US20030056247A1

US20030056247A1 - Phytochrome regulated transcription factor for control of higher plant development

Info

Publication number: US20030056247A1
Application number: US10/084,553
Authority: US
Inventors: Elaine Tobin; Lin Sun; Zhi-yong Wang
Original assignee: University of California San Diego UCSD
Current assignee: University of California San Diego UCSD
Priority date: 1997-04-18
Filing date: 2002-02-25
Publication date: 2003-03-20
Also published as: CA2288976A1; EP0990029A1; EP0990029A4; US6388172B1; WO1998048007A1; AU7126398A; JP2002501381A

Abstract

The present invention involves the isolation and characterization of the first discovered phytochrome-regulated transcriptional factor, a protein designated CCA1 which binds to the promoter region of the chlorophyll binding protein gene (Lhcb1*3) of Arabidopsis. The Lhcb1*3 gene of Arabidopsis is known to be regulated by phytochrome in etiolated seedlings where a brief illumination by red light results in a large increase in the level of mRNA from this gene. A DNA binding activity, designated CA-1, that interacts with the promoter region of Lhcb1*3 was previously discovered in cellular extracts. This binding activity was used to obtain a cDNA clone for a transcription factor that binds specifically to the Lhcb1*3 promoter. Modification of the expression of CCA1 using techniques of genetic engineering results in unexpected changes in the timing of plant flowering. When CCA1 is overexpressed, it appears that the normal circadian rhythms of the plant are disrupted. The plants take a significantly longer time to reach flowering even in the presence of day length conditions that normally induce flowering. Thus, a method of extending vegetative growth and delaying flowering is provided.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention concerns the field of genetic engineering and more particularly the discovery of a unique light-regulated transcription factor that can be used to control the flowering time of plants.

2. Background of the Invention

The Sun is the primary source of energy on the Earth. It is obvious that impinging solar energy warms our atmosphere and drives the Earth's climate. Perhaps less obvious is that virtually all biological energy including the “fossil” fuels that power our civilization are solar in origin. Solar energy is captured for biological use by photosynthesis, a metabolic process that occurs in green plants. During photosynthesis light energy is captured in various chemical compounds that provide food for all nonphotosynthetic organisms.

Since green plants essentially “feed” on light, it comes as no surprise that these organisms are exquisitely sensitive to light. Many people are aware that plants grow towards a light source in an effort to receive sustaining illumination. However, a green plant's responsiveness to light is much more complex than merely growing towards a light source. Plants contain complex systems for actually measuring the duration of day and night lengths so as to synchronize their growth and lifecycles with the seasons. It is these timing processes that cause chrysanthemums to flower in the autumn and other ornamental and crop plants to flower and fruit at characteristic times. Clearly, the ability to accurately control flowering to promote it or delay it as necessary would be of great economic value. In the middle decades of this century a tremendous amount of biological research was carried out in search of the ever elusive “flowering hormone” or florigen which, for a time, was the holy grail of plant physiology. Although the quest for florigen ended in failure, much was learned about how plants perceive and respond to environmental factors such as the seasonal changes in day length.

Although green plants have multiple light receptors, the protein-pigment phytochrome has been shown to be the primary receptor by which plants track day length and orchestrate a number of light-regulated responses. Phytochrome is a chromoprotein formed by combining a linear tetrapyrolle pigment with an apoprotein. As such it shows some similarities to phycobiliproteins which are accessory pigments of certain algae and photosynthetic bacteria. Phytochrome has the somewhat unusual property of existing in two different photochemically interconvertible forms know as Pr (phytochrome-red) and Pfr (Phytochrome-far red). Phytochrome is synthesized in the Pr form which has an absorption maximum in the red region of the optical spectrum. Numerous experiments have shown that the Pr form of phytochrome is essentially inactive in terms of eliciting changes in plant metabolism. However, when Pr absorbs red light (R), it is rapidly converted into the active Pfr form. Pfr has an absorption maximum in the far red (near infrared) portion of the optical spectrum. Absorption of far red light (FR) induces a back conversion of Pfr to inactive Pr. This red-far/red interaction provides a powerful test of whether a given plant response is phytochrome mediated. For example, if dark-grown (etiolated) seedlings are briefly exposed to red light, Pfr will be formed and there will be a concomitant response. However, if the red light exposure is quickly followed by a far-red light exposure (which converts Pfr to inactive Pr) the response will be prevented. The reversibility of a red light response by a far-red light exposure is a hallmark of a phytochrome response.

Although much is known about the phytochrome proteins and their encoding genes, relatively little is known about how the Pfr effects a response in the plant. Many plant genes are light-regulated and that at least some of this regulation is controlled or influenced by phytochrome. Among the genes whose expression is either negatively or positively influenced by phytochrome are several that have been shown to be transcriptionally regulated. These genes include those encoding the small subunit of ribulose bisphosphate carboxylase/oxygenase, the major light-harvesting chlorophyll a/b binding-proteins (Lhcb) of Photosystem II, NADPH: protochlorophyllide oxioreductase, ferredoxin and phosphoenolpyruvate carboxylase, all components of photosynthesis. While the promoter regions are known for many of these genes, the transcriptional factors that bind to these nucleic acid regions are generally unknown. Furthermore, the signal transduction pathways connecting Pfr to these transcriptional factors are largely unknown. These matters have been recently reviewed in Tobin, E. M. and Kehoe D. M. “Phytochrome regulated gene expression,” Seminars in Cell Biology 5: 335-46 (1994) to which the reader is directed for more detailed information.

SUMMARY OF THE INVENTION

The present invention involves the isolation and characterization of the first discovered phytochrome-regulated transcriptional factor, a protein designated CCA1 which binds to the promoter region of a chlorophyll binding protein gene (Lhcb1*3) of Arabidopsis. The Lhcb1*3 gene of Arabidopsis is known to be regulated by phytochrome in etiolated seedlings where a brief illumination by red light results in a large increase in the level of mRNA from this gene. Karlin-Neumann, G. A., Sun, L., and Tobin, E. M. Plant Physiol. 88:1323-31 (1988). A DNA binding activity, designated CA-1, that interacts with the promoter region of Lhcb1*3 was discovered in cellular extracts. Sun, L., Doxsee, R. A., Harel, E., and Tobin, E. M., Plant Cell 5: 109-21 (1993) (Sun et al., 1993). The promoter region to which CA-1 binds has been shown to be necessary for normal phytochrome regulation of the Lhcb1*3 gene. Kenigsbuch, D. and Tobin, E. M. Plant Physiol. 108:1023-27 (1995). Modification of the expression of CCA1 using techniques of genetic engineering results in unexpected changes in the timing of plant flowering.

BRIEF DESCRIPTION OF THE DRAWINGS

The objects and features of the present invention, which are believed to be novel, are set forth with particularity in the appended claims. The present invention, both as to its organization and manner of operation, together with further objects and advantages, may best be understood by reference to the following description, taken in connection with the accompanying drawings. [0009]
FIG. 1 gives sequences of the A2 fragment of the Lhcb1*3 promoter and of DNA fragments used in EMSA analysis along with indications of nucleotide modifications that reduce CCA1 binding; in the probe sequences (WT1, m1, m2, m3, m4, and WT2) dashes indicate those nucleotides that are identical to the A2 probe while dots denote gaps introduced to optimize the alignment of conserved sequence elements; [0010]
FIG. 2 shows the complete nucleic acid sequence of CCA1, the genomic clone corresponding to the CCA1 cDNA along with the deduced amino acid sequence of the coding portions of the gene; [0011]
FIG. 3 shows the predicted amino acid sequence of CCA1 from amino acid residue 24 to 75 compared to the repeat sequences of various Myb proteins from animals, plants, and yeast; [0012]
FIG. 4 shows results of low-stringency hybridization of Arabidopsis DNA with a CCA1 probe; [0013]
FIG. 5 shows diagrams of the constructs used for expression of CCA1 polypeptides in [0014] E. coli;
FIG. 6 shows EMSA results for DNA binding activities of polypeptides produced from the constructs of FIG. 5; each reaction included 0.3 ng of [0015] ³²P-labeled A2 DNA probe and 1 μg poly(dI-dC); lanes 1-10, 15, and 16 each contain 1 μg of protein for E. coli either induced (+) or not induced (−) with IPTG; lanes 11 and 13 each contain 50 ng of purified GST-CCA1 fusion protein (Factor Xa [−]); and lanes 12 and 14 each contain 50 ng of purified CCA1 protein released from the fusion protein by Factor Xa cleavage (Factor Xa[+]);
FIG. 7 shows on the left the results of EMSA with the A2 fragment and the amount of proteins and poly (dI-dC) shown above each lane; F., free probe; B, CA-1 protein-DNA complex; B1 and B2, CCA1 protein-DNA complexes; on the right is shown the sequencing gel of the cleaved DNA recovered from the phenanthroline-copper reaction with S lanes representing the G+A chemical sequencing reaction and with the actual sequence of the protected region spelled out; [0016]
FIG. 8 shows gels of the effects of DNA modification on CCA1 binding where partially methylated (gels I and II) or depurinated (gels III and IV) A2 probe was labeled at the 3′ end of either the coding strand (I and III) or the noncoding strand (II and IV) were incubated with CCA1; the free DNA (F), the protein-bound DNA (B) and DNA not incubated with protein (C) were cleaved with piperidine and separated on sequencing gels; arrows mark the positions at which modification of DNA interferes with CCA1-DNA binding; [0017]
FIG. 9[0018] a shows the results of a competition experiment with CCA1;
FIG. 9[0019] b shows the results of a competition experiment comparable to FIG. 9a but using plant extract CA-1 instead of E. coli expressed CCA1;
FIG. 10 shows diagrams of constructs for expression of antisense CCA1 in transgenic plants; 35S is the promoter of the cauliflower mosaic virus; NOS the transcription termination sequence of the nos gene while arrows indicate the sense direction of the CCA1 gene with nucleotide positions numbered; [0020]
FIG. 11 shows results of RNase protection assays for Lhcb1*3 and rbcS-1A RNA; dark grown seedlings were given no light (D), 2 min R (R) or 2 min R followed by 10 min FR (F) 4 hours before harvesting; WT, wild type; [0021] Line 21 transformed with CCA1-AS21 construct; all other lines transformed with CCA1 -AS 1.4; ubq3 gene used as internal control;
FIG. 12 shows the induction of CCA1 RNA in etiolated seedlings which were grown for six days in the dark and then transferred to continuous light; RNA samples were taken either immediately before the transfer (0) or at the specified time after transfer and analyzed on gel blots; [0022]
FIG. 13[0023] a shows the circadian concentrations of CCA1 protein in a wild type plant along with the concomitant response of Lhcb1*3 RNA;
FIG. 13[0024] b shows the circadian concentrations of CCA1 protein and Lhcb1*3 RNA in a transgenic plant where CCA1 is constitutively expressed;
FIG. 14 shows a plot of CCA1 protein level versus hypocotyl length in a number of transgenic plant lines that were transformed with a CCA1 nucleic acid sequence according to the present invention; and [0025]
FIG. 15 shows a plot of CCA1 protein level versus bolting time (in days from seed germination) in a number of transgenic plant lines that were transformed with a CCA1 nucleic acid sequence according to the present invention. [0026]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The following description is provided to enable any person skilled in the art to make and use the invention and sets forth the best modes contemplated by the inventors of carrying out their invention. Various modifications, however, will remain readily apparent to those skilled in the art, since the general principles of the present invention have been defined herein specifically to provide the nucleic acid sequence, amino acid sequence and cloned protein of a phytochrome-regulated transcription factor that shows unexpected effects on development and flowering of plants. [0027]
Plant Materials and Growth Conditions [0028]
[0029] Arabidopsis thaliana ecotype Columbia was used in all experiments except for ecotype Nossen (No-O) which was used in transformation experiments with the antisense CCA1_constructs. The medium used for plant growth (MS2S medium) contained 1X MS salts (GIBCO BRL), 0.05% Mes, pH 5.7, 0.8% Phytagar (GIBCO BRL) and 2% sucrose. Light-grown plants were maintained at 24° C. in a growth chamber with light intensity of 150 μE m⁻²sec⁻¹Growth and light treatments of etiolated seedlings for phytochrome experiments were as described previously (Brusslan, J. A., and Tobin, E. M., Proc. Natl. Acad. Sci. USA 89:7791-95 (1992)). White onions used in nuclear localization experiments were purchased from a local supermarket.
Isolation and Sequence Characterization of CCA1 cDNA and Genomic Clones [0030]
Poly(A) RNA was isolated from leaves of Arabidopsis grown for 3 weeks on soil in continuous white light. A directional cDNA expression library was constructed in λgt22A using the SuperScript Lambda system (Bethesda Research Laboratory, Bethesda, Md.). The library was screened essentially as described by Singh, H., Clerc, R. G., and LeBowitz, J. H., [0031] BioTechniques 7:252-61 (1989), except that NEB buffer (25 mM Hepes-NaOH, pH 7.2, 40 mM KCl, 0.1 mM EDTA, 5 mM β-mercaptoethanol, 10% glycerol, Sun et al. 1993) was used as the binding buffer, and the washing solution was supplemented with 0.25% non-fat milk and 0.1% Triton X-100.
An Arabidopsis λ cDNA expression library was screened with the radiolabeled A2 fragment of the Lhcb1*3 promoter because this fragment had been previously shown to interfere with CA-1 binding activity in plant extracts. Approximately 640,000 unamplified recombinant phage plaques were screened in the first round using double-stranded A2 DNA probe (A2, FIG. 1). The positive plaques from the initial screening were rescreened using both the A2 probe and a mutant probe (ml probe, FIG. 1) that is known to poorly bind to the CA-1 activity (Sun et al. 1993). Two phage clones ([0032] clones 21 and 24) that bound only to the A2 and not to the ml probe were isolated as individual plaques. The cDNA inserts were subcloned into the SalI and NotI restriction sites of pGEM11Zf(−) (Promega, Madison, Wis.). Sequence analysis showed that the two clones overlapped by 470 nucleotides and were partial cDNAs derived from the same mRNA. Clone 24 included a polyadenylated tail of 15 bases and, therefore, encompassed the entire 3′ region of the mRNA. The 5′ end of the mRNA was determined by primer, extension analysis as described by Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K., eds., Current Protocols in Molecular Biology (New York: Greene Publishing Associates and Wiley-Interscience) (1987). The sequence of the oligonucleotide primer corresponded to nucleotides +42 to +22 of CCA1 cDNA clone 21. Fifty fmoles of ³²P-labeled primer was annealed to 45 μg of total RNA. The primer was extended using 9.5 units of Avian Myeloblastosis Virus (AMY) reverse transcriptase (Promega) at 37° C. for 1 hr. Dideoxy sequencing reactions were performed using the same ³²P-labeled oligonucleotide primer and CCA1 clone 21 plasmid DNA. This demonstrated that clone 21 included the complete 5′ region of the transcript. A full-length cDNA clone, designated clone 25, was constructed by joining the 5′ and 3′ fragments of clones 21 and 24, respectively, at the unique PstI site in the overlapping region. The 5′ end of the cDNA clone 24 was removed as a SalI-PstI fragment and replaced with that of clone 21. Sequencing the region spanning the PstI junction of clone 25 confirmed the reconstitution of wild-type sequence. The sequence of clone 25 is presented as SEQ ID NO:3.
A genomic clone corresponding to the CCA1 cDNA was isolated by screening a genomic library of [0033] Arabidopsis thaliana ecotype Columbia in λGEM1 1 (Promega, Madison, Wis.), using the SstI-NotI fragment of CCA1 cDNA clone 24 (corresponding to nucleotides 950-2254 of the full-length cDNA). The sequences of the CCA1 cDNA and overlapping fragments of the genomic clone were determined by the dideoxy chain termination method using a Sequenase kit (United States Biochemicals, Cleveland, Ohio) and double-stranded plasmid DNA. Both strands of the cDNA and genomic DNA were completely sequenced.
The sequence of complete gene is shown in FIG. 2 (also SEQ ID NO:1) along with the predicted amino acid sequence for the CCA1 protein. The gene sequence includes seven introns (the first in the 5′ noncoding region from [0034] nucleotide 190 to 273; the others from nucleotides 361 to 438, 551-638, 701-1179, 1375-1461, 1636-1718, and 2601-2670), 237 nucleotides of 5′ untranslated sequence, and 193 nucleotides of 3′ untranslated sequence. The 1824-nucleotide open reading frame (ORF) encodes a protein of 608 amino acids with a calculated molecular weight of 66,970 and an isoelectric point of 5.6. An ORF of 24 nucleotides is present in the 5′ untranslated region of the transcript and is in phase with the main ORF. Such ORFs have been shown in several cases to be involved in translational regulation of gene expression (Lohmer, S., Maddaloni, M., Motto, M., Salamini, F., and Thompson, R. D., Plant Cell 5:65-73 (1993); Hinnebusch, A. G., Trends in Biochem.Sci. 19:409-1.4 (1994)), and have also been found in other plant transcription factor genes (Singh, K., Dennis, E. S., Ellis, J. G., Llewellyn, D. J., Tokuhisa, J. G., Wahleithner, J. A., and Peacock, W. J., Plant Cell 2:891-903 (1990); Ruberti, I., Sessa, G., Lucchetti, S., and Morelli, G., EMBO J. 10:1787-91 (1991); Carabelli, M., Sessa, G., Baima, S., Morelli, G., and Ruberti, I., Plant J. 4:469-79 (1993); Lohmer et al., 1993).
Sequence Analysis and Data Base Searching [0035]
The protein and DNA sequences were analyzed using the MacVector software (IBI, New Haven, Conn.) and the Genetics Computer Group (Madison, Wis.) software package. The GenBank data base was searched with the amino acid sequence of CCA1by using the BLAST (Altschul, S. F., Gish, W., Miller, W., Myers, B. W., and Lipman, D. J., [0036] J. Mol. Biol. 215:403-10 (1990)) and FASTA programs (Pearson, W. R. and Lipman, D. J. Proc. Natl. Acad. Sci. USA 85:2444-48 (1988)) on the National Center for Biotechnology Information (NCBI) on-line service. Sequence alignment was assembled manually based on the results of data base searches. Genomic sequences of light-harvesting complex apoprotein (Lhc) genes and small subunit of ribulose bisphosphate carboxylase/oxygenase (rbcS) genes were retrieved from the Genbank data base using text search on the NCBI world wide web site. The presence of AATCT sequences in the promoter regions of the genes was detected using the FASTA program and further analyzed visually.
The predicted amino acid sequence of the CCA1 protein has a basic region at the N terminus (K-13 to K-107). Within this region is a sequence similar to the repeat sequence highly conserved in Myb-related proteins. FIG. 3 shows the predicted amino acid sequence of CCA1 from amino acid residue 23 to 75 compared to the repeat sequences of various Myb proteins from animals, plants, and yeast. Within this sequence, there is a limited amino acid identity (16 of 52; 31%) and substantial similarity (29 of 52; 56%) when compared to the third repeat of human c-Myb. The sequence identity includes two of the three conserved tryptophans present in most Myb proteins. The conserved residues also include seven of the 11 residues that are known to be important for forming the hydrophobic core and maintaining the three-dimensional structure of the Myb repeat, which forms a helix-turn-helix structure (Ogata, K., Hojo, H., Aimoto, S., Nakai, T., Nakamura, H., Sarai, A., Ishii, S., and Nishimura, Y., [0037] Proc. Natl. Acad. Sci. USA 89:6428-32 (1992)). However, the amino acid residues of human Myb that actually contact the DNA bases are not conserved in CCA1 (N-183 in hMyb, S in CCA1; K-182 versus R; N-186 versus Q; N-179 versus V) (Ogata, K., Morikawa, S., Nakamura, H., Sekikawa, A., Inoue, T., Kanai H., Sarai, A., Ishii, S., and Nishimura, Y., Cell 79:639-48 (1994)). In contrast to most other Myb proteins that have been characterized, this region is not repeated in the CCA1 protein. No other significant homology to any protein in the data base was found.
DNA and RNA Gel Blot Analyses [0038]
Genomic DNA isolation and DNA gel blotting were performed as described by Brusslan et al. (1993). Membranes were hybridized with P-labeled CCA1 cDNA fragments under high stringency conditions (final washes were at 65° C. in 0.1% SSC, [1X SSC is 0.15 M NaCl, 0.015 M sodium citrate] 0.1% SDS) and then stripped and reprobed under low-stringency conditions (hybridization at 32° C. in buffer containing 50% formamide, 0.25 M NaHPO[0039] ₄, pH 7.2, 0.75 M NaCl, 7% [w/v] SDS, and 1 mM EDTA and final washes at 45° C. in 2X SSC, 0.1% SDS). Total RNA was extracted from Arabidopsis seedlings as described by Brusslan and Tobin (1992). Total RNA was separated on a 1% agarose gel containing formaldehyde and blotted onto ZetaProbe membrane (Bio-Rad, Richmond, Calif.) following the manufacturer's instructions. RNA probes were synthesized by in vitro transcription using linearized plasmid DNA. CCA1 RNA probe was synthesized from CCA1 clone 24. To make the ubq10 RNA probe, a fragment of the 3′ untranslated region of the ubq10 gene (Callis, J., Carpenter, T., Sun, S. W., and Vierstra, R. D., Genetics 139:921-39 (1995)) was amplified by polymerase chain reaction (PCR) using the primers:

5′-CTGTTATGCTTAAGAAGTTCAATGT-3′ and

5′-CCACCCTCGAGTAGAACACTTATTCAT-3′.
The amplified fragment was digested with HindIII and XhoI and cloned into pGEM-11Zf(−). This plasmid DNA was digested with HindIII and used as template for synthesis of ubq10 RNA probe. The Lhcb1*3 RNA probe was made as described by Brusslan and Tobin (1992). The membrane blot was hybridized overnight with the RNA probes in buffer containing 50% formamide, 0.3 M NaCl, 0.05 M NaHPO[0040] ₄, pH 6.5, 1 mM EDTA, 1% SDS, 0.1% Ficoll (type 400), 0.1% polyvinylpyrrolidone, 0.1% BSA, 0.5 mg/mL yeast tRNA and 0.5 mg/mL herring sperm DNA. Hybridization of Lhcb1*3 ubq10 and CCA1 probes was performed at 55° C., 52° C. and 58° C., respectively. Final washes were performed at 65° C. in 0.1X SSC, 0.1% SDS. After hybridization with Lhcb1*3 and ubq10 probes, the blot was stripped by boiling in 0.1X SSC, 0.1% SDS, then hybridized with the CCA1 probe. The blots were imaged and quantified using a Phosphorlmager (Molecular Dynamics, Sunnyvale, Calif.). The measurement of the signal for each probe was adjusted for the uridine content of the probe and the exposure time, and the Lhcb1*3 and CCA1 signals were normalized to the ubq10 signal.
We tested whether CCA1 is a member of a gene family in Arabidopsis by genomic DNA gel blot analysis. The results of low-stringency hybridization of Arabidopsis DNA with a CCA1 probe are shown in FIG. 4. There was a single band of hybridization in the lanes that were digested with EcoRI, SstI, and PstI, which have no cleavage site in the probe fragment. There were two bands in the lane that was digested with HindIII, which has a cleavage site in the probe 188 bp from one end. An identical pattern was seen when the blot was hybridized with the same probe under the high-stringency conditions. The DNA gel blots were also hybridized under both low- and high-stringency conditions with a probe consisting of nucleotides 267-949 of the CCA1 cDNA, which includes the region of similarity to the Myb repeat. This probe gave no evidence for any additional closely related sequences. We conclude that although the CCA1 gene includes a small region with amino acid sequence homology to the Myb repeat, there are no genes that are closely related to CCA1 in the Arabidopsis genome. [0041]
Partial purification of Arabidopsis CA-1 protein, A2 probe labeling, and the electrophoretic mobility shift assays (EMSAs) were carried out as described by Sun et al. (1993). Competitor DNA fragments were prepared by annealing synthetic oligonucleotides. Competition EMSA experiments were performed by adding partially purified plant CA-1 protein or affinity-purified CCA1 polypeptide expressed from pXCA-24 in [0042] E. coli to the DNA binding reaction mixture containing A2 probe and specified amounts of competitor DNA fragments. The dried gels were imaged and quantified using a Phosphorlmager.
The fact that the two cDNA clones isolated in the initial filter-binding screening each contained the sequence similar to the Myb repeat suggested that this region was necessary for DNA binding. We tested this possibility and further characterized the CCA1 protein by expressing the polypeptides encoded by various fragments of the CCA1 cDNA. The CCA1 cDNA clones were fused to the coding sequence for glutathione 5-transferase (GST) and used to produce the polypeptides in [0043] Escherichia coli. These constructs are diagrammed in FIG. 5. The polypeptides produced were tested for their ability to bind to the A2 fragment of the Lhcb1*3 promoter by electrophoretic gel mobility shift assays (EMSA). The polypeptides corresponding to the pXCA-23 and pXCA-24 constructs were produced as isopropyl β-d-thiogalactopyranoside (IPTG)-inducible GST-fusion proteins, and were also tested as purified proteins after cleavage from GST. Those corresponding to the cDNA clones CCA1-21 and CCA1-25 (pXCA-21 and pXCA-25, respectively) contained stop codons in the 5′ untranslated region of the cDNA and, thus, were not produced as fusion proteins. FIG. 6 shows an EMSA using either E. coli extracts (lanes 1 to 10, 15 and 16) or with purified proteins before (lanes 11 and 13) and after (lanes 12 and 14) cleavage from GST. DNA binding activities induced by IPTG were observed for proteins produced from constructs pXCA-21, pXCA-24, and pXCA-25, but binding activity could not be detected for the protein produced from construct pXCA-23, which lacked the N-terminal 82 amino acids In conjunction with the finding that the N-terminal 11 amino acids are not necessary for binding (construct pXCA-24), these experiments demonstrate that the sequence containing amino acid residues 11-82 of CCA1, which includes the region with similarity to the Myb DNA binding domain (amino acids 24-75)., is essential for the DNA binding activity of the CCA1 protein. Therefore, homologous CCA1 proteins from other plant species will share this highly conserved binding domain most likely with 85% or higher homology. The asterisk in the figure marks a nonspecific DNA binding activity. The arrow and triangles denote the positions of the major protein-DNA complexes formed by the GST-CCA1 fusion protein and the non-fusion CCA1 polypeptides, respectively. Lanes 15 and 16 are longer exposures of lanes 5 and 6.
Protein Expression in [0044] Escherichia coli and Purification of GST-CCA1 Proteins
The constructs diagrammed in FIG. 5 were made by cloning the CCA1 cDNA fragments into pGEX-3X (Pharmacia) using polymerase chain reaction (PCR)-aided cloning with the following 5′ primers: 5′-GGCCGGGATCCAATTCGTCGACCCACGCG-3′ for pXCA-21, pXCA-24, pXCA-25 and 5′-TAAAGGGATCCATATGGGTCAAGCGCTAG-3′ for pXCA-23. A 3′ primer (5′-ATAGAATTCTCGAGCTTATGCATGCGG-3′) was used for pXCA-21, pXCA-24, pXCA-25 and pXCA-23. The appropriate plasmid DNA (0.5 μg) was amplified for 10 cycles and the PCR products were digested with EcoRI and BamHI. The cDNAs of [0045] clones 21, 24, and 25 and the 483 to 2254-nucleotide region were cloned into pGEX-3X yielded pXCA-21, pXCA-24, pXCA-25 and pXCA-23, respectively. Sequencing of the junction region between the glutathione S-transferase (GST) gene and cDNA confirmed the construction of a translational fusion in pXCA-24 and pXCA-23.
The plasmid constructs were transformed into [0046] E. Coli BL21 (DE3). Protein expression, purification of GST-CCA1 fusion proteins using glutathione agarose, and purification of CCA1 polypeptides by cleavage of matrix-bound GST fusion protein with Factor Xa were performed following the procedure of Ausubel et al. (1987). Protein concentrations were determined by the Bradford assay (Bio-Rad, Richmond, Calif.) using BSA as standard.
Phenanthroline-Copper Footprinting [0047]
The A2 fragment was labeled with [0048] ³²P at the 3′ end of the sense strand by end filling (Sun et al., 1993). Footprint experiments were carried out as described by Kuwabara, M. D., and Sigman, D. S., Biochemistry 26:7234-38 (1987). The EMSA reactions were scaled up fivefold; 10⁶cpm of probe and specified amounts of protein and poly(dI-dC) were used in each reaction. After electrophoresis, the gel was treated with phenanthroline-copper, then exposed wet to x-ray film for 40 min. The bands representing free DNA and protein-DNA complexes were excised from the gel. DNA was eluted from the gel slices, recovered by ethanol precipitation, and loaded on an 8% polyacrylamide-urea sequencing gel. The G+A chemical cleavage sequencing reaction was performed as described by Maxam, A. M., and Gilbert, W., Methods Enzymol 65:499-580 (1980).
Partial methylation and depurination of the A2 DNA probe was performed following the procedure of DNA chemical sequencing (Maxam and Gilbert, 1980). Five ng (10[0049] ⁵cpm) of modified DNA probe was incubated with 0.8 μg of affinity purified CCA1 protein in 50 μL NEB buffer containing 5 μg poly(dI-dC) and 10 μg BSA. The protein-bound and free DNA were separated by filtering the mixture through a nitrocellulose membrane (Ausubel et al., 1987). Free and bound fractions of DNA were recovered, and cleaved with piperidine following the DNA chemical sequencing procedure. An aliquot of probes not incubated with protein was also cleaved with piperidine as a control. Equal amounts of radioactivity from each sample were used on an 8% polyacrylamide-urea sequencing gel.
To compare the binding characteristics of the CCA1 protein and CA-1 activity from the plants, we carried out footprint analyses and binding competition experiments using the A2 fragment of Lhcb1*3 promoter as a probe. The results of 1,10-phenanthroline-copper footprinting are shown in FIG. 7. At left is the EMSA that was performed to resolve the free probe and the DNA-protein complexes. Cleaved DNA was recovered from each band after treatment of the gel with phenanthroline-copper and resolved on the sequencing gel shown at right. With increasing amounts of the CCA1 protein purified from [0050] E. coli, complexes (B1 and B2) of different mobilities could be observed. The nucleotides protected from cleavage in each of the complexes can be seen on the sequencing gel on the right. In complex B1, the −92 to −105 region was protected, and in complex B2, regions from −92 to −105 and from −111 to −122 were protected. This result suggests that the two complexes of different mobilities are a result of the presence of two separate binding sites on this fragment, and that the −92 to −105 region is the higher affinity binding site for CCA1. A nearly perfect repeated sequence of AAA^A/_CAATCTA occurs in each of these footprinted regions.
The CA-1 protein-DNA complex obtained with the plant cell extract (FIG. 7, lane 4) showed protected nucleotides in the region from −94 to −105, and a second experiment confirmed these boundaries, demonstrating that CA-1 (from plant extracts) and CCA1 (from the clone expressed in [0051] E. coli) bind to the same region of the Lhcb1*3 promoter.
FIG. 8 show the results of methylation interference and depurination interference experiments performed with the CCA1 protein. The figure shows the interfering nucleotides on sequencing gels, and their position on the A2 fragment of the promoter is shown in FIG. 1, along with the results of footprinting experiments. Interference with the protein-DNA binding by the modification of a base residue is manifested by increased intensity in the lane with the free DNA fraction and reduced intensity in the lane of protein-bound DNA compared to the lane of control DNA that was not incubated with protein. Both methylation and depurination interference assays identified the same nucleotides, and showed that nucleotides within both nearly perfect repeats (AA[0052] ^A/_CAATCTA) interact directly with the CCA1 protein. In FIG. 1 thick and thin lines 12 and 14 show the regions protected by CCA1 and CA-1, respectively, in the footprint assay. Asterisks in the figure indicate nucleotides that interfere with CCA1-DNA binding when methylated; boldface indicates the nucleotides that interfere with binding when depurinated.
FIG. 1 also summarizes the results of the phenanthroline-copper footprinting. We used unlabeled competitor DNAs in the EMSA to compare binding specificities of the CCA1 protein produced in [0053] E. coli and the CA-1 activity from the plant extracts. The wild-type and mutant promoter fragments used as competitors are shown at the bottom of FIG. 1. A representative result of such experiments is shown in FIG. 9a for CCA1 and FIG. 9b for CA-1. The binding of the E. coli-produced CCA1 protein to the probe was efficiently competed for by either a fragment of the A2 probe that contained the repeated sequence or by a promoter fragment (WT2) of another closely related Lhcb gene (Lhcb1*1, originally called AB165; Leutwiler, L. S., Meyerowitz, E. M., and Tobin, E. M., Nucl. Acids Res. 14:4051-64 (1986)) that contains one copy of this sequence (AAAAATCT). The m3 fragment, which had altered nucleotides in the downstream repeat region, was a less effective competitor than was the wild-type (WT1); m1, m2, and m4 fragments, which had alterations in both repeats, showed the least competition.
When plant extracts were used, all the fragments showed some degree of competition, which is likely in part to be the result of low amounts of the CA-1 protein in plant extracts. The results are not directly comparable to those with the purified CCA1 protein because the absolute amounts of the specific binding proteins are not known. Nonetheless, it can be seen that the m2 fragment served as a better competitor for CA-1 than did the ml fragment, whereas the opposite was found with CCA1. Even more striking are the contrasting results with the m4 competitor DNA. This fragment, in which the C residues of both TCT motifs in the two repeats were changed, was even more effective than was the wild type in competing for the CA-1 activity, whereas it was not a particularly good competitor for CCA1. Thus, although both activities interact with the AAAAATCT sequence, there are differences in the importance of individual nucleotides in this sequence for the binding of CA-1 and CCA1. [0054]
The CCA1 protein interacted with two closely spaced binding sites with nearly perfect 10-bp repeated sequences (AAA[0055] ^A/^CAATCTA) in the Lhcb1*3 promoter. Previous results (Sun et al., 1993) and the results of the phenanthroline-copper footprinting (FIG. 7) show that the CA-1 -activity could protect the same nucleic acid sequence as CCA1. There are, however, some differences in the relative importance of specific nucleotides for the binding of the two activities. The binding of CCA1 was more affected by alteration in the TCT sequence than by alterations in the AAAAA, whereas the opposite was observed with the plant extract activity (cf. m3 and m4, FIG. 9). It is possible that the differences observed are due to differences in modifications of the protein in E. coli and plants or that the CA-1 activity in the plant extracts might be associated with an additional protein or proteins which alter the binding characteristics. It is also possible that CA-1 and CCA1 are actually the products of two different genes, or the result of alternative splicing, in which case they may compete for the same binding sites.
Nuclear Localization [0056]
Onion epidermal peels were transformed by biolistic transformation and analyzed for GUS activity, and nuclei localization was as described in Varagona, M. J., Schmidt, R. J., and Raikhiel, N. V., [0057] Plant Cell 4:1213-27 (1992). Histochemical staining was visualized using a Zeiss Axiophot microscope and photographed using Kodak Ektachrome (Elite Series) ASA 400 film.
Transient expression assay in onion epidermal cells tested whether the product of the CCA1 gene was localized to nuclei, as would be expected for a transcription factor. The uidA gene, which encodes, β-glucuronidase (GUS), was fused in frame to the coding sequence of CCA1 so that GUS activity could be used to localize the compartmentation of the CCA1 protein. An XbaI site and a BamHI site were introduced into CCA1 by PCR amplification of [0058] cDNA clone 25 using the 5′ primer
(5′-GAAGTTGTCTAGAGGAGCTAAGTG-3′)
and 3′ primer (5′-ATGTGGA[0059] TCCTTGAGTTTCCAACCGC-3′) (mismatches are underlined). The resulting PCR product was digested with XbaI and BamHI and inserted in pB1221 (Clontech, Palo Alto, Calif.), yielding p35S-CCA1-GUS. This construct contains CCA1_coding sequence as a 1828-bp XbaI-BamHI fragment inserted between the cauliflower mosaic virus 35S promoter and the uidA gene. pMF::GUS and pMF::B::GUS were obtained from N. Raikhel (Michigan State University, East Lansing, Mich.); construction of these plasmids is described in Varagona et al. (1992). This transient assay should result in the expression of GUS activity in individual cells into which the DNA is effectively introduced. When a CaMV 35S::uidA construct (pMF::GUS) was used in this assay, GUS activity was detected throughout the cytoplasm. When a construct (pMF::B::GUS) with the opaque2 gene, which encodes a well characterized transcription factor from maize, fused to the uidA gene was used, GUS activity was detected specifically in nuclei. Similarly, specific nuclear localization was found for the CCA1-GUS (p35S-CCA1-GUS) fusion protein. These results show that the CCA1 protein is targeted to nuclei and are consistent with the function of the CCA1 protein as a transcription factor.
Plant Transformation Antisense [0060]
It has previously been shown that the promoter region to which the CA-1 activity binds is essential for phytochrome regulation of the Lhcb1*3 gene (Kenigsbuch and Tobin, 1995). Therefore, if the product of the CCA1 gene interacts with this promoter in vivo, it might be expected to affect the phytochrome induction of Lhcb1*3 expression. We addressed this possibility by transforming Arabidopsis with portions of the CCA1 gene in an antisense orientation driven by the constitutive [0061] cauliflower mosaic virus 35S promoter (see FIG. 10). The SstI-NotI fragments of CCA1 clones 21 and 24 in pGEM11Zf(−) were cloned into pBluescript KS (Stratagene, La Jolla, Calif.) at the corresponding sites. The resulting plasmids were digested with BamHI and SstI, and the cDNA fragments were cloned into the BamHI and SstI sites of binary vector pB1121 (Clontech), replacing the uidA gene coding sequence. The binary vectors were transformed into Agrobacterium tumefaciens A2260. Arabidopsis ecotype No-O plants were transformed with the above constructs using the Agrobacterium-mediated root transformation procedure described by Valvekens, D., Van Montagu, M., and Van Lijsebetten, M., Proc. Natl. Acad. Sci. USA 85:5536-40 (1988).
For each independent transgenic line, T2 seeds homozygous for the T-DNA insertion were selected by analysis of the segregation of kanamycin resistance, and seedlings from these homozygous seeds were tested for the phytochrome responsiveness of both the endogenous Lhcb1*3 gene and another phytochrome regulated gene, the rbcS-IA gene. FIG. 11 shows that in five of these seven lines, the level of Lhcb1*3 mRNA after the red treatment was substantially lower than that of the wild-type. However, no substantial effect of the antisense construct was seen for rbcS-IA gene expression in the same lines. The mRNA levels for Lhcb1*3 and rbcS-IA were normalized to a ubiquitin RNA (ubq3; cf Brusslan and Tobin, 1992), and the relative expression levels of these two genes for all the lines and treatments are shown below the autoradiogram. The increase of Lhcb1*3 RNA in response to R was reduced in [0062] lines 4, 14, 17, 21, and 34, ranging from 37-53% that of the wild type, but the induction of the rbcS-IA RNA was not comparably affected. The fact that none of the lines we recovered showed strong suppression of CCA1 RNA suggests the possibility that complete loss of function might be highly deleterious. It is also notable that the antisense construct did not affect the mRNA levels in plants that had been given no light treatment or had been given FR following the R treatment. No obvious visible phenotype was apparent in the antisense lines.
We also used the T3 generation of four of the lines to test whether the reduction in Lhcb1*3 RNA correlated with a reduction in levels of CCA1 RNA. In this generation, there was a smaller effect of the antisense construct. It is not unusual for antisense effects to be lost or diminished in subsequent generations. For example, Chamnongpol, S., Willekens, H., Langebartels, C. Van Montagu, M. Inze, D. And Van Camp, W., [0063] Plant J. 10:491-503 (1996) found that seven of eight lines expressing an antisense construct for a catalase gene lost the catalase suppression phenotype in their progeny. The R induction of Lhcb1*3 RNA ranged from 68 to 86% of the wild-type in the T3 seedlings of these four lines, and the levels of CCA1-RNA were 68 to 75% that of the wild-type plants. Although the effect of the antisense constructs was substantially reduced in this generation, the reduction of CCA1 RNA was accompanied by a similar decrease in the induction of Lhcb1*3 RNA by R. Our results demonstrate that the CCA1 protein can have a specific effect on the phytochrome induction of expression of the endogenous Lhcb1*3 gene in vivo, and strongly indicate that this protein is a part of the normal transduction pathway for the phytochrome response of this gene.
Plant Transformation Constitutive Expression [0064]
As demonstrated above, CCA1 is a transcription factor that is intimately involved in the phytochrome-induced regulation of Lhcb1*3 and the corresponding light harvesting chlorophyll binding protein. Somewhat paradoxically CCA1 may be itself phytochrome regulated. FIG. 12 shows the time course of production of CCA1 and Lhcb1*3 RNAs when dark-grown seedlings are transferred into light. CCA1 is induced within 1 hr and peaks at before 2 hr, thereafter decaying away. Lhcb1*3 begins to appear following the peak in CCA1 and continues at a high level as long as the plants remain in the light. This transient CCA1 response might suggest involvement of CCA1 in more complex events. [0065]
One way of exploring the overall effects of the phytochrome control of CCA1 is to remove it from this phytochrome control and ascertain what, if any effects, there are on plant growth and development. This can be achieved by reversing the direction of the CCA1 sequence inserted into the constructs shown in FIG. 10. Now instead of producing an antisense message, the CCA1 gene behind the 35S cauliflower mosaic promoter results in constitutive production of the CCA1 protein. FIG. 13[0066] a shows the circadian variations in wild-type expression of CCA1. Here the plants have been grown under a normal light/dark regime and then transferred into continuous light. The endogenous circadian rhythm of CCA1 production continues even under constant light. This rhythm closely matches the original light/dark period as is indicated by the time line. Note that the CCA1 is fully expressed at the start of the light period (ZT 1). This production decays before the end of the light period so that by the start of the dark period (ZT 12) CCA1 production is essentially absent—even though the plants remain under constant illumination. CCA1 production leads the beginning of the next light period (ZT 25) being apparent by ZT 19. This circadian rhythm was originally entrained to light/dark periods but clearly continues in plants kept in constant light. Lhcb1*3 RNA tracks CCA1 but decays more slowly so that its level reaches a minimum during the dark period some hours after minimal CCA1 levels have been reached. This difference in phase between CCA1 and Lhcb1*3 RNA results from a lag between CCA1 protein synthesis and RNA synthesis and slow turnover of Lhcb1*3 RNA. However, it is possible that this lag is actually part of the “clock” mechanism by which the plant actually maintains it circadian rhythm.
FIG. 13[0067] b shows that when CCA1 is constitutively expressed in a transgenic plant not only does CCA1 levels remain constant(as expected) but Lhcb1*3 RNA levels also remain constant, thereby damping or obscuring the entire circadian rhythm of Lhcb1*3. In the presence of light the expression pattern of CCA1 now controls the expression of. Lhcb1*3. However, if CCA1 is actually part of the “clock” mechanism, overall results may be more profound.
The real question, then, is whether this apparent damping of the circadian rhythm affects only the level of Lhcb1*3 transcription or represents a more widespread influence on the “clock” that controls plant development and whether such an effect is exhibited under normal light/dark regimes as opposed to constant illumination. Generally the phenotype of the 35S-CCA1 plants is normal. One of the only detectable morphological effects appears to be a relationship between hypocotyl length measured at six days and CCA1 level in a given plant. CCA1 levels can vary considerably from one transgenic line to another. FIG. 14 shows a regression analysis of hypocotyl length versus CCA1 level for 14 different transgenic lines. There is a strong correlation (r=0.73) between increased hypocotyl length and increased levels of CCA1 indicating that longer hypocotyls results from CCA1 overexpression-certainly not a result expected from a transcription factor that controls Lhcb1*3 RNA. Normally, hypocotyls are shorter in bright light than in dark-grown plants. Perhaps the constitutive CCA1 expression is interfering with the plant's perception of light versus dark. This does suggest that CCA1 effects go beyond Light harvesting chlorophyll binding protein. [0068]
Much more exciting and totally unexpected is the effect of CCA1 level on days to flowering of Arabidopsis. Normally the plants are influenced both by day length and by days of growth from seed germination. Under short day conditions the plants will show a relatively prolonged growth phase before they bolt and begin to flower. However, under long day conditions the plants very quickly transit from the vegetative phase to the reproductive phase. That is, if seeds are germinated late in the season (as the days are growing longer) even small seedlings quickly begin to flower completely skipping most of the normal vegetative growth phase. This same behavior is shown by a large number of plants. Most gardeners are well aware of the way that radishes, spinach and lettuce will begin to flower as the summer approaches. This behavior is annoying since the palatability of the vegetables is ruined. This behavior is of far greater economic importance with forage crops, in particular pasture grasses such as rye or fescue. The main value of these crops is in the food their vegetative structures provide to domestic animals. When the forage plants begin to flower, their production of vegetative biomass ends and their value as a crop ceases. Until the present invention the only way to deal with this problem was the use of traditional plant breeding methods to select varieties that were slower to flower. This approach has had some little success but the selected varieties could be improved. By a further delay in flowering. [0069]
FIG. 15 plots the number of days to flowering (bolting) from seed germination for a number of different transformed lines that overexpress CCA1. Just as hypocotyl length was related to CCA1 level in FIG. 14, in FIG. 15 bolting time is highly correlated with CCA1 level (r=0.81). As CCA1 level increases, flowering is delayed so that CCA1 affords the first general method for delaying flowering in plants. This is strong evidence that CCA1 is more than just a transcription factor for regulating Lhcb1*3 RNA. Since altering the level of this factor significantly delays flowering and apparently damps circadian rhythms it seems likely that this factor is part of the “clock” mechanism and is intimately involved in the regulation of a large number of “timing” related aspects of plant development. Since the DNA-binding portion (amino acids 24-75) of CCA1 is highly conserved it is very likely that this protein will be effective in a wide variety of plants and that the method of the present invention will modify flowering in virtually all plants. Conversely, while other species may have homologous CCA1 proteins whose sequences vary from that disclosed herein, the method of using those sequences to modify flowering time is identical to that disclosed and claimed herein. Alteration of plant development by transformation of plants with any nucleic acid sequence that translates to a transcription factor showing significant homology to the key binding region (amino acids 24-75) of CCA1 is contemplated by the present invention. [0070]
Now by constitutively expressing CCA1 it is possible to disrupt the plants built in timing system. With this system disturbed, the plant is much less able to respond to increases in day length as the growing season progresses. This results in a significant increase in vegetative growth and accumulated biomass before a transition to the reproductive state occurs. In the case of forage crops this translates to a dramatic yield increase as a given planting continues to produce biomass for a longer time. In the case of seed crops (e.g. rape seed a relative of Arabidopsis) a delay of flowering can translate to larger plants and a larger seed yield as long as flowering is not delayed too long into the growing season. The first phytochrome-regulated transcription factor provided by the present invention represents the first known way to manipulate plant circadian rhythms and hence flowering through genetic engineering. It seems likely that CCA1 will also serve as the key to unlock other aspects of the phytochrome-based timing system in plants. [0071]
Those skilled in the art will appreciate that various adaptations and modifications of the just-described preferred embodiment can be configured without departing from the scope and spirit of the invention. Therefore, it is to be understood that, within the scope of the appended claims, the invention may be practiced other than as specifically described herein. [0072]

Claims

What is claimed is:

1. A transcription factor comprising a member selected from the group consisting of:

(a) a peptide having an amino acid sequence of SEQ ID NO:2;

(b) a peptide having an amino acid sequence identical to a peptide produced by translation of coding portions of nucleic acid sequence Seq ID NO:1;

(c) a peptide having an amino acid sequence identical to a peptide produced by translation of nucleic acid sequence SEQ NO:3;

(d) a peptide having at least 95% sequence homology to peptide (a).

2. An isolated polynucleotide comprising a member selected from the group consisting of:

(a) a polynucleotide having a sequence identical to SEQ ID NO:1; and

(b) a polynucleotide which hybridizes to and which is at least 95% complementary to polynucleotide (a); and

(c) a polynucleotide that is exactly complementary to polynucleotide (b).

3. An isolated polynucleotide comprising a member selected from the group consisting of:

(a) a polynucleotide having a sequence identical to SEQ ID NO:3;

(c) a polynucleotide that is exactly complementary to polynucleotide (b)

4. A method of altering plant development comprising transforming a plant with nucleic acid sequence selected from the group consisting of:

(a) a polynucleotide having a sequence identical to SEQ ID NO:1;

(b) a polynucleotide which hybridizes to and which is at least 95% complementary to polynucleotide (a)

(c) a polynucleotide having a sequence identical to SEQ ID NO:3; and

(d) a polynucleotide which hybridizes to and which is at least 95% complementary to polynucleotide (c).

5. A transgenic plant produced by transforming a plant with a nucleic acid sequence selected from the group consisting of:

(a) a polynucleotide having a sequence identical to SEQ ID NO:1;

(c) a polynucleotide having a sequence identical to SEQ ID NO:3; and

6. A method of altering plant development comprising transforming a plant with a nucleic acid sequence coding for a CCA1 protein, said protein having a domain showing at least 85% homology to amino acids 24-75 of SEQ ID NO:2.

7. A transgenic plant transformed with a nucleic acid sequence coding for a CCA1 protein, said protein having a domain showing at least 85% homology to amino acids 24-75 of SEQ ID NO:2.