HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Daiyasu, H. Articles by Toh, H. Search for Related Content PubMed PubMed Citation Articles by Daiyasu, H. Articles by Toh, H.

Hiromi Daiyasu^1,*,, Tomoko Ishikawa^2,, Kei-ichi Kuma¹, Shigenori Iwai³, Takeshi Todo² and Hiroyuki Toh¹

¹ Bioinformatics Centre, Institute for Chemical Research, Kyoto University, Gokasho, Uji, Kyoto 611-0011, Japan
² Radiation Biology Centre, Kyoto University, Yoshidakonoe-cho, Sakyo-ku, Kyoto 606-8501, Japan
³ Division of Chemistry, Graduate School of Engineering Science, Osaka University, 1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures References

 
A new type of cryptochrome, CRY-DASH, has been recently identified. The CRY-DASH proteins constitute the fifth subfamily of the photolyase/cryptochrome family. CRY-DASHs have been identified from Synechocystis sp. PCC 6803, Vibrio cholerae, and Arabidopsis thaliana. The Synechocystis CRY-DASH was the first cryptochrome identified from bacteria, and its biochemical features and tertiary structure have been extensively investigated. To determine how broadly the subfamily is distributed within living organisms, we searched for new CRY-DASH candidates within several databases. We found five sequences as new CRY-DASH candidates, which are derived from four marine bacteria and Neurospora crassa. We also found many CRY-DASH candidates from the EST databases, which included sequences from fish and amphibians. We cloned and sequenced the cDNAs of the zebrafish and Xenopus laevis candidates, based on the EST sequences. The proteins encoded by the two genes were purified and characterized. Both proteins contained folate and flavin cofactors, and have a weak DNA photolyase activity. A phylogenetic analysis revealed that the seven candidates actually belong to the new type of cryptochrome subfamily. This is the first report of the CRY-DASH members from vertebrates and fungi.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures References

 
Photolyases and cryptochromes constitute a family of flavoproteins, which are widely distributed within the three kingdoms, eubacteria, archaea and eukaryotes (Sancar 1994; Todo et al. 1996; Cashmore et al. 1999). This family has been classified into four subfamilies, based on molecular phylogenetic analyses and functions: class I cyclobutane pyrimidine dimer (CPD) photolyase, class II CPD photolyase, plant cryptochrome (CRY), and animal CRY including (6–4) photolyases (Cashmore et al. 1999; Todo 1999). The functions of the subfamilies differ from each other, despite their structural and sequence similarities.

CPD photolyase is an enzyme that repairs DNA damaged by UV light (Sancar 1990, 1994). The enzyme absorbs a blue light photon, and an electron on the cofactor flavin adenine dinucleotide (FAD) is excited by the energy of the photon. The excited electron is transferred to the CPD bound to the enzyme, and then the damaged DNA is repaired by cleaving the CPD. Therefore the mechanism is called photorepair. CPD photolyases are classified into two subfamilies, from the evolutionary point of view (Yasui et al. 1994; Kanai et al. 1997). The class I CPD photolyase subfamily has been mainly isolated from eubacteria and fungi (Sancar 1990, 1994; Yasui et al. 1994). The class II CPD photolyase subfamily, on the other hand, is widely distributed throughout all organisms (Todo et al. 1994; Kato et al. 1994; Yasui et al. 1994). The tertiary structures of the class I CPD photolyases have been determined by X-ray crystallographic studies (Miki et al. 1993; Park et al. 1995; Komori et al. 2001). The structure consists of two domains connected by a loop. The N-terminal domain is classified in the /ß class, while helices are abundant in the C-terminal domain, according to the structure classification database, SCOP (Murzin et al. 1995). The FAD cofactor is present at the bottom of the cavity in the C-terminal domain. The CPD is thought to be located in the cavity, to receive the excited electron from FAD.

Cryptochrome does not exhibit photorepair activity. The CRYs are roughly classified into two subfamilies, the animal CRYs and the plant CRYs. The two subfamilies are clearly distinguished in the phylogenetic tree, and the divergence of the two subfamilies is considered to be ancient (Kobayashi et al. 2000; Hitomi et al. 2000; Brudler et al. 2003). The plant CRY proteins were originally identified as blue light photoreceptors, which contribute to seedling growth and flowering season regulation (Ahmad & Cashmore 1993). They are widely distributed throughout plant species. The members of the animal CRY subfamily are distributed from insects to vertebrates, and involved in circadian rhythm (Griffin et al. 1999; Kume et al. 1999; Rosato et al. 2001). However, the mechanism of CRY on the regulation of circadian system differs between Drosophila and mice (Froy et al. 2002), these CRYs belong to the same animal CRY subfamily. The molecular mechanisms of CRY functions have remained elusive. CRY proteins show sequence similarity to the class I CPD photolyases, despite the absence of photorepair activity. The (6–4) photolyases, which selectively repair (6–4) photoproducts and have been isolated from higher eukaryotes (Todo 1999), are derived from the animal CRY subfamily (Todo et al. 1996).

We recently identified a novel subfamily within the photolyase/cryptochrome family (Hitomi et al. 2000; Brudler et al. 2003). This novel subfamily is relatively closer to the animal CRY subfamily, rather than the plant CRY subfamily and the CPD photolyase subfamilies. Due to this evolutionary relationship, the members of the new group were considered to be novel cryptochromes (Hitomi et al. 2000; Brudler et al. 2003). The subfamily consists of proteins derived from Synechocystis sp. PCC 6803, Vibrio cholerae, and Arabidopsis thaliana (A. thaliana), and is called the cryptochrome DASH subfamily (Brudler et al. 2003). In this study, we call the member of the new subfamily as CRY-DASH to distinguish from the other members of cryptochrome subfamilies. Another interesting point is that the subfamily includes two bacterial proteins. Thus, the proteins are considered to be the first bacterial cryptochromes. According to this hypothesis, the biochemical functions of the Synechocystis CRY-DASH were investigated (Hitomi et al. 2000; Brudler et al. 2003). The Synechocystis protein showed no photorepair activity in vitro, although a weak CPD photorepair activity was detected in vivo (Hitomi et al. 2000). Instead, it showed nonspecific DNA binding activity, suggesting its function as transcriptional regulator (Brudler et al. 2003). In fact, microarray analyses revealed that the disruption of the Synechocystis CRY-DASH gene affects the expression of several genes (Brudler et al. 2003). Brudler et al. (2003) solved the crystal structure of the Synechocystis CRY-DASH, which shows high similarity to the structures of class I CPD photolyases. The presence of the FAD in the Synechocystis protein was also revealed by the X-ray crystallographic study. A recent study revealed that the A. thaliana CRY-DASH lacks photorepair activity, but has a nonspecific DNA binding activity (Kleine et al. 2003). In addition, Kleine et al. (2003) reported that the A. thaliana protein was able to bind FAD. Thus, the A. thaliana CRY-DASH has similar features to the Synechocystis CRY-DASH. However, the physiological processes in which the CRY-DASHs actually function are not known.

In our previous study (Brudler et al. 2003), only three proteins were included in the CRY-DASH subfamily, as described above. As compared to the other subfamilies of the photolyase/cryptochrome family, the size of this subfamily is small, and the distribution seems to be restricted. Therefore, this subfamily may have been generated by rare evolutionary events, such as horizontal gene transfer. After the submission of the previous work, large amounts of sequence data, including complete genome sequences, have become available. To examine how widely the members of the CRY-DASH subfamily are actually distributed within living organisms, we tried database searching. Our searches yielded five new CRY-DASH candidates with sources from marine bacteria and Neurospora crassa. We also searched through several EST databases, and obtained many CRY-DASH candidates. Unexpectedly, the EST candidates included sequences derived from several fish and Xenopus laevis. To confirm this observation, we cloned and sequenced the cDNAs corresponding to the EST sequences from zebrafish and Xenopus. A phylogenetic analysis revealed that all of the candidates obtained in this study belong to the CRY-DASH subfamily. Subsequent characterization of the purified proteins showed that the two vertebrate CRY-DASHs share the spectroscopic properties with the bacterial CRY-DASH and have a weak CPD photolyase activity.

	Results

Top Abstract Introduction Results Discussion Experimental procedures References

 
Database searches

Several candidates of new CRY-DASH subfamily members were obtained by searching through the nr protein database of NCBI and some genome databases. The candidates included several bacterial ORF products. In addition to the previously reported bacterial CRY-DASHs (Brudler et al. 2003), four ORF products derived from marine bacteria, such as algicidal bacteria (Vibrio parahaemolyticus, Cytophaga hutchinsonii), a cyanobacterium (Trichodesmium erythraeum) and a planctomycete (Pirellula sp.) were obtained. These ORF products either had no annotations or were regarded as photolyases according to the sequence similarities in the databases. Two amino acid sequences were obtained from A. thaliana. One of them was the CRY-DASH protein itself (NCB accession No. gi: 28971609) (Brudler et al. 2003), whereas the other was an ORF product (gi: 3319288), which was annotated as a photolyase/blue light photoreceptor protein. The sequence identity between two proteins is 40.0%. A deletion of about 100 amino acid residues was observed at the C-terminal region of the protein. An ORF product derived from Neurospora crassa (gi: 28927535) was also detected as a new CRY-DASH. The ORF product was found within the complete genome of N. crassa, which was recently determined (Galagan et al. 2003). The protein is simply described as a hypothetical protein in the database, whereas Lee et al. (2003) referred to it as a cryptochrome orthologue, based on the sequence similarity. However, there was no description of a cryptochrome subfamily for the ORF product (Lee et al. 2003). The ORF product from N. crassa has a C-terminal extension of about 200 amino acid residues, with an abundance of Gly residues. The region did not show significant sequence similarity to any other proteins. No CRY-DASH candidate was detected from the other fungi, Saccharomyces cerevisiae, Schizosaccharomyces pombe or Encephalitozoon cuniculi, with complete genome data. In addition, no animal or plant CRY counterpart was detected in these fungal genomes. During this analysis, we noticed that some organisms have no genes encoding members of the photolyase/cryptochrome family in their genomes. We examined the complete genomes of 118 eubacteria, 16 archaea and 7 eukaryotes in this analysis. Most of them encoded several photolyase and cryptochrome homologues. However, no homologue of the protein family was detected within the Plasmodium falciparum (Gardner et al. 2002) and Caenorhabditis elegans (Rachael et al. 1998) or within the whole genome shotgun data of Ciona intestinalis (Dehal et al. 2002).

The same database search procedure was applied to several EST databases, which yielded many novel CRY-DASH candidates. The EST sequences thus obtained did not cover the entire coding regions. Instead, these sequences showed high sequence similarities to the N- or the C-terminal regions of the known CRY-DASHs. We obtained many CRY-DASH candidates from the EST databases of a wide range of plants, including mosses, gymnosperms, monocots and dicots. An EST sequence from a rhodophata, Porphyra yezoensis (gi: 8586831), was also obtained as a novel CRY-DASH candidate. Unexpectedly, we found several vertebrate EST sequences as CRY-DASH candidates. The vertebrate sources of the EST sequences were Danio rerio, Fugu rubripes, Oryzias latipes, and Xenopus laevis. We made a multiple alignment of the N-terminal regions of the photolyase/cryptochrome family, together with the CRY-DASH candidates obtained from the nr protein database, the genome databases and the EST databases. Based on an alignment of about 110 amino acid residues, a preliminary phylogenetic tree was constructed by the neighbour-joining (NJ) method (data are not shown here, but the alignment and the tree are available at the web site, http://timpani.genome.ad.jp/~dash/). In the preliminary tree, all of the sequences of the CRY-DASH candidates formed a large cluster, together with the known members of the CRY-DASH subfamily (Brudler et al. 2003), and the cluster is distinguishable from the animal CRYs and the plant CRYs.

Isolation of the vertebrate CRY-DASH candidates

The database search results and the topology of the preliminary tree supported the idea that the candidates obtained in this study are new CRY-DASH subfamily members. Especially, the presence of CRY-DASH members in vertebrates would be an interesting and important finding, if the detected sequences actually belong to the CRY-DASH subfamily. To confirm this point, the cDNAs of the CRY-DASH candidates from Danio rerio and Xenopus laevis were cloned, based on the EST sequence information. To obtain the full-length CRY-DASH cDNAs, we designed primers based on the nucleotide sequences of the ESTs for nested 5' and 3' RACE reactions. Both the 5' and 3' RACE reactions resulted in multiple clones, which were sequenced. An alignment of the nucleotide sequences with the ESTs produced a contiguous sequence. This zebrafish 5' RACE product sequence appeared to include the full 5' end of the translated coding region, as indicated by the presence of an in-frame stop codon on the 5' side of the initiating methionine. It also encompassed the 3' end of the coding region, as revealed by the presence of the poly(A) tail. The DDBJ accession numbers of the cDNAs from Xenopus and zebrafish are AB120760 and AB120759, respectively. Then amino acid residues were deduced from the determined cDNA sequences. The Xenopus protein consisted of 524 amino acid residues, while the zebrafish protein was of 521 amino acid residues in length.

Phylogenetic relationships

We obtained the full amino acid sequences of seven novel CRY-DASH candidates. Five of them were detected by database searching, and included the ORF products from V. parahaemolyticus, C. hutchinsonii, T. erythraeum, Pirellula sp., and N. crassa. The rest of them were derived from Xenopus and zebrafish, which were cloned and sequenced as described above. The evolutionary relationship of the seven candidates to the members of the photolyase/cryptochrome family was investigated. First, we constructed a multiple alignment, which consisted of the members from all of the subfamilies. Seven CRY-DASH sequence candidates were included in the alignment. However, no EST sequence was used in this study. The PHR2 protein from A. thaliana was excluded from this phylogenetic analysis, because of the deletion in the C-terminal region, as described above. Due to its ancient divergence from the other subfamily (Kanai et al. 1997), the class II photolyase subfamily was not included in this study. The alignment is available at the web site http://timpani.genome.ad.jp/~dash/. Then, an unrooted NJ tree of the photolyase/cryptochrome family was constructed (see Fig. 1). According to the tree topology, the aligned sequences were classified into four groups, which correspond to the animal CRY/(6–4) photolyase subfamily, the CRY-DASH subfamily, the plant CRY subfamily, and the class I CPD photolyase subfamily, respectively. The classification was basically the same as that in the previous study (Hitomi et al. 2000; Brudler et al. 2003). The first three groups were rooted at the nodes A, B and C with 100% bootstrap probabilities, while the root of the class I CPD photolyase subfamily was obscure in this tree. This tree suggests that the CRY-DASH subfamily can be distinguished from the other subfamilies with statistical significance. As reported previously (Hitomi et al. 2000; Brudler et al. 2003), the CRY-DASH subfamily was more closely related to the animal CRY subfamily, rather than the plant CRY subfamily and the class I CPD photolyase subfamily. We also examined the evolutionary relationship of the four subfamilies by the maximum likelihood (ML) method. The results of the ML analysis supported the relationship in the NJ tree (data not shown, but available at http://timpani.genome.ad.jp/~dash/).

View larger version (42K): [in this window] [in a new window]   Figure 1  An unrooted phylogenetic tree of the photolyase/cryptochrome family. Each sequence is indicated by the GI number from NCBI. The source name is shown next to the GI number. A capital letter after each source name indicates the taxonomic category of the source. (V) vertebrates (B) eubacteria (A) archaea (I) insects (P) plants (S) sponges (F) fungi. When the bootstrap probability for the clustering at a node is greater than 70.0% and the node is critical for discussion, the probability is written near the node. The newly detected CRY-DASH proteins are coloured blue. The sequences determined in this study are coloured red.

Characterization of the vertebrate CRY-DASH members

To determine the spectroscopic and biochemical properties of the vertebrate CRY-DASHs, we cloned the zebrafish and Xenopus cry-dash genes. Both proteins were expressed at high level and readily purified. Figure 2A shows the purified proteins as revealed by SDS-PAGE. The proteins were > 80% pure and therefore appropriate for spectroscopic and enzymatic analyses.

View larger version (32K): [in this window] [in a new window]   Figure 2  Absorption and fluorescence spectra of zCRY-DASH. (a) Purification of zebrafish and Xenopus CRY-DASH proteins. Coomassie Blue stained SDS-PAGE gel (10% polyacrylamide) of purified photolyase/cryptochrome protein family was used in this study. Molecular weight marker (lane 1), E. coli CPD photolyase (lane 2), GST-Xl6–4 photolyase (lane 3), GST-zCRY-DASH (lane 4), GST-XlCRY-DASH (lane 5) and GST-Dm CPD photolyase (lane 6). (b) Absorption spectra of zCRY-DASH. The inset shows an expanded scale of the absorption in the 280–600 nm range. (c) Fluorescence spectra of zCRY-DASH. Folate fluorescence using 380 nm excitation for the emission spectrum (-----) and 460 nm emission for the excitation spectrum (——). (d) Fluorescence spectra of zCRY-DASH. Flavin fluorescence using 470 nm excitation for the emission spectrum (-----) and 520 nm emission for the excitation spectrum (——). (e) Absorption and fluorescence emission spectra of the chromophore released from zCRY-DASH by heat denaturation. (f) Fluorescence emission spectra of the chromophore released from zCRY-DASH. Excitation was at 450 nm. The emission spectra were taken at pH 3.0 (——) and pH 8.0 (-----).

To get further evidence that zCRY-DASH contains both MTHF and FAD, fluorescence spectrum of zCRY-DASH was determined (Fig. 2C,D). Typically, enzyme-bound MTHF has a fluorescence emission maximum at 460–480 nm with an excitation maximum at 360–390 nm. Flavin has an emission maximum at 502–520 nm with excitation maxima at 370 nm and 440 nm. Figure 2C shows the fluorescence spectra of zCRY-DASH. When excitation spectrum was recorded for emission at 460 nm, a peak with max = 385 nm was obtained (Fig. 2C). Excitation at 380 nm gave a fluorescence spectrum with a peak at 475 nm (Fig. 2C), consistent with MTHF being the predominant chromophore in near-UV in zCRY-DASH. Excitation of zCRY-DASH with 470 nm yielded a typical flavin emission spectrum (max = 520) (Fig. 2D), consistent with the presence of FAD. However, when the excitation spectrum was determined for 520 nm emission, a major peak at 385 nm and a minor one around 460 nm (Fig. 2D) instead of the 370 nm and 440 nm peaks typical of flavin absorbance. To confirm further the presence of FAD in zCRY-DASH, the proteins were heat-denatured at neutral pH, and, following removal of the denatured protein by centrifugation, the absorption spectrum was recorded (Fig. 2E). The absorption spectrum of the released chromophore showed a typical flavin absorbance with maxima at 350 nm and 450 nm. The fluorescence emission spectra of the released chromophore were taken at different pH values. Excitation at 450 nm resulted in fluorescence emission with a maximum at 530 nm. The fluorescence emission was pH dependent with fivefold-higher fluorescence at pH 3.0 than at pH 8.0 (Fig. 2F). This pH dependency of fluorescence emission is typical for FAD, confirming the presence of FAD in zCRY-DASH. Therefore, the low emission peak at 440 nm of the holoenzyme (Fig. 2D) suggests that, in zCRY-DASH, MTHF is the main chromophore responsible for excitation of FAD, which then decays by fluorescence at 520 nm.

The stoichiometries of the two cofactors relative to the apoenzyme were roughly calculated. The stoichiometry of FAD to apoenzyme is estimated from the ratio of the 440 nm absorption of the denatured enzyme to the absorption of the holoenzyme at 280 nm. The concentration of MTHF is estimated from the absorption at 380 nm of the holoenzyme. The stoichiometries of the cofactors were found to depend on the growth conditions of the cultures. The following values were obtained as the average of two different purification experiments; FAD:MTHF:apoenzyme = 0.7 : 0.8 : 1.0.

Two different activities were known for the photolyase/cryptochrome family; the repressor activity of CLOCK:BMAL mediated transcription and the light-dependent repair activity of UV-induced DNA damage. We examined whether zebrafish and Xenopus CRY-DASH have these activity.

The CLOCK:BMAL heterodimer drives gene expression when the gene is under control of the E-box bearing promoter. Animal CRY inhibits the CLOCK:BMAL-mediated transcription. We tested whether CRY-DASH also can inhibit the CLOCK:BMAL-mediated transcription in zebrafish BRF41 cell (Fig. 3). Although zCRY1a, a member of zebrafish animal-type CRY, inhibits it completely, zebrafish CRY-DASH showed no inhibition. Xenopus CRY-DASH also showed no inhibition (data not shown), demonstrating that vertebrate CRY-DASH and animal-type CRY are functionally different.

View larger version (15K): [in this window] [in a new window]   Figure 3  Transcriptional repressor activity of zCRY1a and zCRY-DASH, determined by the luciferase reporter gene assay. Zebrafish BRF41 cells were transfected with the pzcry3-luc reporter plasmid (50 ng) and the expression vectors shown (200 ng each). Transactivations of the reporter plasmid were examined. Values are mean ± S.E. of three independent experiments. In each experiment, the luciferase activity of the zCLOCK:zBMAL3-containing sample was taken as 100%.

View larger version (16K): [in this window] [in a new window]   Figure 4  Effects of zebrafish and Xenopus CRY-DASH on UV killing and photoreactivation of E. coli. Effects of photoreactivation on the survival of UV-irradiated E. coli SY2 (a) or SY32/pRT2 (b) cells carrying the pGEX4T-2 vector (triangle), pGEX-Dm-CPD (square in a), pGEX-Xl6–4 (square in b), pGEX-zCRY-DASH (circle), or pGEX-Xl-CRY-DASH (diamond). After UV irradiation, E. coli cells plated on LB plates were either kept in darkness (closed symbols) or illuminated with a fluorescent lamp for 1 h (open symbols). The plates were then incubated at 37 °C for 16 h, after which colonies were counted and relative survival was determined. The bars indicate standard errors of three independent experiments.

View larger version (51K): [in this window] [in a new window]   Figure 5  DNA binding and photolyase assays with zebrafish and Xenopus CRY-DASH. (a) The 49 mer oligonucleotide sequence containing a photoproduct. (b) Gel retardation analysis showing CPD-specific binding of zCRY-DASH protein. The 49 bp duplex oligonucleotides (1 nM) non-damaged or having either of CPD, (6–4) photoproduct, or the DEWAR isomer were mixed with 0.1 µg of purified protein then electrophoresed in a non-denaturing gel. (c) Photolyase assay showing the CPD specific repair activity of zCRY-DASH and Xl-CRY-DASH. The 49 bp duplex oligonucleotide (10 nM) non-damaged or having either of CPD, (6–4) photoproduct, or the DEWAR isomer were mixed with 1 µg of purified protein, illuminated with a fluorescent lamp for 1 h on ice, then treated with phenol/chloroform. Samples were treated with MseI then separated in a 10% polyacrylamide gel.

The increased member of CRY-DASH subfamily members made it possible to examine the residue conservation within the subfamily. The sequence alignment of the CRY-DASH subfamily is shown in Fig. 6. We identified the invariant residues from the alignment. About one-third of the invariant sites of the CRY-DASH subfamily were conserved over the photolyase/cryptochrome family. The residues corresponding to the invariant sites were mapped on the tertiary structure of Synechocystis CRY-DASH (Brudler et al. 2003) (Fig. 7). The CRY-DASH structure shares a common folding pattern with those of the photolyases (Miki et al. 1993; Park et al. 1995; Komori et al. 2001), in spite of their functional differences. The mapped residues on the structure were classified into five groups, according to the spatial proximity.

View larger version (125K): [in this window] [in a new window]   Figure 6  A multiple alignment of the CRY-DASH subfamily. The measure over the alignment indicates the site number. A hyphen in the alignment means a gap for an insertion and/or a deletion. The sites corresponding to the -helices and the ß-strands of the Synechocystis CRY-DASH structure are indicated by red and blue bars, respectively. The FAD binding sites are indicated by white colour. The alignment sites that are strongly conserved over the photolyase/cryptochrome family are enclosed by red squares. As discussed in the text, the invariant sites within the CRY-DASH subfamily are classified into five groups, based on the mapping on the structure. The corresponding sites are separately coloured. (see Figure 7 for details).

View larger version (59K): [in this window] [in a new window]   Figure 7  Mapping of the residues corresponding to the invariant sites within the CRY-DASH subfamily on the Synechocystis CRY-DASH structure.The main chains of the protein structure are indicated with white tubes. The CPK model of FAD is coloured blue. The invariant residues belonging to the 1st, 2nd, 3rd, 4th and 5th groups are coloured light-blue, orange, yellow, pink and green, respectively.

The invariant residues of the second group were found at the interface region between the N- and the C-terminal domains, and on the loop connecting the two domains. The conservation of residues may be involved in maintaining the proper arrangement of the two domains for the function.

As described above, in the Synechocystis CRY-DASH protein, a FAD is present at the bottom of the cavity, where the cofactor is non-covalently bound to the C-terminal domain. The conserved residues of the third group were present in the C-terminal domain, and most of them occupied the positions that directly interact with FAD. Most of the sites were conserved not only within the CRY-DASH subfamily, but also throughout the photolyase/cryptochrome family. These data are consistent with the spectroscopic results shown in Fig. 2. Brudler et al. (2003) reported that an Asn residue, at the alignment site 523 in Fig. 6, is also involved in FAD binding in the Synechocystis CRY-DASH. As shown in the figure, this Asn was invariant within the CRY-DASH subfamily. Brudler et al. (2003) also reported that two Trp residues involved in FAD binding in the E. coli photolyase are replaced with other residues in the Synechocystis CRY-DASH (alignment sites 389 and 480). As shown in Fig. 6, the Trp residues were also replaced with other residues in all of the members of the CRY-DASH subfamily. The residue conservation pattern suggests that the members of the CRY-DASH subfamily bound FAD in the same manner as the Synechocystis CRY-DASH.

The conserved residues of the fourth group were clustered on the surface of the tertiary structure, and surrounded the cavity including the FAD. The Synechocystis CRY-DASH (Brudler et al. 2003) and the A. thaliana CRY-DASH (Kleine et al. 2003) have a nonspecific DNA binding activity. As described above, zebrafish CRY-DASH can especially bind the CPD-contained DNA (Fig. 5B). For the CPD photolyase and the Synechocystis CRY-DASH, the positively charged Arg residues corresponding to the alignment sites 331, 396, 484, 486 and 539 could be involved in DNA recognition (Brudler et al. 2003). These five residues were completely conserved within the CRY-DASH subfamily. In addition, the invariant positively charged residues at the alignment sites 328, 535 and 545, which belonged to the fourth group, were also located near the entrance of the cavity in the Synechocystis CRY-DASH structure. Therefore, these eight positively charged residues are considered to play a role in DNA recognition.

Three Trp residues may constitute an electron transfer chain from the protein surface to the FAD cofactor in the photolyase structure (Li & Sancar 1990; Aubert et al. 2000). All of them were conserved in the Synechocystis CRY-DASH (Brudler et al. 2003), which corresponded to the alignment sites 488, 501 and 504 in Fig. 6. The Trp residues at sites 501 and 504 were invariant within the CRY-DASH subfamily, but the Trp at site 448 was present only in six members of the CRY-DASH subfamily. This observation suggests that at least six members of the CRY-DASH subfamily may have redox ability, as indicated for the Synechocystis CRY-DASH by Brudler et al. (2003). The conserved residues of the fifth group were arranged to surround the electron transfer chain.

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures References

 
The database searches and the phylogenetic analyses suggested that more living organisms have CRY-DASH proteins than expected previously. The wide distribution of the CRY-DASH subfamily in living organisms suggests that the members are involved in essential functions, although their actual function remains unknown. One of the surprising results was discovery of novel CRY-DASH subfamily members from zebrafish and Xenopus. Both vertebrate CRY-DASHs were similar to other CRY-DASH members not only in the primary structure but also in the spectroscopic and biochemical properties. Among 10 proteins identified as the CRY-DASH members in our phylogenetic analysis (Fig. 1), two bacterial CRY-DASHs, Synechocystis CRY-DASH (Hitomi et al. 2000) and Vibrio cholerae CRY-DASH (Worthington et al. 2003), have been well characterized in their properties in vitro. The Vibrio CRY-DASH gene was designated as Vccry1 in the report by Worthington et al. (2003), but the gene product was identified as a member of CRY-DASH subfamily in our previous study (Brudler et al. 2003). Both the two vertebrate CRY-DASHs and the two bacterial CRY-DASHs show a striking similarity in their absorption spectrum. All recombinant protein expressed in E. coli exhibited a major peak at 380 nm and a minor absorbance beyond 400 nm, suggesting the presence of MTHF chromophore and flavin in the reduced FADH₂ form. In addition to the spectroscopic properties, both vertebrate CRY-DASHs and Synechocystis CRY-DASH conferred light-dependent UV resistance on photolyase-deficient E. coli cells (Fig. 4A and Hitomi et al. 2000). In vitro analysis showed that the conferred resistance was due to the CPD photolyase activity of the vertebrate CRY-DASH, although their activity is very low (Fig. 5C). Vibrio CRY-DASH-dependent increase of survival was not detected in Vibrio cells (Worthington et al. 2003). In the report, however, for the detection of photolyase activity of Vibrio CRY-DASH, the excision repair proficient Vibrio strain and cell free extract, instead of the purified protein, was used for in vivo assay. It might make the weak photolyase activity undetectable. Our results, together with the previous observations, suggest that a weak CPD photolyase activity is widely observed in CRY-DASH subfamily, although the CPD photolyase activity of the Vibrio CRY-DASH (VcCry1) is unclear and that of the Arabidopsis CRY-DASH was not detected (Kleine et al. 2003). The ancestral protein of photolyase/cryptochrome family is suggested to be CPD photolyase (Kanai et al. 1997). The CRY-DASH subfamily might also diverge from the ancestral CPD photolyase and be still keeping photolyase activity as the evolutionary remains, while other CRY subfamilies, the animal CRY and the plant CRY, have lost the activity. However, the CPD photolyase activity was too weak to regard the activity as a proper function of CRY-DASH subfamily. In addition, almost all organisms in which the CRY-DASH proteins have been identified so far possess another CPD photolyase gene. These genes show higher conservation in the primarily structure with the known CPD photolyases and some of the proteins have much stronger CPD photolyase activity than the CRY-DASH proteins (Hitomi et al. 2000; Kleine et al. 2003; Worthington et al. 2003). Therefore, it is likely that CRY-DASH has another regular activity other than remaining CPD photolyase activity, although we cannot exclude the possibility that CRY-DASH has subsidiary role as the second CPD photolyase.

Synechocystis and Arabidopsis CRY-DASH have a nonspecific DNA binding activity, which was predicted to be indispensable for their function (Brudler et al. 2003; Kleine et al. 2003). However, in the present study we cannot detect such activity in zebrafish and Xenopus CRY-DASH (Fig. 5B, lanes 4 and 5). On the other hand, CPD-binding activity of Xenopus CRY-DASH was also undetectable (Fig. 5B, lane 10), while it showed a CPD-photorepair activity (Fig. 5C, lane 6). These results suggest that in the present gel retardation assay only a stable binding was detected and an unstable binding was undetectable by unknown reason (Fig. 5B). Further alteration of the experimental condition might clarify this point.

As described above, we searched for new CRY-DASH subfamily members within of the several complete genomes available today. However, no CRY-DASH candidate was detected from other animals, such as humans, mice, Drosophila melanogaster, and C. elegans. Thus, CRY-DASH protein distribution in animals seems to be restricted to some aquatic organisms, such as fish and amphibians. The EST data analysis suggested that the CRY-DASH subfamily members are widely distributed within the fish. The evolutionary origin of the restricted distribution, for example, lineage specific gene loss or horizontal gene transfer, will await the further accumulation of animal genome data.

As shown in Fig. 1, the A. thaliana CRY-DASH was close to the Synechocystis CRY-DASH and the T. erythraeum ORF product. This suggests that the plant CRY-DASHs may have originated from chloroplasts or photosynthetic bacteria. As for A. thaliana, the CRY-DASH gene is encoded in the nuclear genome. Therefore, it is thought that an ancestral CRY-DASH gene was transferred from the genome of a chloroplast or photosynthetic symbiont to the nuclear genome of an ancestral plant, as suggested by Kleine et al. (2003). Many cases of gene transfer from plant organelles, such as chloroplasts and mitochondria, to host nuclei have been reported (Martin et al. 2002). As mentioned above, the biological function of the CRY-DASH proteins is not known. However, if the gene for the CRY-DASH protein was transferred from the chloroplast to the nucleus in plants, then the product may be involved in a function related to chloroplast regulation. Actually, Kleine et al. (2003) reported that the N-terminal region of the A. thaliana CRY-DASH (about 40 amino acid residues) acts as a signal sequence to mediate the plant CRY-DASH import into chloroplasts and mitochondria. However, the signal sequence did not show significant sequence similarity to the N-terminal regions of PHR2 and the amino acid sequences deduced from the EST sequences of putative plant CRY-DASHs. We found many CRY-DASH candidates in the plant EST data. The preliminary phylogenetic analysis supported the possibility that the candidates actually belong to the CRY-DASH subfamily. We cannot make any definite statement about the assignment of the candidates at this stage. However, the possibility that plants have CRY-DASHs as a second cryptochrome should be considered, when the CRY functions in plants are examined.

In this study, some ORF products from marine bacteria were identified as CRY-DASH proteins. The sources were marine algicidal microorganisms, algae, and planctomyces. Algicidal microorganisms and planctomyces may have acquired the CRY-DASH gene by horizontal gene transfer, after ingesting algae by endocytosis.

We also detected a new CRY-DASH subfamily member derived from N. crassa. This fungus exhibits the circadian rhythm, and many of the proteins that functions in its oscillatory system have been studied (Loros & Dunlap 2001). However, the cryptochrome homologue of N. crassa was not identified until the genome project was completed (Galagan et al. 2003). Both animals and N. crassa generate circadian rhythm with a common basic regulatory system, an auto-regulatory negative feedback loop. Although the core proteins constituting the feedback loop differ between animals and N. crassa (Van Gelder et al. 2003), some of the N. crassa proteins involved in the circadian regulatory systems, such as wc-1 and wc-2, have the sequence homologous domain (PAS domain) that is commonly observed in the clock proteins from many organisms (Dunlap 1999). Some PAS domain-including proteins interact with the animal CRYs in mice and zebrafish (Griffin et al. 1999; Kume et al. 1999; Ishikawa et al. 2002; Hirayama et al. 2003). At this stage, neither the function of the Neurospora CRY-DASH nor the involve-ment of the protein in circadian rhythm is known. However, the possibility that this novel CRY-DASH may regulate the fungal circadian rhythm should be considered, when the function is examined.

Another interesting question is what is the function of vertebrate CRY-DASH. Animal CRYs act as transcriptional repressor, which is an essential step in the circadian system. On the analogy of animal CRYs it is likely that vertebrate CRY-DASHs also function as the regulator of circadian rhythm. However, even if vertebrate CRY-DASHs are participated in the regulation of circadian rhythm, their role must be different from that of animal CRYs, because the vertebrate CRY-DASHs do not show any transcriptional repressor activity. The presence of the second chromophore (MTHF) and efficient energy transfer from MTHF to FAD in CRY-DASH (Worthington et al. 2003) suggest the function as a photoreceptor. Analyses of CRY-DASH members have just begun. However, they will reveal novel functional aspects of the photolyase/cryptochrome family, and may shed new light on the regulation of circadian rhythm.

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures References

 
Sequence and structural data

The BLAST program (Altschul et al. 1997) was used to search for novel CRY-DASH protein candidates in the nr databases at NCBI (http://www.ncbi.nlm.nih.gov/) and GenomeNet (http://www.genome.ad.jp/dbget/dbget.links.html). The amino acid sequences of Synechocystis sp. PCC 6803 CRY-DASH (gi: 28374085) and A. thaliana CRY-DASH (gi: 28971609) were used as the queries. For the sequence selections, we used the list of the sequences, which were sorted with their scores to the query. The sequences were selected from the top of the list, one after another, until a sequence belonging to a subfamily other than the CRY-DASH subfamily was found. Then, each selected sequence was used as a query for database searching through the nr databases. When the top positions in the search output were occupied by the known CRY-DASHs, the selected sequence was regarded as a novel CRY-DASH candidate. Several genome databases at GOLD (http://wit.integratedgenomics.com/GOLD/) were also searched with BLAST for novel CRY-DASH candidates. The CRY-DASHs from Synechocystis and Arabidopsis were again used as the queries. The second database search was performed at the NCBI. The EST databases at NCBI and GenomeNet were also searched, with the same procedure as described above. BLAST was used for the first database search. Whole genome shotgun data, available at DOE (http://www.jgi.doe.gov/), were also used in our quest CRY-DASH candidates.

The NCB accession GI numbers for the sequence data used in this study are shown in Fig. 1. In addition to the amino acid sequence data, the coordinates for a CRY-DASH derived from Synechocystis (pdb code: 1NP7 [PDB] ) and three photolyases (1QNF [PDB] from Synechococcus sp. PCC 6301, 1DNP [PDB] from E. coli, and 1IQR [PDB] from Thermus thermophilus) were used for the structural alignment and the mapping of the conserved residues.

Sequence and structural alignments

The photolyase/cryptochrome family is classified into five subfamilies (Hitomi et al. 2000; Brudler et al. 2003): animal CRY/(6–4) photolyase, CRY-DASH, plant CRY, class I CPD photolyase, and class II CPD photolyase. The class II CPD photolyases were not included in this study, because of their ancient divergence (Kanai et al. 1997). Only some of the detected members from the class I CPD photolyase subfamily and the plant CRY subfamily were used for the analysis, because of the large number of members. Therefore, several representative sequences were selected from each subfamily. The sequences were divided into four groups, corresponding to the subfamilies. Then, a multiple alignment of each group was constructed with the alignment software, CLUSTAL W 1.81 (Thompson et al. 1994). Finally, the four multiple alignments were piled up using the profile alignment function in CLUSTAL W. The alignment thus obtained was slightly modified by visual inspection, to keep the gaps from interrupting the secondary structure elements as much as possible. A structural alignment among the CRY-DASH and three photolyases was used for the modification. The structural alignment was constructed with a program developed by the authors (Toh 1997; Daiyasu & Toh 2000), based on the double dynamic programming algorithm (Taylor & Orengo 1989).

Cloning of zebrafish and Xenopus CRY-DASH homologues

By conducting BLAST searches of the available EST database, we found two over-lapping ESTs in Xenopus and zebrafish. The GeneBank accession numbers of these ESTs were: Xenopus, CA971785 and BU916216; and zebrafish, AW153390 and CA472873. Since both the Xenopus and zebrafish ESTs were partial, we amplified their 5' and 3' ends by the rapid amplification of cDNA ends (RACE) technique (Frohman et al. 1988). The RACE reactions were carried out by using cDNAs extended from the total RNAs of Xenopus oocytes or zebrafish eyes. The primer sequences were as follows: 3' RACE primer for Xenopus, XlCRYDash907R, GTGGCTTTGAAATACGGCAG and XlCRYDash971R GGAAGAGGGACCCAAAGC; and for zebrafish zCRYDash902R TTGTGGCTTTGAAATACGG and zCRYDash934R TATATGAACGGTCTGCAAG; 5' RACE primer for Xenopus, XlCRYDash99L, CTGGTCTGCGTTCCTGTG, XlCRYDash64L, TCGTTATCGTGCAGCCGC and XlCRYDash48L, CAAGTCGTTCCTCAGCAG; and for zebrafish, zfCRYDash105L, GCCCAGTGAAACACCTCG and zfCRYDash52L, GCAGCCGCAGATCGTTCC. The cDNA sequences uncovered by the 5' and 3' RACE products were amplified by PCR using the cDNA with a set of the following gene-specific primers: for Xenopus, XlCRYDash-RI-ATG, CCAGAATTCATGTGTGTCCCTAGCCG, XlCRYDash1065L, TCCAGTCATTGCCAGTTCC; for zebrafish, zCRYDash-RI-ATG, GCCGAATTCATGTCCGCTTCCCGTAC and zCRYDash1012L, TTCTTCCCTCTTTCCATGC. All PCR products were subcloned into the TA plasmid vector. The sequences of five independent plasmid clones were determined.

Detection of transcriptional repressor activity

Luciferase reporter gene assays of BRF41 cells were done as described elsewhere (Kobayashi et al. 2000; Ishikawa et al. 2002). The full-length coding regions of zebrafish cry-dash and Xenopus cry-dash were ligated into the pcDNA3.1 expression vector (Invitrogen). The resulting plasmids were designated pcDNA-zCRY-Dash and pcDNA-XlCRY-Dash. The expression constructs of zebrafish clock, bmal3 and cry1a were described elsewhere (Kobayashi et al. 2000; Ishikawa et al. 2002). Reporter constructs were made as follows (Kobayashi et al. 2000). A 3700 bp segment of the 5' flanking region of zcry3 gene was isolated from the genomic library and subcloned in the pGL3-Basic vector (Promega), generating pzcry3-luc. Cells (1.5 x 10⁵) were seeded in 12-well plates and transfected the next day with Lipofectamine-Plus (Invitrogen). Each transfection had 200 ng each of pcDNA-zCLOCK1, pcDNA-zBMAL3 and pcDNA-zCRY1a or pcDNA-CRY-Dash, and the reporter plasmid pzcry3-luc (10 ng) and the pRL vector (Promega) 25 ng. The total DNA per well was adjusted to 1 µg by adding pcDNA 3.1 vector as the carrier. Forty-eight hours after transfection cells were harvested, and their firefly and Renilla luciferase activities determined by luminometry. The reporter luciferase activity was normalized for each sample by determining the firefly:Renilla luciferase activity ratios. All experiments were done three times.

Detection of DNA photolyase activity in E. coli

The full-length coding regions of zebrafish CRY-Dash and Xenopus CRY-Dash were ligated into the pGEX4T expression vector (Pharmacia). The resulting plasmids were designated pGEX-zCRY-Dash and pGEX-XlCRY-Dash. E. coli strains, SY2 (uvrA^–, recA^–, phr^–) or SY32 (uvrA^–, recA^–, phr^–) cells that carry plasmid pRT2, were used as the hosts for the expression of zebrafish or Xenopus cry-dash genes. Plasmid pRT2 bears the E. coli phr⁺ gene. Thus, neither the UV-induced CPD nor (6–4) photoproduct is photorepaired in the SY2 host cells but CPDs are photorepaired efficiently in SY32/pRT2 cells. E. coli cells SY2 or SY32/pRT2 were transformed with pGEX-zCRY-Dash or pGEX-XlCRY-Dash. The transformed cells were grown overnight in LB medium with 20 mg/L ampicillin, 10 mg/L kanamycin and 10 mg/L tetracycline at 37 °C. Expression of pGEX-zCRY-Dash or pGEX-XlCRY-Dash was induced by adding isopropyl-1-thio-b-D-galactopyranoside (IPTG) to the medium to a final concentration of 0.1 mM and shaking for 1 h at 37 °C. The cells were plated on LB agar and first UV-irradiated with intensities of 0.1, 0.2, 0.3, 0.6 or 1.2 J/m² and subsequently with daylight fluorescence lamps for 1 h as previously described (Kobayashi et al. 2000). Samples were incubated at 37 °C overnight. Surviving colonies were counted the next day. All experiments, except photoreactivation treatments, were performed under yellow light.

Preparation of GST-fusion protein

E. coli SY2 was transformed with the GST fusion vectors (pGEX-zCRY-Dash and pGEX-XlCRY-Dash). Cells were grown at 25 °C to A₆₀₀ 0.5–0.8 by induction with 0.1 mM IPTG for 12 h then collected by centrifugation. The cell pellets were resuspended in phosphate-buffered saline (PBS), then sonicated. Insoluble material was removed by centrifugation (60 min at 10 000 x g). GST-fusion proteins were purified from the soluble extracts in a glutathione-Sepharose 4B column (Amersham Pharmacia Biotech) according to the manufacture's instructions. GST-fusion protein containing fractions, detected by GST activity, were pooled. The gel filtration column PD-10 (Amersham Pharmacia Biotech) was used to remove glutathione contained in buffer. The final buffer contained 50 mM Tris, pH 8.0. Fractions that contained the GST-fusion protein were determined by SDS-PAGE, and they were pooled. Glycerol was added (10% final concentration) before storage at –20 °C. Purification of E. coli CPD photolyase, GST-Dm CPD photolyase and GST-Xl 6–4 photolyase were done as previously described (Todo et al. 1993, 1994, 1997).

Gel mobility shift and in vitro DNA repair assay

Mobility shift assays and in vitro DNA repair assay were essentially done as previously described (Hitomi et al. 1997) with slight modification (Kobayashi et al. 2000). 49 bp DNAs, which have a single UV photoproduct (either a CPD, (6–4) photoproduct or DEWAR isomer) at the MseI site, were prepared as described elsewhere (Hitomi et al. 1997) and used as the substrate both in the gel shift and in vitro repair assays. For the gel shift assay, 1 µg of partially purified GST-fusion protein was added to the ³²P-labelled DNA substrate (1 nM), and the mixture electrophoresed on a non-denaturing acrylamide gel as previously described (Hitomi et al. 1997). In the in vitro repair assay, the MseI restriction enzyme recognition site, TTAA, which becomes uncleavable due to the photoproduct at the TT site, is restored by the photoreversal reaction, and thus MseI digestion generate 21 base pair fragment (Hitomi et al. 1997). Three micrograms of GST-fusion protein was mixed with ³²P-labelled DNA substrate (10 nM) then illuminated with photoreactivating light. After treatment of the mixture with phenol, the DNA substrate was recovered by ethanol precipitation, digested with MseI then electrophoresed on a denaturing acrylamide gel.

Spectroscopic analysis

The absorption and fluorescence spectra were recorded with a BECKMAN DU-640 spectrophotometer and a Shimazu RF-5300PC spectrofluorometer, respectively. The concentrations of the apoproteins and the cofactors were calculated from their molar extinction coefficients. The theoretical molar extinction coefficient of GST-zCRY-DASH apoenzyme at 280 nm is 155 420 M^–1 cm^–1. The molar extinction coefficient of MTHF at 370–380 nm is 24 495 M^–1 cm^–1 (Johnson et al. 1988; Worthington et al. 2003). To determine the flavin concentration, the zCRY-DASH holoproteins were heated at 95 °C for 5 min in buffer containing 50 mM Tris-HCl, pH 8.0, and the precipitated protein was removed by centrifugation. The absorption spectrum in the 300–600 nm range was recorded, and FAD concentration was calculated from 440 nm absorbance using a molar extinction coefficient of 11 300 M^–1 cm^–1. When MTHF is released from the enzyme, the 5–10 methenyl bridge responsible for the 380 nm absorption is broken at neutral pH to generate 10-formyltetrahydrofolate, which does not absorb at > 300 nm and hence does not contribute to the near UV spectrum of the cofactors (Johnson et al. 1988; Worthington et al. 2003).

Phylogenetic analyses

To investigate the evolutionary relationships of the photolyase/cryptochrome family, molecular phylogenetic trees were constructed by two different approaches, the NJ method (Saitou & Nei 1987) and the ML method (Felsenstein 1981). For the analyses, the sites including gaps were excluded from the multiple alignments.

The genetic distance between every pair of aligned sequences was calculated as an ML estimate (Felsenstein 1996), using the JTT model (Jones et al. 1992) for the amino acid substitutions. Based on the distances, an NJ tree was constructed for all of the sequences included in the alignment. The statistical significance of the NJ tree topology was evaluated by a bootstrap analysis (Felsenstein 1985) with one thousand iterative tree re-constructions.

The evolutionary relationship among the four groups was further examined by the ML method. The JTT-F model was used as the amino acid substitution model in the ML analysis. However, the number of sequences included in the alignment was too large for all of the sequences to be subjected to the ML analysis. Therefore, several representative sequences were selected from each group, and the ML analysis was performed with the restricted sequence data. The statistical significance of the ML tree topology was also evaluated by the bootstrap analysis.

For the phylogenetic analyses, two software packages, PHYLIP 3.5c (Felsenstein 1993), and MOLPHP 2.3b3 (Adachi & Hasegawa 1996), were used. The trees thus obtained were drawn with TreeView (Page 1996).

	Footnotes

 
Communicated by: Eisuke Nishida

Both authors contributed equally to this work.

^* Correspondence: E-mail: daiyasu{at}kuicr.kyoto-u.ac.jp

	References

Top Abstract Introduction Results Discussion Experimental procedures References

 
Adachi, J. & Hasegawa, M. (1996) MOLPHY (Programs for Molecular Phylogenetics), 2.3b3. Tokyo: Institute of Statistical Mathematics.

Ahmad, M. & Cashmore, A.R. (1993) HY4 gene of A. thaliana encodes a protein with characteristics of a blue-light photoreceptor. Nature 366, 162–166.[CrossRef][Medline]

Altschul, S.F., Madden, T.L., Schaffer, A.A., et al. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucl. Acids Res. 25, 3389–3402.[Abstract/Free Full Text]

Aubert, C., Vos, M.H., Mathis, P., Eker, A.P. & Brettel, K. (2000) Intraprotein radical transfer during photoactivation of DNA photolyase. Nature 405, 586–590.[CrossRef][Medline]

Brudler, R., Hitomi, K., Daiyasu, H., et al. (2003) Identification of a new cryptochrome class. Structure, function, and evolution. Mol. Cell 11, 59–67.[CrossRef][Medline]

Cashmore, A.R., Jarillo, J.A., Wu, Y.J. & Liu, D. (1999) Cryptochromes: blue light receptors for plants and animals. Science 284, 760–765.[Abstract/Free Full Text]

Daiyasu, H. & Toh, H. (2000) Molecular evolution of the myeloperoxidase family. J. Mol. Evol. 51, 433–445.[Medline]

Dehal, P., Satou, Y., Campbell, R.K., et al. (2002) The draft genome of Ciona intestinalis: insights into chordate and vertebrate origins. Science 298, 2157–2167.[Abstract/Free Full Text]

Dunlap, J.C. (1999) Molecular bases for circadian clocks. Cell 96, 271–290.[CrossRef][Medline]

Felsenstein, J. (1981) Evolutionary trees from DNA sequences: a maximum likelihood approach. J. Mol. Evol. 17, 368–376.[CrossRef][Medline]

Felsenstein, J. (1985) Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783–791.[CrossRef]

Felsenstein, J. (1993) PHYLIP (phylogeny inference package), Version 3.5c.Seattle: Department of Genetics. University of Washington.

Felsenstein, J. (1996) Inferring phylogenies from protein sequences by parsimony, distance, and likelihood methods. Methods Enzymol. 266, 418–427.[Medline]

Frohman, M.A., Dush, M.K. & Martin, G.R. (1988) Rapid production of full-length cDNAs from rare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc. Natl. Acad. Sci. USA 85, 8998–9002.[Abstract/Free Full Text]

Froy, O., Chang, D.C. & Reppert, S.M. (2002) Redox potential: differential roles in dCRY and mCRY1 functions. Curr. Biol. 12, 147–152.[CrossRef][Medline]

Galagan, J.E., Calvo, S.E., Borkovich, K.A., et al. (2003) The genome sequence of the filamentous fungus Neurospora crassa. Nature 422, 859–868.[CrossRef][Medline]

Gardner, M.J., Hall, N., Fung, E., et al. (2002) Genome sequence of the human malaria parasite Plasmodium falciparum. Nature 419, 498–511.[CrossRef][Medline]

Griffin, E.A. Jr, Staknis, D. & Weitz, C.J. (1999) Light-independent role of CRY1 and CRY2 in the mammalian circadian clock. Science 286, 768–771.[Abstract/Free Full Text]

Hirayama, J., Fukuda, I., Ishikawa, T., Kobayashi, Y. & Todo, T. (2003) New role of zCRY and zPER2 as regulators of sub-cellular distributions of zCLOCK and zBMAL proteins. Nucleic Acids Res. 31, 935–943.[Abstract/Free Full Text]

Hitomi, K., Kim, S.T., Iwai, S., et al. (1997) Binding and catalytic properties of Xenopus (6–4) photolyase. J. Biol. Chem. 272, 32591–32598.[Abstract/Free Full Text]

Hitomi, K., Okamoto, K., Daiyasu, H., et al. (2000) Bacterial cryptochrome and photolyase: characterization of two photolyase-like genes of Synechocystis sp. PCC6803. Nucleic Acids Res. 28, 2353–2362.[Abstract/Free Full Text]

Ishikawa, T., Hirayama, J., Kobayashi, Y. & Todo, T. (2002) Zebrafish CRY represses transcription mediated by CLOCK-BMAL heterodimer without inhibiting its binding to DNA. Genes Cells 7, 1073–1086.[Abstract]

Johnson, J.L., Hamm-Alvarez, S., Payne, G., Sancar, G.B., Rajagopalan, K.V. & Sancar, A. (1988) Identification of the second chromophore of Escherichia coli and yeast DNA photolyases as 5,10-methenyltetrahydrofolate. Proc. Natl. Acad. Sci. USA 85, 2046–2050.[Abstract/Free Full Text]

Jones, D.T., Taylor, W.R. & Thornton, J.M. (1992) The rapid generation of mutation data matrices from protein sequences. Comput. Appl. Biosci. 8, 275–282.

Kanai, S., Kikuno, R., Toh, H., Ryo, H. & Todo, T. (1997) Molecular evolution of the photolyase-blue-light photoreceptor family. J. Mol. Evol. 45, 535–548.[CrossRef][Medline]

Kato, T. Jr, Todo, T., Ayaki, H., et al. (1994) Cloning of a marsupial DNA photolyase gene and the lack of related nucleotide sequences in placental mammals. Nucl. Acids Res. 22, 4119–4124.[Abstract/Free Full Text]

Kleine, T., Lockhart, P. & Batschauer, A. (2003) An Arabidopsis protein closely related to Synechocystis cryptochrome is targeted to organelles. Plant J. 35, 93–103.[CrossRef][Medline]

Kobayashi, Y., Ishikawa, T., Hirayama, J., et al. (2000) Molecular analysis of zebrafish photolyase/cryptochrome family: two types of cryptochromes present in zebrafish. Genes Cells 5, 725–738.[Abstract]

Komori, H., Masui, R., Kuramitsu, S., et al. (2001) Crystal structure of thermostable DNA photolyase: pyrimidine-dimer recognition mechanism. Proc. Natl. Acad. Sci. USA 98, 13560–13565.[Abstract/Free Full Text]

Kume, K., Zylka, M.J., Sriram, S., et al. (1999) mCRY1 and mCRY2 are essential components of the negative limb of the circadian clock feedback loop. Cell 98, 193–205.[CrossRef][Medline]

Lee, K., Dunlap, J.C. & Loros, J.J. (2003) Roles for WHITE COLLAR-1 in circadian and general photoperception in Neurospora crassa. Genetics 163, 103–114.[Abstract/Free Full Text]

Li, Y.F. & Sancar, A. (1990) Active site of Escherichia coli DNA photolyase: mutations at Trp277 alter the selectivity of the enzyme without affecting the quantum yield of photorepair. Biochemistry 29, 5698–5706.[CrossRef][Medline]

Loros, J.J. & Dunlap, J.C. (2001) Genetic and molecular analysis of circadian rhythms in Neurospora. Annu. Rev. Physiol. 63, 757–794.[CrossRef][Medline]

Martin, W., Rujan, T., Richly, E., et al. (2002) Evolutionary analysis of Arabidopsis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of cyanobacterial genes in the nucleus. Proc. Natl. Acad. Sci. USA 99, 12246–12251.[Abstract/Free Full Text]

Miki, K., Tamada, T., Nishida, H., et al. (1993) Crystallization and preliminary X-ray diffraction studies of photolyase (photoreactivating enzyme) from the cyanobacterium Anacystis nidulans. J. Mol. Biol. 233, 167–169.[CrossRef][Medline]

Murzin, A.G., Brenner, S.E., Hubbard, T. & Chothia, C. (1995) SCOP: a structural classification of proteins database for the investigation of sequences and structures. J. Mol. Biol. 247, 536–540.[CrossRef][Medline]

Page, R.D. (1996) TreeView: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12, 357–358.

Park, H.W., Kim, S.T., Sancar, A. & Deisenhofer, J. (1995) Crystal structure of DNA photolyase from Escherichia coli. Science 268, 1866–1872.[Abstract/Free Full Text]

Rachael, A., Simon, B., Karen, B., et al. (1998) Genome sequence of the nematode C. elegans: a platform for investigating biology. Science 282, 2012–2018.[Abstract/Free Full Text]

Rosato, E., Codd, V., Mazzotta, G., et al. (2001) Light–dependent interaction between Drosophila CRY and the clock protein PER mediated by the carboxy terminus of CRY. Curr. Biol. 11, 909–917.[CrossRef][Medline]

Saitou, N. & Nei, M. (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425.

Sancar, A. (1994) Structure and function of DNA photolyase. Biochemistry 33, 2–9.[CrossRef][Medline]

Sancar, G.B. (1990) DNA photolyases: physical properties, action mechanism, and roles in dark repair. Mutat. Res. 236, 147–160.[Medline]

Taylor, W.R. & Orengo, C.A. (1989) Protein structure alignment. J. Mol. Biol. 208, 1–22.[CrossRef][Medline]

Thompson, J.D., Higgins, D.G. & Gibson, T.J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680.[Abstract/Free Full Text]

Todo, T. (1999) Functional diversity of the DNA photolyase/blue light receptor family. Mutat. Res. 434, 89–97.[Medline]

Todo, T., Kim, S.T., Hitomi, K., et al. (1997) Flavin adenine dinucleotide as a chromophore of the Xenopus (6–4) photolyase. Nucl. Acids Res. 25, 764–768.[Abstract/Free Full Text]

Todo, T., Ryo, H., Takemori, H., et al. (1994) High-level expression of the photorepair gene in Drosophila ovary and its evolutionary implications. Mutat. Res. 315, 213–228.[Medline]

Todo, T., Ryo, H., Yamamoto, K., et al. (1996) Similarity among the Drosophila (6–4) photolyase, a human photolyase homolog, and the DNA photolyase-blue-light photoreceptor family. Science <