Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

Phylogenetic characterization of transporter proteins in the cnidarian-dinoflagellate symbiosis

Ashley E. Sprolesa, Nathan L. Kirkb, Sheila A. Kitchenb,1, Clinton A. Oakleya, Arthur R. Grossmanc,Virginia M. Weisb, Simon K. Davya,⁎

a School of Biological Sciences, Victoria University of Wellington, Wellington 6012, New ZealandbDepartment of Integrative Biology, Oregon State University, Corvallis, OR 97331, USAc Department of Plant Biology, The Carnegie Institution for Science, Stanford, CA 94305, USA

A R T I C L E I N F O

Keywords:GlucoseGlycerolNitrogenCoralMembrane transportBioinformatics

A B S T R A C T

Metabolic exchange between cnidarians and their symbiotic dinoflagellates is central to maintaining theirmutualistic relationship. Sugars are translocated to the host, while ammonium and nitrate are utilized by thedinoflagellates (Symbiodinium spp.). We investigated membrane protein sequences of each partner to identifypotential transporter proteins that move sugars into cnidarian cells and nitrogen products into Symbiodiniumcells. We examined the facilitated glucose transporters (GLUT), sodium/glucose cotransporters (SGLT), andaquaporin (AQP) channels in the cnidarian host as mechanisms for sugar uptake, and the ammonium and high-affinity nitrate transporters (AMT and NRT2, respectively) in the algal symbiont as mechanisms for nitrogenuptake. Homologous protein sequences were used for phylogenetic analysis and tertiary structure deductions. Incnidarians, we identified putative glucose transporters of the GLUT family and glycerol transporting AQP pro-teins, as well as sodium monocarboxylate transporters and sodium myo-inositol cotransporters homologous toSGLT proteins. We hypothesize that cnidarians use GLUT proteins as the primary mechanism for glucose uptake,while glycerol moves into cells by passive diffusion. We also identified putative AMT proteins in severalSymbiodinium clades and putative NRT2 proteins only in a single clade. We further observed an upregulation ofexpressed putative AMT proteins in Symbiodinium, which may have emerged as an adaptation to conditionsexperienced inside the host cell. This study is the first to identify transporter sequences from a diversity ofcnidarian species and Symbiodinium clades, which will be useful for future experimental analyses of the host-symbiont proteome and the nutritional exchange of Symbiodinium cells in hospite.

1 Introduction

The cnidarian-dinoflagellate symbiosis is an endosymbiotic re-lationship involving intercellular exchange of many compounds.Dinoflagellates of the genus Symbiodinium reside within cnidarian en-dodermal cells and provide the host with the majority of its fixedcarbon as photosynthetic products, with sugars being the major traf-ficked metabolites (Davy et al., 2012; Gordon and Leggat, 2010; Koppet al., 2015). In exchange, the cnidarian provides nitrogenous com-pounds, generated as by-products from heterotrophic feeding, to thealgal symbionts (Allemand et al., 1998; Davy et al., 2012; Houlbrèqueet al., 2004). Successful metabolite exchange between the host andsymbiont is key to the health and survival of reef-building corals, as itallows the partners to obtain necessary nutrients that are often limitingin oligotrophic oceans (Muscatine et al., 1981; O’Neil and Capone,

2008).Characterization of transported metabolites has long been a focus of

cnidarian-dinoflagellate research. Glucose is now considered to be theprimary carbon product translocated to the host from the symbiont(Burriesci et al., 2012; Hillyer et al., 2016), although evidence fromearly studies pointed to glycerol (Lewis and Smith, 1971; Muscatine,1967; Schmitz and Kremer, 1977). It is now thought that glycerol isprimarily released by Symbiodinium cells in isolation (e.g. in culture),while glucose is the major photosynthetic product translocated from thesymbiont to the host in hospite (Burriesci et al., 2012; Hillyer et al.,2016; Ishikura et al., 1999; Whitehead and Douglas, 2003). High levelsof glycerol have also been found to be released as an osmolyte in re-sponse to stress (Suescún-Bolívar et al., 2012). In the opposite direction,the major nitrogen metabolite transported to the symbiont by the hostis thought to be ammonium, while nitrate is an important nutrient

https://doi.org/10.1016/j.ympev.2017.12.007Received 7 November 2016; Received in revised form 22 October 2017; Accepted 5 December 2017

⁎ Corresponding author.

1 Current address: Department of Biology, Pennsylvania State University, University Park, PA 16802, USA.E-mail address: simon.davy@vuw.ac.nz (S.K. Davy).

Molecular Phylogenetics and Evolution 120 (2018) 307–320

Available online 09 December 20171055-7903/ © 2017 Elsevier Inc. All rights reserved.

T

obtained from seawater (Kopp et al., 2013; Muller-Parker et al., 1994;Pernice et al., 2012). Nitrogenous compounds are used for the synthesisof amino acids that are then incorporated into symbiont proteins (Wangand Douglas, 1998) or translocated back to the animal (Tanaka et al.,2006; Wang and Douglas, 1999).

Although these carbon and nitrogen metabolites are an importantsource of energy and nutrition in the symbiosis, the transport me-chanisms are still poorly understood. All molecules entering or leavinga cell must traverse the plasma membrane, most either by activetransport or passive diffusion through transporter or channel proteins(Lodish et al., 2000). In the cnidarian-dinoflagellate symbiosis, Sym-biodinium cells are housed within a host-derived vacuole called thesymbiosome, which acts as the interface between host and symbiont(Fitt and Trench, 1983; Wakefield et al., 2000; Wakefield and Kempf,2001). Therefore, the exchange of compounds must be facilitated byproteins in the cellular membranes of each of the partners as well as inthe symbiosome interface (Fig. 1).

Insights on nutrient transport can be gained from studies of othersystems. Glucose primarily enters eukaryotic cells through transportersin two families: facilitated glucose transporters (GLUT) and sodium-coupled glucose transporters (SGLT) (Scheepers et al., 2015; Wood andTrayhurn, 2003). Members of both families have been extensivelycharacterized in humans and other animals (Doege et al., 2001, 2000;Escher and Rasmuson-Lestander, 1999; Joost et al., 2002; Wright,2013). While few invertebrate GLUT and SGLT homologs have beendescribed (Escher and Rasmuson-Lestander, 1999; Feng et al., 2013),proteins from both of these families are encoded by genes of the cni-darian Aiptasia pallida (also known as Exaiptasia pallida) (Baumgartenet al., 2015). Previous studies have used proteomics (Oakley et al.,

2016; Peng et al., 2010) and transcriptomics (Lehnert et al., 2014) toidentify proteins involved in the symbiosis; however, transportersspecific to the symbiosome membrane have not been firmly established.Glycerol can be trafficked by aquaporin (AQP) channels, which areintrinsic water channels that transport small molecules (Hara-Chikumaand Verkman, 2006; Rojek et al., 2008). Since most eukaryotes alreadyuse AQP to facilitate the diffusion of water, we predict that cnidarianslikely have AQP members capable of glycerol transport as well.

Currently, three genes encoding ammonium transporters (AMTproteins) and two encoding high-affinity nitrate transporters (NRTproteins) have been identified in Symbiodinium transcriptomes (Brownet al., 2011; Mayfield et al., 2010; Van Den Heuvel, 2012). The recentgenome assembly of Symbiodinium kawagutii (Clade F) has also revealedgenes encoding AMT and NRT family members (Lin et al., 2015).However, these studies have only focused on a single Symbiodinium typeand did not consider the large genetic diversity within the genus.Currently, the Symbiodinium genus is classified into nine divergentclades (A–I), with each (except clade E) further divided into sub-cladesor “types” (e.g., A3, B1, D1a) and some described species (Coffroth andSantos, 2005; Lajeunesse et al., 2012; Pochon and Gates, 2010). Geneticdistances between these clades were found to be greater than the dis-tances separating different species of free-living dinoflagellates(Krueger et al., 2015; Sampayo et al., 2009; Santos et al., 2002;Takabayashi et al., 2004). Therefore, analysis of these proteins from adiversity of Symbiodinium types may reveal differences in their con-servation.

The goal of this study was to identify and characterize potentialmembrane transport mechanisms employed by both partners of thecnidarian-dinoflagellate symbiosis by performing informatics/

HostSymbiosome

Symbiont

Chloroplast

Sugars FixedCarbon

NH 4+

NO3-

DIC

Amino Acids

Amino Acids B

A

B

B

B

A

Fig. 1. Proposed membrane transport processes of the major exchanged metabolites in the cnidarian-dinoflagellate symbiosis (A) Dissolved inorganic carbon (HCO3− and CO2) obtained

from the seawater or the host is used by the algal symbionts for photosynthesis. Carbon is fixed during photosynthesis, and a portion of the products are translocated to the host in theform of sugars, which must be facilitated through integral membrane proteins in the algal cell membrane and symbiosome membrane. (B) Nitrate and ammonium are either obtained fromthe seawater or excreted as host waste products, then translocated to the symbiont through integral membrane proteins in the symbiosome membrane and symbiont cell membrane. Thealgae use the ammonium and nitrate for amino acid synthesis, with a portion of amino acids being translocated back to the host.

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

308

phylogenetic investigations of well-described sugar and nitrogentransport proteins that are highly conserved among eukaryotes. Weinvestigated five protein families: three families of secondary activetransporters, GLUT, SGLT, and NRT2, and two families of channelproteins, AQP and AMT. We used a phylogenetic approach to identifycnidarian homologs to GLUT, SGLT, and AQP, and Symbiodiniumhomologs to AMT and NRT2. We then performed protein structuralpredictions, conservation analyses, and structural domain identifica-tions to help establish whether or not these proteins function in meta-bolite exchange within the cnidarian-dinoflagellate symbiosis. Insightsgained from these analyses will help inform further functional in-vestigations of the symbiosome membrane and enhance our under-standing of nutrient exchange in the cnidarian-dinoflagellate symbiosis.

2. Materials and methods

2.1. Identification of cnidarian and Symbiodinium transporter proteins

Separate analyses were completed to categorize the major trans-porters of the cnidarian host and Symbiodinium symbiont. Cnidarian andSymbiodinium transcriptomes and genomes were downloaded frompublically available sources (Table A1), and each was formatted into asearchable database. Reference sequences for each protein family wereobtained from UniProt (Table A2); these included Homo sapiens GLUT,SGLT, and AQP, and Arabidopsis thaliana AMT and NRT2 (Table A2).Putative orthologues to the reference sequences were identified in thecnidarian and Symbiodinium databases by reciprocal BLAST (Telford,2007; Voolstra et al., 2011) with a bit-score cutoff of 80 for any givenalignment. To identify cnidarian sugar transporters, H. sapiens referenceproteins (Table A2) were used as BLAST input queries to search cni-darian databases; A. thaliana reference proteins (Table A2) were simi-larly used as BLAST input queries for searching the Symbiodinium da-tabases for nitrate and ammonium transporters. The best scoringmatches were then converted to predicted proteins using custom Perlscripts (obtained from Eli Meyer, https://github.com/Eli-Meyer).

2.2. Retrieval of protein homologs

Sequences from the cnidarian and Symbiodinium databases withhigh sequence similarity to the reference proteins were retrieved andused for phylogenetic analyses. Those protein sequences were also usedto search the NCBI non-redundant (nr) database using BLASTp, re-trieving sequences with an E value cutoff of ≤1×10−5 and excludinghypothetical proteins (Datasheets A1–A5). The top five annotatedBLAST hits for each sequence were used as additional homologs in thephylogenetic analyses (Datasheets A1–A5).

2.3. Sequence alignment and phylogenetic analysis

Protein family sequences were aligned using MAFFT v. 7.017 (de-fault parameters) (Katoh and Standley, 2013). Alignments weremanually edited in Aliview (Larsson, 2014) to eliminate false positivematches (Bucka-Lassen et al., 1999) by removing protein fragments(sequences of< 150 amino acids) and sequences that did not align withany of the defined motifs in Figs. S1–S5: the GLUT GR, GRR, and PETKmotifs (Fig. S1), all of which are considered necessary for sugar trans-port function (Joost and Thorens, 2001); the SGLT Y, GG, and D motifs(Fig. S2), which are the most conserved (≥77%) between all pro- andeukaryotic taxa (Wright and Turk, 2004); the two NPA motifs and one Dresidue of AQP (Fig. S3), since NPA motifs form the aqueous pore andrepresent the AQP signature while the D residue is present in aqua-glyceroporins and widens the pore opening allowing the transfer oflarger solutes (Agre et al., 2002); the AMT D, AG, VH, and G motifs (Fig.S4), which are responsible for ammonium recognition and transportfunction (Marini et al., 2006); the NRT2 conserved F residue along withthe proposed NNP signature A/G-G-W-G/A-N/D-M/L-G and the

structurally important GG and GAGG glycine clusters (Fig. S5) (Forde,2000; Trueman et al., 1996).

Phylogenetic analysis of each protein family was conducted usingmaximum-likelihood inference in RAxML v8.0.26 (Stamatakis, 2014).The PROTGAMMAAUTO function was used to empirically (highestlikelihood score of starting parsimony tree) determine the best sub-stitution model for each dataset; all datasets in this study were bestsuited to using the AUTO model. The topology with the highest like-lihood score for each protein tree was selected from 100 replicate trees,and then branch node support was calculated using 100 bootstraps.Protein homologs from the most ancestral species were used to rooteach tree. Phylogenetic inference was then used to draw conclusions onthe protein identities of novel cnidarian and Symbiodinium sequences(Eisen, 1998; Eisen et al., 1997; Sjölander, 2004; Zmasek and Eddy,2001).

2.4. Structural predictions and characterization of proteins

Additional sequence analyses were conducted to further char-acterize the identified cnidarian and Symbiodinium proteins. Thenumber of transmembrane (TM) helices and subcellular localization foreach protein were predicted using the MemPype webserver, whichoutperforms other localization methods with an accuracy rate of 70%(Pierleoni et al., 2011), and protein domains were determined using theInterProScan (Quevillon et al., 2005) in Geneious v. 7.1.9. One proteinsequence per gene family was also selected for protein structural ana-lysis. Nematostella vectensis sequences with highest percent similarity tothe reference sequences were used for analysis of the cnidarian proteinsbecause they were closest in amino acid length and had the highestsimilarity to the reference proteins. Furthermore, the N. vectensisgenome had the most complete annotations. A sequence from Symbio-dinium C was used for AMT structural analysis to further examine thehigh AMT protein copies found in this symbiont type and because atleast one AMT sequence from this type was grouped with each eu-karyotic clade (Fig. 6, clades 2–5), while a sequence from Symbiodiniumclade A was chosen for NRT2 structural analysis since it was the onlyclade with sequences retrieved for this protein family. The three-di-mensional structure and protein function were predicted using I-TASSER (Roy et al., 2010; Zhang, 2008). I-TASSER aligns predictionswith the most similar protein structures from the Protein Data Bank(http://www.rcsb.org/pdb) and evaluates the structural similarityusing a TM-score ranging from 0 to 1, where 1 indicates a perfectmatch. Ligands and ligand-binding sites are also predicted as part of I-TASSER using the COACH algorithm (Yang et al., 2013), which eval-uates the confidence of the prediction using a C-score ranging from 0 to1, where 1 indicates 100% confidence in the prediction. To determinethe degree of structural conservation within each cnidarian and Sym-biodinium protein family, all newly identified sequences were alignedwithin their family using MAFFT v. 7.017 and the alignment was sub-mitted to the ConSurf server (Ashkenazy et al., 2010) along with thecorresponding predicted 3D structure as a template to display residueconservation. Conservation scores for each residue were calculatedusing the Bayesian method with the best evolutionary substitutionmodel determined empirically as WAG (Mayrose et al., 2004), andscores were subsequently depicted onto the protein structures usingPolyview 3D (Porollo and Meller, 2007).

3. Results and discussion

3.1. Cnidarian GLUT proteins

A total of 23 cnidarian sequences were identified as GLUT homo-logs, representing putative glucose transporters (Datasheet A1).Sequence lengths were highly variable, with some sequences containingfewer than the 12 TM domains that are typical of proteins within themajor facilitator superfamily (MFS), which includes GLUT proteins

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

309

Fig. 2. Phylogenetic tree of cnidarian GLUT sequences and closest homologs from other organisms. The tree was rooted by the yeast S. cerevisiae homolog. The number of amino acidsubstitutions is indicated by the scale bar, and bootstrap support for nodes> 50% is represented by shaded circles. Cnidarian sequences are shown in bold text. Four main clades emerged:(1) OCT (2) Class 1 and 2 GLUT proteins; (3) GLUT6 and GLUT8; and (4) GLUT10, GLUT12, and HMIT. Clades containing only vertebrate sequences are collapsed to enhance tree clarity.The sequence used for structural analysis is marked by a yellow star (N. vectensis_4 GLUT8).

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

310

(Reddy et al., 2012). These sequences may represent incomplete pro-teins that were not excluded from our analysis because they contain theconserved protein motifs (Datasheet A1). Only cnidarian sequencesfrom GLUT classes 1 and 2 were identified. All sequences include do-mains for MFS sugar transporters, while some also contain domainsassociated with specific GLUT members.

Our phylogenetic analysis of the GLUT proteins resolved the se-quences into four main clades of solute transporters (Fig. 2): (1) organiccation transporters (OCT); (2) class 1 and 2 GLUT proteins; (3) GLUT6and GLUT8; and (4) GLUT10, GLUT12, and HMIT. One divergent pro-tein from Porites australiensis (Fig. 2) branches from near the root of thetree rather than clustering with any of the four main clades. This pro-tein shares the highest sequence similarity with the yeast glucosetransporter RGT2 (Datasheet A1). Since the P. australiensis tran-scriptome was generated from RNA derived from the entire holobiontrather than just host material, it is likely that this deduced protein re-presents RNA contamination from the algal symbiont or bacteria(Shinzato et al., 2014). Robust support is only observed for the OCTclade, with the other major clades being less well resolved. OCT isanother MFS protein family that is closely related to the sugar porterfamily (Saier et al. 1999) (Fig. 2, clade 1). These proteins facilitatediffusion of organic cations such as neurotransmitters, amino and fattyacids, and nucleotides in vertebrates, but they have not been well stu-died in invertebrate systems (Ciarimboli, 2008; Eraly et al., 2004).

The clustering of the GLUT sequences resulted in a branching orderthat follows previously constructed metazoan GLUT phylogenies(Wilson-O’Brien et al., 2010). Class 3 GLUT proteins are ancestral to theother GLUTs, and they have diverged into three well supported sub-clades (GLUT6 and GLUT8; GLUT10 and GLUT12; and HMIT), whileclass 1 and 2 proteins form a well-supported clade as sister groups thatevolved more recently (Scheepers et al., 2015; Wilson-O’Brien et al.,2010). Overall, we identified cnidarian homologs to all class 1 and class3 GLUT members, except for GLUT6 (Datasheet A1).

Ten cnidarian putative GLUT8 proteins were identified (Fig. 2).GLUT8 is of particular interest in the cnidarian-dinoflagellate symbiosisbecause it has a high affinity for glucose (Uldry and Thorens, 2004) andexhibits increased mRNA levels in symbiotic A. pallida relative to thosein aposymbiotic anemones of the same species (Lehnert et al., 2014). Asit is localized to endosomal compartments in mice, GLUT8 is predictedto function as a hexose transporter in intracellular membranes(Augustin et al., 2005). Four of the identified cnidarian GLUT8 proteinsin this study were predicted to be localized to internal membranes,while six were predicted to exist in the plasma membrane (DatasheetA1). Internal GLUT proteins may reside in the symbiosome membrane,which is an internal vacuolar membrane of host origin that is derivedfrom an early endosome (Fitt and Trench, 1983; Rands et al., 1993;Wakefield and Kempf, 2001). Three copies of GLUT8 were found in N.vectensis and two in P. australiensis. All copies in the non-symbioticcnidarian N. vectensis were predicted to be localized to the plasmamembrane. The two copies of GLUT8 in the symbiotic cnidarian, P.australiensis, had different predicted localizations based on the Mem-Loci algorithm; one is likely in the plasma membrane while the otherappears to reside in internal membranes. These analyses raise thepossibility that symbiotic cnidarians have a distinct GLUT8 that is as-sociated with the symbiosome membrane where it facilitates movementof glucose from the symbiont to the host. The protein predicted to beinternally located does form a highly supported group with one of theN. vectensis cell membrane proteins. Perhaps the internally localized P.australiensis protein is one of the most ancestral forms of the symbio-some-associated transporter as it still maintains high similarity to thosein the cell membrane of aposymbiotic animals, or perhaps this parti-cular isoform can be expressed in either the cell or internal membranessince prediction probabilities were over 50% for both (Datasheet A1).

The deduced 3D structure of the cnidarian GLUT8 protein closelyresembles the crystal structure of human GLUT1 (TM score= 0.89,Fig. 3). All ten structural hits from the Protein Data Bank were GLUT

proteins from other species ranging from bacteria to vertebrates (Da-tasheet A1, Tab C). Among the cnidarian GLUT sequences, conservationwas highest for residues on the inner surface of the translocation pore,where the active binding sites are presumably located (Fig. 3). Thepredicted ligand was maltose (Fig. 3), a disaccharide composed of twoglucose molecules (Berg et al., 2002). However, confidence of thisprediction was low (C-score= 0.11), so we cannot make definitiveconclusions about the solutes transported by the cnidarian GLUT pro-tein. Overall, the results of the structural analysis support the hypoth-esis that the identified cnidarian proteins are highly conserved homo-logs of human GLUT proteins.

3.2. Cnidarian SGLT proteins

A total of 23 cnidarian sequences were discovered as putativehomologs to SGLT proteins. While most sequences contain the 12 TMhelices indicative of the amino acid-polyamine- organocation super-family (APC), to which the SGLTs belong (Saier, 2000), three sequencesappear to be protein fragments (Datasheet A2). Only one cnidariansequence was identified as likely to encode for a protein that primarily

E. NRT2

C. AQP

B. SMIT

D. AMT

TM-score: 0.78

Sodium, C-score: 0.21

Galactose, C-score: 0.19

TM-score: 0.90 C-score: 0.46

TM-score: 0.84

Methylamine, C-score: 0.1

Ammonia, C-score: 0.09

TM-score: 0.73 C-score: 0.06

1 9A. GLUT

TM-score: 0.891 C-score: 0.11

Maltose

Glycerol

Monoacylglycerol

(caption on next page)

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

311

transports glucose (N. vectensis SLGT4) rather than myo-inositol. Thissequence clustered with two other cnidarian SGLT sequences that werepreviously identified in A. pallida, and this group was nested inside aclade of cnidarian SMIT proteins. Another group of SMIT proteins re-solved into a sister clade of the vertebrate SMIT clade. Additionally, onecnidarian sequence was identified as a choline transporter (CHT) withhigh similarity to the A. pallida CHT protein. The majority of cnidariansequences were identified as putative sodium myo-inositol transporters(SMIT) or sodium-coupled monocarboxylate transporters (SMCT)(Fig. 4). All of the identified SGLT homologs were predicted to be lo-calized to either the plasma membrane or internal membranes andcontained the sodium solute symporter domain (Datasheet A2).

All proteins in the three main clades (CHT, SMCT, and SGLT) arepart of the SLC5 gene family and function as sodium ion symporters.Recovered cnidarian proteins did not share high sequence similarity tothe SGLT1-5 reference proteins, rather they were closely related toSMCT or SMIT proteins (Fig. 4). The SMCT1 protein (encoded bySLC5A8) couples sodium transport with that of short-chain fatty acids,and is associated with tumor suppression in humans (Gupta et al., 2006;Miyauchi et al., 2004). Cnidarian SMCT proteins were primarily pre-dicted to be localized to internal membranes, with fewer in the plasmamembrane. Additionally, multiple SMCT sequences were found in twoof the symbiotic coral species, Fungia scutaria and P. australiensis, as wellas in the aposymbiotic anemone N. vectensis (Datasheet A2). Fatty acidsare a major component of the metabolites exchanged between partnersin the cnidarian-dinoflagellate symbiosis (Dunn et al., 2012; Hillyeret al., 2016; Imbs et al., 2014). Therefore, the role of SMCT in symbioticcnidarians may be to transport host-derived fatty acids to the sym-bionts, if isoforms of the protein can be expressed in the symbiosome.

The clade of SGLT family proteins has high bootstrap support, andthe tree shows SGLT and SMIT sequences clustering together withinvertebrate and invertebrate sub-clades, where one invertebrate clade is

sister to the vertebrate SMITs and another consists entirely of cnidarianproteins (Fig. 4). The SMIT proteins resolving into two different cladesmay be due to differently functioning SMIT isoforms, where thosebranching from the vertebrate clade may act more similarly to thevertebrate homologs while proteins in the cnidarian SMIT/SGLT groupmay have more derived functions and different ligand transportingcapabilities. Nine cnidarian proteins in total were identified as SMITs,which are closely related to SGLTs of the SLC5 gene family. Theseproteins function similarly to proton myo-inositol transporters (HMIT),primarily transporting myo-inositol (Bourgeois et al., 2005). SMIT1 canalso transport a variety of sugars, including glucose, although it does sowith low affinity (Wright, 2013), while SMIT2 can transport D-chiro-inositol (Coady et al., 2002). Myo-inositol is important for the growth ofmammalian cells and is a key component in phospholipid biosynthesisand eukaryotic signaling pathways (Holub, 1986). The identification ofboth HMIT and SMIT transporters in cnidarians suggests an un-characterized role for myo-inositol in cnidarian metabolism or cellstructure.

A SMIT protein sequence from N. vectensis (N. vectensis_5 SMIT) wasused for 3D structural analysis (Fig. 3). The predicted tertiary structureof the protein most closely resembles that of the K294A mutant of theVSGLT protein from Escherichia coli (TM score= 0.78). Overall residueconservation was low, with the most conserved residues lining theprotein pore (Fig. 3). The ligands were predicted to be sodium (C-score= 0.21) and galactose (C-score= 0.19), which only weakly sup-ports the idea that this protein functions as a sodium-coupled sugartransporter (Fig. 3). Proteins in the SGLT family are known to transportgalactose, some with a high affinity (Zhao and Keating, 2007); how-ever, experimental analysis is needed to confirm this function in cni-darians.

3.3. Metazoan AQP proteins

A total of 16 cnidarian sequences were identified as AQP homologs,all with at least five TM domains typical of AQPs (Datasheet A3). Incnidarians, the majority of the AQP homologs were predicted to belocalized to internal membranes (Datasheet A3). All sequences con-tained domains present in members of the major intrinsic protein familyand in AQP transporters, while some also had domains associated withspecific AQPs (Datasheet A3). Our phylogenetic analysis resulted in theemergence of two sister clades: (1) aquaglyceroporins; and (2) classicalaquaporins (Fig. 5). All sequences identified as aquaglyceroporinscontained the conserved aspartate (D) residue that signifies a largerprotein pore (Agre et al., 2002).

Aquaglyceroporin proteins all share a common ancestor, but theseproteins were shown to have diverged into vertebrate and invertebratelineages before the emergence of the different protein members. Thecnidarian aquaglyceroporins shared high sequence similarity withAQP3 or AQP9 and were labelled as such. However, since divergence ofvertebrate and invertebrate lineages occurred before diversification ofAQP isoforms within each taxonomic clade, functions of the AQP pro-tein members could be different between invertebrates and vertebrates.These sub-clades of exclusively cnidarian sequences are highly sup-ported and also include two previously identified AQP3-like sequencesfrom A. pallida. In mammals, aquaglyceroporins can transport a di-versity of solutes other than glycerol, including urea and water (Rojeket al., 2008). AQP3 primarily transports water and glycerol, whileAQP9 can also transport larger molecules, such as lactate, purines, andpyrimidines (Ishibashi et al., 2011). Experimental evidence is needed toconfirm which of these small solutes the cnidarian AQP proteinstransport.

The sub-clade of metazoan classical aquaporins contains cnidariansequences sharing a common ancestor with vertebrate and invertebrateAQP1, as well as vertebrate AQP4 and AQP0 (Fig. 5). While AQP1 andAQP4 are water-only channels, AQP0 has a variety of functions in otherorganisms (Agre et al., 2002). AQP0 is the lens major intrinsic protein,

Fig. 3. Protein structural analysis of selected sequences from each gene family. Predicted3D structures (left), conservation analysis (middle), and predicted ligands (right) of se-lected sequences from cnidarian GLUT (A), SMIT (B), and AQP (C) proteins andSymbiodinium AMT (D) and NRT2 (E) proteins. Predicted structures of each protein arecolored by structure (alpha helix= pink, coil=white, with yellow arrows showing se-quence direction) and aligned with the top matching structure from the Protein DataBank, depicted by a cyan mesh ribbon. Conservation scores are rendered onto the proteinmodels using a color gradient, from blue (low conservation, score= 1) to red (highconservation, score=9), with yellow regions indicating insufficient data for analysis.Ligand predictions show the ligand at the predicted binding sites with transparent proteinmodels colored by secondary structure (helices= red, coils= blue). (A, left) The cni-darian GLUT protein, as deduced from the amino acid sequence of N. vectensis_4 GLUT8aligned with the human GLUT1 protein (PDB ID=4pyp). (A, middle) The N. vectensisGLUT8 predicted structure colored by degree of residue conservation from all newlyidentified cnidarian GLUT sequences. (A, right) The N. vectensis GLUT8 protein showingpredicted ligand maltose (C-score= 0.11). (B, left) The SGLT protein as deduced from theN. vectensis_5 SMIT sequence aligned with the K294A mutant of VSGLT (PDB ID=2xq2).(B, middle) The N. vectensis SMIT protein colored by degree of residue conservation fromall newly identified cnidarian SGLT sequences. (B, right) The N. vectensis SMIT proteinshowing predicted ligands sodium (top, C-score=0.21) and galactose (bottom, C-score= 0.19). (C, left) The cnidarian AQP protein as deduced from the N. vectensis_4AQP3 sequence aligned with the E. coli glycerol facilitator GLPF (PDB ID=1fx8). (C,middle) The N. vectensis AQP3 protein colored by degree of residue conservation from allnewly identified cnidarian AQP sequences. (C, right) The N. vectensis AQP proteinshowing predicted ligand glycerol (C-score=0.46). (D, left) The Symbiodinium AMTprotein as deduced from the Symbiodinium C1_25 AMT sequence aligned with the S.cerevisiae methylammonium/ammonium permease Mep2 (PDB ID=5aexA). (D, middle)The Symbiodinium AMT protein colored by degree of residue conservation from all newlyidentified Symbiodinium AMT sequences. (D, right) The Symbiodinium AMT proteinshowing predicted ligands methylamine (top, C-score= 0.1) and ammonia (bottom, C-score= 0.09). (E, left) The Symbiodinium NRT protein as deduced from theSymbiodiniumA_24 NRT sequence aligned with the E. coli glycerol-3-phosphate transporterGlpT (PDB ID=1pw4). (E, middle) The Symbiodinium NRT2 protein with degree of re-sidue conservation from all newly identified Symbiodinium NRT2 sequences. (E, right) TheSymbiodinium NRT2 protein showing predicted ligand monoacylglycerol (C-score= 0.06). (For interpretation of the references to colour in this figure legend, thereader is referred to the web version of this article.)

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

312

Fig. 4. Phylogenetic tree of cnidarian SGLT sequences and closest homologs from other organisms. The tree was rooted by the bacterial sodium solute symporter (SSS) homolog fromAcanthamoeba castellanii. The number of amino acid substitutions is indicated by the scale bar, and bootstrap support for nodes> 50% is represented by shaded circles. Cnidariansequences are shown in bold text. Three main clades were resolved: (1) choline transporters (CHT), (2) sodium-coupled monocarboxylate transporters (SMCT), and (3) sodium/glucosecotransporters (SGLT and SMIT). The cluster of vertebrate SMIT proteins is collapsed to enhance tree clarity. The sequence used for structural analysis is marked by a yellow star (N.vectensis_5 SMIT). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

313

originally found in lens fiber cells of vertebrate eyes and later dis-covered to function in increasing the water permeability of animal cells(Gorin et al., 1984; Mulders et al., 1995). Its suspected structural role incell adhesion make AQP0 unique in the aquaporin family (Fotiadiset al., 2000). Cell adhesion proteins are important in the cnidarian-dinoflagellate symbiosis for cell-cell interactions, such as signaling be-tween partners (Reynolds et al., 2000); cnidarian AQP0 may function inthis capacity.

One AQP9 sequence from N. vectensis (N. vectensis_4 AQP3) was usedfor structural analysis of the cnidarian AQP protein (Fig. 3). The de-duced structure was very similar to that of the E. coli glycerol facilitator,GLPF (Fig. 3, TM score= 0.90). Additional hits from the Protein DataBank were protein structures of aquaglyceroporins from other species(Datasheet A3, Tab C). Furthermore, the predicted ligand (Fig. 3) wasglycerol (C-score= 0.46), and the most highly conserved residuesamong all cnidarian AQPs were those lining the protein pore (Fig. 3).Overall, the structural analysis supports the prediction that these pro-teins function as glycerol channels.

3.4. Symbiodinium AMT proteins

A total of 40 Symbiodinium sequences were putatively identified as

AMT proteins. Most of these sequences contained 10–12 TM domains,which is typical of AMT proteins (Thomas and Mullins, 2000; vonWiren et al., 2000), while five sequences were shorter, containing onlyseven to nine TM helices. The sequences were primarily identified asgeneral AMT proteins rather than specific members or isoforms andwere predicted to be localized to either the cell membrane or internalmembranes (Datasheet A4, Tab D). All sequences contained the domainfor ammonium transport (Datasheet A4, Tab D).

Phylogenetic analysis showed the divergence of five main AMTclades: (1) bacterial cyclases; (2) metazoan-like AMT proteins; (3)diaphoretickes AMT proteins; (4) alveolate AMT proteins; and (5)bacterial-like AMT proteins (Fig. 6). The clade of bacterial cyclases isnot well supported and includes the enzymes adenylate cylase andguanylate cyclase exclusively from bacterial species.

Several clades of AMT proteins were identified in the phylogeneticanalysis. The first clade shows AMT proteins diverging into three sub-clades, although not all are well resolved (Fig. 6). Clade 2 has robustsupport and shows metazoan AMT proteins emerging alongside acluster of exclusively Symbiodinium proteins. This clade includes four A.pallida sequences in the metazoan cluster and five of the Symbiodiniumsequences that contain the domain characteristic of AMT3. This findingindicates that horizontal gene transfer may have occurred between

Fig. 5. Phylogenetic tree of cnidarian AQP sequences and closest homologs from other organisms. The tree was rooted by the bacterial homolog from Desulfuromonas acetoxidans. Thenumber of amino acid substitutions is indicated by the scale bar, and bootstrap support for nodes> 50% is represented by shaded circles. Cnidarian sequences are shown in bold text. Theproteins resolved into two main clades: (1) aquaglyceroporins and (2) classical aquaporins. The sequence used for structural analysis is marked by a yellow star (N. vectensis_4 AQP3).*putative Fragaria chiloensis NIP contaminant. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

314

Fig. 6. Phylogenetic tree of Symbiodinium AMT sequences and closest homologs from other organisms. The tree was rooted by the bacterial homolog from a Spirochaeta sp. The number ofamino acid substitutions is indicated by the scale bar, and bootstrap support for nodes> 50% is represented by shaded circles. Symbiodinium sequences are shown in bold text. Theproteins resolved into five main clades: (1) bacterial cyclases, (2) metazoan-like ammonium transporters, (3) plant-like ammonium transporters, (4) chromalveolate ammoniumtransporters, and (5) bacterial-like ammonium transporters. The sequence used for structural analysis is marked by a yellow star (Symbiodinium C1_25 AMT). (For interpretation of thereferences to colour in this figure legend, the reader is referred to the web version of this article.)

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

315

Symbiodinium and the last common ancestor of this metazoan group.Evidence of horizontal gene transfer between symbiotic partners hasalso been discovered between Aiptasia and Symbiodinium by researcherswho assembled the Aiptasia genome (Baumgarten et al., 2015). Twenty-nine Aiptasia gene products were found to best align with Symbiodiniumproteins, rather than metazoan proteins, and clustered together in aphylogenetic analysis (Baumgarten et al., 2015).

Clade 3 is only weakly supported overall and includes proteins fromorganisms within the expansive diaphoretickes super-group (Adl et al.,2012). Nevertheless, a well-supported sub-clade shows a SymbiodiniumAMT sequence (SymbiodiniumC1_12) grouping with vascular plant andgreen algal proteins. This Symbiodinium sequence also had high se-quence similarity to the AMT1-3, which in plants has a high affinity forthe uptake of both methylammonium and ammonium (von Wiren et al.,2000). While clade 4 is also lacking adequate bootstrap support, itcontains a highly supported sub-clade that includes a large amount ofexpressed Symbiodinium C1 AMT proteins in relation to other alveolates(Fig. 6). While expression of multiple AMT proteins is observed foralmost all alveolate species in the analysis, Symbiodinium C1 was foundto express a total of 28 proteins, while other alveolate species have onlyone to four expressed AMT proteins. Symbiodinium C1 is known as ageneralist symbiont that occurs worldwide, forming associations with avariety of invertebrate hosts (LaJeunesse, 2005; Stat et al., 2009).Consequently, the upregulation of AMT proteins in this Symbiodiniumtype may have evolved as an adaptation to survive the varying nitrogenconditions experienced in different host cell environments.

The second lineage including Symbiodinium AMT proteins is a well-supported clade containing a mixture of unresolved sequences frombacteria, chlorophyta, and alveolata (Fig. 6, clade 5). Overall, ouranalysis of the Symbiodinium AMT proteins and their homologs hasrevealed a large diversity of these transporters within this dinoflagellatespecies, which may have arisen through multiple evolutionary events.

For structural analysis of the Symbiodinium AMT proteins, one se-quence from Symbiodinium clade C (SymbiodiniumC1_25) that fallswithin the alveolate cluster was used. The predicted 3D structure mostclosely resembles the structure of the Saccharomyces cerevisiae methy-lammonium permease, Mep2 (TM score= 0.84, Fig. 3). Residues liningthe inside of the protein channel are highly conserved (Fig. 3), sug-gesting their importance in transport function. Ligand predictions(Fig. 3) had very low confidence scores but favored methylamine (C-score= 0.1) and ammonia (C-score= 0.09). The information gainedfrom this structural analysis provides preliminary evidence that thesehighly conserved proteins serve as ammonium channels.

3.5. Symbiodinium NRT2 proteins

Only four sequences in total were identified as Symbiodinium NRT2homologs, all of which were from Symbiodinium clade A, although asequence from clade C and an unspecified Symbiodinium sequencewithin the NCBI database were also found to be related to NRT2. Thefour sequences had high similarity with one another (74–79%) andranged in length from 549 to 613 amino acids with 11–12 TM helices,typical of MFS transporters (Datasheet A5). Half of these NRT2 homo-logs were predicted to be localized to the plasma membrane and theother half to internal membranes, while all sequences contained thedomains for MFS and were related to high affinity nitrate transporter2.3 (Datasheet A5, Tab D).

Phylogenetic analysis of the NRT2 proteins showed divergence of:(1) eukaryotic; and (2) bacterial nitrate transporters (Fig. 7). Surpris-ingly, very few NRT homologs were recovered from the Symbiodiniumdatabases, and no other alveolate NRT2 homologs were identified(Fig. 7). NRT2 genes are highly regulated by environmental nitrateconcentrations. As part of the high-affinity uptake system, the genes areupregulated during nitrogen starvation (< 0.5mM) and repressedwhen nitrate levels are high (> 0.5mM) (Lezhneva et al., 2014;Okamoto et al., 2003). If the Symbiodinium NRT2 protein level is

regulated in the same way, then the gene would be repressed when thecells are grown in the commonly used f/2 culture medium, since itcontains 0.882mM nitrate (Guillard, 1975). However, as copies of theNRT2 genes were not discovered in the Symbiodinium genomes sur-veyed, identification of the protein may indicate that this gene has ei-ther been lost in some derived species of this genus but retained by themore ancestral Symbiodinium clade A, or that the proteins discoveredwere from contaminants in the clade A transcriptome.

The presence of nitrate transporters in only Symbiodinium clade Acould be due to the free-living nature of some dinoflagellate typeswithin this clade (Hirose et al., 2008; Reimer et al., 2010), allowing thecells to more readily utilize nitrate from seawater. Ammonium is ex-creted by the host, while nitrate is not, and ammonium has been pre-dicted to be the primary nitrogen source for symbiotic Symbiodiniumcells (Kopp et al., 2013; Pernice et al., 2014). Many Symbiodinium typesthat have evolved to be strictly symbiotic may have lost the need forhigh-affinity nitrate transport, while those with a partially or whollyfree-living lifestyle may have retained it.

While we had initially hypothesized that the identifiedSymbiodinium NRT2 homologs are nitrate transporters, the structuralanalysis indicates otherwise. The predicted 3D structure of the putativeSymbiodinium NRT2 sequence (Symbiodinium A-24 NRT) was most si-milar to that of the glycerol-3-phosphate transporter from E. coli (TMscore= 0.73, Fig. 3), rather than any nitrate transporters. Overallconservation was low, and one of the TM helices was only found in asingle sequence (insufficient data, Fig. 3). The ligand was predicted asmonoacylglycerol (Fig. 3) but with low confidence (C-score= 0.06).While the identified potential NRT2 homologs resemble integralmembrane transport proteins, they are very likely MFS transporterswith a different function.

4. Conclusions

This study identified sugar transport proteins from cnidarians andnitrogen transport proteins from their dinoflagellate symbionts.Homology searches identified previously undescribed cnidarian GLUT,SGLT, and AQP homologs and Symbiodinium AMT and NRT2 homologs.Phylogenetic and structural analyses demonstrated high conservation ofmost of the investigated protein families, suggesting that the char-acterized transport functions of the reference sequences are likely re-tained in the identified cnidarian and Symbiodinium sequences.

Considerably more cnidarian sequences were identified in the GLUTfamily than in the SGLT family, signifying that GLUT proteins could beinvolved in the transport of glucose between the symbiont and cni-darian host. Furthermore, potentially related transporters of the SLC5gene family known to transport other compounds, such as mono-carboxylates and myo-inositol, were identified in the cnidarians.Protein sequences for cnidarian aquaglyceroporins were also dis-covered, suggesting a potential mechanism for glycerol translocationinto cnidarian cells.

Many of the sugar transporters discussed here have predicted lo-calizations to internal membranes, which are defined in the MemPypepipeline as “the endoplasmic reticulum, the nuclear membranes, theGolgi apparatus, the vesicles, the vacuoles, the lysosomes, the peroxi-some, the microsomes, and the endosome” (Pierleoni et al., 2011).Since the symbiosome is derived from an early endosome (Fitt andTrench, 1983; Wakefield and Kempf, 2001), we presume that symbio-some-localized proteins would fall into the “internal membrane” cate-gory. Transporters predicted as localized to only internal membraneswere OCT and AQP9 (Datasheets A2–A3). Investigations of these pro-teins might identify them as being associated exclusively with thesymbiosome, which would be helpful in the establishment of specificmarker proteins for identification of symbiosome membranes.

Additionally, more Symbiodinium AMT sequences were identifiedthan NRT sequences, which is likely due to ammonium being the pri-mary nitrogen source for symbiotic dinoflagellates (D’Elia et al., 1983;

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

316

Grover et al., 2002). Based on our findings, we suggest the use of AMTtransporters to investigate the nutritional conditions experienced bysymbionts in hospite. Since expression of these proteins is known to beregulated by environmental ammonium concentrations, AMT trans-porter expression patterns by Symbiodinium cells in hospite could revealinformation about the nutrient conditions experienced within symbiotichost cells. It has previously been hypothesized that the host can reg-ulate intracellular symbiont populations by limiting their growth andnutrient supply (Falkowski et al., 1993; Jones and Yellowlees, 1997;Muscatine and Pool, 1979; Wooldridge, 2010). In plants and variousalgal species, AMT isoforms are transcriptionally regulated by en-vironmental conditions, in a manner similar to that for genes encodingNRT2 proteins, where high-affinity isoforms are expressed during ni-trogen starvation and low-affinity isoforms are expressed under amplenutrient availability (Gazzarrini et al., 1999; González-Ballester et al.,2004; Lezhneva et al., 2014). Therefore, verification of regulated ex-pression in Symbiodinium and analysis of AMT proteins during coloni-zation experiments may help to establish whether or not the host im-poses nitrogen limitations on the symbiont population.

The resources used to compile cnidarian and Symbiodinium data-bases in this study included mostly transcriptomic data. Molecularphylogenies conducted on proteins rather than DNA are more accurateat establishing relationships between lineages, since proteins are themolecules on which natural selection acts (Salemi and Vandamme(2003)). While transcriptomes contain highly useful information aboutthe proteins being transcribed during the times of sampling, they re-present only a snapshot of an organism’s genome and not a compre-hensive view of its total coding capacity. The results of this study can beinterpreted as putative identifications of symbiont and host transport

proteins, while the absence of any proteins in our datasets does notnecessarily indicate their absence from the organism. All new proteinidentifications described here are inferred from homology and requireverification through experimental analyses. While the results cannot beused to deduce the full range of cnidarian and Symbiodinium proteinsfrom each of these five transporter families, they do give us a previouslyunseen glimpse of a subset of transporter proteins that are likely used inthe cnidarian-dinoflagellate symbiosis. Using currently available ge-netic resources, we have taken the first steps towards uncovering me-chanisms used in the exchange of carbon and nitrogen between thepartner organisms in the cnidarian-Symbiodinium symbiosis. With theemergence of new genomic resources and proteomic-based experi-ments, we are likely to develop a more complete picture of transporterfunctions and mechanisms in symbiotic associations.

Acknowledgements

We thank Eli Meyer for the use of his Perl scripts and acknowledgethe Center for Genome Research and Biocomputing at OSU for pro-viding computational resources for this project. Financial support forthis research was provided by the Marsden Fund of the Royal Society ofNew Zealand, grant number 1202, to S.K.D., V.M.W., and A.R.G. Thefunding source had no involvement in study design; in the collection,analysis and interpretation of data; in the writing of the report; or in thedecision to submit the article for publication. We thank the library ofscience and medical illustrations (http://www.somersault1824.com/science-illustrations/), creative commons license CC BY-NC-SA 4.0(https://creativecommons.org/licenses/by-nc-sa/4.0/), for objects usedto create Fig. 1. We also thank two anonymous reviewers for their

Fig. 7. Phylogenetic tree of Symbiodinium NRT2sequences and closest homologs from other or-ganisms. The tree was rooted by the clade ofbacterial sequences. The number of amino acidsubstitutions is indicated by the scale bar, andbootstrap support for nodes> 50% is representedby shaded circles. Symbiodinium sequences areshown in bold text. Proteins resolved into twoclades: (1) eukaryotic and (2) bacterial nitratetransporters. The sequence used for proteinstructural analysis is marked with a yellow star(Symbiodinium A-24 NRT). (For interpretation ofthe references to colour in this figure legend, thereader is referred to the web version of this ar-ticle.)

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

317

insightful and helpful comments.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, in theonline version, at https://doi.org/10.1016/j.ympev.2017.12.007.

References

Adl, S.M., Simpson, A.G.B., Lane, C.E., Lukes, J., Bass, D., Bowser, S.S., Brown, M., Burki,F., Dunthorn, M., Hampl, V., Heiss, A., Hoppenrath, M., Lara, E., IeGall, L., Lynn,D.H., McManus, H., Mitchell, E.A.D., Mozley-Stanridge, Parfrey, L.W., Pawlowski, J.,Rueckert, S., Shadwick, L., Schoch, C., Smirnov, A., Spiegel, F.W., 2012. The revisedclassification of eukaryotes. J. Eukaryote Microbiol. 59, 429–493. http://dx.doi.org/10.1111/j.1550-7408.2012.00644.x.The.

Agre, P., King, L.S., Yasui, M., Guggino, W.B., Ottersen, O.P., Fujiyoshi, Y., Engel, A.,Nielsen, S., 2002. Aquaporin water channels–from atomic structure to clinical med-icine. J. Physiol. 542, 3–16. http://dx.doi.org/10.1113/jphysiol.2002.020818.

Allemand, D., Furla, P., Benazet-Tambutte, S., 1998. Mechanisms of carbon acquisitionfor endosymbiont photosynthesis in Anthozoa. Can. J. Bot. 76, 925–941.

Ashkenazy, H., Erez, E., Martz, E., Pupko, T., Ben-Tal, N., 2010. ConSurf 2010: calculatingevolutionary conservation in sequence and structure of proteins and nucleic acids.Nucleic Acids Res. 38, 529–533. http://dx.doi.org/10.1093/nar/gkq399.

Augustin, R., Riley, J., Moley, K.H., 2005. GLUT8 contains [DE]XXXL[LI] sorting motifand localizes to a late endomosal/lysosomal compartment. Traffic 6, 1196–1212.http://dx.doi.org/10.1111/j.1600-0854.2005.00354.x.

Baumgarten, S., Simakov, O., Esherick, L.Y., Liew, Y.J., Lehnert, E.M., Michell, C.T., Li, Y.,Hambleton, E.A., Guse, A., Oates, M.E., Gough, J., Weis, V.M., Aranda, M., Pringle,J.R., Voolstra, C.R., 2015. The genome of Aiptasia, a sea anemone model for coralsymbiosis. Proc. Natl. Acad. Sci. 112, 201513318. http://dx.doi.org/10.1073/pnas.1513318112.

Berg, J., Tymoczko, J., Stryer, L., 2002. Complex carbohydrates are formed by linkage ofmonosaccharides. In: Freeman, W. (Ed.), Biochemistry. New York.

Bourgeois, F., Coady, M.J., Lapointe, J.-Y., 2005. Determination of transport stoichio-metry for two cation-coupled myo-inositol cotransporters: SMIT2 and HMIT. J.Physiol. 563, 333–343. http://dx.doi.org/10.1113/jphysiol.2004.076679.

Brown, T., Coffroth, M.A., Lasker, H., Brazeau, D., 2011. Differential gene expression of aputative ammonium transporter in the algal symbiont, Symbiodinium. Submitt. toEMBL/GenBank/DDBJ databases.

Bucka-Lassen, K., Caprani, O., Hein, J., 1999. Combining many multiple alignments inone improved alignment. Bioinformatics 15, 122–130. http://dx.doi.org/10.1093/bioinformatics/15.2.122.

Burriesci, M.S., Raab, T.K., Pringle, J.R., 2012. Evidence that glucose is the majortransferred metabolite in dinoflagellate-cnidarian symbiosis. J. Exp. Biol. 215,3467–3477. http://dx.doi.org/10.1242/jeb.070946.

Ciarimboli, G., 2008. Organic cation transporters. Xenobiotica 38, 936–971. http://dx.doi.org/10.1080/00498250701882482.

Coady, M.J., Wallendorff, B., Gagnon, D.G., Lapointe, J.Y., 2002. Identification of a novelNa+/myo-inositol cotransporter. J. Biol. Chem. 277, 35219–35224. http://dx.doi.org/10.1074/jbc.M204321200.

Coffroth, M., Santos, S., 2005. Genetic diversity of symbiotic dinoflagellates in the genusSymbiodinium. Protist 156, 19–34. http://dx.doi.org/10.1016/j.protis.2005.02.004.

D’Elia, C., Domotor, S., Webb, K., 1983. Nutrient uptake kinetics of freshly isolatedzooxanthellae. Mar. Biol. 167, 157–167.

Davy, S.K., Allemand, D., Weis, V.M., 2012. Cell biology of cnidarian-dinoflagellatesymbiosis. Microbiol. Mol. Biol. Rev. 76, 229–261. http://dx.doi.org/10.1128/MMBR.05014-11.

Doege, H., Bocianski, A., Scheepers, A., Axer, H., Eckel, J., Joost, H.G., Schurmann, A.,2001. Characterization of human glucose transporter (GLUT) 11 (encoded bySLC2A11), a novel sugar-transport facilitator specifically expressed in heart andskeletal muscle. Biochem. J. 449, 443–449.

Doege, H., Schürmann, A., Bahrenberg, G., Brauers, A., Joost, H.G., 2000. GLUT8, a novelmember of the sugar transport facilitator family with glucose transport activity. J.Biol. Chem. 275, 16275–16280. http://dx.doi.org/10.1074/jbc.275.21.16275.

Dunn, S.R., Thomas, M.C., Nette, G.W., Dove, S.G., 2012. A lipidomic approach to un-derstanding free fatty acid lipogenesis derived from dissolved inorganic carbonwithin cnidarian-dinoflagellate symbiosis. PLoS One 7. http://dx.doi.org/10.1371/journal.pone.0046801.

Eisen, J.A., 1998. Phylogenomics: improving functional predictions for uncharacterizedgenes by evolutionary analysis. Genome Res. 8, 163–167. http://dx.doi.org/10.1101/gr.8.3.163.

Eisen, J.A., Kaiser, D., Myers, R.M., 1997. Gastrogenomic delights: a movable feast. Nat.Med. 3, 1076–1078. http://dx.doi.org/10.1038/nm1097-1076.

Eraly, S.A., Monte, J.C., Nigam, S.K., 2004. Novel slc22 transporter homologs in fly,worm, and human clarify the phylogeny of organic anion and cation transporters.Physiol. Genom. 18, 12–24. http://dx.doi.org/10.1152/physiolgenomics.00014.2004.

Escher, S.A., Rasmuson-Lestander, A., 1999. The Drosophila glucose transporter gene:cDNA sequence, phylogenetic comparisons, analysis of functional sites and secondarystructures. Hereditas 130, 95–103.

Falkowski, P.G.P., Dubinsky, Z., Muscatine, L., Mccloskey, L., 1993. Population control insymbiotic corals. Bioscience 43, 606–611.

Feng, Y., Williams, B.G., Koumanov, F., Wolstenholme, A.J., Holman, G.D., 2013. FGT-1

is the major glucose transporter in C. elegans and is central to aging pathways.Biochem. J. 456, 219–229. http://dx.doi.org/10.1042/BJ20131101.

Fitt, W.K., Trench, R.K., 1983. Endocytosis of the symbiotic dinoflagellate Symbiodiniummicroadriaticum Freudenthal by endodermal cells of the scyphistomae of Cassiopeiaxamachana and resistance of the algae to host digestion. J. Cell Sci. 64, 195–212.

Forde, B.G., 2000. Nitrate transporters in plants: structure, function and regulation.Biochim. Biophys. Acta 1465, 219–235.

Fotiadis, D., Hasler, L., Müller, D.J., Stahlberg, H., Kistler, J., Engel, A., 2000. Surfacetongue-and-groove contours on lens MIP facilitate cell-to-cell adherence. J. Mol. Biol.300, 779–789. http://dx.doi.org/10.1006/jmbi.2000.3920.

Gazzarrini, S., Lejay, L., Gojon, A., Ninnemann, O., Frommer, W.B., von Wirén, N., 1999.Three functional transporters for constitutive, diurnally regulated, and starvation-induced uptake of ammonium into Arabidopsis roots. Plant Cell 11, 937–948. http://dx.doi.org/10.1105/tpc.11.5.937.

González-Ballester, D., Camargo, A., Fernández, E., 2004. Ammonium transporter genesin Chlamydomonas: the nitrate-specific regulatory gene Nit2 is involved in Amt1; 1expression. Plant Mol. Biol. 56, 863–878. http://dx.doi.org/10.1007/s11103-004-5292-7.

Gordon, B.R., Leggat, W., 2010. Symbiodinium-invertebrate symbioses and the role ofmetabolomics. Mar. Drugs 8, 2546–2568. http://dx.doi.org/10.3390/md8102546.

Gorin, M.B., Yancey, S.B., Cline, J., Revel, J.P., Horwitz, J., 1984. The major intrinsicprotein (MIP) of the bovine lens fiber membrane: characterization and structurebased on cDNA cloning. Cell 39, 49–59. http://dx.doi.org/10.1016/0092-8674(84)90190-9.

Grover, R., Maguer, J.-F., Reynaud-Vaganay, S., Ferrier-Pagès, C., 2002. Uptake of am-monium by the scleractinian coral Stylophora pistillata: effect of feeding, light, andammonium concentrations. Limnol. Oceanogr. 47, 782–790. http://dx.doi.org/10.4319/lo.2002.47.3.0782.

Guillard, R.L., 1975. Culture of phytoplankton for feeding marine invertebrates. In:Smith, W.L., Chanley, M.H. (Eds.), Culture of Marine Invertebrate Animals. SpringerUS, Boston, MA, pp. 29–60. http://www.10.1007/978-1-4615-8714-9_3.

Gupta, N., Martin, P.M., Prasad, P.D., Ganapathy, V., 2006. SLC5A8 (SMCT1)-mediatedtransport of butyrate forms the basis for the tumor suppressive function of thetransporter. Life Sci. 78, 2419–2425. http://dx.doi.org/10.1016/j.lfs.2005.10.028.

Hara-Chikuma, M., Verkman, A.S., 2006. Physiological roles of glycerol-transportingaquaporins: the aquaglyceroporins. Cell. Mol. Life Sci. 63, 1386–1392. http://dx.doi.org/10.1007/s00018-006-6028-4.

Hillyer, K.E., Tumanov, S., Villas-Bôas, S., Davy, S.K., 2016. Metabolite profiling ofsymbiont and host during thermal stress and bleaching in a model cnidarian-dino-flagellate symbiosis. J. Exp. Biol. 219, 516–527. http://dx.doi.org/10.1242/jeb.128660.

Hirose, M., Reimer, J.D., Hidaka, M., Suda, S., 2008. Phylogenetic analyses of potentiallyfree-living Symbiodinium spp. isolated from coral reef sand in Okinawa, Japan. Mar.Biol. 155, 105–112. http://dx.doi.org/10.1007/s00227-008-1011-2.

Holub, B.J., 1986. Metabolism and function of myo-inositol and inositol phospholipids.Annu. Rev. Nutr. 32, 59–74.

Houlbrèque, F., Tambutté, E., Richard, C., Ferrier-Pagès, C., 2004. Importance of a micro-diet for scleractinian corals. Mar. Ecol. Prog. Ser. 282, 151–160. http://dx.doi.org/10.3354/meps282151.

Imbs, A.B., Yakovleva, I.M., Dautova, T.N., Bui, L.H., Jones, P., 2014. Diversity of fattyacid composition of symbiotic dinoflagellates in corals: evidence for the transfer ofhost PUFAs to the symbionts. Phytochemistry 101, 76–82. http://dx.doi.org/10.1016/j.phytochem.2014.02.012.

Ishibashi, K., Kondo, S., Hara, S., Morishita, Y., 2011. The evolutionary aspects ofaquaporin family. Am. J. Physiol. Regul. Integr. Comp. Physiol. 300, R566–R576.http://dx.doi.org/10.1152/ajpregu.90464.2008.

Ishikura, M., Adachi, K., Maruyama, T., 1999. Zooxanthellae release glucose in the tissueof a giant clam, Tridacna crocea. Mar. Biol. 133, 665–673. http://dx.doi.org/10.1007/s002270050507.

Jones, R., Yellowlees, D., 1997. Regulation and control of intracellular algae (=zoox-anthellae) in hard corals. Proc. R. Soc. B Biol. Sci. 352, 457–468.

Joost, H.-G., Thorens, B., 2001. The extended GLUT-family of sugar/polyol transporterfacilitators: nomenclature, sequence characteristics, and potential function of itsnovel members (Review). Mol. Membr. Biol. 18, 247–256. http://dx.doi.org/10.1080/09687680110090.

Joost, H., Bell, G.I., Best, J.D., Birnbaum, M.J., Charron, M.J., Chen, Y.T., Doege, H.,James, D.E., Lodish, H.F., Moley, K.H., Moley, J.F., Mueckler, M., Rogers, S., Schu, A.,Seino, S., Thorens, B., 2002. Nomenclature of the GLUT/SLC2A family of sugar/polyol transport facilitators. Am. J. Physiol. Endocrinol. Metab. 282, E974–E976.

Katoh, K., Standley, D.M., 2013. MAFFT multiple sequence alignment software version 7:improvements in performance and usability. Mol. Biol. Evol. 30, 772–780. http://dx.doi.org/10.1093/molbev/mst010.

Kopp, C., Domart-Coulon, I., Escrig, S., Humbel, B.M., Hignette, M., Meibom, A., 2015.Subcellular investigation of photosynthesis-driven carbon and nitrogen assimilationand utilization in the symbiotic reef coral Pocillopora damicornis. MBio 6, 1–9. http://dx.doi.org/10.1128/mBio. 02299-14.Editor.

Kopp, C., Pernice, M., Domart-coulon, I., Djediat, C., Spangenberg, J.E., Alexander,D.T.L., Hignette, M., Meziane, T., 2013. Highly dynamic cellular-level response ofsymbiotic coral to a sudden increase in environmental nitrogen. Am. Soc. Microbiol.4, 1–9. http://dx.doi.org/10.1128/mBio. 00052-13.Invited.

Krueger, T., Fisher, P.L., Becker, S., Pontasch, S., Dove, S., Hoegh-Guldberg, O., Leggat,W., Davy, S.K., 2015. Transcriptomic characterization of the enzymatic antioxidantsFeSOD, MnSOD, APX and KatG in the dinoflagellate genus Symbiodinium. BMC Evol.Biol. 15, 1–20. http://dx.doi.org/10.1186/s12862-015-0326-0.

LaJeunesse, T.C., 2005. “Species” radiations of symbiotic dinoflagellates in the Atlanticand Indo-Pacific since the Miocene-Pliocene transition. Mol. Biol. Evol. 22, 570–581.

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

318

http://dx.doi.org/10.1093/molbev/msi042.Lajeunesse, T.C., Parkinson, J.E., Reimer, J.D., 2012. A genetics-based description of

Symbiodinium minutum sp. nov. and S. psygmophilum sp. nov. (Dinophyceae), twodinoflagellates symbiotic with cnidaria. J. Phycol. 48, 1380–1391. http://dx.doi.org/10.1111/j.1529-8817.2012.01217.x.

Larsson, A., 2014. AliView: a fast and lightweight alignment viewer and editor for largedatasets. Bioinformatics 30, 3276–3278. http://dx.doi.org/10.1093/bioinformatics/btu531.

Lehnert, E.M., Mouchka, M.E., Burriesci, M.S., Gallo, N.D., Schwarz, J.A., Pringle, J.R.,2014. Extensive differences in gene expression between symbiotic and aposymbioticcnidarians. G3 (Bethesda) 4, 277–295. http://dx.doi.org/10.1534/g3.113.009084.

Lewis, D., Smith, D., 1971. The autotrophic nutrition of symbiotic marine coelenterateswith special reference to hermatypic corals. I. Movement of photosynthetic productsbetween the symbionts. Proc. R. Soc. London B. Biol. Sci. 178, 111–129.

Lezhneva, L., Kiba, T., Feria-Bourrellier, A.-B., Lafouge, F., Boutet-Mercey, S., Zoufan, P.,Sakakibara, H., Daniel-Vedele, F., Krapp, A., 2014. The Arabidopsis nitrate transporterNRT2.5 plays a role in nitrate acquisition and remobilization in nitrogen-starvedplants. Plant J. 230–241. http://dx.doi.org/10.1111/tpj.12626.

Lin, S., Cheng, S., Song, B., Zhong, X., Lin, X., Li, W., Li, L., Zhang, Y., Zhang, H., Ji, Z.,Cai, M., Zhuang, Y., Shi, X., Lin, L., Wang, L., Wang, Z., Liu, X., Yu, S., Zeng, P., Hao,H., Zou, Q., Chen, C., Li, Y., Wang, Y., Xu, C., Meng, S., Xu, X., Wang, J., Yang, H.,Campbell, D.A., Sturm, N.R., Dagenais-Bellefeuille, S., Morse, D., 2015. TheSymbiodinium kawagutii genome illuminates dinoflagellate gene expression and coralsymbiosis. Science 350, 691–694. http://dx.doi.org/10.1126/science.aad0408.

Lodish, H.F., Berk, A., Zipursky, S., 2000. Overview of membrane transport proteins. In:Freeman, W.H. (Ed.), Molecular Cell Biology. New York.

Marini, A.M., Boeckstaens, M., Benjelloun, F., Chérif-Zahar, B., André, B., 2006.Structural involvement in substrate recognition of an essential aspartate residueconserved in Mep/Amt and Rh-type ammonium transporters. Curr. Genet. 49,364–374. http://dx.doi.org/10.1007/s00294-006-0062-5.

Mayfield, A.B., Hsiao, Y., Fan, T.Y., Chen, C.-S., 2010. Investigating the impacts of epi-sodic upwelling on the physiology of the common Indo-Pacific reef coral Seriatoporahystrix. Submitted to EMBL/GenBank/DDBJ databases.

Mayrose, I., Graur, D., Ben-Tal, N., Pupko, T., 2004. Comparison of site-specific rate-inference methods for protein sequences: empirical Bayesian methods are superior.Mol. Biol. Evol. 21, 1781–1791. http://dx.doi.org/10.1093/molbev/msh194.

Miyauchi, S., Gopal, E., Fei, Y.J., Ganapathy, V., 2004. Functional identification ofSLC5A8, a tumor suppressor down-regulated in colon cancer, as a Na+-coupledtransporter for short-chain fatty acids. J. Biol. Chem. 279, 13293–13296. http://dx.doi.org/10.1074/jbc.C400059200.

Mulders, S.M., Preston, G.M., Deen, P.M.T., Guggino, W.B., van Os, C.H., Agre, P., 1995.Water channel properties of major intrinsic protein of lens. J. Biol. Chem. 270,9010–9016.

Muller-Parker, G., Mccloskey, L.R., Hoegh-Guldberg, O., McAuley, P.J., 1994. Effect ofammonium enrichment on animal and algal biomass of the coral Pocillopora dami-cornis. Pacific Sci. 48, 273–283.

Muscatine, L., 1967. Glycerol excretion by symbiotic algae from corals and Tridacna andits control by the host. Science 156, 516–519.

Muscatine, L., Mccloskey, L.R., Marian, R.E., 1981. Estimating the daily contribution ofcarbon from zooxanthellae to coral animal respiration. Limonol. Oceanogr. 26,601–611.

Muscatine, L., Pool, R.R., 1979. Regulation of number of intracellular algae. Proc. R. Soc.B Biol. Sci. 204, 131–139.

O’Neil, J.M., Capone, D.G., 2008. Nitrogen in the Marine Environment. Elsevier http://www.10.1016/B978-0-12-372522-6.00021-9.

Oakley, C.A., Ameismeier, M.F., Peng, L., Weis, V.M., Grossman, A.R., Davy, S.K., 2016.Symbiosis induces widespread changes in the proteome of the model cnidarianAiptasia. Cell. Microbiol. 18, 1009–1023. http://dx.doi.org/10.1111/cmi.12564.

Okamoto, M., Vidmar, J.J., Glass, A.D.M., 2003. Regulation of NRT1 and NRT2 genefamilies of Arabidopsis thaliana: responses to nitrate provision. Plant Cell Physiol. 44,304–317. http://dx.doi.org/10.1093/pcp/pcg036.

Peng, S.-E., Wang, Y.-B., Wang, L.-H., Chen, W.-N.U., Lu, C.-Y., Fang, L.-S., Chen, C.-S.,2010. Proteomic analysis of symbiosome membranes in Cnidaria-dinoflagellate en-dosymbiosis. Proteomics 10, 1002–1016. http://dx.doi.org/10.1002/pmic.200900595.

Pernice, M., Dunn, S.R., Tonk, L., Dove, S., Domart-Coulon, I., Hoppe, P., Schintlmeister,A., Wagner, M., Meibom, A., 2014. A nanoscale secondary ion mass spectrometrystudy of dinoflagellate functional diversity in reef-building corals. Environ.Microbiol. 1–29. http://dx.doi.org/10.1111/1462-2920.12518.

Pernice, M., Meibom, A., Van Den Heuvel, A., Kopp, C., Domart-Coulon, I., Hoegh-Guldberg, O., Dove, S., 2012. A single-cell view of ammonium assimilation in coral-dinoflagellate symbiosis. ISME J. 6, 1314–1324. http://dx.doi.org/10.1038/ismej.2011.196.

Pierleoni, A., Indio, V., Savojardo, C., Fariselli, P., Martelli, P.L., Casadio, R., 2011.MemPype: a pipeline for the annotation of eukaryotic membrane proteins. NucleicAcids Res. 39, 375–380. http://dx.doi.org/10.1093/nar/gkr282.

Pochon, X., Gates, R., 2010. A new Symbiodinium clade (Dinophyceae) from soritid for-aminifera in Hawai’i. Mol. Phylogenet. Evol. 56, 492–497. http://dx.doi.org/10.1016/j.ympev.2010.03.040.

Porollo, A., Meller, J., 2007. Versatile annotation and publication quality visualization ofprotein complexes using POLYVIEW-3D. BMC Bioinformatics 8, 316. http://dx.doi.org/10.1186/1471-2105-8-316.

Quevillon, E., Silventoinen, V., Pillai, S., Harte, N., Mulder, N., Apweiler, R., Lopez, R.,2005. InterProScan: protein domains identifier. Nucleic Acids Res. 33, 116–120.http://dx.doi.org/10.1093/nar/gki442.

Rands, M., Loughman, B., Douglas, A.E., 1993. The symbiotic interface in an alga-

invertebrate symbiosis. Proc. R. Soc. B Biol. Sci. 253, 161–165.Reddy, V.S., Shlykov, M.A., Castillo, R., Sun, E.I., Saier, M.H., 2012. The major facilitator

superfamily (MFS) revisited. FEBS J. 279, 2022–2035. http://dx.doi.org/10.1111/j.1742-4658.2012.08588.x.

Reimer, J.D., Shah, M.M.R., Sinniger, F., Yanagi, K., Suda, S., 2010. Preliminary analysesof cultured Symbiodinium isolated from sand in the oceanic Ogasawara Islands, Japan.Mar. Biodivers. 40, 237–247. http://dx.doi.org/10.1007/s12526-010-0044-1.

Reynolds, W.S., Schwarz, J.A., Weis, V.M., 2000. Symbiosis-enhanced gene expression incnidarian-algal associations: cloning and characterization of a cDNA, sym32, en-coding a possible cell adhesion protein. Comp. Biochem. Physiol. A. Mol. Integr.Physiol. 126, 33–44. http://dx.doi.org/10.1016/S0742-8413(00)00099-2.

Rojek, A., Praetorius, J., Frøkiaer, J., Nielsen, S., Fenton, R.A., 2008. A current view of themammalian aquaglyceroporins. Annu. Rev. Physiol. 70, 301–327. http://dx.doi.org/10.1146/annurev.physiol.70.113006.100452.

Roy, A., Kucukural, A., Zhang, Y., 2010. I-TASSER: a unified platform for automatedprotein structure and function prediction. Nat. Protoc. 5, 725–738. http://dx.doi.org/10.1038/nprot.2010.5.

Saier, M.H., 2000. A functional-phylogenetic classification system for transmembranesolute transporters. Microbiol. Mol. Biol. Rev. 64, 354–411. http://dx.doi.org/10.1128/MMBR.64.2.354-411.2000.

Saier, M.H., Eng, B.H., Fard, S., Garg, J., Haggerty, D.A., Hutchinson, W.J., Jack, D.L., Lai,E.C., Liu, H.J., Nusinew, D.P., Omar, A.M., 1999. Phylogenetic characterization ofnovel transport protein families revealed by genome analysis. Biochim. Biophys.Acta, Rev. Biomembr. 1422, 1–56.

Salemi, M., Vandamme, A.-M., 2003. The Phylogenetic Handbook: A Practical Guide toDNA and Protein Phylogeny. Cambridge University Press.

Sampayo, E.M., Dove, S., LaJeunesse, T.C., 2009. Cohesive molecular genetic data de-lineate species diversity in the dinoflagellate genus Symbiodinium. Mol. Ecol. 18,500–519. http://dx.doi.org/10.1111/j.1365-294X.2008.04037.x.

Santos, S.R., Taylor, D.J., Kinzie, R.A., Hidaka, M., Sakai, K., Coffroth, M.A., 2002.Molecular phylogeny of symbiotic dinoflagellates inferred from partial chloroplastlarge subunit (23S)-rDNA sequences. Mol. Phylogenet. Evol. 23, 97–111. http://dx.doi.org/10.1016/S1055-7903(02)00010-6.

Scheepers, A., Joost, H.-G., Schürmann, A., 2015. The glucose transporter families SGLTand GLUT: molecular basis of normal and aberrant function. JPEN. J. Parenter.Enteral Nutr. 28, 364–371. http://dx.doi.org/10.1177/0148607104028005364.

Schmitz, K., Kremer, B., 1977. Carbon fixation and analysis of assimilates in a coral-dinoflagellate symbiosis. Mar. Biol. 313, 305–313.

Shinzato, C., Inoue, M., Kusakabe, M., 2014. A snapshot of a coral “holobiont”: a tran-scriptome assembly of the scleractinian coral, Porites, captures a wide variety of genesfrom both the host and symbiotic zooxanthellae. PLoS One 9. http://dx.doi.org/10.1371/journal.pone.0085182.

Sjölander, K., 2004. Phylogenomic inference of protein molecular function: advances andchallenges. Bioinformatics 20, 170–179. http://dx.doi.org/10.1093/bioinformatics/bth021.

Stamatakis, A., 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysisof large phylogenies. Bioinformatics 30, 1312–1313. http://dx.doi.org/10.1093/bioinformatics/btu033.

Stat, M., Pochon, X., Cowie, R.O.M., Gates, R.D., 2009. Specificity in communities ofSymbiodinium in corals from Johnston Atoll. Mar. Ecol. Prog. Ser. 386, 83–96. http://dx.doi.org/10.3354/meps08080.

Suescún-Bolívar, L.P., Iglesias-Prieto, R., Thomé, P.E., 2012. Induction of glycerolsynthesis and release in cultured Symbiodinium. PLoS One 7, e47182. http://dx.doi.org/10.1371/journal.pone.0047182.

Takabayashi, M., Santos, S.R., Cook, C.B., 2004. Mitochondrial DNA phylogeny of thesymbiotic dinoflagellates (Symbiodinium, Dinophyta). J. Phycol. 40, 160–164. http://dx.doi.org/10.1046/j.1529-8817.2004.03097.x.

Tanaka, Y., Miyajima, T., Koike, I., Hayashibara, T., Ogawa, H., 2006. Translocation andconservation of organic nitrogen within the coral-zooxanthella symbiotic system ofAcropora pulchra, as demonstrated by dual isotope-labeling techniques. J. Exp. Mar.Biol. Ecol. 336, 110–119. http://dx.doi.org/10.1016/j.jembe.2006.04.011.

Telford, M.J., 2007. Phylogenomics. Curr. Biol. 17, 945–946.Thomas, G.H., Mullins, J.G.L., 2000. Membrane topology of the Mep/Amt family of

ammonium transporters. Mol. Microbiol. 37, 331–344.Trueman, L.J., Richardson, A., Forde, B.G., 1996. Molecular cloning of higher plant

homologues of the high-affinity nitrate transporters of Chlamydomonas reinhardtii andAspergillus nidulans. Gene 175, 223–231.

Uldry, M., Thorens, B., 2004. The SLC2 family of facilitated hexose and polyol trans-porters. Pflugers Arch. Eur. J. Physiol. 447, 480–489. http://dx.doi.org/10.1007/s00424-003-1085-0.

Van Den Heuvel, A., 2012. The Effects of Exogenous Nitrogen on the Coral/DinoflagellateSymbiosis: Effect of Nitrogen on Staghorn Coral. University of Queensland.

Von Wiren, N., Gazzarrini, S., Gojon, A., Frommer, W.B., 2000. The molecular physiologyof ammonium uptake and retrieval. Curr. Opin. Plant Biol. 3, 254–261. http://dx.doi.org/10.1016/S1369-5266(00)80074-6.

Voolstra, C.R., Sunagawa, S., Matz, M.V., Bayer, T., Aranda, M., Buschiazzo, E., Desalvo,M.K., Lindquist, E., Szmant, A.M., Coffroth, M.A., Medina, M., 2011. Rapid evolutionof coral proteins responsible for interaction with the environment. PLoS One 6,e20392. http://dx.doi.org/10.1371/journal.pone.0020392.

Wakefield, T.S., Farmer, M., Kempf, S.C., 2000. Revised description of the fine structureof in situ “Zooxanthellae” genus Symbiodinium. Biol. Bull. 199, 76–84.

Wakefield, T.S., Kempf, S.C., 2001. Development of host- and symbiont-specific mono-clonal antibodies and confirmation of the origin of the symbiosome membrane in acnidarian-dinoflagellate symbiosis. Biol. Bull. 200, 127–143.

Wang, J., Douglas, A., 1999. Essential amino acid synthesis and nitrogen recycling in analga–invertebrate symbiosis. Mar. Biol. 135, 219–222.

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

319

Wang, J., Douglas, A.E., 1998. Nitrogen recycling or nitrogen conservation in an alga-invertebrate symbiosis? J. Exp. Biol. 201, 2445–2453.

Whitehead, L.F., Douglas, A.E., 2003. Metabolite comparisons and the identity of nu-trients translocated from symbiotic algae to an animal host. J. Exp. Biol. 206,3149–3157. http://dx.doi.org/10.1242/jeb.00539.

Wilson-O’Brien, A.L., Patron, N., Rogers, S., 2010. Evolutionary ancestry and novelfunctions of the mammalian glucose transporter (GLUT) family. BMC Evol. Biol. 10,152. http://dx.doi.org/10.1186/1471-2148-10-152.

Wood, I.S., Trayhurn, P., 2003. Glucose transporters (GLUT and SGLT): Expanded familiesof sugar transport proteins. Br. J. Nutr. 89, 3–9. http://dx.doi.org/10.1079/BJN2002763.

Wooldridge, S.A., 2010. Is the coral-algae symbiosis really “mutually beneficial” for thepartners? Bioessays 32, 615–625. http://dx.doi.org/10.1002/bies.200900182.

Wright, E.M., 2013. Glucose transport families SLC5 and SLC50. Mol. Aspects Med. 34,183–196. http://dx.doi.org/10.1016/j.mam.2012.11.002.

Wright, E.M., Turk, E., 2004. The sodium/glucose cotransport family SLC5. Pflugers Arch.Eur. J. Physiol. 447, 510–518. http://dx.doi.org/10.1007/s00424-003-1063-6.

Yang, J., Roy, A., Zhang, Y., 2013. Protein-ligand binding site recognition using com-plementary binding-specific substructure comparison and sequence profile align-ment. Bioinformatics 29, 2588–2595. http://dx.doi.org/10.1093/bioinformatics/btt447.

Zhang, Y., 2008. I-TASSER server for protein 3D structure prediction. BMC Bioinformatics9, 40. http://dx.doi.org/10.1186/1471-2105-9-40.

Zhao, F.-Q., Keating, A.F., 2007. Functional properties and genomics of glucose trans-porters. Curr. Genom. 8, 113–128. http://dx.doi.org/10.2174/138920207780368187.

Zmasek, C.M., Eddy, S.R., 2001. ATV: display and manipulation of annotated phyloge-netic trees. Bioinformatics 17, 383–384. http://dx.doi.org/10.1093/bioinformatics/17.4.383.

A.E. Sproles et al. Molecular Phylogenetics and Evolution 120 (2018) 307–320

320