The Common Function of a Novel Subfamily of B-Box Zinc Finger Proteins with Reference to Circadian-Associated Events in Arabidopsis thaliana

Transcription

1 Bioscience, Biotechnology, and Biochemistry ISSN: 91-1 (Print) (Online) Journal homepage: The Common Function of a Novel Subfamily of B-Box Zinc Finger Proteins with Reference to Circadian-Associated Events in Arabidopsis thaliana Takeshi KUMAGAI, Shogo ITO, Norihito NAKAMICHI, Yusuke NIWA, Masaya MURAKAMI, Takafumi YAMASHINO & Takeshi MIZUNO To cite this article: Takeshi KUMAGAI, Shogo ITO, Norihito NAKAMICHI, Yusuke NIWA, Masaya MURAKAMI, Takafumi YAMASHINO & Takeshi MIZUNO () The Common Function of a Novel Subfamily of B-Box Zinc Finger Proteins with Reference to Circadian-Associated Events in Arabidopsis thaliana, Bioscience, Biotechnology, and Biochemistry, 7:, , DOI: 1.171/bbb.1 To link to this article: Published online: May 1. Submit your article to this journal Article views: 397 View related articles Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at Download by: [ ] Date: 3 November 17, At: 9:

2 Biosci. Biotechnol. Biochem., 7 (), , The Common Function of a Novel Subfamily of B-Box Zinc Finger Proteins with Reference to Circadian-Associated Events in Arabidopsis thaliana Takeshi KUMAGAI, 1 Shogo ITO, 1;y Norihito NAKAMICHI, Yusuke NIWA, 1 Masaya MURAKAMI, 1 Takafumi YAMASHINO, 1 and Takeshi MIZUNO 1 1 Laboratory of Molecular Microbiology, School of Agriculture, Nagoya University, Chikusa-ku, Nagoya -1, Japan Division of Biological Science, Graduate School of Science, Nagoya University, Chikusa-ku, Nagoya -, Japan Received January 1, ; Accepted February, ; Online Publication, June 7, [doi:1.171/bbb.1] Downloaded by [ ] at 9: 3 November 17 Over 1, genes encoding putative transcription factors have been identified in the Arabidopsis genome sequence, however, their physiological functions are not yet fully understood. In this study, a small subfamily of double B-box zinc finger (DBB, DOUBLE B-BOX) genes, encoding eight putative transcription factors, were characterized with reference to the circadian rhythm and the early photomorphogenic regulation of hypocotyl elongation in response to light signals. Among these, it was found that the transcriptions of five DBB genes were under the control of circadian rhythm. To gain insight into the physiological roles of these putative transcription factors, forward and reverse genetic studies were carried out. The results suggested that they are commonly implicated in light signal transduction during early photomorphogenesis, however, their functions are not totally redundant, as judged by the fact that their circadian-expression profiles (or phases) were distinctive from each other, and by the fact that some DBBs (named,,, and ) were apparently implicated in light signal transduction in a negative manner, whereas another (named DBB3) was implicated in a positive manner with regard to light-induced inhibition of elongation of hypocotyls. We also found that homologous B-box zinc finger genes are widely conserved in higher plants (e.g., Oryza sativa). Taking this altogether, it is probable that in addition to previously characterized bzip-type transcription factors (e.g., HY and HYH) and bhlh-type transcription factors (e.g., PIF and PIF/PIL), a set of B-box zinc finger transcription factors should also be taken into consideration for a better understanding of the complex molecular mechanisms underlying the early photomorphogenic development of Arabidopsis thaliana. Key words: Arabidopsis thaliana; circadian rhythm; photomorphogenesis; transcription factor; zinc finger protein Over 1, transcription factor genes have been identified in the Arabidopsis genome sequence, representing approximately % of the total number of genes, and these transcription factor genes can be grouped into a certain number of different families. 1) Among them, the zinc finger family of proteins contains a large number of members, and these are further classified into several distinct subfamilies. The B-box zinc finger family is such an example (see SMART, smart.embl-heidelberg.de/smart/show motifs.pl), of which the best-characterized member is named CO (CONSTANS).,3) The CO protein has two tandem (or double) B-box zinc finger domains at its N-terminal part, and also has a common C-terminal CCT (CO/ COL/TOC1) motif. ) In Arabidopsis thaliana, there are many proteins that have the same structural design as CO, and these proteins were collectively designated CONSTANS-LIKE (COL1 to COL1). 3) The characteristic nature of CO has been well documented with special reference to its role in the photoperiodic control of flowering time.,) One of the important facts is that the transcription of CO is under the control of circadian rhythm. 7) This event is important for CO to play a critical role in the photoperiodic control of flowering time. An inspection of current Arabidopsis databases revealed that the expression profiles of many other COL genes are also diurnally oscillated (COL1, COL, COL, COL, COL9, COL1, and COL1 at least). ) Some COL genes have been characterized by reference to functions associated with the control of flowering time. 9,1) We have been studying the mechanisms underlying the circadian clock, and also clock-controlled flowering time. 11,1) During the course of such studies, an inspection of current Arabidopsis databases revealed that this model plant has still many more homologous genes (at least additional 1 members) encoding proteins containing the B-box zinc finger domain. In contrast y To whom correspondence should be addressed. Tel: ; Fax: ; nqi33717@nifty.com Abbreviations: CO/COL, constans/constans-like; CT, circadian time; DBB, double B-box; HY, long hypocotyl ; FT, flowering locus T; HYH, HY homolog; PRR, pseudo-response regulator; TOC1, timing of cab expression 1

3 Downloaded by [ ] at 9: 3 November 17 1 T. KUMAGAI et al. to the CO family protein, however, they have no C-terminal CCT motif. Although the exact biochemical nature of the B-box-type zinc finger domain is not yet known, it is likely that proteins containing this domain serve as transcriptional regulators, directly or indirectly. For instance, CO regulates the transcription of FT (FLOWERING LOCUS T). 3,13) These B-box zinc finger proteins might play roles in highly divergent ways, and these proteins might also be relevant to certain clockcontrolled events. Nevertheless, most of these new members remain elusive. Keeping this background in mind, in this study we focused on a subfamily of such as-yet uncharacterized B-box zinc finger proteins of Arabidopsis thaliana, collectively designated DBB (DOUBLE B-BOX) genes (see Fig. 1). A B AT3G19 AT3G11 AT1G19 COL1 AT1G ATG7 ATG731 ATG1 B-box COL11 COL1 B-box B-box COL9 COL1 COL1 COL13 COL1 COL CO (Arabidopsis) Hd1(Rice) Os9g79 (OsDBB1) ATG13 () ATG39 () AT1G7 () ATG397 (DBB) DBB AT1G7 (DBB3) B-box B-box AT1G () ATG313().1 ATG1(DBB) CO COL1 CCT COL COL COL3 COL COL COL7 Osg1 DBB3 (OsDBB3a) Osg (OsDBB3b) Os1g (OsDBB3c).1 Osg93 (Os) Osg (Os) Fig. 1. Phylogenic Tree of Arabidopsis Genes Encoding Proteins That Contain the B-Box-Type Zinc Finger Domain. A, By employing about 3 amino acid sequences containing the B-box-type zinc finger domain, a non-rooted neighbor-joining phylogenic tree with branch length was constructed with the program CLUSTALW. Other details are given in the text. B, A comparative phylogenic tree containing the circadian clock-controlled Arabidopsis DBB genes and the homologous rice OsDBB genes was also constructed. For to the nomenclature for each rice gene, see the text. First these members of the DBB subfamily were characterized with special reference to circadian rhythm. We found that many of these genes were under the control of the circadian clock at the level of transcription. Evidence was also presented that highly homologous proteins are conserved in rice. Through forward and reverse genetic studies, it was finally found that many of the Arabidopsis DBB genes were commonly implicated in light signal transduction pathways during early photomorphogenesis. Materials and Methods Plant materials and growth conditions. Arabidopsis thaliana umbia was used in this study, unless otherwise noted. Plants were grown in soil and/or on MS medium (Murashige and Skoog salts with 1% sucrose and.3% gellan-gum). They were incubated in growth chambers under fluorescent white light of 9 1 mmol m s 1 and 1 1 mmol m s 1 on a 1 h light/ h dark cycle or a 1 h light/1 h dark cycle. Oryza sativa (Nipponbare) were grown in a growth chamber at C on MS medium (Murashige and Skoog salts with 1% sucrose and.% agar) under fluorescent white light ( 7 nm, 1 mmol m s 1 ). Construction of DBBs-ox transgenic plants. The entire coding sequence of the DBBs cdna was cloned by PCR-aided cloning into the psk1 vector 1) at the XbaI-NotI site downstream of the CaMV 3S promoter. The primer sets used in this study are given in Table 1. The resulting plasmid constructs were transferred via Agrobacterium tumefaciens strain EHA11 Arabidopsis to plants () by the conventional vacuum infiltration transformation method. 1) Examination of light response in early photomorphogenesis. To determine the light response in early photomorphogenesis of plants, seeds were sown on gellan-gum (.3%)-plates containing MS salts without sucrose, then they were kept at C for h in the dark. After that, the seeds were exposed to white light for 3 h in order to enhance germination, followed by incubation at C for 1 h in the dark. Plants were grown for h or 7 h under continuous light across a varied range of fluence rates or in the dark. As light sources for continuous irradiation, light-emitting diodes were used: (for red light, STICK-mR, max ¼ nm at 3 mmol m s 1 ; far-red, STICK-mFR max ¼ 73 nm at mmol m s 1 ; blue, STICK-mB max ¼ 7 nm at mmol m s 1, Tokyo Rika, Japan, as described previously. 1) Preparation of RNA, northern blotting analysis, and real time RT-PCR. For northern blotting analysis, total RNA from A. thaliana was purified by ATA methods, as described previously. 17) Total RNA from O. sativa was isolated by the method described by Carpenter and Simon. 1) Several double-stranded 3 P-labeled DNA

4 Plant Clock-Controlled B-Box Zinc Finger Proteins 11 Table 1. Oligo-Nucleotide Primers Used in This Study Downloaded by [ ] at 9: 3 November 17 probes were used to detect the specific mrnas. The probes used in northern blotting were amplified by PCR with an appropriate set of primers. The primer sets are given in Table 1. the 3 P-labeled probes were prepared by the Megaprime DNA Labeling System (GE Healthcare Bio-Sciences Chalfont St Giles UK). For RT-PCR, total RNA samples were purified with an RNeasy kit (Qiagen K. K., Tokyo Japan). To synthesize cdna, 1 mg of RNA was reverse-transcribed with ReverTea Ace (Toyobo, Osaka, Japan) and oligo-dt primer. The synthesized cdnas were amplified with gene-specific primers for 3 cycles. Sequence analysis. RiceBLAST ( dna.affrc.go.jp/cgi-bin/robo-blast/blast.cgi) and KOME ( for homologs of Arabidopsis clock-associated genes. The predicted protein sequences were aligned with CLUSTALW, ( The same program was used to analyze protein phylogeny by the neighbor-joining method. The phylogenic tree was displayed with TreeView ( zoology.gla.ac.uk/rod/treeview.html). SMART ( coot.embl-heidelberg.de/smart) was used to analyze domains and motifs. Results For amplification of the probes used in Northern hybridization and real-time PCR analysis : -AGGCGATAAGCCTAATCATG/ -GAGATATAACTCTGAAGAATCG : -CTAAAGAAAACAATACGAGGGAC/ -GAATTTCAGAGATTAGCTTACATTC DBBa: -GTTGAGGATTTCCTCGATTC/ -TACATGAAACATGTGCATTAATGG DBBb: -AATCAGTATCGAGCGAGGTC/ -CAACAAGTTGGTTTGGATTCATG DBB3: -AGCTTCCTCTCGTAAGTAGC/ -CTAACAAAGAACACCGACTC DBB: -CTCTATGTGATATCTGCCAG/ -TCCGGAACCATGATGTTGTG : -TCTTCTGCTACTCCACTTCC/ -GTATCAAGTTTGTAGCATCAAAGC : -AGAAGCAGAGGCTTGAGTAC/ -CTCAAGCTTCTATTACGTAATCTG OsDBB1: -GGATGTTAGCGTCAACAATC/ -AAATGATGGGTCTATTAATC OsDBB3c: -TGCCGGAAGTTCACTGGGCC/ -TGCAGCCCCTAATTAATCAG Os: -AGGGCTCTCCTATTGGGTTC/ -GTACATCATAGTTCATGCCC Os: -TCCACAACCAGGCGCCCAAG/ -GCTCAAATATGTCACAGTTG For detection of transcripts by RT-PCR : -AACCCTTTAGTCTCTCCATC/ -AGTCCCAGTTGGTCAGCATG : -GGCTGATAACTCCTGTTTTG/ -CTCTGCTGATTAGATGGGTC-3 For cloning of cdnas : -ATTCTAGAATGCGAATTTTGTGTG/ -ATGAATAGGCGGCCGCCTATTCATGCTCGTG : -ATTCTAGAATGCGGATTTTGTGCG/ -AGAAGTGAGCGGCCGCTCACTTCTCAGACTC DBB3: -ATTCTAGAATGAAGATTCAGTGTAACG/ -GGTTCTAGGCGGCCGCCTAGAACCGTCGCCGCCG : -ATTCTAGAATGAAGATACAGTGTGATG/ -TTGGCTAAGCGGCCGCTTAGCCAAGATCAGGGAC : -ATTCTAGAATGAAGATACAATGTGATG/ -TAGGCTAAGCGGCCGCTTAGCCTAGGTCGGGGAC A subfamily of DBB (DOUBLE B-BOX) zinc finger genes As mentioned earlier, CO is the representative of B-box zinc finger proteins in Arabidopsis thaliana. By the use of the amino acid sequence of the B-box zinc finger region of CO, a BLAST search was conducted with the Arabidopsis AGI protein database ( We identified a number of inferred amino acid sequences, each of which contained a B-box zinc finger domain. At least 3 candidates were identified, and they were characterized through CLUSTALW ( top-j.html) (Fig. 1A). The resulting phylogenic tree provided insight into the classification of these B-box zinc finger proteins. They were classified into three distinct subgroups. Each member of the largest representative group (designated the CO/COL subfamily) contains two tandem B-box zinc finger domains, followed by a CCT (CO/COL/TOC1) motif, as mentioned earlier (Fig. 1, see schematic illustration). The members of another group resembled CO in that they contained a double B-box zinc finger domain at the N- terminal end, but they had no CCT motif. The members of the third group contained only a single B-box-type zinc finger domain with no CCT motif. Since the CO/ COL subfamily proteins have been characterized, 3) we focused our attention on the second subfamily, consisting of eight members (see Fig. 1, the shaded area). The coding genes were designated DBB (DOUBLE B-BOX), as follows: (Atg13), DDB1b (Atg39), DBB (Atg397), DBB3 (At1g7), and DBB (Atg1), in addition to (At1g), (Atg313), and (At1g7). and / stand for SALT TOLERANCE and SALT TOLER- ANCE HOMOLOGES respectively. They were characterized previously to some extent, as discussed below. 19 ) The amino acid sequences of the and gene products are highly homologous to each other, not only in their B-box domains, but also in their entire C-terminal sequences. The same is true for the SHO and gene product, otherwise, the C-terminal amino acid sequences of these DBB proteins are quite

5 1 T. KUMAGAI et al. A DBB3 DBB3 Transcripts in 1 3 Time in LL (h) References in B DBB3 Transcripts in prr9/prr7/prr 1 3 Time in LL (h) References in prr9/prr7/prr DBB3 1 3 Time in LL (h) C /APX3.. cca1 lhy -1-1 Time (h) Downloaded by [ ] at 9: 3 November 17 Fig.. Northern Blot Hybridization Analysis of Transcripts of the DBB Family Genes. A and B, Wild-type () and prr9 prr7 prr triple mutant plants were grown under LD (1 h light/1 h dark) for d, and then they were released to continuous light (LL). RNA samples were prepared at 3 h intervals, as schematically indicated (open and hatched rectangles indicate subjective day and night respectively). They were subjected to Northern blot hybridization analysis with the indicated DNA probes. For these genes, rhythmic expression profiles are clearly seen in (panel A), but not in prr9 prr7 prr, indicating that the expression of these genes oscillates diurnally, except for (panel B). These results are consistent with those inferred from the microarray datasets of diurnally oscillated genes with Afymerix GeneChip Arabidopsis ATH1 K arrays, representing nearly all protein-coding transcripts of the reference plant. ) The amounts of were examined as loading references. Other details are given in Materials and Methods. The same experiments were repeated twice with the same results. C, Characterization of expression profiles of in the clock-defective mutant grown under a light/dark cycle (1 h light/1 h dark). In addition to wild-type, the clock-defective mutant used was the cca1 lhy double mutant. ) RNA samples were prepared at different timing, as indicated (the dark period is indicated by a shaded background). The transcript levels of were quantified by real-time PCR with appropriate sets of primers (see Table 1). The data were normalized with an appropriate internal reference (the APX3 transcript encoding a ascorbate peroxidase 3), by taking the maximal value in to be 1. Other experimental procedures were the same as those described previously. ) different from each other. Based on this, we characterized these DBB subfamily genes together by comparing them with each other by conventional forward and reverse genetic approaches. DBB subfamily genes are highly conserved in divergent plants First, we wanted to know whether the DBB subfamily genes are conserved in other higher plants. The results of an inspection with the rice genomic database (Oryza sativa japonica) showed that this monocotyledonous model plant also has a set of highly homologous DBBrelated gene, ( robo-blast/) (Fig. 1B). The rice Hd1 gene has been identified as an ortholog of the Arabidopsis CO gene. 3) In this comparative phylogenic tree, it was found that the Hd1 and CO genes together belong to an out-group. Hence, we designated these rice DBB-related genes as OsDBB-series, as indicated in the phylogenic tree. These results are compatible with the idea that DBB subfamily genes commonly play fundamental roles in higher plants. This assumption was further supported by the fact that a certain number of highly homologous genes were also found in the entire genomic databases of Populus trichocarpa (woody poplar, jgi-psf.org/poptr1/poptr1.home.html) and Vitis vinifera (fruit crop grapevine, Poptr1.home.html) (data not shown). Hence we further examined in detail the functions of the clock-controlled DBB subfamily of zinc finger genes in Arabidopsis thaliana. Transcriptions of several DBB genes were under the control of a circadian clock In accordance with our primary interest in DBB subfamily genes, first we examined to determine whether the transcriptions of these genes are under the control of a circadian clock. Wild-type plants (umbia, ) were grown for d under light/dark (LD) (1 h light/1 h dark) cycle conditions, and then they were released to continuous white light (LL) conditions. The previously characterized clock-defective prr9 prr7 prr triple loss-of-function mutant plants were also employed as reference. ) RNA samples were prepared at 3 h intervals for d to use in Northern blotting hybridization analysis, with special reference to freerunning oscillation profiles of these DBB genes at the level of transcription. The analyses were carried out by the use of appropriate DNA probes specific to,,, DBB3,, and (Fig. ). We did not obtain data for DBB or DBB, because the expression of DBB was lower than the detectable

6 Downloaded by [ ] at 9: 3 November 17 level under our hybridization conditions, and also because the specific cdna probe for DBB could not be amplified from the plants even after several attempts, suggesting that the expression levels of these genes in plants are extremely low. Otherwise, the expression profiles were successfully examined for the remaining six DBB genes (Fig. ). The results showed that all of them, except for, exhibited a robust free-running circadian rhythm in the wild-type plants (Fig. A), and it is interesting that the phases of rhythms (i.e., the positions of peaks) were considerably different from each other. No such robust oscillation profiles were observed for the RNA samples, which were isolated from the clockdefective prr9 prr7 prr triple mutant plants (Fig. B). The results suggest that the rhythms of these DBB genes are controlled by a circadian clock, in which the PRR9/ 7/ (PSEUDO-RESPONSE REGULATOR) genes play essential roles. 1) In addition to the clock-associated PRR genes, two homologous genes, CCA1 and LHY (CIRCADIAN CLOCK-ASSOCIATED 1 and LATE ELONGATED HYPOCOTYL), also played a critical clock-associated role(s). Hence the diurnal expression profiles of were also examined in both the wild-type and the cca1 lhy double mutant.,) With this alternative clock-defective mutant, RNA samples were prepared at intervals from 11-d-old young seedlings grown under a light/dark (LD) cycle (1 h light/1 h dark). The quantitative results of real-time PCR analyses of transcripts showed that the diurnal expression profiles of in the clock-defective cca1 lhy mutant were markedly different from those in the wild-type (Fig. C). The same analyses were carried out for and DBB3, with the same results (data not shown). These results support the thesis that these DBB genes were under the control of circadian clock under both the continuous light (LL) and the light/dark (LD) cycle condition. To extend this line of examination, RNA samples were prepared from rice seedlings (Oryza sativa, Nipponbare), grown under LL growth conditions, followed by Northern blotting hybridization analysis, with the rice OsDBB genes as probes (Fig. 1B). Expression of the OsDBB1, OsDBB3c, Os, and Os genes also oscillated in LL at the level of transcription (Fig. 3). Notably, the phases of their peaks were different from each other, as observed for Arabidopsis DBB genes (see Fig. ). The OsDBB3a and OsDBB3b genes expressed continuously (data not shown). Taking their expression patterns together, we concluded that /b corresponds to the single clock-controlled OsDBB1 gene (Os9g79), that OsDBB3c (Os1g) is the counterpart of DBB3, and that, and are orthologous to Os (Osg) and Os (Osg93) respectively (Fig. 1B). More importantly, both for Arabidopsis and rice certain members of DBB genes are commonly under the control of circadian clock. Plant Clock-Controlled B-Box Zinc Finger Proteins 13 OsDBB1 OsDBB3c Os Os 1 3 Time in LL (h) References OsDBB1 OsDBB3c Os Os 1 3 Time in LL (h) Fig. 3. Characterization of Homologous DBB Genes Found in Rice. Northern blot hybridization analysis with rice mrna prepared from rice seedlings under light/dark cycle conditions, as indicated (closed rectangle, night; open rectangle, daytime; hatched rectangle, subjective night). Other details were the same as those given in the legend to Fig. 1. The first conclusion of this study is that five out of eight Arabidopsis DBB genes are under the control of a circadian clock, and these genes are highly conserved in rice. These findings tempted us to examine further the functions of these five Arabidopsis clock-controlled and characteristic zinc finger genes (,, DBB3,, and ). The gene was not characterized, because its expression was not under the control of circadian rhythm. The expression profiles of DBB and DBB remain to be examined more closely by a sensitive RT-PCR method. Isolation of a set of T-DNA insertion mutants for the DBB genes Obviously, a direct approach to defining the physiological function of a given DBB gene is to characterize each loss-of-function (or null) mutant. To this end, we isolated homozygous T-DNA insertion mutants for the respective genes, as schematically illustrated in Fig.. The isolated homozygous mutants (or alleles) were dbb1a-1 (SAIL-79-E1), dbb1a- (SALK-19), dbb1b-1 (SALK-9), dbb3-1 (SALK-137), sto-1 (SALK-773), sto- (FLAG-D3), and sth-1 (GABI-91B). They are derivatives from the ecotype umbia (), except for sto- from Wassilewskija (Ws). In these mutant plants, no detectable amount of the corresponding full-length transcript was seen in RT-PCR analysis, except for dbb1a-1 and dbb1a- (two independent progenies were tested for each, see Fig. ). The dbb1a-1 and dbb1a- alleles isolated in this study appear to be undesirable for genetic studies (no other candidate of dbb1a mutant is available to the public at present). All the homozygous mutant plants grew well on MS agar-plates and in soil under standard laboratory growth conditions, and eventually they set reproductive flowers and seeds. We did not detect any overt phenotype for these mutant plants. 7 7

7 1 T. KUMAGAI et al. Downloaded by [ ] at 9: 3 November 17 dbb1a-1 DBB3 ATG WT L1 L3 dbb3-1 ATG WT L1 L dbb1b-1 WT L3 L WT L1 L sto- sto-1 ATG WT L1 L ATG ATG WT L1 L sth-1 Act WT L1 L dbb1a- Act TAA TAG TAG TGA B-box1 B-box Primers Fig.. A Set of T-DNA Insertion Mutants of DDB Genes. To characterize the functions of the clock-controlled DBB family genes, a set of candidate seeds of T-DNA insertion mutants were obtained, mainly from The Salk Institute Genomic Analysis Laboratory (San Diego, CA, USA) and The Arabidopsis Biological Resource Center (ABRC, umbus, OH, USA). They were established as homozygous mutant lines after the insertion sites were confirmed by nucleotide sequencing. The positions of primers were setting appropriately to screen each mutant allele. Rectangles indicate exons, and horizontal lines indicate introns. The coding regions for B-box-type zinc fingers are also indicated. Another replicated experiment was carried out, with essentially the same results (data not shown). TAA Some DBB genes might be implicated in early photomorphogenesis Some previous reports have suggested that the,, and genes are implicated in a light signal transduction pathway.,7) A hallmark assay examining the light signal transduction in young seedlings is the inhibition assay of elongation of hypocotyls during early photomorphogenesis at given fluence rates of red, farred, and blue light (or white light). Here, such assays were done for the set of dbb mutants. The results for red light sensitivity showed that the sto-1 and sto- mutant seedlings displayed detectable phenotypes as to the light sensitivity (Fig. ). The sth-1 mutant also showed a similar phenotype, but the others did not (Fig. ). Furthermore, the sto-1 and sto- mutant seedlings consistently showed phenotypes of hypersensitivity, not only to red, but also to far-red light (Fig. ). They were also hypersensitive to blue light, although the phenotypes were subtle (Fig. ). Such an alteration of far-red light and blue light sensitivity was less evident in the case of sth1-1 (data not shown). Otherwise, the other WT () dbb1a-. WT () dbb3-1 1 WT (Ws) WT () sto- sth-1 p <. p <. 1 Red light ( µmol m - s -1 ) mutant seedlings did not display any overt phenotype in this respect (the dbb1a T-DNA insertion mutants must be very leaky). As judged by the phylogenic profiles (Fig. 1), we concluded that these homologous DBB gene products might play redundant roles. To solve this problem in general, it might be necessary to construct multiple lossof-function dbb mutant lines. In fact, we have isolated a sto-1 sth-1 double mutant, together with a dbb1a- dbb1b-1 double mutant. Nevertheless, no new overt phenotype has been observed for them. Hence, it might be necessary to do more laborious genetics by creating multiple loss-of-function mutants, as by isolating a desirable loss-of-function allele for. We attempted to apply another conventional genetic strategy by establishing transgenic lines constitutively expressing (or miss-expressing) each of these clock-controlled DBB genes, as described below. WT () dbb1b-1 1 WT () sto-1 p <. 1 1 Fig.. Red Light Responses of a Set of Homozygous Mutants. Seedlings (Six types of mutants together with ) were grown under a given fluence rate of continuous red light as indicated, and in the dark for. The resulting lengths of the hypocotyls were quantitatively examined. Error bars represent the SE values of n >, and asterisks highlight significant differences (P < :). Other experimental details are given in Materials and Methods. The same experiments were repeated twice, with the same results.

8 Plant Clock-Controlled B-Box Zinc Finger Proteins 1 Downloaded by [ ] at 9: 3 November 17 WT () WT () sto-1 sto-1 p <. p <. 1 Far-red light ( µmol m - s -1 ) WT (Ws) sto- p <. 1 Far-red light ( µmol m - s -1 ) Isolation of transgenic lines each overexpressing a given DBB gene To this end, a series of Cauliflower Mosaic Virus (CaMV) 3S-promoter::DBB cdna composite transgenes were constructed and introduced into the wildtype umbia ecotype (). Stable homozygous transgenic lines were established for,, DBB3,, and (Fig. 7A; at least two independent lines were isolated for each). It was confirmed that a transcript was indeed over accumulated in any given transgenic line. More importantly, the clock-controlled transcripts were constitutively miss-expressed, regardless of the timing of preparation of the RNA samples (i.e., morning or evening; compare the pairs of lanes denoted by CT, circadian time). These transgenic lines were designated -ox (overexpresser), -ox, DBB3-ox, -ox, and -ox. The DBB genes were implicated in light signal transduction Next we examined the DBB-ox transgenic lines for their overt phenotypes. The homozygous (T3) seeds germinated well on MS agar-plates, and the young seedlings grew normally under standard laboratory 1 Blue light ( µmol m - s -1 ) WT (Ws) sto- p <. 1 Blue light ( µmol m - s -1 ) Fig.. Far-Fed and Blue Light Responses of the sto Homozygous Mutants. Mutant seedlings, together with, were grown under a given fluence rate of continuous far-red or blue light as indicated, and in the dark for d. The resulting lengths of the hypocotyls were quantitatively examined. Error bars represent the SE values of n >, and asterisks highlight significant differences (P < :). Other experimental details are given in Materials and Methods. Another replicated experiment was carried out, with the same results. A B DBB3-ox L L7 CT1 CT CT1 CT CT1 CT DBB3 -ox L1 L CT1 CT3 CT1 CT3 CT1 CT3 -ox L L9 CT CT1 CT CT1 CT CT1 3mm -ox L3 L CT1 CT CT1 CT CT1 CT -ox -ox DBB3-ox -ox L L CT1 CT CT1 CT CT1 CT -ox 7-d-old -ox Fig. 7. Isolation of Transgenic Lines Overexpressing the Respective DBB Transcripts. A, Northern blot hybridization analysis was carried out to detect overexpressed DBB transcripts. Two independent lines were established for each. To confirm miss-expression of the DBB genes in a manner independent of circadian rhythm, RNA samples were prepared at two different circadian times (CT = light on = morning). These overexpressers were designated -ox, -ox, DBB3-ox, -ox, and -ox. B, These established homozygous transgenic lines were germinated and grown on MS agar-plates for 7 d under light/dark cycle conditions, and representatives of the resulting seedlings were photographed for each. conditions. When a set of DBB-ox seedlings together with were germinated under white light and grown under standard dark/light conditions (1 h light/1 dark, 1 1 mmol m s 1 ), we noticed that the lengths of hypocotyls were different considerably (Fig. 7B). For instance, the average lengths of hypocotyls of -ox, -ox, -ox, and -ox seedlings were significantly longer than that of the wild-type. The phenotype of -ox was complementary to the earlier observation for the sto-1 and sto- lossof-function mutant seedlings (Figs. and ). In other words, the -ox seedlings were hyposensitive to light signals, while the sto loss-of-function mutant seedlings were hypersensitive. Under the same growth conditions, the hypocotyls of DBB3-ox were considerably shorter that that of. These phenotypes are indicative of light-dependent signal transduction during early photomorphogenesis, suggesting that these DBB genes might be commonly implicated in a light signal transduction pathway. To confirm the above phenotypes, the sensitivities to red light and far-red light during early photomorphogenesis of these DBB-ox seedlings were quantitatively examined and compared (Fig. ). The results indicated

9 1 T. KUMAGAI et al. Downloaded by [ ] at 9: 3 November ox 1 1 -ox 1 1 DBB3-ox 1 1 -ox 1 1 -ox DBB3-ox 1 1 -ox -ox ox 1 1 -ox Red light ( µmol m - s -1 ) Far-red light ( µmol m - s -1 ) Fig.. Red and Far-Red Responses of the Set of DBB-ox Transgenic Lines. Transgenic seedlings, together with, were grown at a wide range of fluence rates of continuous red or far-red light as indicated, and in the dark for 3 d. The resulting lengths of the hypocotyls were quantitatively examined. Error bars represent the SE values of n >. Another replicated experiment was carried out, with the same results. CCA1 -ox -ox DBB3-ox -ox -ox DBB3-ox that the -ox, -ox, -ox, and -ox seedlings were hyposensitive to both red and far-red light, as compared with in the seedlings, while the DBB3-ox seedlings were hypersensitive to both red and far-red light under the same conditions. The altered responses of these transgenic seedlings to blue light were less evident (data not shown). DBB genes act in output pathways downstream of the circadian clock Finally, it was of interest to determine whether the DBB genes act within the circadian clock per se, because certain clock-components are often involved in early photomorphogenesis.,9) To this end, -ox, -ox, and DBB3-ox transgenic plants were grown under a light/dark cycle (1 h light/1 h dark), and then they were released to continuous light to determine the rhythmic expression profiles of the clock component gene, CCA1 (Fig. 9). The results of Northern blot hybridization analyses showed that the rhythmic profiles of CCA1 in these transgenic plants were indistinguishable from that in wild-type. The results suggest that the DBB genes tested are not clock-associated genes, as judged by the general idea that overexpression of any clock-associated gene in plants results in severe perturbation of the clock function per se. 3) Rather, it is possible that DBB genes act in an output pathways downstream of the circadian clock. Discussion Time in LL (h) Fig. 9. Characterization of DBB Genes with Reference to the Circadian Clock. Together with wild-type, -ox, -ox, and DBB3- ox transgenic plants were grown under a light/dark cycle (1 h light/1 h dark), and then, they were released to continuous light. RNA samples were prepared at intervals 3 h, and they were analyzed by Northern blot hybridization analysis with a probe appropriate to CCA1 (solid rectangle, dark; open rectangle, day; shaded rectangle, subjective night). As loading references, the amounts of were analyzed. Other experimental details, including the probe for CCA1, were the same as described previously. 1) Taking the various results of this study together, it was found that many members of the DBB family of zinc finger genes were under the control of a circadian clock at the level of transcription. More importantly,

10 Downloaded by [ ] at 9: 3 November 17 they are implicated as putative transcriptional factors in light signal transduction. More specifically, the results suggested several tentative conclusions: (i) Arabidopsis thaliana has a small DBB subfamily consisting of eight members, each of which contains a CO-related double B-box zinc finger domain (Fig. 1). (ii) Among these, the transcription of five DBB family genes are under the control of circadian rhythm (Fig. ), (iii) they are commonly implicated in light signal transduction during early photomorphogenesis (Figs.,, and ), (iv) Their functions, however, are not completely redundant, as judged by the fact that their circadian-expression profiles were distinctive from each other (Fig. ), also by the fact that some DBB genes ( and,, and ) are implicated in light signal transduction, apparently in a negative manner with regard to light signal transduction through promotion of the elongation of hypocotyls, whereas another (DBB3) acts in a positive way in this sense (Fig. ). (v) Finally, genes homologous to Arabidopsis DBBs are widely conserved in higher plants (Fig. 1). Characterization of the physiological functions of DBB genes is at very early stage. Based on the findings of this study, however, it is of interest to study the DBB subfamily members with special reference to light signal transduction in plants. Among these DBB genes, the gene was originally identified as one that can complement the growth defect of a yeast calcineurindeficient mutant, which shows a phenotype of salinity sensitivity. Hence, it has long been thought that the gene might also be implicated in salt tolerance in plants. 1,) Another previous observation is that the protein interacts with COP1 (CONSTITUTIVE PHOTOMORPHOGENIC 1), which regulates early photomorphogenic responses by acting as a component of E3 ligase. 31) Then it was found that the gene participates in a light signal transduction pathway. ) The results of this study support this recent view. More importantly, the results further suggest that, not only but also other members of the DBB subfamily are implicated in the early photomorphogenesis in response to red, far-red, and blue light, as summarized above. The nature of the DBB family of putative transcription factors is reminiscent of the well-characterized photomorphogenic transcription factor HY (LONG HYPOCOTYL ), a bzip DNA-binding protein. 3) HY is a representative target of COP1, 33) which negatively regulated photomorphogenic processes in the dark, as explained above. HY acts as a positive regulator in light signal transduction. It was recently found that (Fig. 1) has the ability to interact with HY both in yeast and plants, and it is possible is implicated as a positive regulator in photomorphogenic developments. 19) Some DBB protein ( and,, and ) are implicated in light signal transduction apparently in a negative manner, whereas others ( and DBB3) act in a positive way (Fig. ). It is tempting to speculate that these DBB proteins might Plant Clock-Controlled B-Box Zinc Finger Proteins 17 interact with HY, thereby enhancing and/or inhibiting HY activity. It is further tempting to propose that some DBB zinc finger proteins, together with the HY bzip protein, play complex roles as positive/negative transcriptional regulators that are involved in the mechanisms underlying the early photomorphogenesis of higher plants in natural environments, which is currently a major subject in plant biology. 3) The expressions of many DBB family genes are under the control of circadian rhythm (or clock) (Fig. ). Furthermore, the timing (or phases) of expression is different form each other. We showed that some DBB genes act in output pathways downstream of the circadian clock (Figs. and 9). Currently, it is believed that the circadian clock regulates the rhythmic elongation rate of hypocotyls during early photomorphogenesis. ) Accordingly, certain Arabidopsis clock mutants (e.g., toc1 and prr) show altered phenotypes with reference to early photomorphogenesis.,9,3) In investigating the molecular mechanisms underlying such clock-controlled photomorphogenic responses, Nozue et al. (7) found that two basic helix-loop-helix (bhlh) transcription factor genes are crucially implicated, viz., PHYTOCHROME-INTERACTING FACTOR (PIF) and PIF/PIL. ) They suggested that these two clock-controlled bhlh genes integrate both the clock and the light signals to regulate the diurnal rhythmic growth of hypocotyls (or stems). This is a good example of a clock-controlled transcription factor playing a central role in the diurnally regulated photomorphogenesis of plant development. Hence, it is possible that the clock-controlled DBB family of zinc finger proteins is commonly implicated in a complex mechanism underlying the diurnal regulation of photomorphogenesis. In conclusion, in addition to the previously characterized bzip transcription factors (HY and HYH) and bhlh transcription factors (PIF and PIF/PIL), we propose that a set of B-box zinc finger proteins (including the newly identified /b and DBB3) must be taken into the consideration for a better understanding of the complex molecular mechanisms underlying the early photomorphogenic development of Arabidopsis thaliana. Furthermore, not only /b and DBB3, but also HY and HYH, PIF, and PIL, are all under the control of robust circadian rhythm at the level of transcription (our unpublished data). 3,37) Hence it is of interest to characterize the functions of /b and DBB3 within the cop1, hy, and/or pif/ mutant backgrounds to gain insight into the genetic linkages between the newly identified DBB family genes and well-known photomorphogenic genes. Acknowledgments This study was supported by Grant-in-Aid for Young Scientists (B) (Grant no. 177, to T.Y.) from the Ministry of Education, Culture, Sports, Science, and

11 Downloaded by [ ] at 9: 3 November 17 1 T. KUMAGAI et al. Technology of Japan. N.N. and S.I. were supported by the Japan Society for the Promotion of Science Research Fellowship for Young Scientists. References 1) Riechmann, J. L., Heard, J., Martin, G., Reuber, L., Jiang, C., Keddie, J., Adam, L., Pineda, O., Ratcliffe, O. J., Samaha, R. R., Creelman, R., Pilgrim, M., Broun, P., Zhang, J. Z., Ghandehari, D., Sherman, B. K., and Yu, G., Arabidopsis transcription factors: genome-wide comparative analysis among eukaryotes. Science, 9, 1 11 (). ) Putterill, J., Robson, F., Lee, K., Simon, R., and Coupland, G., The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors. Cell,, 7 7 (199). 3) Robson, F., Costa, M. M., Hepworth, S. R., Vizir, I., Pineiro, M., Reeves, P. H., Putterill, J., and Coupland, G., Functional importance of conserved domains in the flowering-time gene CONSTANS demonstrated by analysis of mutant alleles and transgenic plants. Plant J.,, (1). ) Griffiths, S., Dunford, R. P., Coupland, G., and Laurie, D. A., The evolution of CONSTANS-like gene families in barley, rice, and Arabidopsis. Plant Physiol., 131, 1 17 (3). ) Onouchi, H., Igeno, M. I., Perilleux, C., Graves, K., and Coupland, G., Mutagenesis of plants overexpressing CONSTANS demonstrates novel interactions among Arabidopsis flowering-time genes. Plant Cell, 1, 9 (). ) Samach, A., Onouchi, H., Gold, S. E., Ditta, G. S., Schwarz-Sommer, Z., Yanofsky, M. F., and Coupland, G., Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science,, (). 7) Imaizumi, T., and Kay, S. A., Photoperiodic control of flowering: not only by coincidence. Trends Plant Sci., 11, (). ) Blasing, O. E., Gibon, Y., Gunther, M., Hohne, M., Morcuende, R., Osuna, D., Thimm, O., Usadel, B., Scheible, W. R., and Stitt, M., Sugars and circadian regulation make major contributions to the global regulation of diurnal gene expression in Arabidopsis. Plant Cell, 17, (). 9) Cheng, X. F., and Wang, Z. Y., Overexpression of COL9, a CONSTANS-LIKE gene, delays flowering by reducing expression of CO and FT in Arabidopsis thaliana. Plant J., 3, 7 7 (). 1) Datta, S., Hettiarachchi, G. H., Deng, X. W., and Holm, M., Arabidopsis CONSTANS-LIKE3 is a positive regulator of red light signaling and root growth. Plant Cell, 1, 7 (). 11) Mizuno, T., Plant response regulators implicated in signal transduction and circadian rhythm. Curr. Opin. Plant Biol., 7, 99 (). 1) Mizuno, T., and Nakamichi, N., Pseudo-response regulators (PRRs) or true oscillator components (TOCs). Plant Cell Physiol.,, 77 (). 13) Wenkel, S., Turck, F., Singer, K., Gissot, L., Le Gourrierec, J., Samach, A., and Coupland, G., CONSTANS and the CCAAT box binding complex share a functionally important domain and interact to regulate flowering of Arabidopsis. Plant Cell, 1, (). 1) Kojima, S., Banno, H., Yoshioka, Y., Oka, A., Machida, C., and Machida, Y., A binary vector plasmid for gene expression in plant cells that is stably maintained in Agrobacterium cells. DNA Res.,, 7 1 (1999). 1) Makino, S., Matsushika, A., Kojima, M., Yamashino, T., and Mizuno, T., The APRR1/TOC1 quintet implicated in circadian rhythms of Arabidopsis thaliana: I. Characterization with APRR1-overexpressing plants. Plant Cell Physiol., 3, 9 (). 1) Sato, E., Nakamichi, N., Yamashino, T., and Mizuno, T., Aberrant expression of the Arabidopsis circadian-regulated APRR gene belonging to the APRR1/TOC1 quintet results in early flowering and hypersensitiveness to light in early photomorphogenesis. Plant Cell Physiol., 3, (). 17) Makino, S., Kiba, T., Imamura, A., Hanaki, N., Nakamura, A., Suzuki, T., Taniguchi, M., Ueguchi, C., Sugiyama, T., and Mizuno, T., Genes encoding pseudoresponse regulators: insight into His-to-Asp phosphorelay and circadian rhythm in Arabidopsis thaliana. Plant Cell Physiol., 1, (). 1) Carpenter, C. D., and Simon, A. E., Preparation of RNA. Methods Mol. Biol.,, 9 (199). 19) Datta, S., Hettiarachchi, C., Johansson, H., and Holm, M., SALT TOLERANCE HOMOLOG, a B-box protein in Arabidopsis that activates transcription and positively regulates light-mediated development. Plant Cell, 19, 3 3 (7). ) Indorf, M., Cordero, J., Neuhaus, G., and Rodriguez- Franco, M., Salt tolerance (), a stress-related protein, has a major role in light signalling. Plant J., 1, 3 7 (7). 1) Lippuner, V., Cyert, M. S., and Gasser, C. S., Two classes of plant cdna clones differentially complement yeast calcineurin mutants and increase salt tolerance of wild-type yeast. J. Biol. Chem., 71, 19 1 (199). ) Nagaoka, S., and Takano, T., Salt tolerance-related protein binds to a Myb transcription factor homologue and confers salt tolerance in Arabidopsis. J. Exp. Bot.,, (3). 3) Yano, M., Katayose, Y., Ashikari, M., Yamanouchi, U., Monna, L., Fuse, T., Baba, T., Yamamoto, K., Umehara, Y., Nagamura, Y., and Sasaki, T., Hd1, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS. Plant Cell, 1, 73 (). ) Nakamichi, N., Kita, M., Ito, S., Yamashino, T., and Mizuno, T., PSEUDO-RESPONSE REGULATORS, PRR9, PRR7 and PRR, together play essential roles close to the circadian clock of Arabidopsis thaliana. Plant Cell Physiol.,, 9 (). ) Niwa, Y., Ito, S., Nakamichi, N., Mizoguchi, T., Niinuma, K., Yamashino, T., and Mizuno, T., Genetic linkages of the circadian clock-associated genes, TOC1, CCA1 and LHY, in the photoperiodic control of flowering time in Arabidopsis thaliana. Plant Cell Physiol.,, (7). ) Mizoguchi, T., Wright, L., Fujiwara, S., Cremer, F.,

12 Downloaded by [ ] at 9: 3 November 17 Lee, K., Onouchi, H., Mouradov, A., Fowler, S., Kamada, H., Putterill, J., and Coupland, G., Distinct roles of GIGANTEA in promoting flowering and regulating circadian rhythms in Arabidopsis. Plant Cell, 17, 7 (). 7) Khanna, R., Shen, Y., Toledo-Ortiz, G., Kikis, E. A., Johannesson, H., Hwang, Y. S., and Quail, P. H., Functional profiling reveals that only a small number of phytochrome-regulated early-response genes in Arabidopsis are necessary for optimal deetiolation. Plant Cell, 1, (). ) Nozue, K., Covington, M. F., Duek, P. D., Lorrain, S., Fankhauser, C., Harmer, S. L., and Maloof, J. N., Rhythmic growth explained by coincidence between internal and external cues. Nature,, 3 31 (7). 9) Ito, S., Nakamichi, N., Nakamura, Y., Niwa, Y., Kato, T., Murakami, M., Kita, M., Mizoguchi, T., Niinuma, K., Yamashino, T., and Mizuno, T., Genetic linkages between circadian clock-associated components and phytochrome-dependent red light signal transduction in Arabidopsis thaliana. Plant Cell Physiol.,, (7). 3) Matsushika, A., Murakami, M., Ito, S., Nakamichi, N., Yamashino, T., and Mizuno, T., Characterization of Circadian-associated pseudo-response regulators: I. Comparative studies on a series of transgenic lines misexpressing five distinctive PRR Genes in Arabidopsis thaliana. Biosci. Biotechnol. Biochem., 71, 7 3 Plant Clock-Controlled B-Box Zinc Finger Proteins 19 (7). 31) Holm, M., Hardtke, C. S., Gaudet, R., and Deng, X. W., Identification of a structural motif that confers specific interaction with the WD repeat domain of Arabidopsis COP1. EMBO J.,, (1). 3) Oyama, T., Shimura, Y., and Okada, K., The Arabidopsis HY gene encodes a bzip protein that regulates stimulus-induced development of root and hypocotyl. Genes Dev., 11, (1997). 33) Osterlund, M. T., Hardtke, C. S., Wei, N., and Deng, X. W., Targeted destabilization of HY during lightregulated development of Arabidopsis. Nature,, (). 3) Nozue, K., and Maloof, J. N., Diurnal regulation of plant growth. Plant Cell Environ., 9, 39 (). 3) Mas, P., Alabadi, D., Yanovsky, M. J., Oyama, T., and Kay, S. A., Dual role of TOC1 in the control of circadian and photomorphogenic responses in Arabidopsis. Plant Cell, 1, 3 3 (3). 3) Yamashino, T., Matsushika, A., Fujimori, T., Sato, S., Kato, T., Tabata, S., and Mizuno, T., A Link between circadian-controlled bhlh factors and the APRR1/ TOC1 quintet in Arabidopsis thaliana. Plant Cell Physiol.,, 19 9 (3). 37) Fujimori, T., Yamashino, T., Kato, T., and Mizuno, T., Circadian-controlled basic/helix-loop-helix factor, PIL, implicated in light-signal transduction in Arabidopsis thaliana. Plant Cell Physiol.,, 17 1 ().