Gene 555 (2015) 73–79

Contents lists available at ScienceDirect

Gene journal homepage: www.elsevier.com/locate/gene

Distinct evolutionary rate in the eye field transcription factors found by estimation of ancestral protein structure Ai Kamijyo a, Kei Yura a,b,c, Atsushi Ogura d,⁎ a

Graduate School of Humanities and Sciences, Ochanomizu University, 2-1-1 Otsuka, Bunkyo, Tokyo 112-8610, Japan Center for Informational Biology, Ochanomizu University 2-1-1 Otsuka, Bunkyo, Tokyo 112-8610, Japan National Institute of Genetics, 1111 Yata, Mishima, Shizuoka 411-8540, Japan d Department of Computer Bio-Science, Nagahama Institute of BioScience and Technology, 1266 Tamura, Nagahama, Shiga 526-0829, Japan b c

a r t i c l e

i n f o

Article history: Received 3 July 2014 Received in revised form 16 September 2014 Accepted 2 October 2014 Available online 7 October 2014 Keywords: Ancestral structure estimation Evolution of eye development Transcription factor

a b s t r a c t Eye-field transcription factors (EFTFs) are a set of genes that compose a regulatory network for eye development in animals, which are highly conserved among various animal phyla. To investigate the processes of conservation and diversification of the transcription factors for eye development, we examined the structural changes in the EFTF proteins by estimating the ancestral sequences with the available genome information. Among the different types of EFTFs, we selected otx2, tbx3, rx1, pax6, six3/6, lhx2 and nr2e1 because they are highly conserved in bilaterian animals. We searched the genome sequences of representative animal phyla for EFTF protein sequences. With deduced ancestral sequences and three-dimensional structures of EFTFs, we traced the evolutionary changes in amino acid residues and found that the DNA-binding domains were always more conserved than other regions, and that the other regions showed distinct evolutionary rates. The EFTF rx1, which resides at the pivotal part of the EFTF network, had a faster evolutionary rate than the others. These results indicated that the evolutionary rates of each protein in the EFTF network, which were expected to be consistent with each other to maintain the interactions in the network, were not constant among or within the factors, but rather, varied to a significant extent. © 2014 Published by Elsevier B.V.

1. Introduction Organisms with eyes appear sporadically on the tree of life, which suggests that the evolution of eyes occurred many times independently in various animals (Salvini-Plawen and Mayr, 1977). The differences in the morphology of eye types support the independent origins of those eyes. However, the acquisition of the eye might have occurred once in Cnidaria, which is one of the most primitive groups of animals that possess eyes, and for which the basic pattern for eye formation and development was established (Halder et al., 1995; Piatigorsky, 2003). Therefore, the evolution of the eye was a single event that originated from a prototype eye in an ancestral species. The present diversity of eye types was the result of modifications from the prototype eyes via differential activation of genes and changes in their regulatory mechanisms (Gehring and Ikeo, 1999; Rivera et al., 2010, Datta et al., 2011). The molecular evolution of the eye can be resolved once the fundamental genes for animal eye formation and a common type of photoreceptor cell are discovered in animals (Halder et al., 1995). Abbreviations: EFTF, eye field transcription factors; Md, the ratio of residue substitution in DNA-binding domains; Mn, the ratio of residue substitution in non-DNA-binding domains; PDB, Protein Data Bank. ⁎ Corresponding author. E-mail address: [email protected] (A. Ogura).

http://dx.doi.org/10.1016/j.gene.2014.10.003 0378-1119/© 2014 Published by Elsevier B.V.

Eye-field transcription factors (EFTFs) are a set of genes (here, we considered the EFTFs otx2, tbx3, rx1, pax6, six3, six6, lhx2, and nr2e1) that are involved in the morphogenesis of eyes and that are extensively studied in frogs and mice (Oliver et al., 1995; Zuber, 2003). The EFTFs are synchronously expressed in the eye-field during its specification and form a transcription regulation cascade (Zuber, 2003; Sernagor et al., 2006). Previous studies characterized the transcription network among EFTFs in frogs and proposed a model for the interactions during eye-field specification (Fig. 1) (Zuber, 2003; Sernagor et al., 2006). These studies showed that the neural plate, which is formed in response to neural inducers such as noggin, generated the regions that developed into the eyes. A transcription factor, otx2 with a homeodomain, was required for forebrain and midbrain specifications. The early-expressed EFTFs, i.e., tbx3, rx1, pax6, six3, and lhx2 (the latter four contain a homeodomain DNA-binding region) in coordination specify the eye-field within the presumptive forebrain (Mathers et al., 1997; Kawakami et al., 2000; Gehring and Ikeo, 1999; Wilson, 2002; Pan et al., 2006; Srivastava et al., 2010). The EFTFs nr2e1, with a zincfinger motif, and six6, with a homeodomain, play a role in the late specification of eye formation (Sernagor et al., 2006; Kawakami et al., 2000; Cheng et al., 2008). The EFTFs are also required for the induction of functional eyes at inauthentic locations (Mathers et al., 1997; Wawersik and Maas, 2000; Tucker et al., 2001).

74

A. Kamijyo et al. / Gene 555 (2015) 73–79

showed that there were organisms without a complete set of EFTF genes. Nemertea was the phylum with the least EFTFs, with only the pax6 gene, although it has rhabdomeric and subepidermally situated eyes. This apparent inconsistency in the relation between the existence of the EFTF network and the eyes suggested that the eye formation in this group was regulated by a different set of genes, or that the genome sequencing was incomplete. By contrast, some invertebrate phyla without eyes, such as Placozoa, did have orthologous genes for part of the EFTF network. This suggested that the entire set of EFTFs was required for vertebrate eye formation. The profile also showed that the duplication of six3/6 happened at the diversification of the vertebrates, and that this event evidently led to the completion of the EFTF gene set, which confirmed the results of Kawakami et al. (2000). The complete set of EFTF genes was established in the common ancestor of vertebrates. We, hereafter, lumped six3 and six6 as six3/6 in the current analyses. 2.2. Sequence conservation of the DNA-binding domain and diversification of the nonDNA-binding domain

Fig. 1. The EFTF network. The proteins in the top half of the network (upper-stream) are involved in the spatiotemporal development of the eyes and the proteins in the bottom half (lower-stream) are involved in the development of the components of the eyes.

These lines of evidence suggest that EFTFs are candidates for being the fundamental genes of animal eye formation. Moreover, the regulatory network of EFTFs originated in the first organism with eyes. However, homologues for EFTFs are only reported in model organisms, such as Drosophila melanogaster, Xenopus laevis, Oryzias latipes and Homo sapiens (Zuber, 2003; Mathers et al., 1997; Wawersik and Maas, 2000; Quiring et al., 1994; Seo et al., 1998), and homologues in invertebrates are rarely reported. Examination of the distribution of the homologues among the animal phyla could be a source of information to address the evolution of the EFTF network. The existence of plausible orthologues of the set of known EFTF genes provides a clue to the foundation of the network, and the examination of the evolutionary rate of the proteins provides a clue to the stability of the interactions in the network. Unveiling these pieces of information will help to reveal the possible evolutionary path of the eye. In this study, we addressed how the EFTFs and their network developed and evolved in animals, and we examined the evolution of EFTF protein structures and the EFTF network in various animal phyla. We searched for homologous proteins in animal genomes and tabulated the existence of EFTFs in different phyla. Computational analyses were performed on evolutionary rates and three-dimensional (3D) structures of EFTF proteins to trace the evolution of the EFTF network. We found that (i) the DNA-binding domains of the EFTFs were well conserved, and that the nonDNA-binding regions that interacted with other molecules often contained highly diversified residues, and that (ii) the evolutionary rates differed in an upper-stream and a lower-stream within the EFTF network. The results suggest that EFTFs emerged as a combination of, at least, two modules of networks, which managed a different set of roles in the development of the eye. 2. Results 2.1. The orthologue profile of EFTFs in various animal phyla The EFTF sequences in different animal phyla are summarized in Table 1 and represent the orthologue profile. The orthologue profile

With the sequences in the orthologue profile and a topology tree of the species, we inferred the ancestral sequences of each node of the tree, and we traced amino acid changes in the proteins of each site. The trace showed that the ratio of residue substitution in DNA-binding domains (Md) was almost always lower than that in nonDNA-binding domains (Mn) in every branch of the phylogenetic tree of EFTF proteins (Suppl. Table 1). For example, the Mn of Pax6 was five-fold higher than the Md of Pax6. To understand the balance between Md and Mn rates in animals, we calculated the mean of Mn/Md for all EFTFs for each branch on the tree and depicted the value in log-scaled color (Fig. 2). In Fig. 2, the deeper color of the branch indicates a larger Mn/Md ratio, and when the branch is white, Mn and Md are equal. After the bilaterian split, at the branch of deuterostomes, Mn was far higher than Md (Fig. 2). The Mn/Md of the vertebrate branch was not the highest, but the substitution rates of the nonDNA-binding domains accelerated before the divergence of vertebrates and the urochordata. The differences in the substitution rates of DNA-binding and nonDNA-binding domains were substantiated with 3D protein structures. We obtained a reasonable model structure for each ancestral EFTF protein domain and pbx1, a reference protein, and located the functionally critical substitutions of amino acids in three dimensions. The mapping of the residue-wise substitution rates of EFTF proteins in 3D structures further emphasized the skewed distribution of the highly substituted residues in the different domains (Fig. 3, the other seven results are shown in Supplemental Fig. 2). The substituted sites were spatially clustered in nonDNA-binding domains in all EFTF proteins (red in Fig. 3), whereas the DNA-binding domains were well-conserved (blue in Fig. 3). The difference in substitution rates was evident particularly for nr2e1 in which the DNA-binding domain was more conserved than the activation (nonDNA-binding) domain. 2.3. Distinct substitution rate on the DNA-binding domain of Rx1 The comparison of Mn and Md among EFTFs showed notable differences (Fig. 4). The box plot of Mns showed a slight decrease from otx2 to six3/6 in the distribution range (Fig. 4a), though statistical significance was not found. The box plot of Mds showed a significant difference in the distributions (p b 4.4 × 10−5 by the Kruskal Wallis test), particularly of rx1 (Fig. 4b). Hence, the DNA-binding domain of rx1 experienced a significantly accelerated evolution. When the set of proteins in an EFTF network were divided into two at rx1, one group contained otx2, tbx3 and rx1, which corresponded to a subset of proteins in the upper-stream of the EFTF network, and the other group contained pax6, lhx2, nr2e1 and six3/6, which corresponded to a subset of proteins in the lower-stream of the network. The former was a group

A. Kamijyo et al. / Gene 555 (2015) 73–79

75

Table 1 Gene profile of EFTFs in Animalia. A circle indicates, at least, one amino acid sequence similar to human EFTF proteins was found in the proteome of each species by reciprocal BLAST.

Protostome

Animal

Deuterostome

Chordate

Category

Species examples

Otx2

0

Rx1

Pax6

Lhx2

Nr2e1

Six3

Six6

Pbx1

Eye structure

Vertebrata

Human Homo sapiens

Camera type

Urochordata

Ascidian Ciona intestinalis

Eye spot

Cephalochordata

Amphioxus Branchiostoma floridae

–

Eye spot

–

Hemichordata

Acorn worm Saccoglossus kowalevskii

ND

Echinodermata

Sea urchin Strongglocentrotus purpuratus

Eye spot

Arthropoda

Fly Drosophila melanogaster

Compound eye

Nematoda

Caenorhabditis elegans

ND

Annelida

Ragworm Platynereis dumerilii

Ocellus

Mollusca

Squid Idiosepius paradoxus

Camera type Eye spot Ocellus

Limpet Lottia gigantea

Nemertea

Ribbon worm Lineus sanguineus

Platyhelminthes

Planaria Dugesia japonica

–

–

–

–

–

–

–

Cnidaria

Hydra Hydra vulgaris

Placozoa

Trichoplax adhaerens

Rhombozoa

Dicyema acuticephalum

Eye spot

Cup eye

–

ND

–

with a relatively high Mn, and the latter was a group with a relatively low Mn (Fig. 4a). The Md and Mn of pbx1, a reference protein, were almost the same as those of the lower-stream proteins, and -1.2

Tbx3

1.2

Vertebrata Cephalochordata Urochordata Hemichordata Echinodermata Cephalopoda Gastropoda Annelida Nemertea Platyhelminthes Rhombozoa Arthropoda Nematoda Cnidaria Placozoa

Fig. 2. The Mn/Md in each branch of the species tree. The tree was drawn schematically based on the consensus taxonomic relationships. The branches are colored in the logscaled mean of Mn/Md of all EFTFs. A red branch indicates that Mn is greater than Md, and a blue branch indicates that Mn is smaller than Md. The arrow represents the branch of deuterostomes after the bilaterian split.

–

–

–

–

–

–

ND –

–

ND

therefore, these results suggested that the evolution of nonDNA binding domains of the upper-stream proteins in the EFTF network was accelerated. The division between the upper-stream and the lower-stream coincided with the functional differences of the encoded proteins. The upper-stream genes encode proteins for the formation of retinal progenitor cells (Bailey et al., 2004; Horton et al., 2008). For instance, rx1 is important for the regulation of pax6 expression, a master control gene for eye morphology in retinal progenitors during the early retinal neurogenesis (Gehring and Ikeo, 1999; Nelson et al., 2009). The lower-stream genes encode proteins for the regulation of late retinal and corneal cell fate decisions and for the development of the forebrain and of eye related structures (Kawakami et al., 2000; Yu et al., 2000; Mathers and Jamrich, 2000; Tétreault et al., 2009; Cheng et al., 2008). Thus, the upper-stream genes determine the place and time of the eye specification, and the lower-stream genes develop tissues of the eyes such as the lens and retina. A comparison of Md and Mn between upper-stream and lower-stream genes found a higher Mn in upperstream genes (Fig. 3(a)). Furthermore, the significant elevation in the Md of rx1 was also noted (Fig. 3(b)). These findings demonstrated that the genes that specify the location of eyes have more rapid substitutions on nonDNA-binding domains than the domains in the proteins that build the eye structures. A difference in conservation was observed between the two DNA-binding domains of Rx1, and this suggested that one DNA-binding domain located near the N-terminus experienced high evolutionary pressure.

76

A. Kamijyo et al. / Gene 555 (2015) 73–79

a

b

Fig. 3. The substitution rates of residues in (a) nr2e1 and in (b) rx1. The three-dimensional structures of ancestral proteins are colored in accordance with the substitution rate defined in the Materials and methods section. Highly mutated residue is reddish in color and conserved residue is bluish in color. The portion that is depicted in three dimensions is shown in the diagram below the structure. Either the DNA interface or the opposite side of the DNA interface is shown in (i) and the flip side is shown in (ii).

3. Discussion The slow evolution in DNA-binding domains of all EFTF proteins except rx1 suggested that the combination between the EFTF proteins and the promoters of the activated genes were kept constant during the evolution of EFTFs. By contrast, the rapid evolution of nonDNA-binding domains, where other proteins interact, suggested extensive modification of the surface of the protein for protein–protein interactions. The modification might have followed either of the following two scenarios. First, the EFTFs switched the proteins that interacted with the domain, and consequently, the diversification of the activation domains. This type of scenario was observed in the interactions between the Src Homology 3 (SH3) domain and kinase in fungal species, where interacting proteins were switched and the signal transduction network was rewired (Sun et al., 2012). Another possibility was that the EFTF activation domain, part of the nonDNA-binding domain, coevolved with the

proteins that interact with the activation domain. Coevolution in the interface was observed in the network of germline development in Drosophila, for example. The combination of Sevenless and Bride of Sevenless, the ligand of Sevenless, is known to coevolve; correlated evolutionary change of amino acid residues in the extracellular domains of both Sevenless and Bride of Sevenless was observed (Bao and Friedrich, 2008). A similar mechanism could have operated on the EFTF proteins. The original EFTF activation domain could bind to one or more target proteins at various stages and in various tissues. To maintain the connectivity in the network, the target proteins of the activation domains could also modify their surface structures to compensate for the changes in the EFTF activation domains. In this study, we found that a full set of EFTF genes was established in the common ancestor of the vertebrates. The EFTF genes in animalia are not always required for the development of eyes because there are nonvertebrates with only a part of the EFTF gene set. In the EFTF

A. Kamijyo et al. / Gene 555 (2015) 73–79

77

a

b

Fig. 4. Box plot diagram with the median (the line inside the box), the interquartile range (box) and the range (whisker) comparing the (a) Mn and (b) Md of each EFTF protein and pbx1. The values outside the whisker are plotted as individual outliers.

genes in animalia, the DNA-binding domains were generally highly conserved, but nonDNA-binding domains were relatively diversified. The conservation in the DNA-binding domains suggested that the relationships between the target DNA sequences and the DNA-binding domain of the EFTFs were conserved. The diversification of the nonDNA-binding domains suggested that either a coevolution between proteins that interact with the domains or a frequent switching of interaction partners occurred. The exception in the high degree of conservation of DNAbinding domains was found in rx1, a transcription factor in the middle of the EFTF network. In the development of the vertebrate eye, rx1 induces the retinal region by suppressing genes that normally specify the telencephalon and that are highly diversified among animals (Chuang and Raymond, 2001). This suggested that a rapid evolutionary rate of rx1 changed an induction of the retinal region by switching targets for DNA-binding domains. In this study, we examined the evolutionary processes of the EFTF proteins and found that rx1 likely played a pivotal role in the EFTF network. The results suggest that the EFTFs emerged as a network that consisted of an upper-stream and a lower stream divided by rx1, which managed a different set of roles in eye development, namely spatiotemporal specification and component assembly, respectively, and that rx1 functioned as a pivot for the combination.

4. Materials and methods 4.1. Sequence retrieval and ancestral sequence estimation EFTFs are defined by many preceding studies utilizing various animals. Here, we selected eight transcription factor genes as EFTFs because they are highly conserved among bilaterian animals. We also used pbx1 as an out-group in this study. For the phyla to be studied, we selected one representative species of each animal phyla to avoid information bias by considering availability and quality of genome sequences. We first collected EFTF proteins from GenBank (Benson et al., 2013) and Ensembl (Flicek et al., 2012). For the species for which EFTF genes were not found, we selected putative EFTF orthologues by the reciprocal ‘best-hit’ BLAST search (Altschul et al., 1997) against JGI (http:// www.jgi.doe.gov/) and Ensembl genome databases. We used human EFTF proteins as queries, and the E-value threshold was set to 10−10. Multiple sequence alignment was conducted with ClustalW (Larkin et al., 2007), and a phylogenetic tree for each EFTF protein was constructed by the maximum likelihood method based on the topology tree adopted from the widely accepted species tree (Pawlowski et al., 1996; Peterson et al., 2004; Hedges et al., 2006; Blair, 2009). In the present analyses, the same species relationship for all the EFTFs was

78

A. Kamijyo et al. / Gene 555 (2015) 73–79

required, but the gene trees were not always the same as the trees for species relations; hence, we used the species tree for all EFTF phylogenetic analyses and achieved the same topology for all of the trees. We used the MEGA5 program with the option of “Analyze User Tree by Maximum Likelihood” (Tamura et al., 2011) for the calculation (Supplemental Fig. 1). We applied the Jones–Taylor–Thornton (JTT) model and uniform evolutionary rates. We then estimated the ancestral sequences of each node in the phylogenetic tree using GASP software with the default settings (Edwards, 2004). For a reference for the evolutionary rates of EFTF proteins, we also searched for pbx1 sequences from the databases. Pbx1 has a homeodomain and is involved in cell differentiation (Huang et al., 2009), as are EFTF proteins, and pbx1 can, therefore, be a reference for measuring the specific differences in eye related proteins.

Acknowledgments This work was supported by the Platform for Drug Discovery, Informatics, and Structural Life Science from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan to KY, and the Program to Disseminate Tenure-Tracking System of the MEXT, and the Grant-in-Aid for Young Scientists (B) (Grant number: 24770002) to AO. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2014.10.003. References

4.2. Three-dimensional structure estimation The three-dimensional structures of EFTF proteins were predicted using the comparative modeling method. A template 3D structure for each EFTF ancestral sequence was searched for in the Protein Data Bank (PDB) (Berman et al., 2003) using BLAST with default parameters. The template structures for each EFTF protein were as follows: Drosophila paired protein (PDB ID: 1fjl) for otx2, human tbx3 (1h6f) for tbx3, human pax6 paired domain (6pax) for the N-terminal side of the DNA-binding domains of rx1 and pax6, Drosophila aristaless and clawless homeodomains (3a01) for the C-terminal side of the DNA-binding domains of rx1, pax6, and lhx2, human pbx1 (1b72) for six3/six6, mouse Lhx4 (3mmk) for the N-terminal domain of lhx2, and human retinoic acid receptor RxR-alpha (3dzy) for nr2e1. The template-target alignment was conducted with ALAdeGAP (Hijikata et al., 2011), and the MODELLER program performed the coordinate building (Sánchez and Sali, 1997). The modeled structures with the optimum DOPE energy were selected as the predicted structures.

4.3. Measures for evolutionary differences in protein 3D structure The evolutionary distance was measured. Amino acid residues were classified into six different groups: aliphatic (I, V, L, and M), aromatic (F, W, and Y), cysteine, small (Pro, Gly, Ala, Thr, and Ser), acidic (Glu, Asp, Gln, and Asn), and basic (Lys, Arg, and His). The distance R among these groups was defined based on the differences of the chemical properties of the groups in the range from one to four and the differences in the corresponding residues between the ancestral sequence and the current sequence. A pair of the same amino acid residues was scored zero, and a substitution within the group was scored one. A score of five was assigned to a gap. The coloring Vi on 3D structure of the residue at position i was  N k then based onV i ¼ 1 N ∑ Rðai ; bi Þ2, where N is the number of the current k¼1

sequences, ai is the residue of the ancestral sequence at position i, bi is that of the current sequence, and k is the number given to each sequence. Residue i in the structure was colored in accordance with the value of V from blue (small V) to red (large V).

Contributions AO proposed the concept. KY and AO designed the research. AK and KY conducted the analyses. All authors wrote the manuscript and prepared the figures. All authors reviewed the manuscript.

Competing financial interests The authors declare that they have no competing financial interests.

Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman, D.J., 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389. Bailey, T.J., El-Hodiri, H., Zhang, L., Shah, R., Mathers, P.H., Jamrich, M., 2004. Regulation of vertebrate eye development by Rx genes. Int. J. Dev. Biol. 48, 761–770. Bao, R., Friedrich, M., 2008. Fast co-evolution of sevenless and bride of sevenless in endopterygote insects. Dev. Genes Evol. 218, 215–220. Benson, D.A., Cavanaugh, M., Clark, K., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J., Sayers, E.W., 2013. GenBank. Nucleic Acids Res. 41, D36–D42. Berman, H.M., Henrick, K., Nakamura, H., 2003. Announcing the worldwide Protein Data Bank. Nat. Struct. Biol. 10, 980. Blair, J.E., 2009. Animals (Metazoa). In: Blair Hedges, S., Kumar, S. (Eds.), The Timetree of Life. Oxford University Press, pp. 223–230. Cheng, D., Xia, Q., Duan, J., Wei, L., Huang, C., Li, Z., Wang, G., Xiang, Z., 2008. Nuclear receptors in Bombyx mori: insights into genomic structure and developmental expression. Insect Biochem. Mol. Biol. 1–8. Chuang, J.C., Raymond, P.A., 2001. Zebrafish genes rx1 and rx2 help define the region of forebrain that gives rise to retina. Dev. Biol. 231 (1), 13–30. Datta, R.R., Cruickshank, T., Kumar, J.P., 2011. Differential selection within the Drosophila retinal determination network and evidence for functional divergence between paralog pairs. Evol. Dev. 13 (1), 58–71 (Jan-Feb). Edwards, R., 2004. GASP: Gapped Ancestral Sequence Prediction for Proteins. BMC Bioinformatics. Flicek, P., Ahmed, I., Amode, M.R., et al., 2012. Ensembl 2013. Nucleic Acids Res. 41, D48–D55. Gehring, W.J., Ikeo, K., 1999. Pax 6: mastering eye morphogenesis and eye evolution. Trends Genet. 15, 371–377. Halder, G., Callaerts, P., Gehring, W.J., 1995. New perspectives on eye evolution. Curr. Opin. Genet. Dev. 5, 602–609. Hedges, S.B., Dudley, J., Kumar, S., 2006. TimeTree: a public knowledge-base of divergence times among organisms. Bioinformatics 22, 2971–2972. Hijikata, A., Yura, K., Noguti, T., Go, M., 2011. Revisiting gap locations in amino acid sequence alignments and a proposal for a method to improve them by introducing solvent accessibility. Proteins 79, 1868–1877. Horton, A.C., Mahadevan, N.R., Minguillon, C., Osoegawa, K., Rokhsar, D.S., Ruvinsky, I., Jong, P.J., Logan, M.P., Gibson-Brown, J.J., 2008. Conservation of linkage and evolution of developmental function within the Tbx2/3/4/5 subfamily of T-box genes: implications for the origin of vertebrate limbs. Dev. Genes Evol. 218, 613–628. Huang, Y, Huang, K., Boskovic, G., Dementieva, Y., Denvir, J., Primerano, D.A., Zhu, G.Z., 2001. Proteomic and genomic analysis of PITX2 interacting and regulating networks. FEBS Lett. 18, 583 (4), 13–30. Kawakami, K., Sato, S., Ozaki, H., Ikeda, K., 2000. Six family genes—structure and function as transcription factors and their roles in development. Bioessays 22, 616–626. Larkin, M.A., Blackshields, G., Brown, N.P., et al., 2007. Clustal W and Clustal X version 2.0. Bioinformatics 23, 2947–2948. Mathers, P.H., Jamrich, M., 2000. Regulation of eye formation by the Rx and pax6 homeobox genes. Cell. Mol. Life Sci. 57, 186–194. Mathers, P.H., Grinberg, A., Mahon, K.A., Jamrich, M., 1997. The: Rx: homeobox gene is essential for vertebrate eye development: article: nature. Nature 387, 603–607. Nelson, S.M., Park, L., Stenkamp, D.L., 2009. Retinal homeobox 1 is required for retinal neurogenesis and photoreceptor differentiation in embryonic zebrafish. Dev. Biol. 328, 24. Oliver, G., Mailhos, A., Wehr, R., Copeland, N.G., Jenkins, N.A., Gruss, P., 1995. Six3, a murine homologue of the sine oculis gene, demarcates the most anterior border of the developing neural plate and is expressed during eye development. Development 121, 4045–4055. Pan, Y., Nekkalapudi, S., Kelly, L.E., El-Hodiri, H.M., 2006. The Rx-like homeobox gene (Rx-L) is necessary for normal photoreceptor development. Invest. Ophthalmol. Vis. Sci. 47, 4245. Pawlowski, J., Montoya-Burgos, J.I., Fahrni, J.F., Wüest, J., Zaninetti, L., 1996. Origin of the Mesozoa inferred from 18S rRNA gene sequences. Mol. Biol. Evol. 13, 1128–1132. Peterson, K.J., Lyons, J.B., Nowak, K.S., Takacs, C.M., Wargo, M.J., McPeek, M.A., 2004. Estimating metazoan divergence times with a molecular clock. Proc. Natl. Acad. Sci. U. S. A. 101, 6536. Piatigorsky, J., 2003. Gene sharing, lens crystallins and speculations on an eye/ear evolutionary relationship. Integr. Comp. Biol. 43, 492–499.

A. Kamijyo et al. / Gene 555 (2015) 73–79 Quiring, R., Walldorf, U., Kloter, U., Gehring, W.J., 1994. Homology of the eyeless gene of Drosophila to the small eye gene in mice and aniridia in humans. Science 265, 785–789. Rivera, A.S., Pankey, M.S., Plachetzki, D.C., Villacorta, C., Syme, A.E., Serb, J.M., Omilian, A.R., Oakley, T.H., 2010. Gene duplication and the origins of morphological complexity in pancrustacean eyes, a genomic approach. BMC Evol. Biol. 10, 123 (Apr 30). Salvini-Plawen, L., Mayr, E., 1977. On the evolution of photoreceptors and eyes. In: Hecht, M.K. (Ed.), Evolutionary Biology 10. Plenum, pp. 207–263. Sánchez, R., Sali, A., 1997. Evaluation of comparative protein structure modeling by MODELLER-3. Proteins (Suppl. 1), 50–58. Seo, H.C., Drivenes, Ellingsen S., Fjose, A., 1998. Expression of two zebrafish homologues of the murine Six3 gene demarcates the initial eye primordia. Mech. Dev. 73, 45–57. Sernagor, E., Eglen, S., Harris, W.A., Wong, R., 2006. Retinal development. Cambridge University Press, Cambridge. Srivastava, M., Larroux, C., Lu, D.R., Mohanty, K., Chapman, J., Degnan, B.M., Rokhsar, D.S., 2010. Early evolution of the LIM homeobox gene family. BMC Biol. 8, 4. Sun, M.G.F., Sikora, M., Costanzo, M., Boone, C., Kim, P.M., 2012. Network evolution: rewiring and signatures of conservation in signaling. PLoS Comput. Biol. 8 (3), e1002411.

79

Tamura, K.K., Peterson, D.D., Peterson, N.N., Stecher, G.G., Nei, M.M., Kumar, S.S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739. Tétreault, N., Champagne, M.-P., Bernier, G., 2009. The LIM homeobox transcription factor Lhx2 is required to specify the retina field and synergistically cooperates with Pax6 for Six6 trans-activation. Dev. Biol. 327, 541–550. Tucker, P., Laemle, L., Munson, A., Kanekar, S., Oliver, E.R., Brown, N., Schlecht, H., Vetter, M., Glaser, T., 2001. The eyeless mouse mutation (ey1) removes an alternative start codon from the Rx/rax homeobox gene. Genesis 31, 43–53. Wawersik, S., Maas, R.L., 2000. Vertebrate eye development as modeled in Drosophila. Hum. Mol. Genet. 9, 917–925. Wilson, V., 2002. The T-box Family. Genome Biol. Yu, R.T., Chiang, M.Y., Tanabe, T., Kobayashi, M., Yasuda, K., Evans, R.M., Umesono, K., 2000. The orphan nuclear receptor Tlx regulates Pax2 and is essential for vision. Proc. Natl. Acad. Sci. U. S. A. 97, 2621–2625. Zuber, M.E., 2003. Specification of the vertebrate eye by a network of eye field transcription factors. Development 130, 5155–5167.