin of Myxozoa within Cnidaria. Proceedings of the National Academy of Sciences USA, 112(48), 14912–14917. doi:10.1073/pnas.1511468112Find this resource:

Charette, S. J., & Cosson, P. (2007). A LYST/beige homolog is involved in biogenesis of Dictyostelium secretory lysosomes. Journal of Cell Science, 120(Pt 14), 2338–2343. doi:10.1242/jcs.009001Find this resource:

Chen, G. Q., Cui, C., Mayer, M. L., & Gouaux, E. (1999). Functional characterization of a potassium-selective prokaryotic glutamate receptor. Nature, 402(6763), 817–821. doi:10.1038/45568Find this resource:

Chiu, J. C., Brenner, E. D., DeSalle, R., Nitabach, M. N., Holmes, T. C., & Coruzzi, G. M. (2002). Phylogenetic and expression analysis of the glutamate-receptor-like gene family in Arabidopsis thaliana. Molecular Biology and Evolution, 19(7), 1066–1082.Find this resource:

Colbourne, J. K., Pfrender, M. E., Gilbert, D., Thomas, W. K., Tucker, A., Oakley, T. H., … Boore, J. L. (2011). The ecoresponsive genome of Daphnia pulex. Science, 331(6017), 555–561. doi:10.1126/science.1197761Find this resource:

Cottrell, G. A. (1997). The first peptide-gated ion channel. Journal of Experimental Biology, 200(Pt 18), 2377–2386.Find this resource:

Craxton, M. (2004). Synaptotagmin gene content of the sequenced genomes. BMC Genomics, 5(1), 43. doi:10.1186/1471-2164-5-43Find this resource:

Craxton, M. (2010). A manual collection of Syt, Esyt, Rph3a, Rph3al, Doc2, and Dblc2 genes from 46 metazoan genomes—an open access resource for neuroscience and evolutionary biology. BMC Genomics, 11, 37. doi:10.1186/1471-2164-11-37Find this resource:

de Mendoza, A., Suga, H., Permanyer, J., Irimia, M., & Ruiz-Trillo, I. (2015). Complex transcriptional regulation and independent evolution of fungal-like traits in a relative of animals. Elife, 4, e08904. doi:10.7554/eLife.08904Find this resource:

Dulon, D., Safieddine, S., Jones, S. M., & Petit, C. (2009). Otoferlin is critical for a highly sensitive and linear calcium-dependent exocytosis at vestibular hair cell ribbon synapses. Journal of Neuroscience, 29(34), 10474–10487. doi:10.1523/JNEUROSCI.1009-09.2009Find this resource:

Emes, R. D., & Grant, S. G. (2012). Evolution of synapse complexity and diversity. Annu Reviews in Neuroscience, 35, 111–131. doi:10.1146/annurev-neuro-062111-150433Find this resource:

Emes, R. D., Pocklington, A. J., Anderson, C. N., Bayes, A., Collins, M. O., Vickers, C. A., … Grant, S. G. (2008). Evolutionary expansion and anatomical specialization of synapse proteome complexity. Nature Neuroscience, 11(7), 799–806. doi:10.1038/nn.2135Find this resource:

Estevez-Herrera, J., Dominguez, N., Pardo, M. R., Gonzalez-Santana, A., Westhead, E. W., Borges, R., & Machado, J. D. (2016). ATP: The crucial component of secretory vesicles. Proceedings of the National Academy of Sciences USA, 113(28), E4098–4106. doi:10.1073/pnas.1600690113Find this resource:

Feuda, R., Dohrmann, M., Pett, W., Philippe, H., Rota-Stabelli, O., Lartillot, N., … Pisani, D. (2017). Improved modeling of compositional heterogeneity supports sponges as sister to all other animals. Current Biology, 27(24), 3864–3870 e3864. doi:10.1016/j.cub.2017.11.008Find this resource:

Fredriksson, R., Nordstrom, K. J., Stephansson, O., Hagglund, M. G., & Schioth, H. B. (2008). The solute carrier (SLC) complement of the human genome: Phylogenetic classification reveals four major families. FEBS Letters, 582(27), 3811–3816. doi:10.1016/j.febslet.2008.10.016Find this resource:

Gilliham, M., Campbell, M., Dubos, C., Becker, D., & Davenport, R. (2006). The Arabidopsis thaliana Glutamate-like Receptor Family (AtGLR). In F. Baluška, S. Mancuso, & D. Volkmann (Eds.), Communication in plants (pp. 187–204). Berlin, Heidelberg: Springer-Verlag.Find this resource:

Golubovic, A., Kuhn, A., Williamson, M., Kalbacher, H., Holstein, T. W., Grimmelikhuijzen, C. J., & Grunder, S. (2007). A peptide-gated ion channel from the freshwater polyp Hydra. Journal of Biological Chemistry, 282(48), 35098–35103. doi:10.1074/jbc.M706849200Find this resource:

Gould, S. J., & Vrba, E. S. (1982). Exaptation—a missing term in the science of form. Paleobiology, 8(1), 4–15.Find this resource:

Grant, S. G. (2009). A general basis for cognition in the evolution of synapse signaling complexes. Cold Spring Harbor Symposia of Quantitative Biology, 74, 249–257. doi:10.1101/sqb.2009.74.033Find this resource:

Grant, S. G. (2016). The molecular evolution of the vertebrate behavioural repertoire. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 371(1685), 20150051. doi:10.1098/rstb.2015.0051Find this resource:

Grunder, S., & Assmann, M. (2015). Peptide-gated ion channels and the simple nervous system of Hydra. Journal of Experimental Biology, 218(Pt 4), 551–561. doi:10.1242/jeb.111666Find this resource:

Henry, J. P., Botton, D., Sagne, C., Isambert, M. F., Desnos, C., Blanchard, V., … Gasnier, B. (1994). Biochemistry and molecular biology of the vesicular monoamine transporter from chromaffin granules. Journal of Experimental Biology, 196, 251–262.Find this resource:

Henry, J. P., Gasnier, B., Desnos, C., Scherman, D., Krejci, E., & Massoulie, J. (1994). The catecholamine transporter of adrenal medulla chromaffin granules. Annals of the NY Academy of Sciences, 733, 185–192.Find this resource:

Hernandez-Nicaise, M. L. (1973). The nervous system of ctenophores. III. Ultrastructure of synapses. Journal of Neurocytology, 2(3), 249–263.Find this resource:

Herring, B. E., Silm, K., Edwards, R. H., & Nicoll, R. A. (2015). Is aspartate an excitatory neurotransmitter? Journal of Neuroscience, 35(28), 10168–10171. doi:10.1523/JNEUROSCI.0524-15.2015Find this resource:

Hiasa, M., Miyaji, T., Haruna, Y., Takeuchi, T., Harada, Y., Moriyama, S., … Moriyama, Y. (2014). Identification of a mammalian vesicular polyamine transporter. Science Reports, 4, 6836. doi:10.1038/srep06836Find this resource:

Horisberger, J. D., & Chraibi, A. (2004). Epithelial sodium channel: A ligand-gated channel? Nephron Physiology, 96(2), p37–41. doi:10.1159/000076406Find this resource:

Hou, Z., & Cao, J. (2016). Comparative study of the P2X gene family in animals and plants. Purinergic Signaling, 12(2), 269–281. doi:10.1007/s11302-016-9501-zFind this resource:

Hu, J., Adler, K., Farah, C. A., Hastings, M. H., Sossin, W. S., & Schacher, S. (2017). Cell-specific PKM isoforms contribute to the maintenance of different forms of persistent long-term synaptic plasticity. Journal of Neuroscience, 37(10), 2746–2763. doi:10.1523/JNEUROSCI.2805-16.2017Find this resource:

Hu, J., Ferguson, L., Adler, K., Farah, C. A., Hastings, M. H., Sossin, W. S., & Schacher, S. (2017). Selective erasure of distinct forms of long-term synaptic plasticity underlying different forms of memory in the same postsynaptic neuron. Current Biology, 27(13), 1888–1899 e1884. doi:10.1016/j.cub.2017.05.081Find this resource:

Immke, D. C., & McCleskey, E. W. (2003). Protons open acid-sensing ion channels by catalyzing relief of Ca2+ blockade. Neuron, 37(1), 75–84.Find this resource:

Jackman, S. L., Turecek, J., Belinsky, J. E., & Regehr, W. G. (2016). The calcium sensor synaptotagmin 7 is required for synaptic facilitation. Nature, 529(7584), 88–91.Find this resource:

Jekely, G., Paps, J., & Nielsen, C. (2015). The phylogenetic position of ctenophores and the origin(s) of nervous systems. Evodevo, 6, 1. doi:10.1186/2041-9139-6-1Find this resource:

Kaeser-Woo, Y. J., Younts, T. J., Yang, X., Zhou, P., Wu, D., Castillo, P. E., & Sudhof, T. C. (2013). Synaptotagmin-12 phosphorylation by cAMP-dependent protein kinase is essential for hippocampal mossy fiber LTP. Journal of Neuroscience, 33(23), 9769–9780. doi:10.1523/JNEUROSCI.5814-12.2013Find this resource:

Kass-Simon, G., & Pierobon, P. (2007). Cnidarian chemical neurotransmission: An updated overview. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 146(1), 9–25. doi:10.1016/j.cbpa.2006.09.008Find this resource:

Kassabov, S. R., Choi, Y. B., Karl, K. A., Vishwasrao, H. D., Bailey, C. H., & Kandel, E. R. (2013). A single Aplysia neurotrophin mediates synaptic facilitation via differentially processed isoforms. Cell Report, 3(4), 1213–1227. doi:10.1016/j.celrep.2013.03.008Find this resource:

Kim, H., Kwon, H., Kim, S., Kim, M. K., Botella, M. A., Yun, H. S., & Kwon, C. (2016). Synaptotagmin 1 negatively controls the two distinct immune secretory pathways to powdery mildew fungi in Arabidopsis. Plant Cell Physiology, 57(6), 1133–1141. doi:10.1093/pcp/pcw061Find this resource:

King, N., Westbrook, M. J., Young, S. L., Kuo, A., Abedin, M., Chapman, J., … Rokhsar, D. (2008). The genome of the choanoflagellate Monosiga brevicollis and the origin of metazoans. Nature, 451(7180), 783–788. doi:10.1038/nature06617Find this resource:

Kuner, T., Seeburg, P. H., & Guy, H. R. (2003). A common architecture for K+ channels and ionotropic glutamate receptors? Trends in Neuroscience, 26(1), 27–32.Find this resource:

Lampinen, M., Pentikainen, O., Johnson, M. S., & Keinanen, K. (1998). AMPA receptors and bacterial periplasmic amino acid-binding proteins share the ionic mechanism of ligand recognition. EMBO Journal, 17(16), 4704–4711. doi:10.1093/emboj/17.16.4704Find this resource:

Larsson, M., Sawada, K., Morland, C., Hiasa, M., Ormel, L., Moriyama, Y., & Gundersen, V. (2012). Functional and anatomical identification of a vesicular transporter mediating neuronal ATP release. Cerebral Cortex, 22(5), 1203–1214. doi:10.1093/cercor/bhr203Find this resource:

Lee, J., Guan, Z., Akbergenova, Y., & Littleton, J. T. (2013). Genetic analysis of synaptotagmin C2 domain specificity in regulating spontaneous and evoked neurotransmitter release. Journal of Neuroscience, 33(1), 187–200. doi:10.1523/JNEUROSCI.3214-12.2013Find this resource:

Lee, J. H., Kang, G. B., Lim, H. H., Jin, K. S., Kim, S. H., Ree, M., … Eom, S. H. (2008). Crystal structure of the GluR0 ligand-binding core from Nostoc punctiforme in complex with L-glutamate: Structural dissection of the ligand interaction and subunit interface. Journal of Molecular Biology, 376(2), 308–316. doi:10.1016/j.jmb.2007.10.081Find this resource:

Lee, P. N., Callaerts, P., & de Couet, H. G. (2009). Culture of Hawaiian bobtail squid (Euprymna scolopes) embryos and observation of normal development. Cold Spring Harbor Protocol, 2009(11), pdb prot5323. doi:10.1101/pdb.prot5323Find this resource:

Liebeskind, B. J., Hillis, D. M., & Zakon, H. H. (2015). Convergence of ion channel genome content in early animal evolution. Proceedings of the National Academy of Sciences USA, 112(8), E846–851. doi:10.1073/pnas.1501195112Find this resource:

Lingueglia, E., Champigny, G., Lazdunski, M., & Barbry, P. (1995). Cloning of the amiloride-sensitive FMRFamide peptide-gated sodium channel. Nature, 378(6558), 730–733. doi:10.1038/378730a0Find this resource:

Liu, Z., Chen, Z., Shang, C., Yan, F., Shi, Y., Zhang, J., … Cao, P. (2017). IGF1-dependent synaptic plasticity of mitral cells in olfactory memory during social learning. Neuron, 95(1), 106–122 e105. doi:10.1016/j.neuron.2017.06.015Find this resource:

Luo, F., Bacaj, T., & Sudhof, T. C. (2015). Synaptotagmin-7 is essential for Ca2+-triggered delayed asynchronous release but not for Ca2+-dependent vesicle priming in retinal ribbon synapses. Journal of Neuroscience, 35(31), 11024–11033. doi:10.1523/JNEUROSCI.0759-15.2015Find this resource:

Martinac, B., Saimi, Y., & Kung, C. (2008). Ion channels in microbes. Physiology Review, 88(4), 1449–1490. doi:10.1152/physrev.00005.2008Find this resource:

Masuda, T., Ozono, Y., Mikuriya, S., Kohro, Y., Tozaki-Saitoh, H., Iwatsuki, K., … Inoue, K. (2016). Dorsal horn neurons release extracellular ATP in a VNUT-dependent manner that underlies neuropathic pain. Nature Communications, 7, 12529. doi:10.1038/ncomms12529Find this resource:

Matsuoka, H., Harada, K., Nakamura, J., Fukuda, M., & Inoue, M. (2011). Differential distribution of synaptotagmin-1, -4, -7, and -9 in rat adrenal chromaffin cells. Cell Tissue Research, 344(1), 41–50. doi:10.1007/s00441-011-1131-8Find this resource:

Maycox, P. R., Deckwerth, T., Hell, J. W., & Jahn, R. (1988). Glutamate uptake by brain synaptic vesicles. Energy dependence of transport and functional reconstitution in proteoliposomes. Journal of Biological Chemistry, 263(30), 15423–15428.Find this resource:

Meyerhoff, O., Muller, K., Roelfsema, M. R., Latz, A., Lacombe, B., Hedrich, R., … Becker, D. (2005). AtGLR3.4, a glutamate receptor channel-like gene is sensitive to touch and cold. Planta, 222(3), 418–427. doi:10.1007/s00425-005-1551-3Find this resource:

Miyaji, T., Echigo, N., Hiasa, M., Senoh, S., Omote, H., & Moriyama, Y. (2008). Identification of a vesicular aspartate transporter. Proceedings of the National Academy of Sciences USA, 105(33), 11720–11724. doi:10.1073/pnas.0804015105Find this resource:

Moroz, L. L. (2009). On the independent origins of complex brains and neurons. Brain Behavior and Evolution, 74(3), 177–190. doi:10.1159/000258665Find this resource:

Moroz, L. L. (2015). Convergent evolution of neural systems in ctenophores. Journal of Experimental Biology, 218(Pt 4), 598–611. doi:10.1242/jeb.110692Find this resource:

Moroz, L. L., Kocot, K. M., Citarella, M. R., Dosung, S., Norekian, T. P., Povolotskaya, I. S., … Kohn, A. B. (2014). The ctenophore genome and the evolutionary origins of neural systems. Nature, 510(7503), 109–114. doi:10.1038/nature13400Find this resource:

Moroz, L. L., & Kohn, A. B. (2015). Unbiased view of synaptic and neuronal gene complement in Ctenophores: Are there Pan-neuronal and Pan-synaptic genes across Metazoa? Integrative Comparative Biology, 55(6), 1028–1049. doi:10.1093/icb/icv104Find this resource:

Moroz, L. L., & Kohn, A. B. (2016). Independent origins of neurons and synapses: insights from ctenophores. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences Royal Society, 371(1685), 20150041. doi:10.1098/rstb.2015.0041Find this resource:

Moss, A. G., & Tamm, S. L. (1986). Electrophysiological control of ciliary motor responses in the ctenophore Pleurobrachia. Journal of Comparative Physiology A, 158(3), 311–330.Find this resource:

Moss, A. G., & Tamm, S. L. (1987). A calcium regenerative potential controlling ciliary reversal is propagated along the length of ctenophore comb plates. Proceedings of the National Academy of Sciences USA, 84(18), 6476–6480.Find this resource:

Mousavi, S. A., Chauvin, A., Pascaud, F., Kellenberger, S., & Farmer, E. E. (2013). Glutamate receptor-like genes mediate leaf-to-leaf wound signalling. Nature, 500(7463), 422–426. doi:10.1038/nature12478Find this resource:

Nakanishi, N., Stoupin, D., Degnan, S. M., & Degnan, B. M. (2015). Sensory flask cells in sponge larvae regulate metamorphosis via calcium signaling. Integrative Comparative Biology, 55(6), 1018–1027. doi:10.1093/icb/icv014Find this resource:

Namikawa, H., Mawatari, S. F., & Clader, D. R. (1993). Reproduction, planula development, and substratum selection in three species of Stylactaria. Journal of Natural History, 27, 521–533.Find this resource:

Ni, J., Yu, Z., Du, G., Zhang, Y., Taylor, J. L., Shen, C., … Wu, Y. (2016). Heterologous expression and functional analysis of rice glutamate receptor-like family indicates its role in glutamate triggered calcium flux in rice roots. Rice (NY), 9(1), 9. doi:10.1186/s12284-016-0081-xFind this resource:

Nithianantharajah, J., Komiyama, N. H., McKechanie, A., Johnstone, M., Blackwood, D. H., St Clair, D., … Grant, S. G. (2013). Synaptic scaffold evolution generated components of vertebrate cognitive complexity. Nature Neuroscience, 16(1), 16–24. doi:10.1038/nn.3276Find this resource:

Pang, Z. P., Bacaj, T., Yang, X., Zhou, P., Xu, W., & Sudhof, T. C. (2011). Doc2 supports spontaneous synaptic transmission by a Ca(2+)-independent mechanism. Neuron, 70(2), 244–251. doi:10.1016/j.neuron.2011.03.011Find this resource:

Philippe, H., Derelle, R., Lopez, P., Pick, K., Borchiellini, C., Boury-Esnault, N., … Manuel, M. (2009). Phylogenomics revives traditional views on deep animal relationships. Current Biology, 19(8), 706–712. doi:10.1016/j.cub.2009.02.052Find this resource:

Putnam, N. H., Butts, T., Ferrier, D. E., Furlong, R. F., Hellsten, U., Kawashima, T., … Rokhsar, D. S. (2008). The amphioxus genome and the evolution of the chordate karyotype. Nature, 453(7198), 1064–1071. doi:10.1038/nature06967Find this resource:

Putnam, N. H., Srivastava, M., Hellsten, U., Dirks, B., Chapman, J., Salamov, A., … Rokhsar, D. S. (2007). Sea anemone genome reveals ancestral eumetazoan gene repertoire and genomic organization. Science, 317(5834), 86–94. doi:10.1126/science.1139158Find this resource:

Qi, Z., Stephens, N. R., & Spalding, E. P. (2006). Calcium entry mediated by GLR3.3, an Arabidopsis glutamate receptor with a broad agonist profile. Plant Physiology, 142(3), 963–971. doi:10.1104/pp.106.088989Find this resource:

Ramesh, S. A., Tyerman, S. D., Gilliham, M., & Xu, B. (2017). gamma-Aminobutyric acid (GABA) signalling in plants. Cellular and Molecular Life Sciences, 74(9), 1577–1603. doi:10.1007/s00018-016-2415-7Find this resource:

Riesgo, A., Farrar, N., Windsor, P. J., Giribet, G., & Leys, S. P. (2014). The analysis of eight transcriptomes from all poriferan classes reveals surprising genetic complexity in sponges. Molecular Biology and Evolution, 31(5), 1102–1120. doi:10.1093/molbev/msu057Find this resource:

Riviere, G., Klopp, C., Ibouniyamine, N., Huvet, A., Boudry, P., & Favrel, P. (2015). GigaTON: an extensive publicly searchable database providing a new reference transcriptome in the pacific oyster Crassostrea gigas. BMC Bioinformatics, 16, 401. doi:10.1186/s12859-015-0833-4Find this resource:

Roux, I., Safieddine, S., Nouvian, R., Grati, M., Simmler, M. C., Bahloul, A., … Petit, C. (2006). Otoferlin, defective in a human deafness form, is essential for exocytosis at the auditory ribbon synapse. Cell, 127(2), 277–289. doi:10.1016/j.cell.2006.08.040Find this resource:

Ryan, J. F. (2014). Did the ctenophore nervous system evolve independently? Zoology (Jena), 117(4), 225–226. doi:10.1016/j.zool.2014.06.001Find this resource:

Ryan, J. F., Pang, K., Schnitzler, C. E., Nguyen, A. D., Moreland, R. T., Simmons, D. K., … Baxevanis, A. D. (2013). The genome of the ctenophore Mnemiopsis leidyi and its implications for cell type evolution. Science, 342(6164), 1242592. doi:10.1126/science.1242592Find this resource:

Ryan, T. J., Kopanitsa, M. V., Indersmitten, T., Nithianantharajah, J., Afinowi, N. O., Pettit, C., … Komiyama, N. H. (2013). Evolution of GluN2A/B cytoplasmic domains diversified vertebrate synaptic plasticity and behavior. Nature Neuroscience, 16(1), 25–32. doi:10.1038/nn.3277Find this resource:

Sagne, C., & Gasnier, B. (2008). Molecular physiology and pathophysiology of lysosomal membrane transporters. Journal of Inherited Metabolic Disease, 31(2), 258–266. doi:10.1007/s10545-008-0879-9Find this resource:

Saheki, Y., & De Camilli, P. (2017). The extended-synaptotagmins. Biochimica et Biophysica Acta, 1864(9), 1490–1493. doi:10.1016/j.bbamcr.2017.03.013Find this resource:

Sakamoto, S., Miyaji, T., Hiasa, M., Ichikawa, R., Uematsu, A., Iwatsuki, K., … Moriyama, Y. (2014). Impairment of vesicular ATP release affects glucose metabolism and increases insulin sensitivity. Science Report, 4, 6689. doi:10.1038/srep06689Find this resource:

Sakarya, O., Armstrong, K. A., Adamska, M., Adamski, M., Wang, I. F., Tidor, B., … Kosik, K. S. (2007). A post-synaptic scaffold at the origin of the animal kingdom. PLoS One, 2(6), e506. doi:10.1371/journal.pone.0000506Find this resource:

Scappaticci, A. A., & Kass-Simon, G. (2008). NMDA and GABA B receptors are involved in controlling nematocyst discharge in hydra. Comparative Biochemistry and Physiology: Part A: Molecular and Integrative Physiology, 150(4), 415–422. doi:10.1016/j.cbpa.2008.04.606Find this resource:

Schonn, J. S., Maximov, A., Lao, Y., Sudhof, T. C., & Sorensen, J. B. (2008). Synaptotagmin-1 and -7 are functionally overlapping Ca2+ sensors for exocytosis in adrenal chromaffin cells. Proceedings of the National Academy of Sciences USA, 105(10), 3998–4003. doi:10.1073/pnas.0712373105Find this resource:

Schuchert, P. (1993). Trichoplax-adhaerens (phylum Placozoa) has cells that react with antibodies against the neuropeptide rfamide. Acta Zoologica, 74(2), 115–117.Find this resource:

Sebe-Pedros, A., Irimia, M., Del Campo, J., Parra-Acero, H., Russ, C., Nusbaum, C., … Ruiz-Trillo, I. (2013). Regulated aggregative multicellularity in a close unicellular relative of metazoa. Elife, 2, e01287. doi:10.7554/eLife.01287Find this resource:

Sekito, T., Fujiki, Y., Ohsumi, Y., & Kakinuma, Y. (2008). Novel families of vacuolar amino acid transporters. IUBMB Life, 60(8), 519–525. doi:10.1002/iub.92Find this resource:

Shin, O. H., Xu, J., Rizo, J., & Sudhof, T. C. (2009). Differential but convergent functions of Ca2+ binding to synaptotagmin-1 C2 domains mediate neurotransmitter release. Proceedings of the National Academy of Sciences USA, 106(38), 16469–16474. doi:10.1073/pnas.0908798106Find this resource:

Shinzato, C., Shoguchi, E., Kawashima, T., Hamada, M., Hisata, K., Tanaka, M., … Satoh, N. (2011). Using the Acropora digitifera genome to understand coral responses to environmental change. Nature, 476(7360), 320–323. doi:10.1038/nature10249Find this resource:

Shomrat, T., Graindorge, N., Bellanger, C., Fiorito, G., Loewenstein, Y., & Hochner, B. (2011). Alternative sites of synaptic plasticity in two homologous “fan-out fan-in” learning and memory networks. Current Biology, 21(21), 1773–1782. doi:10.1016/j.cub.2011.09.011Find this resource:

Shomrat, T., Turchetti-Maia, A. L., Stern-Mentch, N., Basil, J. A., & Hochner, B. (2015). The vertical lobe of cephalopods: An attractive brain structure for understanding the evolution of advanced learning and memory systems. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 201(9), 947–956. doi:10.1007/s00359-015-1023-6Find this resource:

Shrestha, P., Orvis, J., Tallon, L. J., Mahurkar, A., Fraser, C. M., & Abrams, T. W. (2015). Strategy for achieving a complete cns transcriptome for Aplysia, a model system in learning and memory. Society for Neuroscience Abstracts, 262.04.Find this resource:

Si, K., Choi, Y. B., White-Grindley, E., Majumdar, A., & Kandel, E. R. (2010). Aplysia CPEB can form prion-like multimers in sensory neurons that contribute to long-term facilitation. Cell, 140(3), 421–435. doi:10.1016/j.cell.2010.01.008Find this resource:

Si, K., Giustetto, M., Etkin, A., Hsu, R., Janisiewicz, A. M., Miniaci, M. C., … Kandel, E. R. (2003). A neuronal isoform of CPEB regulates local protein synthesis and stabilizes synapse-specific long-term facilitation in Aplysia. Cell, 115(7), 893–904.Find this resource:

Silvestro, D., & Michalak, I. (2012). raxmlGUI: A graphical front-end for RAxML. Organisms Diversity & Evolution, 12(4), 335–337.Find this resource:

Simakov, O., Kawashima, T., Marletaz, F., Jenkins, J., Koyanagi, R., Mitros, T., … Gerhart, J. (2015). Hemichordate genomes and deuterostome origins. Nature, 527(7579), 459–465. doi:10.1038/nature16150Find this resource:

Simakov, O., Marletaz, F., Cho, S. J., Edsinger-Gonzales, E., Havlak, P., Hellsten, U., … Rokhsar, D. S. (2013). Insights into bilaterian evolution from three spiralian genomes. Nature, 493(7433), 526–531. doi:10.1038/nature11696Find this resource:

Smith, C. L., Pivovarova, N., & Reese, T. S. (2015). Coordinated feeding behavior in Trichoplax, an animal without synapses. PLoS One, 10(9), e0136098. doi:10.1371/journal.pone.0136098Find this resource:

Smith, C. L., Varoqueaux, F., Kittelmann, M., Azzam, R. N., Cooper, B., Winters, C. A., … Reese, T. S. (2014). Novel cell types, neurosecretory cells, and body plan of the early-diverging metazoan Trichoplax adhaerens. Current Biology, 24(14), 1565–1572. doi:10.1016/j.cub.2014.05.046Find this resource:

Srivastava, M., Begovic, E., Chapman, J., Putnam, N. H., Hellsten, U., Kawashima, T., … Rokhsar, D. S. (2008). The Trichoplax genome and the nature of placozoans. Nature, 454(7207), 955–960. doi:10.1038/nature07191Find this resource:

Srivastava, M., Simakov, O., Chapman, J., Fahey, B., Gauthier, M. E., Mitros, T., … Rokhsar, D. S. (2010). The Amphimedon queenslandica genome and the evolution of animal complexity. Nature, 466(7307), 720–726. doi:10.1038/nature09201Find this resource:

Stamatakis, A. (2014). RAxML version 8: A tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics, 30(9), 1312–1313. doi:10.1093/bioinformatics/btu033Find this resource:

Sudhof, T. C. (2012). Calcium control of neurotransmitter release. Cold Spring Harbor Perspectives in Biology, 4(1), a011353. doi:10.1101/cshperspect.a011353Find this resource:

Sugita, S., Han, W., Butz, S., Liu, X., Fernandez-Chacon, R., Lao, Y., & Sudhof, T. C. (2001). Synaptotagmin VII as a plasma membrane Ca(2+) sensor in exocytosis. Neuron, 30(2), 459–473.Find this resource:

Takamori, S., Holt, M., Stenius, K., Lemke, E. A., Gronborg, M., Riedel, D., … Jahn, R. (2006). Molecular anatomy of a trafficking organelle. Cell, 127(4), 831–846. doi:10.1016/j.cell.2006.10.030Find this resource:

Takeuchi, T., Harada, Y., Moriyama, S., Furuta, K., Tanaka, S., Miyaji, T., … Hiasa, M. (2017). Vesicular polyamine transporter mediates vesicular storage and release of polyamine from mast cells. Journal of Biological Chemistry, 292(9), 3909–3918. doi:10.1074/jbc.M116.756197Find this resource:

Talavera, G., & Castresana, J. (2007). Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Systematic Biology, 56(4), 564–577. doi:10.1080/10635150701472164Find this resource:

Traynelis, S. F., Wollmuth, L. P., McBain, C. J., Menniti, F. S., Vance, K. M., Ogden, K. K., … Dingledine, R. (2010). Glutamate receptor ion channels: structure, regulation, and function. Pharmacology Review, 62(3), 405–496. doi:10.1124/pr.109.002451Find this resource:

Turchetti-Maia, A., Stern-Mentch, N., Nesher, N., Shomrat, T., & Hochner, B. (2016). A novel LTP mechanism that involves persistent self-activation of nitric oxide synthase (NOS). Society for Neuroscience Abstracts, 685.01.Find this resource:

Wang, H., Han, S., Siao, W., Song, C., Xiang, Y., Wu, X., … Zhao, H. (2015). Arabidopsis synaptotagmin 2 participates in pollen germination and tube growth and is delivered to plasma membrane via conventional secretion. Molecular Plant, 8(12), 1737–1750. doi:10.1016/j.molp.2015.09.003Find this resource:

Watanabe, S., & Boucrot, E. (2017). Fast and ultrafast endocytosis. Current Opinion in Cell Biology, 47, 64–71. doi:10.1016/ this resource:

Westfall, J. A. (1996). Ultrastructure of synapses in the first-evolved nervous systems. Journal of Neurocytology, 25(12), 735–746.Find this resource:

Westfall, J. A., & Kinnamon, J. C. (1978). A second sensory—motor—interneuron with neurosecretory granules in Hydra. Journal of Neurocytology, 7(3), 365–379.Find this resource:

Westfall, J. A., & Kinnamon, J. C. (1984). Perioral synaptic connections and their possible role in the feeding behavior of Hydra. Tissue Cell, 16(3), 355–365.Find this resource:

Winters, G. C., Kohn, A. B., Polese, G., Dicosmo, A., Hochner, B., Stern, N., … Moroz, L. L. (2016). Molecular organization of octopus brains reveals first insight into unique memory center signaling. Society for Neuroscience Abstracts, 642.04.Find this resource:

Winters, G. C., Kohn, A. B., Stern, N., Hochner, B., Walters, E. T., Crook, R. J., & Moroz, L. L. (2015). Cephalopod neural transcriptomes reveal unique strategies for memory centers. Society for Neuroscience Abstracts, 205.07.Find this resource:

Wood, M. W., VanDongen, H. M., & VanDongen, A. M. (1995). Structural conservation of ion conduction pathways in K channels and glutamate receptors. Proceedings of the National Academy of Sciences USA, 92(11), 4882–4886.Find this resource:

Yang, X., Pei, J., Kaeser-Woo, Y. J., Bacaj, T., Grishin, N. V., & Sudhof, T. C. (2015). Evolutionary conservation of complexins: From choanoflagellates to mice. EMBO Reports, 16(10), 1308–1317. doi:10.15252/embr.201540305Find this resource:

Yang, Z., & Rannala, B. (2012). Molecular phylogenetics: principles and practice. Nature Reviews Genetics, 13(5), 303–314. doi:10.1038/nrg3186Find this resource:

Yoshihara, M., Guan, Z., & Littleton, J. T. (2010). Differential regulation of synchronous versus asynchronous neurotransmitter release by the C2 domains of synaptotagmin 1. Proceedings of the National Academy of Sciences USA, 107(33), 14869–14874. doi:10.1073/pnas.1000606107Find this resource:

Yu, A., Alberstein, R., Thomas, A., Zimmet, A., Grey, R., Mayer, M. L., & Lau, A. Y. (2016). Molecular lock regulates binding of glycine to a primitive NMDA receptor. Proceedings of the National Academy of Sciences USA, 113(44), E6786–E6795. doi:10.1073/pnas.1607010113Find this resource:

Yunes, R. A., Poluektova, E. U., Dyachkova, M. S., Klimina, K. M., Kovtun, A. S., Averina, O. V., … Danilenko, V. N. (2016). GABA production and structure of gadB/gadC genes in Lactobacillus and Bifidobacterium strains from human microbiota. Anaerobe, 42, 197–204. doi:10.1016/j.anaerobe.2016.10.011Find this resource:

Zepeda, E. A., Veline, R. J., & Crook, R. J. (2017). Rapid associative learning and stable long-term memory in the squid Euprymna scolopes. Biological Bulletin, 232(3), 212–218. doi:10.1086/693461Find this resource: