Evolution of the Rho guanine nucleotide exchange factors Kalirin and Trio and their gene expression in Xenopus development

Gene Expression Patterns 32 (2019) 18–27

Contents lists available at ScienceDirect

Gene Expression Patterns journal homepage: www.elsevier.com/locate/gep

Evolution of the Rho guanine nucleotide exchange factors Kalirin and Trio and their gene expression in Xenopus development

T

Marie-Claire Kratzera,c, Laura Englanda, David Apelb,c, Monika Hasselb,c, Annette Borchersa,c,∗ a

Philipps‐Universität Marburg, Faculty of Biology, Molecular Embryology, Marburg, Germany Philipps‐Universität Marburg, Faculty of Biology, Morphology and Evolution of Invertebrates, Marburg, Germany c DFG Research Training Group, Membrane Plasticity in Tissue Development and Remodeling, GRK 2213, Philipps-Universität Marburg, Marburg, Germany b

A R T I C LE I N FO

A B S T R A C T

Keywords: Rho guanine nucleotide exchange factors Trio Kalirin Xenopus development Neural crest Placodes

Guanine nucleotide exchange factors (GEFs) activate Rho GTPases by accelerating their GDP/GTP exchange. Trio and its paralog Kalirin (Kalrn) are unique members of the Rho-GEFs that harbor three catalytic domains: two functional GEF domains and a serine/threonine kinase domain. The N-terminal GEF domain activates Rac1 and RhoG GTPases, while the C-terminal GEF domain acts specifically on RhoA. Trio and Kalrn have an evolutionary conserved function in morphogenetic processes including neuronal development. De novo mutations in TRIO have lately been identified in patients with intellectual disability, suggesting that this protein family plays an important role in development and disease. Phylogenetic and domain analysis revealed that a Kalrn/Trio ancestor originated in Prebilateria and duplicated in Urbilateria to yield Kalrn and Trio. Only few taxa outside the vertebrates retained both of these highly conserved proteins. To obtain first insights into their redundant or distinct functions in a vertebrate model system, we show for the first time a detailed comparative analysis of trio and kalrn expression in Xenopus laevis development. The mRNAs are maternally transcribed and expression increases starting with neurula stages. Trio and kalrn are detected in mesoderm/somites and different neuronal populations in the neural plate/tube and later also in the brain. However, only trio is expressed in migrating neural crest cells, while kalrn expression is detected in the cranial nerves, suggesting distinct functions. Thus, our expression analysis provides a good basis for further functional studies.

1. Introduction Rho GTPases are important regulators of cellular function and control a large variety of biological processes relevant to embryonic development as well as tissue homeostasis. Rho GTPases function as molecular switches and alternate between an active GTP-bound and an inactive GDP-bound state. Rho guanine nucleotide exchange factors (GEFs) catalyze the exchange of GDP to GTP thereby activating Rho GTPases. The majority of Rho GEF proteins are characterized by a Dblhomolog (DH) domain and an adjacent Pleckstrin-homology (PH) domain (Schmidt and Hall, 2002). In addition, proteins of the Dbl family contain catalytic and protein-protein interaction domains controlling protein activity and localization thereby supporting integration into various signaling pathways. Trio and its paralog Kalirin (Kalrn) are unique members of the Dbl family as they have two GEF domains of distinct specificity. Trio was originally identified as an interactor of the transmembrane protein

∗

tyrosine phosphatase LAR (leukocyte-antigen-related protein) and named Trio as it displays three enzymatic domains: two functional GEF domains and a serine/threonine kinase (STK) domain (Debant et al., 1996). The N-terminal DH-PH unit (GEF1 domain) activates Rac1 and RhoG GTPases, while the C-terminal DH-PH unit (GEF2 domain) acts specifically on RhoA (Bellanger et al., 1998; Blangy et al., 2000; Debant et al., 1996). In addition to the GEF domains Trio displays an N-terminal putative lipid-transfer SEC14 domain, a variable number of spectrin-like domains, two Src-homology 3 (SH3) domains, one immunoglobuline-like (Ig) domain, and a C-terminal STK domain. Kalrn has a very similar domain structure as Trio, but differs by an additional fibronectin-like (FN3) domain placed between the Ig-like domain and the STK domain. Thus, Trio and Kalrn are characterized by distinct protein or lipid binding domains facilitating interactions with distinct proteins and phospholipids. This is also reflected in the name Kalirin, derived from the Hindu goddess Kali, who has multiple arms (Alam et al., 1997). Indeed, Kalrn was first described as an interaction partner

Corresponding author. Philipps-University Marburg, Faculty of Biology, Molecular Embryology, Karl-von-Frisch-Str. 8, 35043, Marburg, Germany. E-mail address: [email protected] (A. Borchers).

https://doi.org/10.1016/j.gep.2019.02.004 Received 13 December 2018; Received in revised form 26 February 2019; Accepted 26 February 2019 Available online 04 March 2019 1567-133X/ © 2019 Elsevier B.V. All rights reserved.

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

Abbreviations a b ba bl cg cn e ec fb ha hb i Ig in FN l le m ma

mb me mn n-tub nc no np npb nt op ov pg pom rb s sc STK tg tn tp vn

animal brain branchial arches blastopore lip cement gland cranial nerves eye ectoderm forebrain hyoidal arch hindbrain intermediate domain of primary neurons immunoglobulin interneuron fibronectin lateral domain of primary neurons lens medial domain of primary neurons mandibular arch

of peptidylglycine α-amidating monooxygenase (PAM), an integral membrane protein known to be involved in the biosynthesis of neuropeptides (Alam et al., 1997). Subsequently Kalrn was also found to interact with Huntingtin-associated protein 1 (HAP1) (Colomer et al., 1997). Likewise, a number of interaction partners have been identified for Trio (reviewed in (Schmidt and Debant, 2014; van Rijssel and van

midbrain mesoderm motorneuron neural β-tubulin neural crest notochord neural plate neural plate border neural tube olfactory placode otic vesikel pineal gland periocular mesenchyme Rohon-Beard neuron somites sclerotome serine/threonine kinase trigeminal ganglion trigeminal nerve (V) trigeminal placode vestibulocochlear nerve

Buul, 2012)). For example, Trio interacts with FAK, different cadherins, and proteins involved in actin dynamics – like Filamin, CARMIL and Tara – indicating a role in the control of cell motility (Backer et al., 2007; Bellanger et al., 2000; Charrasse et al., 2007; Kashef et al., 2009; Medley et al., 2003; Timmerman et al., 2015; Vanderzalm et al., 2009; Yano et al., 2011). Thus, Trio and Kalrn likely act as signal integrators

Fig. 1. Graphical summary of the protein domain structures of Kalrn and Trio using an idealized animal tree of life. The tree reflects the currently accepted phylogeny of animals. Different protein domains are indicated by color. (The domain architecture of different Kalrn/Trio proteins in respect to protein length (in amino acids) is depicted in Supplementary Fig. 2). Species abbreviation Ap: Acanthaster planci; Aq: Amphimedon queenslandica; Bb: Branchiostoma belcheri; Ce: Caenorhabditis elegans; Ct: Capitella teleta; Dm: Drosophila melanogaster; Fo: Frankliniella occidentalis; Hv: Hydra vulgaris; Lg: Lottia gigantea; Lg: Sk: Saccoglossus kowalevskii; Ta: Trichoplax adhaerens; Tc: Tribolium castaneum; Xl: Xenopus laevis; Xt: Xenopus tropicalis.//indicate a gap in the protein domain structure, as two shorter fragments were assembled for graphical representation. 19

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

Trio during contact inhibition of locomotion, a process controlling the directional migration of neural crest cells (Moore et al., 2013). Although Trio has been shown to play an important role in Xenopus neural crest migration, the temporal and spatial expression of trio and kalrn in embryonic development has not been characterized yet. Thus, the results of this study will provide a good starting point for future functional studies.

relaying input of extrinsic signals to Rho GTPases thereby affecting various biological processes. Comparison between vertebrates and invertebrates identified common functions of Trio and Kalrn in embryonic development and disease. For example Trio has an evolutionary conserved role in the development of the nervous system (reviewed in (Schmidt and Debant, 2014)). Loss-of-function mutations in C. elegans unc-73 and Drosophila trio affect axon guidance (Awasaki et al., 2000; Bateman et al., 2000; Liebl et al., 2000; Steven et al., 1998), while mouse loss-of-function mutations disturb neuronal migration and morphogenesis leading to defects in brain organization (Backer et al., 2018; O'Brien et al., 2000; Peng et al., 2010). Moreover, de novo mutations of human TRIO were lately identified in patients with complex neurodevelopmental disorders including intellectual disability, microcephaly, autism spectrum disorders and schizophrenia (Ba et al., 2016; Katrancha et al., 2017; Pengelly et al., 2016; Sadybekov et al., 2017); a number of these mutations clustered in the DH1 domain of the GEF1 domain, which is required for activation of Rac1. Like Trio, Kalrn is also expressed in the brain (Alam et al., 1997; McPherson et al., 2002), where it affects synapse formation and dendritic spine morphogenesis, processes relevant for learning and memory. Furthermore, Kalrn has been implicated in neuropsychiatric and neurological disorders like Schizophrenia, Alzheimer's and Huntington's disease (reviewed in (Mandela and Ma, 2012)). In addition to a function in neural development and morphogenesis, Trio and Kalrn have also been implicated in muscle development. Trio loss-of-function in the mouse caused defects in myogenesis likely due to aberrant myoblast alignment and fusion (Charrasse et al., 2007; O'Brien et al., 2000). Kalrn isoforms are also expressed in mouse skeletal muscle and loss of function affects neuromuscular development and sarcomere length (Mandela et al., 2012). Furthermore, the C. elegans ortholog UNC-73 plays a role in pharynx and vulval muscle development (Steven et al., 2005). Thus, Trio and Kalrn functions in muscle and neural development may be evolutionarily conserved, however their evolutionary origin is so far unclear. Here we perform a comparative analysis of Trio and Kalrn evolution and complement this with a comprehensive study of trio and kalrn expression in Xenopus development. Previously, Trio has been shown to affect Xenopus neural crest migration downstream of Cadherin-11 (Kashef et al., 2009). Furthermore, the polarity protein PAR3 inhibits

2. Results and discussion 2.1. Phylogenetic and domain analysis of kalrn and trio proteins Annotated protein sequences of human Trio (Hs TRIO) and its Kalrn paralog (Hs KALRN) were used as a query to identify similar Trio and Kalrn sequences (Supplementary Table 1) searching public databases. Protein sequences with the highest percentage of similarity were selected for members of the Deuterostomia, Protostomia and Prebilateria. As representatives of vertebrates we used rat, mouse, chick, Xenopus, and zebrafish; for Cephalochordata, Branchiostoma belcheri, for Hemichordata, Saccoglossus kowalewskii, and for Echinodermata, Acanthaster planci, were used. Within the Protostomia, Caenorhabditis elegans, Drosophila melanogaster, Tribolium castaneum and several more basally branching insects served as representatives of the Ecdysozoa. Further, Lottia gigantea and Capitella teleta were used as members of the Lophotrochozoa and Hydra vulgaris, Trichoplax adhaerens and Amphimedon queenslandica as prebilaterian taxa. Our database search and sequence alignment revealed a generally conserved protein structure and identified the functionally important domains of Kalrn and/or Trio in Eumetazoa (Fig. 1 and alignment in Supplementary Fig. S1). In vertebrates, the typical domain structure of Kalrn comprises one N-terminal SEC14 domain, seven SPEC domains, followed by the GEF1 (DH1-PH1) and GEF2 (DH1-PH1) domains, which are separated by an SH3 domain. In addition to these domains Kalirins harbor a second SH3, one Ig, one FN3 and an STK domain. In contrast, vertebrate Trio proteins lack the 7th SPEC and the FN3 domain. Further, we verified a conspicuous C-terminal truncation of Trio in the Protostomia, which had been previously identified (Schmidt and Debant, 2014). In Prebilateria, a core sequence for a putative Kalrn/ Trio-related protein is predicted for the sponge Amphimedon

Fig. 2. Bayesian phylogenetic tree of vertebrate and invertebrate Kalrn and Trio. Species abbreviation as in Fig. 1. Accession numbers are given in Supplementary Table 1. 20

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

purposes. It is interesting to note, that vertebrates retained this evolutionary flexibility by using C-terminal splice sites to modify Trio (Schmidt and Debant, 2014).

queenslandica comprising two SPEC and two Rho GEF domains (Fig. 1). For the placozoan Trichoplax one Kalrn Trio-related sequence is predicted, in which both SH3 domains, and the FN3 domains are missing, while the C-terminal STK domain exists. In contrast, the Hydra Kalrnlike sequence lacks part of the STK domain (incomplete sequence) but otherwise corresponds to Kalrn of other Eumetazoa. Phylogenetic analysis (Fig. 2) was carried out using only the eumetazoan sequences, because inclusion of the Trichoplax and sponge sequences disrupted the tree topology (not shown), likely due to their highly diverging sequence and domain structures (Fig. 1). The eumetazoan Kalrn and Trio proteins as well as the protostomian Kalrn-like and Trio-like proteins were placed with convincing support along the tree of animal evolution (Fig. 2), with the exception of UNC-73, the Kalrn-like protein of C. elegans. Placement of this protein outside the Ecdysozoa and at the base of Bilateria does not correspond to the accepted animal phylogeny. It might be due to a long-branch attraction (many changes indicate an ancestral position) caused by the quickly evolving nematode genome (Memar et al., 2018). The domain features of Kalrn and Trio along the evolutionary tree of animals are summarized graphically in an idealized phylogenetic tree (Fig. 1) conforming to the currently accepted animal phylogeny (Telford et al., 2015). Almost all analyzed eumetazoan proteins contain the SEC14 domain (only missing in Saccoglossus and Xenopus tropicalis Kalrn) and a variable number of SPEC domains. The Hydra and C. elegans (UNC-73) sequences feature a C-terminal FN3 domain, which identifies them as Kalirins. Since the Trichoplax Kalrn-like protein sequence already includes the STK domain, it was probably lost in C. elegans. The Hydra sequence is incomplete, and its predicted partial STK domain could not be elongated beyond the stop codon by genomic and EST database searches. A conspicuous difference between the protoand deuterostomian Trio proteins is the truncation of the C-terminus comprising the second SH3, the Ig and the STK domain (Fig. 1) as previously reported (Schmidt and Debant, 2014). Taken together, our data provide evidence for a prebilaterian existence of a Kalrn/Trio ancestor and its early duplication in Urbilateria followed by lineage-specific and unpredictable loss of either Kalrn or Trio during animal evolution. In detail, based on the domain analysis in Fig. 1, we propose the following model: A single Kalrn/Trio ancestor existed in Prebilateria, which might have consisted of a (SPEC)2-(RhoGEF)2 core sequence as indicated by the predicted Amphimedon Kalrn/ Trio-related sequence. The presence of the STK domain in Trichoplax indicates that its origin antedates the last common ancestor of Eumetazoa. The sequence encoding the domains between Rho-GEF2 and STK domain was acquired at the latest in the last common ancestor of Cnidaria and Bilateria and resulted in an ancestral Kalrn. This ancestral Kalrn duplicated in Urbilateria to yield Kalrn and Trio. Although Kalrn and Trio are highly conserved, their further evolution comprises unpredictable loss or retention events in the deuterostomian and protostomian lineages. Trio was lost in the basally positioned Deuterostomia (Echinodermata, Hemi-and Cephalochordata) and it lost its FN3 domain in the last common ancestor of all vertebrates. In a protostomianlineage-dependent manner, the whole C-terminal gene sequence of Trio was deleted. Thereby, the Protostomia acquired a C-terminally truncated Trio-like protein in addition to Kalrn. Either both were retained, like in some ancestral insects including Tribolium (Coleoptera), Athalia rosae (XP_012266030.1, XP_012266033.1) and Cephus cinctus (XP_015585808.1, XP_015585812.1) (both Hymenoptera), or one of them was lost. The thrips Frankliniella occidentalis (Thysanoptera), for example, retained Kalrn, just like the basal Deuterostomia Acanthaster planki (Echinodermata), Saccoglossus kowalewsii (Hemichordata) and Branchiostoma (Cephalochordata). In contrast, the hymenopteran Diachasma alloeum (XP_015125607.1) as well as Drosophila, the snail Lottia gigantea (Lophotrochozoa) and the polychaete annelid Capitella teleta lost Kalrn and retained only the truncated Trio-like protein. The dynamic loss of domains in Trio indicates a high flexibility of the gene encoding this signaling element, favorable for adaptive

2.2. Xenopus Trio and Kalrn paralogs The Xenopus Trio and Xenopus Kalrn were clearly assigned as orthologs of the Trio and Kalrn protein families (Fig. 2). Concerning the Xenopus paralogs, we assembled the protein-encoding sequences from genomic sequences for the diploid Xenopus tropicalis and the allotetraploid Xenopus laevis. For Xenopus tropicalis a Trio ortholog (Xt Trio) was identified and two predicted Kalrn protein fragments, one consisting of four SPEC domains and the GEF1 domain, the other of the GEF2 domain with adjacent SH3 domain and Ig, FN3, and STK domain. As both predicted Kalrn fragments localize to the same scaffold, they likely represent one protein and were therefore assembled (referred to as Xt Kalrn1+2) for phylogenetic analysis. For the allotetraploid Xenopus laevis (Session et al., 2016) the homologous genes – distinguished by the abbreviation L or S (Matsuda et al., 2015) – were used. Genomic analysis identifies one trio ortholog (Xl trio) and two kalrn paralogs, which localize to the S- (Xl kalrn.S) or the L-form (Xl kalrn.L) of chromosome 9. Thus, the two species of Xenopus most commonly used for research express trio and kalrn, whereby Xenopus laevis features two kalrn alloalleles. It is interesting to note that the domain variations in Xenopus Kalirins (SEC14, SPEC and SH3 domains) correspond to the variations observed during evolution of Kalrn/Trio (Fig. 1). 2.3. Temporal and spatial expression of trio and kalrn in Xenopus laevis development To analyze the temporal gene expression pattern of trio and its kalrn paralog in Xenopus laevis development total RNA was isolated from different stages of development and analyzed by RT-PCR (Fig. 3). The trio and kalrn primers were designed to amplify fragments of 462 bp and 460 bp, respectively, ranging from the most C-terminal SPEC domain to the DH motif of the first GEF domain. The kalrn primers were designed to amplify specifically the kalrn.L or kalrn.S isoforms. As an internal control, primers for histone H4 were used to assess RNA quality and to allow for semi-quantitative comparison. Trio expression was detected at all stages analyzed. Maternal trio transcripts were detected at early cleavage and blastula stages (Fig. 3, upper lane). Zygotic expression was

Fig. 3. Temporal gene expression pattern of Xenopus laevis trio and kalrn. Expression levels were determined by RT-PCR at different stages of Xenopus laevis development. Amplification of Histone H4 (H4) was used as an internal control. Reactions performed in the absence of reverse transcriptase (-RT) served as negative control. The temporal expression pattern of trio (upper lane), kalrn.L (middle lane) and kalrn.S (lower lane) are shown. 21

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

Fig. 4. Spatial gene expression pattern of trio. Trio gene expression was analyzed by whole mount in situ hybridization. Embryonic stages are indicated. The detailed expression pattern is described in the text. A Embryo at cleavage stage, lateral view. B Blastula stage embryo, lateral view. C Gastrula stage embryo, blastopore lip is indicated. D Transverse section of embryo in C; section plane is indicated in C. E Control in situ hybridization using a sense trio probe; gastrula stage embryo. F,G Neurula stage embryos, dorsal view. H Tailbud stage embryo, dorsal view. I,J Transverse sections of embryo shown in H. Section planes are indicated in H. Arrowhead in I marks staining between the meso- and endoderm likely representing migrating NC cells. K-M Tailbud stage embryos, lateral view. Sense control at stage 28 is indicated in the insert in L. N Higher magnification of the head of the embryo shown in L. O Double in situ hybridization showing trio (blue) and twist (red) co-expression in cranial NC cells. P, Q Transverse sections of tadpole embryos analyzed by trio/twist double in situ hybridization. Section planes are indicated in O. Abbreviations: a: animal, b: brain; ba: branchial arches; bl: blastopore lip; cg: cement gland; e: eye; ec: ectoderm; ha: hyoidal arch; ma: mandibular arch; me: mesoderm; nc: neural crest; no: notochord; np: neural plate; npb: neural plate border; nt: neural tube; ov: otic vesicle; s: somites; pom: periocular mesenchyme; sc: sclerotome; tg: trigeminal ganglion; v: vegetal. Scale bar: 500 μm (I,J: 200 μm).

22

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

Fig. 5. Spatial gene expression pattern of kalrn. Kalrn gene expression was analyzed by whole mount in situ hybridization. Embryonic stages are indicated. The detailed expression pattern is described in the text. A Embryo at cleavage stage, lateral view. B Blastula stage embryo, lateral view. C Gastrula stage embryo, blastopore lip is indicated. D Transverse section of embryo in C; section plane is indicated in C. E Sense control of gastrula stage embryo. F-H,K Neurula stage embryos, dorsal view. I,J Transverse sections of embryo shown in H. Section planes are indicated in H. K Anterior view of embryo shown in H. M-N Tailbud stage embryos, lateral view. O,P Higher magnification of the embryo heads shown in M,N, respectively; the mandibular branch (Vmd) and the ophthalmic branch (Vop) of the trigeminal nerve (V), the facial VII, the glossopharyngeal (IX), and the vagus (X) nerve are indicated in P. Q Control in situ hybridization using a sense trio probe; tailbud stage embryo. Abbreviations: a: animal; bl: blastopore lip; cn: cranial nerves; fb: forebrain; hb: hindbrain; i: intermediate domain of primary neurons; in: interneuron; l: lateral domain of primary neurons; le: lens; m: medial domain of primary neurons; mb: midbrain; me: mesoderm; mn: motor neuron; no: notochord; np: neural plate; npb: neural plate border; nt: neural tube; op: olfactory placode; pg: pineal gland; rb: Rohon-Beard neuron; s: somites; sc: sclerotome; tg: trigeminal ganglion; tn: trigeminal nerve; tp: trigeminal placode; v: vegetal; vn: vestibulo cochlear nerve. Scale bar: 500 μm (I,J: 200 μm).

23

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

(Fig. 5Q). At stage 38, kalrn is strongly expressed in the brain and the cranial nerves including the trigeminal (V), the facial (VII), the glossopharyngeal (IX), and the vagus (X) nerve (Fig. 5N,P). All these data are consistent with a similar analysis of zebrafish showing a significant anterior kalrn expression at stages of brain development (Sarangdhar et al., 2017). In summary, kalrn expression is most pronounced in fish and frog neural tissue – particularly in different neurons, the cranial nerves and all major brain areas. Taken together our expression analysis suggests that Kalrn and Trio share functional roles in vertebrate neuronal and brain development as well as somite development. These roles are likely evolutionary conserved, as mutant studies in C. elegans, Drosophila and mice revealed defects in axon guidance, neuronal migration and/or brain organization and myogenesis (Awasaki et al., 2000; Backer et al., 2018; Bateman et al., 2000; Liebl et al., 2000; Mandela et al., 2012; O'Brien et al., 2000; Peng et al., 2010; Steven et al., 1998; Steven et al., 2005). Furthermore, our gene expression analysis demonstrates that both paralogs are expressed in the eye at different stages of development. While trio is relatively broadly expressed in the eye, kalrn expression is first detected in the lens and then later throughout the eye. It was previously shown that Trio is required for apical constriction and epithelial invagination during chick lens morphogenesis (Plageman et al., 2011), in addition the trio and kalrn expression patterns in Xenopus may indicate a function at later steps of eye development. Interestingly, the role in neural crest development is likely limited to the Trio paralog, as we did not detect expression of kalrn in migrating cranial neural crest cells (Kashef et al.,

low at gastrula stages and increased at neurula and tadpole stages. Both isoforms of kalrn are maternally expressed at low levels (Fig. 3, middle and lower lane). No significant expression of kalrn.S and kalrn.L is detected at gastrula stages. Starting with neurula stages expression of both isoforms increased and the highest levels were reached at tadpole stages. In summary, trio and kalrn orthologs are expressed from early cleavage stages up to tadpole stages indicating a role in early development. The spatial expression pattern of Xenopus laevis trio and kalrn genes was analyzed by in situ hybridization. Trio is maternally expressed in the animal region of cleavage and blastula stage embryos (Fig. 4A and B). Zygotic expression is detected in the marginal zone during gastrulation (Fig. 4C and D), while in situ hybridization using a sense probe did not detect any specific staining (Fig. 4E). Transverse sections through the blastopore lip of stage 10.5 embryos show trio expression in the involuting dorsal mesoderm and the anterior ectoderm (Fig. 4D). At stage 13, trio is expressed in the neural plate (Fig. 4F). This expression persists at later neurula stages (Fig. 4G). At stage 22, trio expression increases in the neural plate, the developing somites, and the anterior head region including the cranial neural crest cells (Fig. 4H). Transverse sections reveal trio transcripts in the deep layer of the forming neural tube. Staining is also detected in the notochord, the somites and between the meso- and endoderm likely representing migrating NC cells (Fig. 4I and J). Posterior transverse sections show trio expression in the sclerotome (Fig. 4J). At stage 23, trio transcripts are expressed in the head, the cement gland, the periocular mesenchyme, the trigeminal ganglion and migrating cranial neural crest cells (Fig. 4K). In the trunk region trio is expressed in the neural tube and the developing somites and continues to be expressed there at later stages. In addition, tailbud stage 28 embryos express trio in the branchial, hyoidal and mandibular arches, the brain, the eye anlage, the otic vesicle, and the trigeminal nerve (Fig. 4L,N), while no signal is detected in the sense control (Fig. 4L, insert). Trio expression in the head – in particular in the brain and branchial arches – increases at later tadpole stages as seen here in a stage 38 embryo (Fig. 4M). As it was previously shown that Trio plays a role in Xenopus neural crest migration (Kashef et al., 2009; Moore et al., 2013), we used double in situ hybridization to further analyze trio expression in cranial neural crest cells. A twist in situ hybridization probe (red) was used to mark cranial neural crest cells, while trio transcripts are stained in blue. Indeed, purple cells co-expressing twist and trio can be identified in whole embryos as well as sections (Fig. 4O–Q) confirming that trio is expressed in migrating cranial neural crest cells. Thus, trio expression in Xenopus laevis is consistent with functions in myogenesis, organization of neural tissue as well as neural crest development. Kalrn is also maternally expressed and kalrn transcripts are detected in the animal pole of stage 6 and stage 8/9 embryos (Fig. 5A and B). Expression levels decrease at gastrula stages, when kalrn is only weakly expressed in the region of the blastopore lip of stage 10.5 embryos in a thin layer of the mesoderm (Fig. 5C and D); sense controls at this stage do not show any staining (Fig. 5E). Significant kalrn expression is detected at neurula stages. At stage 13, kalrn transcripts are localized in the neural plate and predominately in the medial, intermediate, and lateral longitudinal domains of primary neurons (Fig. 5F). The neuronal expression was confirmed by double in situ hybridization using kalrn (blue) and n-tubulin (red) antisense probes (Fig. 6). Kalrn expression in neurons is also detected at later neurula stages (Fig. 5G and H). Transverse sections through a stage 21 embryo show kalrn expression in motor, inter- and Rohon-Beard neurons and a weaker expression in the somites and the sclerotome of forming somites (Fig. 5I and J). In addition, an anterior view of stage 21 embryos reveals expression in the trigeminal placode and the olfactory placode (Fig. 5K). This expression persists at stage 23 (Fig. 5L). At stage 28, kalrn transcripts are detected in the trigeminal nerve, the vestibule cochlear nerve, the pineal gland, and the olfactory placode (Fig. 5M,O), as well as the neural tube. In contrast, sense controls at tailbud stages do not show any staining

Fig. 6. Kalrn is co-expressed with n-tubulin in different neurons. Kalrn (blue) and n-tubulin (red) co-expression was analyzed by double whole mount in situ hybridization. A Neurula embryo (stage 13) analyzed by in situ hybridization using a kalrn antisense probe. B Stage 13 embryo analyzed by kalrn (blue) and n-tubulin (red) double in situ hybridization. C Transverse section of the embryo seen in A. D Transverse section of the embryo seen in B shows that the purple areas expressing kalrn and n-tubulin significantly overlap with the kalrn expression in C. Abbreviations: i: intermediate domain of primary neurons; l: lateral domain of primary neurons; m: medial domain of primary neurons; ntub: n-tubulin; no: notochord; np: neural plate. Scale bar: A,B: 500 μm; C,D: 200 μm. 24

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

Kalrn1) and “PREDICTED: kalirin, RhoGEF kinase isoform X1“ (XP_012825853.1, named here Xt Kalrn2). Xt Kalrn1 represents a protein fragment consisting of four SPEC domains and the GEF1 domain, Xt Kalirin2 is a fragment including the GEF2 domain with adjacent SH3 domain and Ig, FN and serine/threonine kinase domain. Using the Xenopus tropicalis 9.2 Genome-Database (Xenbase) it was determined that both fragments localize to scaffold_232. The gene region coding for Xt kalrn1 lies upstream of the Xt kalrn2. Taken together, both fragments likely represent one common protein, although the GEF1 SH3 coding region and the SEC domain are missing. Thus, for phylogenetic analysis both proteins were assembled (Xt Kalrn1+2). The identified protein sequences of rat (Rn, Rattus norvegicus), mouse (Mm, Mus musculus), chicken (Gg, Gallus gallus), Xenopus laevis (Xl), Xenopus tropicalis (Xt), zebrafish (Dr, Danio rerio), the cephalochordate Branchiostoma belcheri (Bb), the echinoderm Acanthaster planci (Ap), the hemichordate Saccoglossus kowalevskii (Sk), and sequences of several corresponding Ecdysozoa proteins (Dm, Drosophila melanogaster; Tc, Tribolium castaneum; C. elegans (UNC-73)), of Lophotrochozoa like the annelid Capitella teleta (Ct) and the mollusc Lottia gigantea (Lg) as well as of the prebilaterian cnidarian Hydra vulgaris (Hv), Trichoplax adhaerens (Ta, Placozoa) and Amphimedon queenslandica (Aq, Porifera) were aligned with the human sequences using the multiple sequence alignment program ClustalX 2.1 (Jeanmougin et al., 1998; Larkin et al., 2007; Thompson et al., 1997) with a BLOSUM30 alignment matrix. The multiple sequence alignment was edited with Genedoc 0.7.1. and converted with MEGA to construct a midpoint rooted phylogenetic tree using a Metropolis-coupled Markov Chain Monte Carlo (MrBayes 3.2, ngen = 1000000, samplefreq = 100). FigTree v1.4.2. was used for visualization. Protein domains were predicted by SMART (Simple Modular Architecture Research Tool (http:// smart.embl.de/) and by the NCBI (National Center for Biotechnology Information) “Identify Conserved Domains” tool. Protein-domainstructure was visualized by IBS v1.0.3 (Illustrator for Biological Sequences, CUCKOO Workgroup).

2009; Moore et al., 2013). As previous studies mainly focused on the expression of one paralog, trio or kalrn, in specific brain areas, this comparative study gives new insight into the developmental role of these unique members of the RhoGEF protein family. For future functional studies it is interesting, that Xenopus trio is expressed in tissues and organs that are also affected in human disease. For example, de novo mutations of TRIO have been identified in patients with a combination of neurodevelopmental, skeletal and craniofacial abnormalities (Ba et al., 2016; Pengelly et al., 2016). In general, Xenopus provides a time- and cost-efficient animal model system to study the function of candidate genes with relevance to human disease (Blum and Ott, 2018). Amphibians are closer related to humans than zebrafish, but also allow the direct observation of embryonic development including live-cell imaging. Thus, for example the function of Trio in neural crest development – which may be implicated by the craniofacial defects observed in patients with mutation in TRIO – can be analyzed at high temporal and spatial resolution. Moreover, antisense Morpholinomediated knockdown allows to reduce protein levels and is thereby ideally suited to model congenital human syndromes, which are caused by heterozygous hypomorphic alleles. Finally, if a Xenopus Trio disease model is established, novel mutations identified in patients can be evaluated by analyzing their ability to rescue the loss of function phenotype or by characterizing their overexpression phenotypes. Thus, the expression patterns presented here provide a solid basis to analyze the function of Trio and Kalrn in development and disease. 2.4. Conclusion Using in silico analysis we identified Xenopus laevis trio and kalrn paralogs, and showed by phylogenetic analysis that they likely originate from a single Kalrn/Trio-like protein in the last common ancestor of Bilateria. Interesting is the dynamic, unpredictable loss and retention of the kalrn and trio genes in invertebrates as well as the loss of the Cterminal Trio domains in Protostomia. This feature might indicate a functional adaptability, which is retained in vertebrates by splice variants modifying the C-terminus. Moreover, as exemplified by the Xenopus laevis Kalrn variants, several domains within Kalrn may be lost, when a duplicate is available thereby generating tools for potential novel functions. Temporal and spatial expression analysis of trio and kalrn in Xenopus laevis shows that these proteins are likely controlling distinct steps in embryonic development. They may play redundant roles in muscle development, however, their distinct gene expression in migrating neural crest cells (Trio) or the cranial nerves (Kalrn) suggests non-redundant functions. Considering their evolutionary conserved protein structure this expression analysis provides a good starting point for further functional studies.

3.2. Xenopus expression analysis and histology Xenopus laevis embryos were obtained and cultured according to standard protocols and staged using the normal table of Nieuwkoop and Faber (Nieuwkoop and Faber, 1956). All procedures were performed according to the EU Directive 2010/63/EU for animal experiments. For analysis of the temporal expression pattern by RT-PCR total RNA from five embryos at the indicated stages was extracted using illustra™ RNAspin Mini Kit (GE Healthcare, Buckinghamshire). Reverse transcription was performed using Revert Aid Reverse Transcriptase (Thermo Fisher Scientific, Waltham, USA) with 1 μg total RNA and random hexamer primers (Thermo Fisher Scientific, Waltham, USA). cDNA from three different female frogs was pooled and amplified using the following primers: 5‘ TACCTTTCCACGCACACCTC 3‘ (trio-forward); 5‘ GCACTCCCGAAGATCACGAA 3’ (trio-reverse); 5′ AGGAAGCAGGGG AGCATTAT 3’ (kalrn.LS-forward); 5′ GCAAGTCCCTCACATACGCT 3’ (kalrn.L-reverse); 5′ AAGTCCCTCACATACGCCTT 3’ (kalrn.S-reverse). H4 primers served as control. Amplified products were analyzed by agarose gel electrophoresis and documented using the LI-COR Odyssey®Fc Imaging System. To analyze the spatial trio and kalrn gene expression patterns in Xenopus development albino embryos were cultured to different stages, fixed in MEMFA (3.7% formaldehyde, 0.1M MOPS, 2 mM EGTA, 1 mM MgSO4) and processed for in situ hybridization according to standard protocols (Harland, 1991). Double in situ hybridization was performed as previously published (Koestner et al., 2008) using the following published plasmids: Twist (Hopwood et al., 1989), N-Tubulin (Richter et al., 1988). Sense controls were performed for all stages analyzed to monitor the specificity of the antisense probe. For kalrn in situ hybridization a published plasmid (Kashef et al., 2009) that was at the time annotated as Xenopus Trio was used, because sequence analysis

3. Experimental procedures 3.1. Database search and phylogenetic analysis Based on the annotated human amino acid sequence of TRIO (NP_009049.2) and KALRN (NP_001019831.2) similar/homolog sequences in Xenopus laevis and major model organisms were searched with the Basic Local Alignment Search Tool (BLAST) in the NCBI database (https://www.ncbi.nlm.nih.gov/). Only sequences with 60% query cover or more were chosen for phylogenetic analysis. NCBI Blast of the human Kalrn sequence in Xenopus laevis revealed two different Kalrn proteins. Using the Xenopus laevis J-strain 9.2 Genome-Database (Xenbase) the KALRN protein, annotated as “kalirin RhoGEF kinase L homeolog” (NP_001121262.1, too short), was identified as a part of the full-length Kalrn localized at Xenopus laevis chromosome 9L. The second annotated Kalrn protein is a “PREDICTED: kalirin-like” (XP_018093229.1) localized at Xenopus laevis chromsome 9S. For Xenopus tropicalis, two Kalrn proteins were identified by NCBI Blast search: “PREDICTED: kalirin-like” (XP_017945506.1, named here Xt 25

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

using current genomic data (X. laevis 9.2 (Vize and Zorn, 2017)) identified this construct as the Kalrn paralog. The Trio construct used for preparation of trio antisense RNA was cloned from cDNA of stage 26 Xenopus embryos. A 687 bp cDNA fragment of trio (containing part of the 5′UTR and part of the ORF) was amplified by PCR (5′ TAATTGGC GGAGCGTGTGAC 3’ (forward primer); 5′CTGGAGTTAGCTGGGAAGGA 3’ (reverse primer)), cloned into TOPO vector (Thermo Fisher Scientific, Waltham, USA) and used for the preparation of sense and antisense RNA. For sectioning embryos were fixed in MEMFA and embedded in 4% low melting agarose (Roth) in 0.1 MBS buffer. 40 μm sections were prepared using a vibratome (Leica VT1200 S, Leica Biosystems, Wetzlar, Germany) and the coverslip was mounted with Mowiol (Merck KGaA, Darmstadt, Germany). Images were taken using a binocular microscope (Leica M165FC with a DFC450C Camera).

like polypeptide, with a rac1 guanine nucleotide exchange factor domain. Hum. Mol. Genet. 6, 1519–1525. Debant, A., Serra-Pages, C., Seipel, K., O'Brien, S., Tang, M., Park, S.H., Streuli, M., 1996. The multidomain protein Trio binds the LAR transmembrane tyrosine phosphatase, contains a protein kinase domain, and has separate rac-specific and rho-specific guanine nucleotide exchange factor domains. Proc. Natl. Acad. Sci. U. S. A. 93, 5466–5471. Harland, R.M., 1991. In situ hybridization: an improved whole-mount method for Xenopus embryos. Methods Cell Biol. 36, 685–695. Hopwood, N.D., Pluck, A., Gurdon, J.B., 1989. A Xenopus mRNA related to Drosophila twist is expressed in response to induction in the mesoderm and the neural crest. Cell 59, 893–903. Jeanmougin, F., Thompson, J.D., Gouy, M., Higgins, D.G., Gibson, T.J., 1998. Multiple sequence alignment with Clustal X. Trends Biochem. Sci. 23, 403–405. Kashef, J., Kohler, A., Kuriyama, S., Alfandari, D., Mayor, R., Wedlich, D., 2009. Cadherin-11 regulates protrusive activity in Xenopus cranial neural crest cells upstream of Trio and the small GTPases. Genes Dev. 23, 1393–1398. Katrancha, S.M., Wu, Y., Zhu, M., Eipper, B.A., Koleske, A.J., Mains, R.E., 2017. Neurodevelopmental disease-associated de novo mutations and rare sequence variants affect TRIO GDP/GTP exchange factor activity. Hum. Mol. Genet. 26, 4728–4740. Koestner, U., Shnitsar, I., Linnemannstons, K., Hufton, A.L., Borchers, A., 2008. Semaphorin and neuropilin expression during early morphogenesis of Xenopus laevis. Dev. Dynam. 237, 3853–3863. Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., McGettigan, P.A., McWilliam, H., Valentin, F., Wallace, I.M., Wilm, A., Lopez, R., Thompson, J.D., Gibson, T.J., Higgins, D.G., 2007. Clustal W and clustal X version 2.0. Bioinformatics 23, 2947–2948. Liebl, E.C., Forsthoefel, D.J., Franco, L.S., Sample, S.H., Hess, J.E., Cowger, J.A., Chandler, M.P., Shupert, A.M., Seeger, M.A., 2000. Dosage-sensitive, reciprocal genetic interactions between the Abl tyrosine kinase and the putative GEF trio reveal trio's role in axon pathfinding. Neuron 26, 107–118. Mandela, P., Ma, X.M., 2012. Kalirin, a key player in synapse formation, is implicated in human diseases. Neural Plast. 728161 2012. Mandela, P., Yankova, M., Conti, L.H., Ma, X.M., Grady, J., Eipper, B.A., Mains, R.E., 2012. Kalrn plays key roles within and outside of the nervous system. BMC Neurosci. 13, 136. Matsuda, Y., Uno, Y., Kondo, M., Gilchrist, M.J., Zorn, A.M., Rokhsar, D.S., Schmid, M., Taira, M., 2015. A new nomenclature of Xenopus laevis chromosomes based on the phylogenetic relationship to Silurana/Xenopus tropicalis. Cytogenet. Genome Res. 145, 187–191. McPherson, C.E., Eipper, B.A., Mains, R.E., 2002. Genomic organization and differential expression of Kalirin isoforms. Gene 284, 41–51. Medley, Q.G., Buchbinder, E.G., Tachibana, K., Ngo, H., Serra-Pages, C., Streuli, M., 2003. Signaling between focal adhesion kinase and trio. J. Biol. Chem. 278, 13265–13270. Memar, N., Schiemann, S., Hennig, C., Findeis, D., Conradt, B., Schnabel, R., 2018. Twenty million years of evolution: the embryogenesis of four Caenorhabditis species are indistinguishable despite extensive genome divergence. Dev. Biol. https://doi. org/10.1016/j.ydbio.2018.12.022. Moore, R., Theveneau, E., Pozzi, S., Alexandre, P., Richardson, J., Merks, A., Parsons, M., Kashef, J., Linker, C., Mayor, R., 2013. Par3 controls neural crest migration by promoting microtubule catastrophe during contact inhibition of locomotion. Development 140, 4763–4775. Nieuwkoop, P.D., Faber, J., Hubrecht Laboratory Utrecht, 1956. Normal Table of Xenopus laevis (Daudin) : a Systematical and Chronological Survey of the Development from the Fertilized Egg till the End of Metamorphosis. North-Holland Pub. Co., Amsterdam, pp. 243. O'Brien, S.P., Seipel, K., Medley, Q.G., Bronson, R., Segal, R., Streuli, M., 2000. Skeletal muscle deformity and neuronal disorder in Trio exchange factor-deficient mouse embryos. Proc. Natl. Acad. Sci. U. S. A. 97, 12074–12078. Peng, Y.J., He, W.Q., Tang, J., Tao, T., Chen, C., Gao, Y.Q., Zhang, W.C., He, X.Y., Dai, Y.Y., Zhu, N.C., Lv, N., Zhang, C.H., Qiao, Y.N., Zhao, L.P., Gao, X., Zhu, M.S., 2010. Trio is a key guanine nucleotide exchange factor coordinating regulation of the migration and morphogenesis of granule cells in the developing cerebellum. J. Biol. Chem. 285, 24834–24844. Pengelly, R.J., Greville-Heygate, S., Schmidt, S., Seaby, E.G., Jabalameli, M.R., Mehta, S.G., Parker, M.J., Goudie, D., Fagotto-Kaufmann, C., Mercer, C., Debant, A., Ennis, S., Baralle, D., 2016. Mutations specific to the Rac-GEF domain of TRIO cause intellectual disability and microcephaly. J. Med. Genet. 53, 735–742. Plageman Jr., T.F., Chauhan, B.K., Yang, C., Jaudon, F., Shang, X., Zheng, Y., Lou, M., Debant, A., Hildebrand, J.D., Lang, R.A., 2011. A Trio-RhoA-Shroom3 pathway is required for apical constriction and epithelial invagination. Development 138, 5177–5188. Richter, K., Grunz, H., Dawid, I.B., 1988. Gene expression in the embryonic nervous system of Xenopus laevis. Proc. Natl. Acad. Sci. U. S. A. 85, 8086–8090. Sadybekov, A., Tian, C., Arnesano, C., Katritch, V., Herring, B.E., 2017. An autism spectrum disorder-related de novo mutation hotspot discovered in the GEF1 domain of Trio. Nat. Commun. 8, 601. Sarangdhar, M.A., Chaubey, D., Bhatt, A., Km, M., Kumar, M., Ranjan, S., Pillai, B., 2017. A novel long non-coding RNA, durga modulates dendrite density and expression of kalirin in zebrafish. Front. Mol. Neurosci. 10, 95. Schmidt, A., Hall, A., 2002. Guanine nucleotide exchange factors for Rho GTPases: turning on the switch. Genes Dev. 16, 1587–1609. Schmidt, S., Debant, A., 2014. Function and regulation of the Rho guanine nucleotide exchange factor Trio. Small GTPases 5, e29769. Session, A.M., Uno, Y., Kwon, T., Chapman, J.A., Toyoda, A., Takahashi, S., Fukui, A.,

Author Contribution A.B. and M.K. conceived and designed the experiments. M.K. and D.A. performed the phylogenetic analysis. M.K. and L.E. carried out all other experiments. All authors analyzed and discussed the data. The manuscript was written by A.B., M.H. and M.K. and critically edited by all authors. Acknowledgements The authors thank Christiane Rohrbach for technical assistance and Jubin Kashef for providing plasmids. The project was supported by a grant of the German Research Foundation to A.B. (KA 4104/1-2) and the DFG Research Training Group (GRK 2213, Membrane Plasticity in Tissue Development and Remodeling). Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.gep.2019.02.004. References Alam, M.R., Johnson, R.C., Darlington, D.N., Hand, T.A., Mains, R.E., Eipper, B.A., 1997. Kalirin, a cytosolic protein with spectrin-like and GDP/GTP exchange factor-like domains that interacts with peptidylglycine alpha-amidating monooxygenase, an integral membrane peptide-processing enzyme. J. Biol. Chem. 272, 12667–12675. Awasaki, T., Saito, M., Sone, M., Suzuki, E., Sakai, R., Ito, K., Hama, C., 2000. The Drosophila trio plays an essential role in patterning of axons by regulating their directional extension. Neuron 26, 119–131. Ba, W., Yan, Y., Reijnders, M.R., Schuurs-Hoeijmakers, J.H., Feenstra, I., Bongers, E.M., Bosch, D.G., De Leeuw, N., Pfundt, R., Gilissen, C., De Vries, P.F., Veltman, J.A., Hoischen, A., Mefford, H.C., Eichler, E.E., Vissers, L.E., Nadif Kasri, N., De Vries, B.B., 2016. TRIO loss of function is associated with mild intellectual disability and affects dendritic branching and synapse function. Hum. Mol. Genet. 25, 892–902. Backer, S., Hidalgo-Sanchez, M., Offner, N., Portales-Casamar, E., Debant, A., Fort, P., Gauthier-Rouviere, C., Bloch-Gallego, E., 2007. Trio controls the mature organization of neuronal clusters in the hindbrain. J. Neurosci. 27, 10323–10332. Backer, S., Lokmane, L., Landragin, C., Deck, M., Garel, S., Bloch-Gallego, E., 2018. Trio GEF Mediates RhoA Activation Downstream of Slit2 and Coordinates Telencephalic Wiring, vol.145 Development. Bateman, J., Shu, H., Van Vactor, D., 2000. The guanine nucleotide exchange factor trio mediates axonal development in the Drosophila embryo. Neuron 26, 93–106. Bellanger, J.M., Astier, C., Sardet, C., Ohta, Y., Stossel, T.P., Debant, A., 2000. The Rac1and RhoG-specific GEF domain of Trio targets filamin to remodel cytoskeletal actin. Nat. Cell Biol. 2, 888–892. Bellanger, J.M., Lazaro, J.B., Diriong, S., Fernandez, A., Lamb, N., Debant, A., 1998. The two guanine nucleotide exchange factor domains of Trio link the Rac1 and the RhoA pathways in vivo. Oncogene 16, 147–152. Blangy, A., Vignal, E., Schmidt, S., Debant, A., Gauthier-Rouviere, C., Fort, P., 2000. TrioGEF1 controls Rac- and Cdc42-dependent cell structures through the direct activation of rhoG. J. Cell Sci. 113 (Pt 4), 729–739. Blum, M., Ott, T., 2018. Xenopus: an undervalued model organism to study and model human genetic disease. Cells Tissues Organs 1–11. Charrasse, S., Comunale, F., Fortier, M., Portales-Casamar, E., Debant, A., GauthierRouviere, C., 2007. M-cadherin activates Rac1 GTPase through the Rho-GEF trio during myoblast fusion. Mol. Biol. Cell 18, 1734–1743. Colomer, V., Engelender, S., Sharp, A.H., Duan, K., Cooper, J.K., Lanahan, A., Lyford, G., Worley, P., Ross, C.A., 1997. Huntingtin-associated protein 1 (HAP1) binds to a Trio-

26

Gene Expression Patterns 32 (2019) 18–27

M.-C. Kratzer, et al.

evolution. Curr. Biol. 25, R876–R887. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G., 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876–4882. Timmerman, I., Heemskerk, N., Kroon, J., Schaefer, A., van Rijssel, J., Hoogenboezem, M., van Unen, J., Goedhart, J., Gadella Jr., T.W., Yin, T., Wu, Y., Huveneers, S., van Buul, J.D., 2015. A local VE-cadherin and Trio-based signaling complex stabilizes endothelial junctions through Rac1. J. Cell Sci. 128, 3514. van Rijssel, J., van Buul, J.D., 2012. The many faces of the guanine-nucleotide exchange factor trio. Cell Adhes. Migrat. 6, 482–487. Vanderzalm, P.J., Pandey, A., Hurwitz, M.E., Bloom, L., Horvitz, H.R., Garriga, G., 2009. C. elegans CARMIL negatively regulates UNC-73/Trio function during neuronal development. Development 136, 1201–1210. Vize, P.D., Zorn, A.M., 2017. Xenopus genomic data and browser resources. Dev. Biol. 426, 194–199. Yano, T., Yamazaki, Y., Adachi, M., Okawa, K., Fort, P., Uji, M., Tsukita, S., 2011. Tara up-regulates E-cadherin transcription by binding to the Trio RhoGEF and inhibiting Rac signaling. J. Cell Biol. 193, 319–332.

Hikosaka, A., Suzuki, A., Kondo, M., van Heeringen, S.J., Quigley, I., Heinz, S., Ogino, H., Ochi, H., Hellsten, U., Lyons, J.B., Simakov, O., Putnam, N., Stites, J., Kuroki, Y., Tanaka, T., Michiue, T., Watanabe, M., Bogdanovic, O., Lister, R., Georgiou, G., Paranjpe, S.S., van Kruijsbergen, I., Shu, S., Carlson, J., Kinoshita, T., Ohta, Y., Mawaribuchi, S., Jenkins, J., Grimwood, J., Schmutz, J., Mitros, T., Mozaffari, S.V., Suzuki, Y., Haramoto, Y., Yamamoto, T.S., Takagi, C., Heald, R., Miller, K., Haudenschild, C., Kitzman, J., Nakayama, T., Izutsu, Y., Robert, J., Fortriede, J., Burns, K., Lotay, V., Karimi, K., Yasuoka, Y., Dichmann, D.S., Flajnik, M.F., Houston, D.W., Shendure, J., DuPasquier, L., Vize, P.D., Zorn, A.M., Ito, M., Marcotte, E.M., Wallingford, J.B., Ito, Y., Asashima, M., Ueno, N., Matsuda, Y., Veenstra, G.J., Fujiyama, A., Harland, R.M., Taira, M., Rokhsar, D.S., 2016. Genome evolution in the allotetraploid frog Xenopus laevis. Nature 538, 336–343. Steven, R., Kubiseski, T.J., Zheng, H., Kulkarni, S., Mancillas, J., Ruiz Morales, A., Hogue, C.W., Pawson, T., Culotti, J., 1998. UNC-73 activates the Rac GTPase and is required for cell and growth cone migrations in C. elegans. Cell 92, 785–795. Steven, R., Zhang, L., Culotti, J., Pawson, T., 2005. The UNC-73/Trio RhoGEF-2 domain is required in separate isoforms for the regulation of pharynx pumping and normal neurotransmission in C. elegans. Genes Dev. 19, 2016–2029. Telford, M.J., Budd, G.E., Philippe, H., 2015. Phylogenomic insights into animal

27