CA2177217A1 - Compositions and methods for production of male-sterile plants - Google Patents

Abstract

The present invention provides plant calcium/calmodulin-dependent protein kinase (CCaMK) nucleic acids, polypeptides, antibodies, and related methods. CCaMK genes are expressed in anthers in a developmental stage-specific manner. Suppression of CCaMK expression, e.g., by an antisense transgene, results in male-sterility.

Description

21772~7 _ AED:en 4630-45115.app 5/23/96 - 1 -COMPOSITIONS AND METHODS FOR PRO~-llON OF
MALE-STERILE PLANTS

Cro~s Reference to Related Cases This application claims the benefit of U.S.
Provisional Application No. 60/014,743, filed March 28, 1996, incorporated herein by reference.

Acknowledqment of Government Sup~ort This invention was made with government support under National Science Foundation grant number DCB 91-4586. The government has certain rights in this invention.

Technical Field This invention relates to plant calcium/calmodulin-dependent protein kinases, particularly anther-specific calcium/calmodulin-dependent protein kinases.

Back4 o~-d of the Invention Calcium and calmodulin regulate diverse cellular processes in plants (Poovaiah and Reddy, CRC Crit. Rev.
Plant Sci. 6:47-103, 1987, and CRC Crit. Rev. Plant Sci.
12:185-211, 1993; Roberts and Harmon, Annu. Rev. Plant Physiol . Plant Mol . Biol . 43:375-414, 1992; Gilroy and Trewavas, BioEssays 16:677-682, 1994). Transient changes in intracellular Ca2+ concentration can affect a number of physiological processes through the action of calmodulin (CaM), a ubiquitous Ca2+-binding protein.
Ca2+/calmodulin-regulated protein phosphorylation plays a pivotal role in amplifying and diversifying the action of Ca2+-mediated signals (Veluthambi and Poovaiah, Science 223:167-169, 1984; Schulman, Curr. Opin. in Cell. Biol. 5:247-253, 1993). Extracellular and intracellular signals regulate the activity of protein kinases, either directly or through second messengers.
These protein kinases in turn regulate the activity of their substrates by phosphorylation (Cohen, Trends . , AED:en 4630-45115.app 5/23/96 --2-Biochem. Sci. 17:408-413, 1992; Stone and Walker, Plant Physiol. 108:451-457, 1995).
In animals, Ca2+/calmodulin-dependent protein kinases play a pivotal role in cellular regulation (Colbran and Soderling, Current Topics in Cell. Reg.
31:181-221, 1990; Hanson and Schulman, Annu. Rev.
Biochem. 61:559-601, 1992; Mayford et al., Cell 81:891-904, 1995). Several types of CaM-dependent protein kinases (CaM kinases, phosphorylase kinase, and myosin light chain kinase) have been well characterized in m~mm~l ian systems (Fujisawa, BioEssays 12:27-29, 1990;
Colbran and Soderling, Current Topics in Cell. Reg.
31:181-221,1990; Klee, Neurochem. Res. 16:1059-1065, 1991; Mochizuki et al., J. Biol. Chem. 268:9143-9147, 1993).
Although little is known about Ca2+/calmodulin-dependent protein kinases in plants (Poovaiah et al., in Progress in Plant Growth Regulation, Karssen et al., eds., Dordrecht, The Netherlands: Kluwer Academic Publishers, 1992, pp. 691-702; Watillon et al., Plant Physiol. 101:1381-1384, 1993), Ca2+-dependent, calmodulin-independent protein kinases (CDPKs) have been identified (Harper et al., Science 252:951-954, 1991;
Roberts and Harmon, Annu. Rev. Plant Physiol. Plant Mol.
Biol. 43:375-414, 1992).
Male gametophyte formation in the anther is a complex developmental process involving many cellular events. During microsporogenesis, microsporocytes undergo meiosis to form tetrads of microspores that are surrounded by a callose wall composed of ~-1,3-glucan.
The callose wall is subsequently degraded by callase, which is secreted by cells of the tapetum (Steiglitz, Dev. Biol. 57:87-97, 1977), a specialized anther tissue that produces a number of proteins and other substrates that aid in pollen development or become a component of the pollen outer wall (Paciani et al., Plant Syst. Evol.
149:155-185, 1985; Bedinger, Plant Cell 4:879-887, 1992;
Mariani et al., Nature 347:737-741, 1990). The timing , , AED:en 4630-45115.app 5/23/96 --3-of callase secretion is critical for microspore development. Male sterility has been shown to result from premature or delayed appearance of callase (Worral et al., Plant Cell 4:759-771, 1992; Tsuchiya et al., Plant Cell Physiol. 36:487-494, 1995).
Induction of male sterility in plants can provide significant cost savings in hybrid plant production, enable production of hybrid plants where such production was previously difficult or impossible, and allow the production of plants with reduced pollen formation to reduced the tendency of such plants to elicit allergic reactions or to extend the life of flowers that senesce upon pollination (e.g., orchids).
Several strategies have been developed for the production of male-sterile plants (Goldberg et al., Plant Cell 5:1217-1229, 1993), including: selective destruction of the tapetum by fusing the ribonuclease gene to a tapetum-specific promoter, TA29 (Mariani et al., Nature 347:737-741, 1990); premature dissolution of the callose wall in pollen tetrads by fusing glucanase gene to tapetum-specific A9 or Osg6B promoters (Worrall et al., Plant Cell 4:759-771, 1992; Tsuchiya et al., Plant Cell Physiol. 36:487-494, 1995); antisense inhibition of flavonoid biosynthesis within tapetal cells (Van der Meer et al., Plant Cell 4:253-262, 1992);
tapetal-specific expression of the Agrobacterium rhizogenes rolB gene (Spena et al., Theor. Appl. Genet.
84:520-527, 1992); and overexpression of the mitochondrial gene atp9 (Hernould et al., Proc. Natl.
Acad. Sci. USA 90:2370-2374, 1993).

SUMMARY OF THE lNV~.. ~lON
Genes encoding plant calcium/calmodulin-dependent protein kinases (CCaMKs) have been cloned and sequenced.
Expression of CCaMK genes is highly organ- and developmental stage-specific. When CCaMK antisense constructs were expressed in plants, the plants were rendered male-sterile. The availability of CCaMK cDNA

. , en 4630-45115.app 5/23/96 --4--and genomic DNA sequences makes possible the production of a wide variety of male-sterile plants, including monocotyledonous, dicotyledonous, and other plant varieties. CCaMK promoters are also useful for targeted expression of heterologous genes, as is described in greater detail below.
Accordingly, the present invention provides isolated nucleic acids based on the cloned CCaMK
sequences. Nucleic acids that include at least 15 contiguous nucleotides of a native lily or tobacco CCaMK
gene and hybridize specifically to a CCaMK sequence under stringent conditions are useful, for example, as CCaMK-specific probes and primers. CCaMK promoter sequences are useful for the expression of heterologous genes in anthers of transgenic plants in a developmental stage-specific manner.
Isolated CCaMK nucleic acids can be expressed in host cells to produce recombinant CCaMK polypeptide or fragments thereof, which in turn can be used, for example, to raise CCaMK-specific antibodies that are useful for CCaMK immunoassays, for purification of CCaMK
polypeptides, and for screening expression libraries to obtain CCaMK homologs from other plant species. The native CCaMK sequence can be altered, e.g., by silent and conservative substitutions, to produce modified forms of CCaMK that preferably retain calcium/
calmodulin-dependent protein kinase activity.
Alternately, CCaMK polypeptides can be obtained from plant tissue by standard protein purification techniques, including the use of CCaMK-specific antibodies.
The foregoing and other objects and advantages of the invention will become more apparent from the following detailed description and accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 shows the nucleotide and deduced amino acid , .
AED: en 4630-45115 . app 5/23/96 --5--sequences of lily CCaMK. Diagnostic sequences (GKGGFS, DLKPEN, SIDYVSPE) for serine/threonine kinases are underlined. Sequences corresponding to two PCR primers (DLKPEN and FNARRKL) are indicated by arrows. The calmodulin-binding domain is double-underlined, Ca2+-binding EF-hand motifs are boxed, the putative autophosphorylation sites (RXXS/T) are indicated by asterisks, and the hatched region indicates the putative biotin-binding site (LKAMKMNSLI).
FIG. 2 shows a comparison of the deduced amino-acid sequence of the C-terminal region (amino acid residues 338-520) of lily CCaMK to neural visinin-like Ca2+-binding proteins. Conserved amino acids are boxed; Ca2+-binding domains (I-III) are indicated by solid lines;
putative autophosphorylation site is indicated by an asterisk; and the putative biotin-binding site (B) is indicated by a hatched box. Abbreviations: Rahcl, rat hippocalcin (Gen2:Ratp23K); Ravl3, rat neural visinin-like protein (Gen2:Ratnvp3); Bovll, bovine neurocalcin (Genl:Bovpcaln); Ravll, rat neural visinin-like protein (Gen2:Ratnvpl); Chvll, chicken visinin-like protein (Gen2:Ggvilip); Ravl2, rat neural visinin-like protein (Gen2:Ratnvp2); Drfrl, Drosophi1a frequenin (Gen2:Drofreq).
FIG. 3 is a schematic representation of structural features of the lily CCaMK polypeptide.
FIG. 4A shows SDS-polyacrylamide gel electrophoresis (SDS-PAGE) of lily CCaMK polypeptide expressed in E. coli after induction by IPTG. Lanes: 1, IPTG-induced cell extract; 2, uninduced cell extract.
Protein size is marked on the left.
FIG. 4B shows binding of lily CCaMK protein (250 ng) to 35S-labeled calmodulin (50 nM) in the presence of either 5 mM EGTA or 1 mM CaCl2. The histogram shows radioactivity (cpm) on the nitrocellulose filter.
Columns: 1, 5 mM EGTA; 2, 1 mM CaCl2; 3, 1 mM CaCl2 plus 2.5 ~M unlabeled calmodulin. An autoradiogram is shown on top of each corresponding column.

. , AED:en 4630-45115.app 5/23/96 - 6-FIG. 4C shows Ca2+-binding to lily CCaMK. 1, bovine serum albumin (2 ~g); 2, calmodulin (2 ~g); 3, CCaMK (2 ~g).
FIG. 5 shows results of an RNase protection assay using total RNA (20 ~g) from various parts of lily.
Lanes: 1, leaf; 2, stem; 3, anthers from phase II; 4, sepals and petals from phase III; 5 anthers from phase III; 6, sepals and petals from phase III; 7, yeast tRNA
control.
FIG. 6 shows a Southern blot of lily genomic DNA
digested with restriction enzymes and probed with a lily CCaMK cDNA probe. Lanes: 1, DraI; 2, EcoRI; 3, EcoRV;
4, HindIII, 5, PstI; 6, XbaI; 7, XhoI.
FIG. 7A shows Ca2+/calmodulin-dependent phosphorylation of histone IIAS by lily CCaMK in the presence of 0.5 mM CaCl2 and increasing amounts of calmodulin (~M). CCaMK activity is presented as nmol phosphate/min/mg CCaMK.
FIG. 7B shows the time course of phosphorylation of histone IIAS by lily CCaMK in the presence of 2.5 mM
EGTA (-) or 0.5 mM CaCl2 (~) or 0.5 mM CaCl2 and 1 ~M
calmodulin (O). CCaMK activity is represented as nmol of phosphate per mg CCaMK.
FIG. 8 shows a saturation curve of 35S-calmodulin binding to lily CCaMK. The amount of bound calmodulin at each point is represented as percent of maximal binding. InQet: Scatchard plot analysis (bound/free and bound calmodulin are expressed as B/F and B, respectively).
FIG. 9A shows the results of calmodulin binding assays using wild-type and truncated forms of lily CCaMK
in order to determine the calmodulin binding site.
Left: Schematic diagram of wild-type and truncation mutants of CCaMK used for 35S-calmodulin binding assays.
Right: Autoradiograms showing calmodulin binding corresponding to the wild-type and mutant CCaMKs.
FIG. 9B shows a comparison of amino acid sequences surrounding the putative calmodulin-binding sites of , ~ I
AED:en 4630-45115.app 5/23/96 --7--lily CCaMK and ~ subunit of mammalian calmodulin kinase II (CaMKII).
FIG. 10 shows calmodulin binding to synthetic peptides from the calmodulin-binding region (residues 5 311-340). A. Schematic diagram of wild-type and truncation mutants of CCaMK used for 35S-calmodulin binding assays are shown on the left. The mutants 1-356 and 1-322 represent CCaMK lacking the visinin-like domain and both visinin-like and calmodulin-binding domains. The autoradiogram is shown on the right of each diagram (boxed area). The radioactivity (cpm) of bound 35S-calmodulin was 11,600 for wild-type, 12,500 for the mutant 1-356 and 99 for the mutant 1-322, - respectively. B. Comparison of amino acid sequences surrounding the putative calmodulin-binding sites of CCaMK and the ~ subunit of CaMK II. C. Calmodulin binding to the synthetic peptides in a gel mobility-shift assay. Non-denaturing gel electrophoresis performed in the presence of 0.5 mM CaCl2 using calmodulin alone (lane 1) or a mixture of calmodulin and each of the following peptides: CCaMK 328-340 (lane 2);
CCaMK 322-340 (lane 3); 317-340 (lane 4); and 311-340 (lane 5). The bands of calmodulin (free CaM) or CaM-peptide complex were visualized by Coomassie Brilliant Blue.
FIG. 11 shows a helical wheel projection of calmodulin-binding sequences in lily CCaMK (left) and animal CaMKII~ (right). Hydrophobic amino acid residues are boxed. Basic amino acid residues are marked with (+).
FIG. 12 shows a time course of autophosphorylation of lily CCaMK in the presence of 2.5 mM EGTA (-) or 0.5 mM CaCl2 (~) or 0.5 mM CaCl2 and 1 ~M calmodulin (O).
The autophosphorylation is presented as pmol 32p incorporated per 21.4 pmol of CCaMK.
FIG 13A shows the effect of calmodulin on Ca2+-dependent autophosphorylation of lily CCaMK in the presence of CaCl2 (0.5 mM) and increasing concentrations ,~ , AED:en 4630-45115.app 5/23/96 - 8--of calmodulin. Lane 1, +CaCl2, (0.5 mM); lanes 2-6, +CaCl2 (0.5 mM) and 60, 120, 240, 360, and 480 nM of calmodulin respectively.
FIG. 13B shows phosphoamino acid analysis of autophosphorylated lily CCaMK (200 ng) either in the presence of 2.5 mM EGTA (-Ca), 0.5 mM CaCl2 (+Ca), or 0.5 mM CaCl2 plus 1 ~M calmodulin (+Ca/CaM). The positions of phosphoserine (S) and phosphothreonine (T) are marked.
FIG. 14A shows amino acid sequences of the three EF-hand motifs in the visinin-like domain of lily CCaMK.
Six Ca2+-ligating residues denoted as x, y, z, -y, -x, -z, respectively, are marked. Site-directed mutants were prepared by substituting the amino acid residues at the -x position with alanine (A).
FIG. 14B shows the Ca2+-dependent mobility shift of wild-type CCaMK and CCamKs mutated in the visinin-like domain in the presence of 2.5 mM EGTA (lane 1) or 0.5 mM
CaCl2 (lanes 2-6). Wild-type protein (lanes 1 and 2), proteins mutated in the EF hand I (lane 3), EF-hand II
(lane 4), EF-hand III (lane 5), and all three EF hands (lane 6) are shown.
FIG. 15 shows a comparison of enzyme activity of wild-type (A and B) and a truncated lily CCaMK mutant (1-356) (C and D) with respect to Ca2+-dependent autophosphorylation (A and C) and Ca2+/calmodulin-dependent histone IIAS phosphorylation (B and D). The assays were carried out in the presence of 2.5 mM EGTA
(-Ca), 0.5 mM CaCl2 (+Ca) or 0.5 mM CaCl2 and 1 ~M
calmodulin (+Ca/CaM).
FIG. 16A shows the effect of increasing concentrations of calmodulin on the GS peptide phosphorylation by autophosphorylated lily CCaMK.
FIG. 16B shows the effect of CCaMK
autophosphorylation on Ca2+/calmodulin-dependent and calmodulin-independent activity. Column 1, CCaMK
autophosphorylated in the presence of 0.5 mM CaCl2 and used for Ca2+/calmodulin-dependent GS peptide , AED:en 4630-45115.app 5/23/96 - 9 -phosphorylation (hatched bar). Column 2, unphosphorylated enzyme used for Ca2+/calmodulin-dependent GS peptide phosphorylation (hatched bar).
Solid bars represent the activity of autophosphorylated CCaMK (column 1) and unphosphorylated CCaMK (column 2) in the presence of 2.5 mM EGTA.
FIG. 17 shows the effects of increasing amounts of synthetic peptides derived from the CCaMK autoinhibitory domain (amino acid residues 311-340) on the activity of the constitutive mutant 1-322.
FIG. 18 shows models describing the regulation of CCaMK by Ca2+ and Ca2+/calmodulin (A) and the autoinhibitory domain (B).
FIGS. l9A-19C show cross-sections of lily anthers demonstrating progressive development of tapetal cells and microspores. A. Pollen mother cell stage (2.0 cm bud), 97x. B. Meiosis stage I (2.5 cm bud), 97x. C.
Uninucleate microspore stage (3.5 cm bud), 97x.
Abbreviations: E, epidermis; En, endothecium; M, middle layers; PMC, pollen mother cells; DMS, dividing microspores; UMS, uninucleate microspores; T, tapetal cells.
FIGS. l9D-19E show the results of in situ hybridization showing localization of CCaMK in tapetal cells. Abbreviations: Exine (Ex), microspore (MS) (3.5 cm bud). Magnification: l9D, 865x; l9E, 1600x. Arrows indicate hybridization signals.
FIG. 20 shows Ca2+/calmodulin-dependent phosphorylation of heat-inactivated lily anther proteins in the presence (+) or absence of CCaMK (-) at different stages of development. Numbers on top indicate the sizes of flower buds in cm from which the anthers were used for protein extraction. Molecular weight markers (kDa) are indicated on the left. Arrow indicates 24 kDa protein showing high levels of phosphorylation when buds are 1.0-3.0 cm.
FIG. 21 shows the expression pattern of CCaMK-binding proteins at different stages of lily anther , AED:en 4630-45115.app 5/23/96 --10--development. The numbers indicate the sizes of flower buds in cm from which the anthers were used for protein extraction. Molecular weight markers (kDa) are indicated on the left.
FIG. 22 shows the nucleotide sequence and deduced amino-acid sequence of the tobacco CCaMK cDNA.
FIG. 23A shows a comparison of deduced amino acid sequences of tobacco and lily CCaMKs. Eleven major conserved subdomains of serine/threonine protein kinases are marked. Hatched region indicates calmodulin-binding domain, the three Ca2~-binding EF-hands are boxed.
FIG. 23B is a diagram showing the kinase domain, calmodulin-binding domain, and visinin-like Ca2+-binding domain of the lily and tobacco CCaMK polypeptides.
Three Ca2'-binding sites within the visinin-like binding domain are indicated by Roman numerals I, II, and III.
FIG. 24 shows the results of RT-PCR showing the expression pattern of tobacco CCaMK at different stages of anther development. Numbers on top of each lane indicate bud size in cm, "M" indicates mature anther.
Lower band in each lane shows calmodulin (CaM) control.
FIG. 25 shows a Southern blot of tobacco genomic DNA digested with various restriction enzymes and probed with CCaMK. Lanes: 1, EcoRI; 2, EcoRV; 3, HindIII.
Sizes in kb are shown on the right.
FIG. 26 shows the nucleotide sequence of the promoter region of the tobacco CCaMK genomic clone. The putative TATA box is underlined and the start codon is boxed.
FIG. 27 shows tobacco flowers from wild-type and male sterile antisense plants. A. Wild-type flower before dehiscence. B. Antisense flower before dehiscence. (Note: there are no obvious differences between wild-type and antisense flower prior to dehiscence.) C. Wild-type flower after dehiscence. D.
Antisense flower after dehiscence. Note: the wild-type anthers are white and fluffy, while the antisense anthers are bare. E. Enlarged view of the wild-type AED:en 4630-45115.app 5/23/96 ~

anther, 5x. F. Enlarged view of antisense anther, 5x.
G. Enlarged view of wild-type anther, 20x. H. Enlarged view of antisense anther, 20x.
FIG. 28 shows scanning electron micrographs showing wild-type and antisense pollen grains. A. Wild-type 300x. B. Antisense, 300x. C. Wild-type, 100,000x. D.
Antisense, 100,000x. Note: antisense pollen grains are smaller and shriveled in contrast to wild-type. The whitish granular structures on the right of FIGS. 24B
and 24D are remnants of pollen grains that have failed to develop.
FIG. 29 shows histochemical localization of callose in the outer walls of pollen grains from wild-type and antisense plants. On the left are the bright field images. On the right are the same respective views under blue excitation to highlight callose. A, B. Wild-type pollen grains, 300x. C, D. Antisense pollen grains, 300x. E, F. Wild-type pollen grains, 480x. G, H. Antisense pollen grains, 480x. Note: the pollen wall is not smooth and callose granules are unevenly distributed in antisense plants.
FIG. 30 shows pollen germination. A. Wild-type pollen, 139x. B. Antisense pollen, 139x.
FIG. 31A shows the results of slot-blot analysis demonstrating suppression of CCaMK mRNA in antisense plants (A3, A4, A14, and A17), wild-type plants (WT), and transgenic plants carrying vector alone (C). FIG.
31B shows calmodulin control (the same filter re-hybridized with calmodulin).
FIG. 32 shows GUS and antisense CCaMK constructs for transformation of plants. I. Transcriptional fusion of the tobacco CCaMK promoter to the ~-glucuronidase (GUS) reporter gene. II. Transcriptional fusion of a truncated version of the tobacco CCaMK promoter to GUS.
III. Translational fusion of the tobacco CCaMK promoter to the tobacco CCaMK coding region and GUS. IV.
Transcriptional fusion of the CCaMK promoter to the tobacco CCaMK in an antisense orientation. V.

AED: en 4630 -45115 . app 5/23/96 - 12--Transcriptional fusion of the TA29 promoter to antisense tobacco CCaMK. VI. Transcriptional fusion of the cauliflower mosaic virus (CaMV) 35S promoter to antisense tobacco CCaMK. All constructs include the Agrobacterium tumefaciens nopaline synthase terminator sequence (Nos-ter).

DET~TT-~n DESCRIPTION OF THE PREFERRED EMBODIMENTS
Definitions and Methods The following definitions and methods are provided to better define the present invention and to guide those of ordinary skill in the art in the practice of the present invention. Definitions of common terms in molecular biology may also be found in Rieger et al., Glossar~ of Genetics: Classical and Molecular, 5th edition, Springer-Verlag: New York, 1991; and Lewin, Genes V, Oxford University Press: New York, 1994. The nomenclature for DNA bases as set forth at 37 CFR
1.822 is used. The standard one- and three letter nomenclature for amino acid residues is used.

Nucleic Acidn "CCaMK Gene". The term "CCaMK gene" refers to a native CCaMK nucleic acid sequence or a fragment thereof, e.g., the native lily or tobacco CCaMK cDNA or genomic sequences and alleles and homologs thereof. The term also encompasses variant forms of a native CCaMK
nucleic acid sequence or fragment thereof as discussed below, preferably a nucleic acid that encodes a polypeptide having CCaMK biological activity. Native CCaMK sequences include cDNA sequences and the corresponding genomic sequences (including flanking or internal sequences operably linked thereto, including regulatory elements and/or intron sequences).
"Native". The term "native" refers to a naturally-occurring ("wild-type") nucleic acid or polypeptide.
"Homolog". A "homolog" of a lily or tobacco CCaMK
gene is a gene sequence encoding a CCaMK polypeptide 2~ 7~2~ 7 ABD:en 4630-45115.app 5/23/96 --13--isolated from an organism other than lily or tobacco.
"Isolated". An "isolated" nucleic acid is one that has been substantially separated or purified away from other nucleic acid sequences in the cell of the organism in which the nucleic acid naturally occurs, i.e., other chromosomal and extrachromosomal DNA and RNA, by conventional nucleic acid-purification methods. The term also embraces recombinant nucleic acids and chemically synthesized nucleic acids.
Fraqments, Probes, and Primers. A fragment of a CCaMK nucleic acid is a portion of a CCaMK nucleic acid that is less than full-length and comprises at least a minimum length capable of hybridizing specifically with a native CCaMK nucleic acid under stringent hybridization conditions. The length of such a fragment is preferably at least 15 nucleotides, more preferably at least 20 nucleotides, and most preferably at least 30 nucleotides of a native CCaMK nucleic acid sequence.
Nucleic acid probes and primers can be prepared based on a native CCaMK gene sequence. A "probe" is an isolated DNA or RNA attached to a detectable label or reporter molecule, e.g., a radioactive isotope, ligand, chemiluminescent agent, or enzyme. "Primers" are isolated nucleic acids, generally DNA oligonucleotides 15 nucleotides or more in length, preferably 20 nucleotides or more, and more preferably 30 nucleotides or more, that are annealed to a complementary target DNA
strand by nucleic acid hybridization to form a hybrid between the primer and the target DNA strand, then extended along the target DNA strand by a polymerase, e.g., a DNA polymerase. Primer pairs can be used for amplification of a nucleic acid sequence, e.g., by the polymerase chain reaction (PCR) or other conventional nucleic-acid amplification methods.
Methods for preparing and using probes and primers are described, for example, in Molecular Cloning: A
Laboratory Manual, 2nd ed., vol. 1-3, ed. Sambrook et al., Cold Spring Harbor Laboratory Press, Cold Spring AED:en 4630-45115.app 5/23/96 --14--Harbor, NY, 1989 (hereinafter, "Sambrook et al., 1989");
Current Protocols in Molecular Biology, ed. Ausubel et al., Greene Publishing and Wiley-Interscience, New York, 1987 (with periodic updates) (hereinafter, "Ausubel et al., 1987); and Innis et al., PCR Protocols: A Guide to Methods and Applications, Academic Press: San Diego, 1990. PCR-primer pairs can be derived from a known sequence, for example, by using computer programs intended for that purpose such as Primer (Version 0.5, 1991, Whitehead Institute for Biomedical Research, Cambridge, MA).
Substantial Similarity. A first nucleic acid is "substantially similar" to a second nucleic acid if, when optimally aligned (with appropriate nucleotide insertions or deletions) with the other nucleic acid (or its complementary strand), there is at least about 75%
nucleotide sequence identity, preferably at least about 85% identity, and more preferably at least about 90%
identity. Sequence similarity can be determined by comparing the nucleotide sequences of two nucleic acids using sequence analysis software such as the Sequence Analysis Software Package of the Genetics Computer Group, University of Wisconsin Biotechnology Center, Madison, WI.
Alternatively, two nucleic acids are substantially similar if they hybridize under stringent conditions, as defined below.
Operably Linked. A first nucleic-acid sequence is "operably" linked with a second nucleic-acid sequence when the first nucleic-acid sequence is placed in a functional relationship with the second nucleic-acid sequence. For instance, a promoter is operably linked to a coding sequence if the promoter affects the transcription or expression of the coding sequence.
Generally, operably linked DNA sequences are contiguous and, where necessary to join two protein coding regions, in reading frame.
"Recombinant". A "recombinant" nucleic acid is .~, A13D: en 4630 -45115 . app 5/23/96 --15--made by an artificial combination of two otherwise separated segments of sequence, e.g., by chemical synthesis or by the manipulation of isolated segments of nucleic acids by genetic engineering techniques.
Techniques for nucleic-acid manipulation are well-known (see, e.g., Sambrook et al., 1989, and Ausubel et al., 1987). Methods for chemical synthesis of nucleic acids are discussed, for example, in Beaucage and Carruthers, Tetra. Letts. 22:1859-1862, 1981, and Matteucci et al., ~. Am. Chem. Soc. 103: 3185, 1981.
Chemical synthesis of nucleic acids can be performed, for example, on commercial automated oligonucleotide synthesizers.
Preparation of Recombinant or Chemically 15 Synthesized Nucleic acids; Vectors, Transformation, Host cells. Natural or synthetic nucleic acids according to the present invention can be incorporated into recombinant nucleic-acid constructs, typically DNA
constructs, capable of introduction into and replication 20 in a host cell. Such a construct preferably is a vector that includes a replication system and sequences that are capable of transcription and translation of a polypeptide-encoding sequence in a given host cell. For the practice of the present invention, conventional 25 compositions and methods for preparing and using vectors and host cells are employed, as discussed, inter alia, in Sambrook et al., 1989, or Ausubel et al., 1987. A
variety of well-known promoters or other sequences useful in constructing expression vectors are available 30 for use in bacterial, yeast, m~mm~l ian, insect, amphibian, avian, or other host cells.
A cell, tissue, organ, or organism into which has been introduced a foreign nucleic acid, such as a recombinant vector, is considered "transformed", 35 "transfected", or "transgenic." A "transgenic" or "transformed" cell or organism also includes (1) progeny of the cell or organism and ~2) progeny produced from a breeding program employing such a "transgenic" plant as , ABD:en 4630-45115.app 5/23/96 -16--a parent in a cross and exhibiting an altered phenotype resulting from the presence of the recombinant CCaMK DNA
construct.
Nucleic-Acid Hybridization: "Stringent Conditions";
"Specific". The nucleic-acid probes and primers of the present invention hybridize under stringent conditions to a target DNA sequence, e.g., to a CCaMK gene.
The term "stringent conditions" is functionally defined with regard to the hybridization of a nucleic-acid probe to a target nucleic acid (i.e., to aparticular nucleic-acid sequence of interest) by the hybridization procedure discussed in Sambrook et al., 1989, at 9.52-9.55. See also, Sambrook et al., 1989 at 9.47-9.52, 9.56-9.58; Kanehisa, Nucl. Acids Res. 12:203-213, 1984; and Wetmur and Davidson, ~. Mol. Biol.
31:349-370, 1968.
Regarding the amplification of a target nucleic-acid sequence (e.g., by PCR) using a particular amplification primer pair, "stringent conditions" are conditions that permit the primer pair to hybridize only to the target nucleic-acid sequence to which a primer having the corresponding wild-type sequence (or its complement) would bind and preferably to produce a unique amplification product.
The term "specific for (a target sequence)"
indicates that a probe or primer hybridizes under stringent conditions only to the target sequence in a sample comprising the target sequence.
Nucleic-Acid Amplification. As used herein, "amplified DNA" refers to the product of nucleic-acid amplification of a target nucleic-acid sequence.
Nucleic-acid amplification can be accomplished by any of the various nucleic-acid amplification methods known in the art, including the polymerase chain reaction (PCR).
A variety of amplification methods are known in the art and are described, inter alia, in U.S. Patent Nos.
4,683,195 and 4,683,202 and in PCR Protocols: A Guide to Methods and Applications, ed. Innis et al., Academic AED:en 4630-45115.app 5/23/96 --17--Press, San Diego, 1990.
Methods of Obtaining Alleles and Homologs of Lily and Tobacco CCaMK. Based upon the availability of the lily CCaMK cDNA and tobacco CCaMK cDNA and genomic sequences disclosed herein, alleles and homologs can be readily obtained from a wide variety of plants by cloning methods known in the art, e.g., by screening a cDNA or genomic library with a probe that specifically hybridizes to a native CCaMK sequence under stringent conditions or by PCR or another amplification method using a primer or primers that specifically hybridize to a native CCaMK sequence under stringent conditions.
Cloning of a CCaMK Genomic Sequence. The availability of a CCaMK cDNA sequence enables the skilled artisan to obtain a genomic clone corresponding to the cDNA (including the promoter and other regulatory regions and intron sequences) and the determination of its nucleotide sequence by conventional methods. Both monocots and dicots possess CCaMK genes.
Primers and probes based on the native lily and tobacco CCaMK sequences disclosed herein can be used to confirm (and, if necessary, to correct) the CCaMK
sequences by conventional methods.
Nucleotide-Sequence Variants of Native CCaMK
Nucleic Acids and Amino Acid Sequence Variants of Native CCaMK Proteins. Using the nucleotide and the amino-acid sequence of the CCaMK polypeptides disclosed herein, those skilled in the art can create DNA molecules and polypeptides that have minor variations in their nucleotide or amino acid sequence.
~ Variant" DNA molecules are DNA molecules containing minor changes in a native CCaMK sequence, i.e., changes in which one or more nucleotides of a native CCaMK sequence is deleted, added, and/or substituted, preferably while substantially maintaining a CCaMK biological activity. Variant DNA molecules can be produced, for example, by standard DNA mutagenesis techniques or by chemically synthesizing the variant DNA

2~77217 , .
AED:en 4630-45115.app 5/23/96 --18--molecule or a portion thereof. Such variants preferably do not change the reading frame of the protein-coding region of the nucleic acid and preferably encode a protein having no change, only a minor reduction, or an increase in CCaMK biological function.
Amino-acid substitutions are preferably substitutions of single amino-acid residues. DNA
insertions are preferably of about 1 to 10 contiguous nucleotides and deletions are preferably of about 1 to 30 contiguous nucleotides. Insertions and deletions are preferably insertions or deletions from an end of the protein-coding or non-coding sequence and are preferably made in adjacent base pairs. Substitutions, deletions, insertions or any combination thereof can be combined to arrive at a final construct.
Preferably, variant nucleic acids according to the present invention are "silent" or "conservative"
variants. "Silent" variants are variants of a native CCaMK sequence or a homolog thereof in which there has been a substitution of one or more base pairs but no change in the amino-acid sequence of the polypeptide encoded by the sequence. "Conservative" variants are variants of the native CCaMK sequence or a homolog thereof in which at least one codon in the protein-coding region of the gene has been changed, resulting ina conservative change in one or more amino acid residues of the polypeptide encoded by the nucleic-acid sequence, i.e., an amino acid substitution. A number of conservative amino acid substitutions are listed below.
In addition, one or more codons encoding cysteine residues can be substituted for, resulting in a loss of a cysteine residue and affecting disulfide linkages in the CCaMK polypeptide.

, , AED:en 4630-45115.app 5/23/96 - 19 ~

Original Residue Conservative Substitutions Ala ser Arg lys Asn gln, his Asp glu Cys ser Gln asn Glu asp Gly pro His asn; gln Ile leu, val Leu ile; val Lys arg; gln; glu Met leu; ile Phe met; leu; tyr Ser thr Thr ser Trp tyr Tyr trp; phe Val ile; leu Substantial changes in function are made by selecting substitutions that are less conservative than those listed above, e.g., causing changes in: (a) the structure of the polypeptide backbone in the area of the substitution; (b) the charge or hydrophobicity of the polypeptide at the target site; or (c) the bulk of an amino acid side chain. Substitutions generally expected to produce the greatest changes in protein properties are those in which: (a) a hydrophilic residue, e.g., seryl or threonyl, is substituted for (or by) a hydrophobic residue, e.g., leucyl, isoleucyl, phenylalanyl, valyl or alanyl; (b) a cysteine or proline is substituted for (or by) any other residue; (c) a residue having an electropositive side chain, e.g., lysyl, arginyl, or histadyl, is substituted for (or by) an electronegative residue, e.g., glutamyl or aspartyl;
or (d) a residue having a bulky side chain, e.g., phenylalanine, is substituted for (or by) one not having a side chain, e.g., glycine.

, .
AED:en 4630-45115.app 5/23/96 --20--A CCaMK gene sequence can be modified as follows:
(1) To improve expression efficiency and redirect the targeting of the expressed polypeptide:
For expression in non-plant hosts (or to direct the expressed polypeptide to a different intracellular compartment in a plant host), an appropriate transit or secretion peptide sequence can be added to the protein-coding region of the native gene sequence. In addition, one or more codons can be changed, for example, to conform the gene to the codon usage bias of the host cell or organism for improved expression. Enzymatic stability can be altered by removing or adding one or more cysteine residues, thus removing or adding one or more disulfide bonds.
(2) To alter substrate and liqand bindinq and CCaMK enzymatic activity: One or more amino acid residues in a substrate or ligand binding domain (e.g., the calmodulin-binding domain) can be mutagenized to affect the strength or specificity of the interaction between CCaMK and its polypeptide substrates or to affect control of CCaMK activity by Ca2+ and/or calmodulin. The autoinhibitory domain can also be mutagenized or an autophosphorylation site removed or added to affect CCaMK activity. Further targets, including the visinin-like Ca2+-binding domain or one or more EF-hand motifs thereof, can also be mutagenized.
Nucleic Acids Attached to a Solid Support. The nucleic acids of the present invention can be free in solution or covalently or noncovalently attached by conventional means to a solid support, such as a hybridization membrane (e.g., nitrocellulose or nylon), a bead, etc.

PolYPePtides "CCaMK Protein". The term "CCaMK protein" (or polypeptide) refers to a protein encoded by a CCaMK
nucleic acid, including alleles, homologs, and variants AED:en 4630-45115 .app 5/23/96 --21--of a native CCaMK nucleic acid, for example. A CCaMK
polypeptide can be produced by the expression of a recombinant CCaMK nucleic acid or be chemically synthesized. Techniques for chemical synthesis of polypeptides are described, for example, in Merrifield, J. Amer. Chem. Soc. 85:2149-2156, 1963.
Polypeptide Sequence Homology. Ordinarily, CCaMK polypeptides encompassed by the present invention are at least about 70% homologous to a native CCaMK
polypeptide, preferably at least about 80~ homologous, and more preferably at least about 95~ homologous. Such homology is considered to be "substantial homology,"
although more important than shared amino-acid sequence homology can be the common possession of characteristic structural features and the retention of biological activity that is characteristic of CCaMK, preferably CCaMK catalytic activity.
Polypeptide homology is typically analyzed using sequence analysis software such as the Sequence Analysis Software Package of the Genetics Computer Group, University of Wisconsin Biotechnology Center, Madison, WI). Polypeptide sequence analysis software matches homologous sequences using measures of homology assigned to various substitutions, deletions, substitutions, and other modifications.
"Isolated," "Purified," "Homogeneous"
Polypeptides. A polypeptide is "isolated" if it has been separated from the cellular components (nucleic acids, lipids, carbohydrates, and other polypeptides) that naturally accompany it. Such a polypeptide can also be referred to as "pure" or "homogeneous" or "substantially" pure or homogeneous. Thus, a polypeptide which is chemically synthesized or recombinant (i.e., the product of the expression of a recombinant nucleic acid, even if expressed in a homologous cell type) is considered to be isolated. A
monomeric polypeptide is isolated when at least 60-90 by weight of a sample is composed of the polypeptide, AED:en 4630-45115.app 5/23/96 -22-preferably 95~ or more, and more preferably more than 99~. Protein purity or homogeneity is indicated, for example, by polyacrylamide gel electrophoresis of a protein sample, followed by visualization of a single polypeptide band upon staining the polyacrylamide gel;
high pressure liquid chromatography; or other conventional methods.
Protein Purification. The polypeptides of the present invention can be purified by any of the means known in the art. Various methods of protein purification are described, e.g., in Guide to Protein Purification, ed. Deutscher, Meth. Enzymol. 185, Academic Press, San Diego, 1990; and Scopes, Protein Purification: Principles and Practice, Springer Verlag, New York, 1982.
Variant and Modified Forms of CCaMK
Polypeptides. Encompassed by the CCaMK polypeptides of the present invention are variant polypeptides in which there have been substitutions, deletions, insertions or other modifications of a native CCaMK polypeptide. The variants substantially retain structural characteristics and biological activities of a corresponding native CCaMK polypeptide and are preferably silent or conservative substitutions of one or a small number of contiguous amino acid residues.
A native CCaMK polypeptide sequence can be modified by conventional methods, e.g., by acetylation, carboxylation, phosphorylation, glycosylation, ubiquitination, and labeling, whether accomplished by in vivo or in vitro enzymatic treatment of a CCaMK
polypeptide or by the synthesis of a CCaMK polypeptide using modified amino acids.
Labeling. There are a variety of conventional methods and reagents for labeling polypeptides and fragments thereof. Typical labels include radioactive isotopes, ligands or ligand receptors, fluorophores, chemiluminescent agents, and enzymes. Methods for labeling and guidance in the choice of labels , AED:en 4630-45115.app 5/23/96 --23--appropriate for various purposes are discussed, e.g., in Sambrook et al., 1989 and Ausubel et al., 1987.
Polypeptide Fragments. The present invention also encompasses fragments of a CCaMK polypeptide that lacks at least one residue of a native full-length CCaMK
polypeptide. Preferably, such a fragment retains Ca2+/calmodulin-dependent kinase activity or possession of a characteristic functional domain (e.g., a calmodulin-binding domain, Ca2+-binding EF-hand motif(s), autophosphorylation site(s), etc.), or an immunological determinant characteristic of a native CCaMK polypeptide (and thus able to elicit production of a CCaMK-specific antibody in a mouse or rabbit, for example or to compete with CCaMK for binding to CCaMK-specific antibodies) that is therefore useful in immunoassays for the presence of a CCaMK polypeptide in a biological sample. Such immunologically active fragments typically have a minimum size of 7 to 17 or more amino acids.
The terms "biological activity", "biologically active", "activity" and "active" refer primarily to the characteristic enzymatic activity or activities of a native CCaMK polypeptide, including, but not limited to, Ca2+/calmodulin-dependent kinase activity.
Fusion Polypeptides. The present invention also provides fusion polypeptides including, for example, heterologous fusion polypeptides, i.e., a CCaMK
polypeptide sequence or fragment thereof and a heterologous polypeptide sequence, e.g., a sequence from a different polypeptide. Such heterologous fusion polypeptides thus exhibit biological properties (such as substrate or ligand binding, enzymatic activity, antigenic determinants, etc.) derived from each of the fused sequences. Fusion partners include, for example, ~-glucuronidase, immunoglobulins, beta galactosidase, trpE, protein A, beta lactamase, alpha amylase, alcohol dehydrogenase, yeast alpha mating factor, and various signal and leader sequences which, e.g., can direct the AED:en 4630-45115.app 5/23/96 --24-secretion of the polypeptide. Fusion polypeptides are - preferably made by the expression of recombinant nucleic acids produced by standard techniques.
Polypeptide Sequence Determination. The sequence of a polypeptide of the present invention can be determined by various methods known in the art. In order to determine the sequence of a polypeptide, the polypeptide is typically fragmented, the fragments separated, and the sequence of each fragment determined.
To obtain fragments of a CCaMK polypeptide, the polypeptide can be digested with an enzyme such as trypsin, clostripain, or Staphylococcus protease, or with chemical agents such as cyanogen bromide, o-iodosobenzoate, hydroxylamine or 2-nitro-5-thiocyanobenzoate. Peptide fragments can be separated,e.g., by reversed-phase high-performance liquid chromatography (HPLC) and analyzed by gas-phase sequencing.
Polypeptide Coupling to a Solid Phase Support.
The polypeptides of the present invention can be free in solution or coupled to a solid-phase support, e.g., nitrocellulose, nylon, column packing materials (e.g., Sepharose beads), magnetic beads, or glass wool, by conventionaI methods.
Antibodies The present invention also encompasses polyclonal and/or monoclonal antibodies capable of specifically binding to a CCaMK polypeptide and/or fragments thereof. Such antibodies are raised against a CCaMK polypeptide or fragment thereof and are capable of distinguishing a CCaMK polypeptide from other polypeptides.
An immunological response is usually assayed with an immunoassay. Normally such immunoassays involve some purification of a source of antigen, for example, produced by the same cells and in the same fashion as the antigen was produced.

. . .
AED:en 4630-45115.app 5/23/96 --25-For the preparation and use of antibodies according to the present invention, including various immunoassay techniques and applications, see, e.g., Goding, Monoclonal Antibodies: Principles and Practice, '2d ed, Academic Press, New York, 1986; and Harlow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1988. CCaMK-specific antibodies are useful, for example in:
purifying a CCaMK polypeptide from a biological sample, such as a host cell expressing recombinant a CCaMK
polypeptide; in cloning a CCaMK allele or homolog from an expression library; as antibody probes for protein blots and immunoassays; etc.
CCaMK polypeptides and antibodies can be labeled by joining, either covalently or noncovalently, to a substance which provides for a detectable signal by conventional methods. A wide variety of labels and conjugation techniques are known. Suitable labels include radionuclides, enzymes, substrates, cofactors, inhibitors, fluorescent agents, chemiluminescent agents, magnetic particles, etc.

Plant Transformation and Re~eneration Various nucleic acid constructs that include a CCaMK nucleic acid are useful for producing male-sterile plants. As detailed in the Examples below, transgenic plants containing as a transgene a nucleic acid construct in which a CCaMK nucleic acid is expressed in an antisense orientation are male sterile.
CCaMK nucleic acids can be expressed in plants or plant cells in a sense or antisense orientation under the control of an operably linked promoter that is capable of expression in a cell of a particular plant.
Various promoters suitable for expression of heterologous genes in plant cells are well known, including constitutive promoters (e.g. the CaMV 35S
promoter), organ- or tissue-specific promoters (e.g., the tapetum-specific TA29, A9 or Osg6B promoters), and AED:en 4630-45115.app 5/23/96 --26--promoters that are inducible by chemicals such as methyl jasminate, salicylic acid, or Safener, for example.
In addition to antisense expression of CCaMK in transgenic plants, as discussed below (see also, U.S.
5 Patent No. 5,283,184), the availability of CCaMK genes permits the use of other conventional methods for interfering with CCaMK gene expression, including triplex formation, production of an untranslatable plus-sense CCaMK RNA, etc.
0 A CCaMK promoter can be used to drive the expression of a CCaMK antisense transgene and also to express other nucleic acids in transgenic plants in an organ- and developmental stage-specific manner. For example, a CCaMK promoter can be used to drive the expression in transcriptional or translational fusions of antisense versions of nucleic acids encoding polypeptides necessary for male fertility, e.g., antisense inhibition of flavonoid biosynthesis (Van der Meer et al., Plant Cell 4:253-262, 1992), or to express, in a sense orientation, genes that interfere with male fertility, e.g., ribonuclease (Mariani et al., Nature 347:737-741, 1990); glucanase (Worrall et al., Plant Cell 4:759-771, 1992; Tsuchiya et al., Plant Cell Physiol. 36:487-494, 1995); Agrobacterium rhizogenes rolB (Spena et al., Theor. Appl. Genet. 84:520-527, 1992); and mitochondrial gene atp9 (Hernould et al., Proc. Natl. Acad. Sci. USA 90:2370-2374, 1993).
Any well-known method can be employed for plant cell transformation, culture, and regeneration in the practice of the present invention with regard to a particular plant species. Methods for introduction of foreign DNA into plant cells include, but are not limited to: transfer involving the use of Agrobacterium tumefaciens and appropriate Ti vectors, including binary vectors; chemically induced transfer (e.g., with polyethylene glycol); biolistics; and microinjection.
See, e.g., An et al., Plant Molecular Biology Manual A3:
1-19 ~ 1988.

~ 21 7721 7 AED:en 4630-45115.app 5/23/96 --27--The term "plant" encompasses any higher plant and progeny thereof, including monocots (e.g., lily, corn, rice, wheat, etc.), dicots (e.g., tobacco, potato, apple, tomato, etc.), gymnosperms, etc., and includes parts of plants, including reproductive units of a plant, fruit, flowers, wood, etc.
A "reproductive unit" of a plant is any totipotent part or tissue of the plant from which one can obtain a progeny of the plant, including, for example, seeds, cuttings, tubers, buds, bulbs, somatic embryos, cultured cell (e.g., callus or suspension cultures), etc.
The invention will be better understood by reference to the following Examples, which are intended to merely illustrate the best mode now known for practicing the invention. The scope of the invention is not to be considered limited thereto, however.

EXAMPLES
EXAMPLE 1: Identification of Plant Calcium/Calmodulin-Dependent Protein Kinase Gene Materials and Methods Plant Material. Lily (Lilium longiflorum Thunb cv. Nellie White) plants were grown under greenhouse conditions and various parts were excised and frozen in liquid nitrogen.
PCR and cDNA Library Screening. Three different lily cDNA libraries made from developing anthers, mature and germinating pollen were used for PCR. Degenerate oligonucleotides corresponding to two highly conserved regions of m~mm~l ian Ca2+/calmodulin-dependent protein kinases, DLKPEN and FNARRKL, were used as primers for PCR (Hunter, Cell 50:823-829, 1987). The amplification reaction contained lx PCR buffer (Cetus Corp.), 200 ~M
dNTPs, 50 pmoles of each primer, 1.5 mM MgCl2, 2 ~l cDNA
library (109 pfu/ml), and 2.5 units of Taq DNA polymerase in a 100 ~l total reaction volume. The cycling profile was 30 cycles, each cycle including 94~C for 1 min, 48~C

AED:en 4630-45115.app 5/23/96 - 28--for 1 min, and 72~C for 1 min. The specific PCR product of the expected size (471 bp) was subcloned into pBluescriptII KS+ (Stratagene) and sequenced. This fragment was used to screen the cDNA library (Sambrook et al., 1989) from developing anthers (Kim et al., Plant Mol. Biol. 21:39-45, 1993) to obtain the lily CCaMK cDNA
clone.
Sequence Analysis. The sequencing of the lily CCaMK cDNA was carried out using the Sanger dideoxy-nucleotide chain-termination method (Sanger et al., Proc. Natl. Acad. Sci. USA 74:5463-5467, 1977). A
search of the GenBank/EMBL databases (March, 1994) was done using GCG version 7.0 software (Devereaux et al., Nucl. Acids Res. 12:387-395, 1984).
Expression of the CCaMK Gene. An RNase protection assay (Sambrook et al., 1989) was performed using total RNA (20 ~g) from various parts of lily.
Total RNA was isolated from leaf, stem, and various organs from immature flower (Verwoerd et al., Nucl.
Acids Res. 17:2362, 1989). A 612 bp fragment of the CCaMK coding region (nucleotides 1010-1621) was subcloned into pBluescriptII KS+ plasmid (Stratagene) and used as a template for making the 32P-labeled RNA probe.
Southern Analysis. 5 ~g of lily genomic DNA was digested with different restriction enzymes and transferred to a nylon membrane, followed by Southern analysis using standard protocols (Sambrook et al., 1989).
Expression of CCaMK in E. coli. The CCaMK
protein was expressed in E. coli using pET3b vector (Studier et al., Meth. Enzymol. 185:60-89, 1990;
Novagen, Inc.). E. coli strain BL21 (DE3)-pLysS was transformed with the pET3b expression vector containing CCaMK cDNA. Bacteria were grown at 35~C in M9 minimal medium supplemented with 2 g/L casamino acid, 100 mg/L
ampicillin, 25 mg/L chloramphenicol. The protein was induced by adding 0.5 mM isopropylthiogalactoside (IPTG) when the OD600 reached 0.5 - 0.7. Three hours after AED:en 4630-45115.app 5/23/96 --29--induction, cells were collected by centrifugation, the protein was then extracted and purified by using calmodulin-affinity Sepharose 4B column according to Hagiwara et al. (.J. Biol. Chem. 266, 16401-16408, 1991).
The quality of the purified protein was checked by SDS-polyacrylamide gel electrophoresis.
PreParation of 35S-Labeled Calmodulin and Calmodulin-Binding Assay. 35S-labeled calmodulin was prepared as described by Fromm and Chua (Plant Mol.
0 Biol. Rep. 10:199-206, 1992) using a calmodulin cDNA
(Jena et al ., Proc. Natl . Acad. Sci . USA 86:3644-3648, 1989) cloned into pET3b expression vector. The CCaMK
protein (250 ng) was electrophoretically transferred to a nitrocellulose filter and incubated in a solution containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 ~
(w/v) nonfat dry milk, and 50 nM 35S-calmodulin (0.5x 106 cpm/llg) plus either 1 mM CaCl2 or 5 mM EGTA (Sikela and Hahn, Proc. Natl . Acad. Sci . USA 84:3038-3042, 1987). A
50-fold excess of unlabeled calmodulin was used as a competitor to show specific binding of calmodulin to CCaMK. Calmodulin binding to CCaMK was quantified by measuring radioactivity in each slot using a liquid scintillation counter.
45Ca-Binding Assay. Calcium binding to CCaMK was studied as described by Maruyama et al. (.J. Biochem.
95:511-519, 1984). Purified CCaMK protein was transferred to Zeta-probe membrane (Bio-Rad) using slot blot apparatus (Millipore) and incubated with buffer containing 10 mM Tris-HCl (pH 7.5), 100 mM KCl, 5 mM
MgCl2, and 10 ~Ci/ml 45Ca for 20 min. The membrane was washed for 5 min in the same buffer without 45Ca and exposed to X-ray film.

Reaults and Diacussion A partial clone of lily CCaMK (471 bp) was obtained from developing anthers of lily by PCR using degenerate oligonucleotide primers corresponding to two highly conserved regions of m~mm~l ian Ca2+/calmodulin-,~, AED:en 4630-45115 .app 5/23/96 --30--dependent protein kinases. This fragment was not amplified when the cDNA libraries made from mature and germinating pollens were used. The nucleotide sequence of the PCR-amplified fragment contained conserved sequences corresponding to catalytic subdomains VI-XI
and part of the calmodulin-binding domain of mammalian CaM KII (Hanks et al., Science 241:42-52, 1988).
A 2514 bp lily cDNA clone was obtained by screening the cDNA library using the PCR amplified fragment as a probe and its nucleotide sequence was determined (FIG. 1). The cDNA codes for a polypeptide of 520 amino acids flanked by a 634 bp 5'-untranslated region and a 317 bp 3'-untranslated region. The lily CCaMK polypeptide contains all eleven major conserved subdomains of the catalytic domain of serine/threonine kinases (Hanks et al., Science 241:42-52, 1988).
Sequence comparisons revealed that CCaMK has high homology to Ca2'/calmodulin-dependent protein kinases, especially in the kinase and the calmodulin-binding domains (amino acid residues 1-338). This region of CCaMK has highest homology to kinases from apple (Gen3:Mdstpkn), rat (Gen2:Ratpk2g), human (Genl:Humccdpkb), and fruitfly (Gen2:Drocdpkb, Gen2:Drocdpkd).
The calmodulin-binding region of CCaMK
(ARRKLRAAAIASVL, residues 325-338) has 79~ similarity to the calmodulin-binding domain (ARRKLKGAILTTML, residues 296-309) of ~-subunit of mammalian CaM KII, a well characterized Ca2~/calmodulin-dependent protein kinase (Colbran et al., Biochem. ~. 258:313-325, 1989) and 43~
and 50~ similarity to the calmodulin-binding domains of CaM KII homologs of yeast and Aspergillus, respectively (Pausch et al., EMBO ~. 10:1511-1522, 1991; Kornstein et al., Gene 113, 75-82, 1992). The helical wheel projection of the calmodulin-binding domain (amino acid residues 325-338) of CCaMK formed a basic amphipathic alpha helix (O'Neill et al., Trends in Biochem. Sci.
15:59-64, 1990), a characteristic feature of calmodulin-AED:en 4630-45115.app 5/23/96 --31--binding sites.
The sequence downstream of the calmodulin-binding region of CCaMK (amino acid residues 339-520) does not have significant homology to known Ca2+/calmodulin-dependent protein kinases. Further analysis of this region revealed the presence of three Ca2+-binding EF-hand motifs that had the highest homology (52-54~ similarity; 32-35~ identity) to a family of genes belonging to visinin-like Ca2+-binding proteins (FIG. 2), which are found mainly in neural tissue (Kuno et al., Biochem. Biophys. Res. Commun. 184:1219-1225, 1992; Kobayashi et al., Biochem. Biophys. Res. Commun.
189:511-517, 1992; Lenz et al., Mol. Brain Res. 15:133-140, 1992; Okazaki et al., Biochem. Biophys. Res.
Commun. 185:147-153, 1992; Pongs et al., Neuron 11:15-28, 1993). Even though four EF-hand motifs are present in the calmodulin-like domain of CDPKs, this domain shared only 25~ identity with the visinin-like domain of CCaMK. Out of the six residues of the EF-hand [positions l(X), 3(Y), 5(Z), 7(-Y), 9(-X), and 12(-Z)]
involved in Ca2+-binding, position 7(-Y) is not conserved in CCaMK. A similar deviation is also observed in visinin-like proteins, wherein the residue at position 9(-X) of the EF-hand motifs of visinin-like proteins (FIG. 2) is not conserved. These differences between the EF-hands of the visinin-like domain of CCaMK and other Ca2+-binding proteins may affect Ca2+-binding and protein-protein interactions.
Frequenin, neurocalcin, hippocalcin, and visinin-like neural Ca2+-binding proteins are members of a novel family of Ca2+ sensitive regulators, each with three Ca2+-binding EF-hand motifs. The presence of such proteins has not been reported in plants. These proteins are activated at nanomolar concentrations of Ca2+. At such low levels, calmodulin-dependent pathways are not activated. Frequenin acts as a Ca2+-sensitive activator of photoreceptor particulate guanylyl cyclase (Pongs et al., Neuron 11:15-28, 1993) and may be AED:en 4630-45115.app 5/23/96 --32--involved in activating protein kinases and phosphatases in response to changes in intracellular Ca2+, similar to the action of calmodulin (Pongs et al., Neuron 11:15-28, 1993).
An unusual feature of CCaMK is the presence of a putative biotin-binding site (LKAMKMNSLI) within the visinin-like domain (FIG. 2). Such a biotin-binding site has not been observed in neural visinin-like proteins. Biotin plays a catalytic role in several essential metabolic carboxylation and decarboxylation reactions (Ch~n~ler et al ., J. Biol . Chem. 263:1013-1016, 1988). CCaMK also contains two consensus motifs, RXXT/S (FIGS. 1 and 2), analogous to the autophosphorylation site of mammalian CaM KII and its homologs (Colbran et al., Curr. Top. Cell. Regul.
31:181-221, 1990).
The structural features of the CCaMK gene indicate that it is a novel chimeric Ca2+- and Ca2+/calmodulin-dependent protein kinase with two discrete regulatory domains, a calmodulin-binding domain and a visinin-like Ca2+-binding domain (FIG. 3). The presence of these distinct domains suggest dual modes of regulation. Furthermore, the presence of a putative biotin-binding site suggests yet another mode of regulation, adding to the functional diversity of CCaMK.
The chimeric feature of the CCaMK gene suggests that it has evolved from a fusion of two genes that are functionally different and phylogenetically diverse in origin.
The functional role of the predicted structural motifs of CCaMK was studied using Ca2+ and calmodulin-binding assays. Recombinant lily CCaMK protein was produced in transformed E. coli (FIG. 4A) and purified by calmodulin-affinity chromatography to near homogeneity, as judged by SDS-PAGE. FIG. 4B shows binding of lily CCaMK protein (250 ng) after transfer to a nitrocellulose filter and incubation with 35S-labeled calmodulin in a buffer containing either 5 mM EGTA or 1 -~D:en 4630-45115.app 5/23/96 - 33-mM CaCl2. Calmodulin binds to CCaMK only in the presence of Ca2+.
To determine the functional role of the EF-hand motifs within the visinin-like domain, 45Ca-binding assays were carried out. FIG. 4C shows that 45Ca binds directly to lily CCaMK protein transferred to a Zeta-probe membrane. When incubated with excess amounts (50-fold) of unlabeled calmodulin, the binding of 35S-labeled calmodulin to CCaMK was effectively reduced, suggesting that calmodulin binding to CCaMK was specific.
Moreover, CCaMK showed a Ca2+-dependent shift in mobility, as revealed by SDS-PAGE. These results suggest that CCaMK has some of the structural properties of both Ca2+-dependent and Ca2+/calmodulin-dependent protein kinases (FIG. 3). The intensity of a calmodulin control is less than CCaMK, possibly as a result of inefficient binding of calmodulin to the membrane (Van Eldik and Wolchok, Biochem. Biophys. Res. Commun.
124:752-759, 1984)).
The CCaMK gene was preferentially expressed during phase III (Wang et al., Am. ~. Bot. 79:118-127, 1992) of anther development, as revealed by a ribonuclease protection assay (FIG. 5). Phases II and III correspond to stages of anther development as described by Wang et al. (Am. J. Bot. 79:118-127, 1992).
The expression of lily CCaMK during phase III suggests the involvement of CCaMK in microsporogenesis. Some EF-hand proteins like calmodulin (Moncrief et al., ~. Mol.
Evol. 30:522-562, 1990) are ubiquitous and are active in diverse tissues. However, visinin-like proteins are restricted to specialized tissues such as neurons.
CCaMK, which has a visinin-like domain, is also expressed in an organ-specific manner.
Genomic Southern analysis revealed that CCaMK is coded by a single gene (FIG. 6). Hybridization at low stringency using the lily CCaMK cDNA probe indicated the presence of a CCaMK homolog in other plants, such as Arabidopsis, apple and tobacco. The tobacco homolog to AED:en 4630-45115.app 5/23/96 --34-lily CCaMK has been cloned and its deduced polypeptide displays structural components similar to lily CCaMK, including calmodulin-binding and visinin-like domains.
The CCaMK-like gene is present in both monocotyledonous and dicotyledonous plants.
The Ca2+-signaling pathway mediated through Ca2+/calmodulin-dependent protein phosphorylation is well established in animals. This report confirms the presence of a novel Ca2+/calmodulin-dependent protein kinase in plants. However, the presence of a visinin-like Ca2+-binding domain in CCaMK adds an additional Ca2+
sensing mechanism and distinguishes CCaMK from all other known Ca2+/calmodulin-dependent protein kinases. The discovery of the CCaMK gene adds a new dimension to the understanding of Ca2+-mediated signal transduction in plants.

EXAMPLE 2: Biochemical Properties of LilY CCaMR
Materials and Methods Materials. Proteinase inhibitors, histone IIAS, IIIS, myelin basic protein (MBP), syntide-2, GS peptide (PLSRTLSVAAKK), MBP peptide (QKRPSQRSKYL) and spinach calmodulin were purchased from Sigma. [~_32p] ATP was obtained from DuPont NEN. Calmodulin-Sepharose 4B and Klenow enzyme were obtained from Pharmacia. Restriction enzymes and biotinylated calmodulin were from Bethesda Research Laboratory.
Expression and Purification of CCaMK. E. coli cells carrying plasmid pET3b (Novagen, Inc.) containing CCaMK cDNA were induced by IPTG as described earlier (Patil et al., Proc. Natl . Acad. sci. USA 92:4797-4801, 1995). IPTG-induced E. coli cells were harvested and suspended in a homogenization buffer (40 mM Tris-HCl, pH
7.6, 1 mM DTT, 2 mM EDTA, 0.1~ Triton-X 100, 1 mM PMSF
and 10 ~g/ml each of leupeptin, pepstatin and antipain).
Cells were broken by freeze-thawing followed by sonication. Subsequent procedures were carried out at 4~C. The cell extract was clarified by centrifugation at AED:en 4630-45115.app 5/23/96 --35--12,000 g for 30 min. Solid ammonium sulfate (50~
saturation) was added to the supernatant and incubated on ice for 1-4 hr. The enzyme was recovered by centrifugation for 30 min at 12,000 g. The pellet was solubilized in column buffer (40 mM Tris-HCl, pH 7.6, 1 mM CaCl2, 1 mM dithiothreitol (DTT), 10~ ethylene glycol, 0.05~ Tween-20, 50 mM NaCl, 1 mM PMSF and 10 ~g/ml each of leupeptin, pepstatin and antipain) and applied onto a calmodulin-Sepharose column, which was previously equilibrated with the column buffer. The column was washed first with column buffer, then with column buffer containing 1 M NaCl. CCaMK was eluted from the column with buffer containing 40 mM Tris (pH 7.6), 1.5 mM EGTA, 10 % ethylene glycol, 0.05 ~ Tween-20, 200 mM NaCl and 1 mM PMSF. Fractions containing CCaMK were pooled and thoroughly dialyzed against buffer containing 40 mM Tris (pH 7.6), 1 mM DTT, 1 mM EDTA and 10~ ethylene glycol.
Gel Electrophoresis. SDS-PAGE was performed according to Laemmli (Nature 237:680-685, 1970). Non-denaturing gel electrophoresis was performed using a 14separating gel in 375 mM Tris-Cl (pH 8.8), 5~ stacking gel in 125 mM Tris-Cl (pH 6.8) and 25 mM Tris-192 mM
glycine electrophoresis buffer (pH 8.3), at 80 V for 8 h. Protein bands were visualized by staining with Coomassie Brilliant Blue.
Calmodulin-Binding Assays. The potato calmodulin PCM6 cDNA (Takezawa et al., Plant Mol. Biol.
27:693-703, 1995) was cloned into the pET3b expression vector, and 35S-labeled calmodulin was prepared as described by Fromm and Chua (Plant Mol. Biol. Rep.
10:199-206, 1992). Wild-type and mutant CCaMK proteins were electrophoretically transferred to nitrocellulose filters and incubated in binding buffer (10 mM Tris-Cl, pH 7.5, 150 mM NaCl, and 1~ (w/v) non-fat dry milk) containing 35S-calmodulin (0.5 x 106 cpm/~g) plus either 1 mM CaCl2 or 5 mM EGTA as previously described by Patil et al . (Proc. Natl. Acad. Sci. USA 92:4797-4801, 1995).
Binding assays using biotinylated calmodulin were AED:en 4630-45115 .app 5/23/96 --36--performed as previously described by Reddy et al. (Plant Sci. 94:109-117, 1993).
Assay for Peptide Binding to Calmodulin.
Synthetic peptides were prepared using an Applied Biosystems peptide synthesizer 431A. Different lengths of synthetic peptides were incubated with 100 pmol (1.7 ~g) of calmodulin in 10 ~L of 20 mM Hepes (pH 7.5) for 5 min and analyzed by non-denaturing PAGE.
Deletion Mutants of CCaMK. The mutant construct 1-356 was created by removing a 0.9 kb BamHI fragment containing the visinin-like domain from the original CCaMK expression plasmid pNY10. The mutant construct 1-322 was created by partial digestion of pNY10 with XbaI
and filling the site with the Klenow fragment of DNA
polymerase I. The resulting construct was then inserted into the pET14b expression vector. The mutant proteins were expressed in E. coli and purified using either a calmodulin-Sepharose column (Pharmacia) or a Ni2+-resin column (Novagen) following manufacturer's instructions.
Site-Directed Mutagenesis and Expression of the Visinin-Like Domain. A 0.9 kb BamHI fragment containing the visinin-like domain of CCaMK was subcloned into M13mpl8 RF and site-directed mutagenesis was performed (Kunkel et al., Meth. Enzymol . 154:367-382, 1987).
Oligonucleotide primers used for the site-directed mutagenesis were 5'-CTCTCATGGCTATAGTTCC-3'for EF-hand I
mutation, 5'-CCTCCTTGGCGATACATCC-3' for EF-hand II
mutation, and 5'-GTCGAACGCGACAA~TCC-3'for EF-hand III
mutation. An Nde I site was created at the position of amino acid residue 358 (Met) using 5'-GGATCCCATCATATGAAATCG-3'. Wild-type and the mutant constructs were then inserted into the pET14b expression vector. All mutant sequences were confirmed by DNA
sequencing using the fmol PCR sequencing kit (Promega).
Protein Kinase Assay. Phosphorylation assays (25 ~L) were carried out at 30~C in 50 mM Hepes (pH 7.6), 1 mM DTT, 10 mM magnesium acetate, 200 ~M [r-32P]ATP, ~ .
AED:en 4630-45115 .app 5/23/96 - 37-(1,500 to 2000 cpm/pmol) in the presence of either 2.5 mM EGTA or indicated amounts of Ca2' and calmodulin.
Protein (0.2 mg/ml), and synthetic peptides (100 ~M) were added in the reaction mixture to study substrate phosphorylation. When protein substrates were used, the reaction was terminated by adding SDS-PAGE sample buffer (Laemmli, Nature 237:680-685, 1970) and analyzed after electrophoresis on 12~ SDS polyacrylamide gels.
Proteins were visualized by staining with Coomassie Brilliant Blue. The gels were dried and subjected to autoradiography. Incorporation of 32p into the substrate was determined by counting the excised protein bands in a liquid scintillation counter. When peptide substrates were used the reaction was terminated by spotting the reaction mixture on P81 phosphocellulose filters (Whatman). The filters were washed in 75 mM phosphoric acid and 32p incorporation was determined (Roskoski, Jr., Meth. Enzymol. 99:3-6, 1983).
Autophosphorylation Assay. The autophosphorylation assay was carried out at 30~C in the presence of 50 mM Hepes, pH 7.5 containing 10 mM
magnesium acetate, 1 mM DTT, 1 mM [~-32P]ATP (300 to 400 cpm/pmol) and either EGTA (2.5 mM), CaCl2 (0.5 mM), or CaCl2 (0.5 mM) plus calmodulin (1 ~M). For time course assays (100 ~L), 1.2 ~g (21.4 pmol) of CCaMK and 1 mM
[ r - 32p] ATP (2,000-3,000 cpm/pmol) were used. Aliquots (10 ~L) were transferred at indicated time points into SDS-PAGE sample buffer to stop the reaction. Aliquots for the zero time point were taken immediately after the addition of CCaMK. The samples were then analyzed by electrophoresis on a 12~ SDS polyacrylamide gel. The amount of phosphate transferred to the enzyme was determined by counting the radioactivity of the excised CCaMK bands in a liquid scintillation counter.
Phosphoamino Acid Analysis. The purified CCaMK
(200 ng) was autophosphorylated in the presence of EGTA
(2.5 mM), or CaCl2 (0.5 mM) or CaCl2 (0.5 mM) plus 1 ~M
calmodulin, and subjected to SDS-PAGE. The gel was AED:en 4630-45115.app 5/23/96 --38--briefly stained with Coomassie Brilliant Blue, and CCaMK
bands were excised and the protein was eluted from the gel. The eluted protein was hydrolyzed with 6 N HCl for 2 h at 110~C and subjected to paper chromatography using propionic acid: 1 M NH40H: isopropyl alcohol (45:17.5:17.5) as a solvent (Cooper, Meth. Enzymol.
99:387-402, 1983). Phosphoserine and phosphothreonine standards (50 mg/ml in 10~ w/v isopropyl alcohol) were visualized by ninhydrin reagent.
Results To study the Ca2+/calmodulin-dependent kinase activity of lily CCaMK, the E. coli-expressed protein was purified. The protein was essentially pure as revealed by SDS-PAGE and was stable at 4~C for a few days. The purified protein was used to phosphorylate different substrates such as casein, histones, myelin basic protein, and synthetic peptides. Histone IIAS was found to be the most reactive protein substrate for CCaMK, and was used for studying calmodulin concentration-dependent protein kinase activity. The addition of increasing amounts of calmodulin in the presence of 0.5 mM Ca2+ stimulated CCaMK activity (FIG.
7A). Kinase activity was saturated at calmodulin concentrations around 1.0 ~M. The concentration of calmodulin required for half-maximal activity (Ka) of CCaMK was approximately 0.2 ~M. The time course studies revealed that histone IIAS phosphorylation was saturated after 10 min in the presence of Ca2+/calmodulin (FIG.
7B). In the presence of 2.5 mM EGTA or 0.5 mM Ca2+
alone, the enzyme has a basal activity that is ten- to fifteen-fold lower than the maximal activity achieved with Ca2+/calmodulin. Among other protein substrates tested, CCaMK phosphorylated histone IIIS and myelin basic protein, but it did not phosphorylate phosvitin, PEP carboxylase, synapsin I, and casein. CCaMK also phosphorylated synthetic peptides such as GS peptide, MBP peptide, and syntide-2. Among these peptides, GS

, AED:en 4630-45115.app 5/23/96 --39--peptide was most efficiently phosphorylated by CCaMK in the presence of Ca2+/calmodulin.
FIG. 8 shows a saturation curve of 35S-calmodulin binding to purified CCaMK to determine the calmodulin-binding affinity of CCaMK. Upon induction of CCaMKexpression in E. coli, 4 pmol of protein was separated by SDS-PAGE, electrophoretically transferred to a nitrocellulose filter, and incubated with different amounts of 35S-labeled calmodulin. After washing in buffer without 35S-calmodulin, the radioactivity of the filter was measured using a liquid scintillation counter. Binding of calmodulin to CCaMK saturated at concentrations above 300 nM. From the saturation curve, the dissociation constant (Kd) of calmodulin for CCaMK
was estimated to be approximately 55 nM. The binding of calmodulin to CCaMK was completely blocked in the presence of 5 mM EGTA. A Scatchard plot of the binding data shows that the binding ratio of calmodulin to CCaMK
is 1:1 (FIG. 8, inset), indicating that CCaMK has a single calmodulin-binding site.
To identify the calmodulin-binding region of CCaMK, truncated mutant constructs were prepared (FIG.
9A). The CCaMK mutant 1-356 lacks the C-terminal domain which has high homology to visinin-like proteins.
Another mutant, CCaMK 1-322, is further truncated but retains all eleven domains conserved in serine/threonine protein kinases (Hanks et al., Science 241:42-52, 1988).
Wild-type CCaMK (1-520), and truncated mutants 1-356 and 1-322 were expressed in E. coli, subjected to SDS-PAGE, and transferred to nitrocellulose filter. Excised bands containing the expressed proteins were assayed for binding to35S-calmodulin in the presence of Ca2+. The radioactivity of bound 35S-calmodulin was 11,600 cpm for wild-type CCaMK, 12,500 cpm for the mutant 1-356, and 99 cpm for the mutant 1-322, respectively. Thus, binding of calmodulin to wild-type and mutant 1-356 CCaMKs were similar; calmodulin did not bind to the mutant CCaMK 1-322 (Fig. 9A, in boxed region), indicating that amino , AED:en 4630-45115.app 5/23/96 -40-acid residues 322-356 (FIG. 9A) are essential for calmodulin-binding to CCaMK. Another mutant' CCaMK 1-341, also binds to calmodulin in the presence of Ca2+.
Similar results were obtained when biotinylated 5 calmodulin was used instead of 35S-calmodulin.
Calmodulin binding to wild-type and mutant CCaMKs was prevented by the addition of 5 mM EGTA, indicating that Ca2+ is required for calmodulin binding. Comparison of amino acid residues of this region of CCaMK
10 corresponding to regions of animal CaMKIIcY revealed high homology (FIG. 9B).
Synthetic peptides from the calmodulin-binding region (amino acid residues 311-340) were used to identify amino acid residues necessary for calmodulin-15 binding by a gel mobility-shift assay using non-denaturing PAGE in the presence of 0.5 mM CaCl2. The bands of calmodulin and calmodulin-peptide complex were visualized by staining with Coomassie Brilliant Blue.
Calmodulin mixed with peptides 311-340, 317-340, and 322-340 migrated above the position of calmodulin alone.
Peptide 328-340 did not affect the mobility of calmodulin (FIG. 10, top), suggesting that the calmodulin-binding site exists between amino acid residues 322-340. Addition of these peptides to 25 calmodulin in the presence of 2.5 mM EGTA did not affect the mobility of calmodulin, suggesting that peptide binding to calmodulin is Ca2+-dependent. Increasing amounts of peptide 322-340 facilitates the gel mobility shift towards the upper, higher molecular weight position (FIG. 10, bottom). Similar results were obtained when peptides 317-340 and 311-340 were used, suggesting that amino acid residues 322-340 have a pivotal role in CCaMK calmodulin binding.
The helical wheel pro]ection revealed that amino acid residues 325-338 of CCaMK form a basic amphiphilic ol helix (O'Neil and DeGrado, Trends Biochem. Sci. 15:59-64, 1990) similar to CaMKIIcY (FIG. 11).
To study autophosphorylation, CCaMK was .
AED:en 4630-45115.app 5/23/96 -41-incubated at 30~C with 10 mM magnesium acetate, 1 mM [~-32p] ATP and 2.5 mM EGTA. In 30 min, approximately 0.098 mol 32P/mol of CCaMK was incorporated. This basal autophosphorylation was induced approximately 3.4 fold 5 in the presence of 0.5 mM CaCl2 (0.339 mol 32P/mol of CCaMK) (FIG. 12). Increasing the incubation time to 60 min did not improve the stoichiometry of Ca2+-dependent autophosphorylation. Ca2+-dependent autophosphorylation was inhibited to basal levels (0.061 mol 32P/mol of CCaMK) by the addition of 1 ~M calmodulin (FIG. 12).
Calmodulin inhibits Ca2+-stimulated autophosphorylation in a concentration dependent manner (FIG. 13A). These results indicate that Ca2+ and calmodulin have opposite effects on autophosphorylation of CCaMK. Phosphoamino acid analysis revealed that CCaMK autophosphorylates at the threonine residue(s) (FIG. 13B), and that autophosphorylation was stimulated by Ca2+ and inhibited by Ca2+/calmodulin.
Apart from the calmodulin-binding domain, CCaMK
has another regulatory domain nearer the C-terminus that has high homology to animal visinin-like proteins. The visinin-like domain of CCaMK contains three EF-hand motifs with conserved Ca2+-ligating amino acid residues (FIG. 14A). To study Ca2+-binding properties of the visinin-like domain of CCaMK, recombinant visinin-like domain protein was expressed in E. coli, using the pET14b expression vector. The visinin-like domain protein was expressed at high levels upon induction with 0.5 mM IPTG. Most of the protein was present in the soluble fraction. The expressed protein was purified using a Ni2+ resin column. Protein eluted from the column with lM imidazole buffer was dialyzed in 50 mM
Tris-Cl (pH 7.5) and used in a Ca2+-dependent mobility shift assay. Electrophoretic mobility of the recombinant visinin-like domain protein on a 14~ SDS-polyacrylamide gel was just above the 20.1 kDa molecular-weight marker in the presence of 2.5 mM EGTA.
Addition of Ca2+ shifted the electrophoretic mobility AED:en 4630-45115.app 5/23/96 -42-toward a lower molecular weight (FIG. 14B), suggesting that Ca2+ binding to the recombinant visinin-like domain protein induces a conformational change.
To verify that the EF-hand motifs in the visinin-like domain are responsible for the Ca2+-dependent mobility shift, site-directed mutants of the visinin-like domain protein were created. Each of the EF-hands (I, II, and III) were mutated by replacing the amino acid residue at the -x position (D417 to A, S453 to A, and T495 to A) in the EF-hands (FIG. 14A), which are known to be primary determinants of the Ca2+
dissociation rate (Renner et al., J. Biol. Chem.
90:6493-6497, 1993). A mutant in which all three EF-hands are mutated was expressed in E. coli, purified, and analyzed by SDS-PAGE in the presence of Ca2+. The visinin-like protein mutated in the EF hand I migrated at a similar position to the wild-type protein, suggesting that this site may not be functional.
However, mutations in EF-hands II and III shifted the mobility of the protein toward the higher molecular weight. The mutant of the EF-hand III migrated to a similar position to the protein in which all three EF
hands are mutated (FIG. 14B). The migration of EF-hand III mutant in the presence of Ca2+ was also similar to the wild-type protein in the absence of Ca2+. These results suggest that Ca2+ binding to the EF-hands II and III contribute to the Ca2+-dependent mobility shift of the visinin-like domain protein. Removal of Ca2+ by EGTA
causes the mobility of all the mutant proteins to shift upward to similar higher molecular weight positions. To study the role of the visinin-like domain in Ca2+-stimulated autophosphorylation, CCaMK mutant 1-356, which lacks the visinin-like domain, was used for autophosphorylation and substrate phosphorylation studies. Autophosphorylation of mutant 1-356 was not stimulated by Ca2+ (FIGS. 15A and 15C), but retained Ca2+/calmodulin-dependent kinase activity at a substantially reduced level (FIGS. 15B and 15D). This , AED:en 4630-45115.app 5/23/96 --43--result indicates that the visinin-like domain is required for Ca2+-stimulated autophosphorylation as well as for maximal substrate phosphorylation.
The significance of Ca2+-stimulated autophosphorylation on substrate phosphorylation by CCaMK was studied using histone IIAS and GS peptide as substrates. In the presence of histone IIAS, calmodulin did not suppress the Ca2+-dependent autophosphorylation of CCaMK, probably due to interaction of histone IIAS
with acidic proteins such as calmodulin and the visinin-like domain of CCaMK. FIG. 16A shows the effect of increasing concentrations of calmodulin on the GS
peptide phosphorylation by autophosphorylated CCaMK.
The rate of phosphorylation of the GS peptide by unphosphorylated CCaMK was stimulated by increasing concentrations of calmodulin, but the maximal stimulation was only 3- to 4-fold as compared to the basal activity. However, when autophosphorylated CCaMK
was used, calmodulin stimulated the rate of phosphorylation of the GS peptide with kinetics similar to histone IIAS (FIG. 16A).
To study the effect of autophosphorylation on kinase activity, the Ca2+/calmodulin-dependent and Ca2+/calmodulin-independent activities of autophosphorylated and unphosphorylated CCaMKs were compared using GS peptide as substrate.
Autophosphorylated CCaMK exhibited approximately five-fold greater Ca2+/calmodulin-dependent kinase activity than unphosphorylated CCaMK. The maximal stimulation of autophosphorylated CCaMK by Ca2+/calmodulin was 20-fold to 25-fold compared to the EGTA control (FIG. 16B).
Ca2+/calmodulin-independent activity was not significantly affected by autophosphorylation. These results show that Ca2+-induced autophosphorylation stimulates Ca2+/calmodulin dependent activity of CCaMK.
A deletion mutant lacking both calmodulin-binding and visinin-like domains (1-322) showed constitutive activity (Ca2+/calmodulin-independent), AED:en 4630-45115.app 5/23/96 --44--suggesting the presence of an autoinhibitory domain.
Synthetic peptides derived from the putative autoinhibitory domain (amino acid residues 311-340) inhibited the activity of the constitutive mutant 1-322 (FIG. 17). Models describing the regulation of CCaMK by Ca2+ and Ca2+/calmodulin and the autoinhibitory domain are shown in FIG. 18. The autoinhibitory domain of CCaMK
has similarity to the autoinhibitory domain of m~mm~l ian calmodulin kinase II (CaMKII) (Brickey et al., J. Biol.
0 Chem. 269:29047-29054, 1994).

Discussion These studies provide biochemical evidence for a Ca2+/calmodulin-dependent protein kinase in plants.
Although several Ca2+/calmodulin-dependent kinases have been characterized from animal systems (Nairn and Picciotto, Semin. Cancer Biol. 5:295-303, 1994), CCaMK
is the only plant kinase whose activity is regulated by both Ca2+ and Ca2+/calmodulin. Among the substrates tested, histone IIAS and synthetic GS peptide are the most efficient phosphate acceptors. CCaMK exhibits a higher Ka value (150-200 nM) for calmodulin (FIGS. 7A
and 16A) compared to CaMKII (20-100 nM) (Schulman, Adv.
Second Messenger Phosphoprotein Res. 22:39-112, 1988) and CaMKIV (26-150 nM) (Kameshita and Fujisawa, J.
Biochem. (Tokyo) 113:583-590, 1993; Enslen et al., J.
Biol. Chem. 269:15520-15527, 1994), indicating that plant kinase requires a higher concentration of calmodulin for its activity. This is probably due to a higher dissociation constant of calmodulin for CCaMK (55 nM) than for animal Ca2+/calmodulin-dependent protein kinases (1-10 nM) (Sikela and Hahn, Proc. Natl. Acad.
Sci. USA 84:3038-3042, 1987). 35S-labeled calmodulin binding and peptide binding assays revealed that the calmodulin-binding site of CCaMK is present between amino acid residues 322-340 (FIG. 9). This region has homology to animal CaMKII, with conserved basic (Arg-325, Arg-326, and Lys-327) as well as hydrophobic (Phe-AED:en 4630-45115.app 5/23/96 --45-323, Ala-325, Ala-332, and Leu-338) amino acid residues.
The visinin-like Ca2+-binding domain, a novel feature of CCaMK, is not known to exist in other protein kinases. The visinin-like domain contains three EF-hand motifs (FIG. 14A) similar to animal visinin-like proteins. Frequenin, neurocalcin, and visinin-like proteins are known to be members of Ca2+-sensitive guanylyl cyclase activators that are involved in cation channel regulation in neuronal tissues (Palczewski et al., Neuron 13:395-404, 1994). Visinin-like proteins typically contain three conserved EF-hand motifs, each with a different affinity for Ca2+ (Pongs et al., Neuron 11:15-28, 1993; Ames et al ., J. Biol . Chem. 270: 4526-4533, 1995). The Ca2+-dependent mobility-shift assay suggests that binding of Ca2+ to the EF-hands II and III
is important for inducing conformational changes in the visinin-like domain of CCaMK (FIG. 14B). Ca2+-induced conformational change in the visinin-like domain may be critical for regulation of CCaMK activity. The CCaMK
mutant 1-356 lacking this domain did not show Ca2+-dependent autophosphorylation. The mutant 1-356 also exhibited reduced activity as compared to the wild-type enzyme, suggesting that the visinin-like domain is required for the maximal activation of CCaMK. It is unlikely that this reduced activity is due to lowered affinity of mutant 1-356 to calmodulin, since the saturation curve of 35S-calmodulin binding for mutant 1-356 indicated that it has a similar Kd (60 nM) for calmodulin. However, it is possible that the visinin-like domain may stabilize the conformation of CCaMK,which is indispensable for its maximal activity.
Phosphoamino acid analysis revealed that CCaMK
autophosphorylation is due to the phosphorylation of the threonine residue(s) (FIG. 13B). Autophosphorylation of CCaMK increased its Ca2+/calmodulin-dependent kinase activity by five-fold (FIG. 16B). Ca2+/calmodulin-dependent autophosphorylation of animal CaMKII at Thr-286 NH2-terminal to the calmodulin-binding site, is known - ' - 2 1 77 2 1 7 AED:en 4630-45115.app 5/23/96 --46--to stimulate Ca2+-independent activity (Colbran and Soderling, Curr. Topics Cell. Regul. 31:181-221, 1990;
Theil et al., Proc. Natl. Acad. Sci. USA 85:6337-6341, 1988; Fong et al., J. Biol. Chem. 264:16759-16763, 1989). In contrast, Ca2+/calmodulin-independent basal autophosphorylation at Thr-305 and 306 within the calmodulin-binding site inactivates CaMKII by inhibiting its ability to bind calmodulin (Leckteig et al., J.
Biol. Chem. 263:19232-19239, 1988; Colbran, J. Biol.
0 Chem. 268:7163-7170, 1993). Although the calmodulin binding region of CCaMK has similarity to the calmodulin-binding region of CaMKII, there are no threonine residues around this area (FIG. 9A). The inhibition of the Ca2+-stimulated CCaMK
autophosphorylation by calmodulin, may be due to the conformational change induced by the calmodulin binding to CCaMK (James et al., Trends Biochem. Sci. 20:38-42, 1995). Inhibition of autophosphorylation by calmodulin is also reported in smooth muscle MLCK (Tokui et al., Biochemistry 34:5173-5179, 1995), in which all three phosphorylated residues are present in proximity to the calmodulin-binding site. The absence of threonine residues around the calmodulin-binding region of CCaMK
suggests that the mechanism of CCaMK regulation by autophosphorylation is different from MLCK and CaMKII.
Signal-induced changes in cytosolic Ca2+
concentration are believed to be important for many cellular processes in plants (Gilroy and Trewavas, Trends Genetics 16:677-682, 1994; Bush, Plant Physiol.
103:7-13, 1993; Gilroy et al., J. Cell. Sci. 106:453-462, 1993). Ca2+ has a dual effect on the stimulation of CCaMK activity. In the presence of calmodulin, Ca2+
binds to calmodulin and stimulates CCaMK activity. In the absence of calmodulin, Ca2+ alone stimulates autophosphorylation of CCaMK which further increases Ca2+/calmodulin-dependent kinase activity (FIG. 16B).
Plants have multiple isoforms of calmodulin and their expression is developmentally regulated and AED:en 4630-45115.app 5/23/96 -47--responsive to environmental signals (Takezawa et al.
Plant Mol . Liol . 27:693-703, 1995; Jena et al., Proc.
Natl. Acad. Sci. USA 86:3644-3648, 1989; Braam and Davis, Cell 60:357-364, 1990). Plant calmodulin mRNA
and protein are also reported to have a relatively rapid turnover rate in the cell (Perera and Zielinski, Plant Physiol. 100:812-819, 1992). Signal-induced expression and rapid turnover suggest that there is a dynamic regulation of calmodulin in vivo. Therefore, it is likely that CCaMK activity is differentially controlled by signal-induced transient changes in free Ca2+
concentration and calmodulin. In plant cells, the Ca2+
concentration required for Ca2+-dependent autophosphorylation and the Ca2+ concentration required for Ca2+/calmodulin-dependent substrate phosphorylation may be different.

EXAMPLE 3: Effects of CCaMK on Male SterilitY
Materials and Methods Plant Material. Tobacco (Nicotiana tabacum cv.
Xanthi) and lily (Lilium longiflorum Thumb cv. Nellie White) plants were grown under normal greenhouse conditions.
In situ Hybridization. Different stages of lily anthers were cut and fixed overnight at 4~C in a solution containing 25~ paraformaldehyde, 1.25~
glutaraldehyde, 50 mM Pipes (pH 7.2). The samples were processed and embedded in LR white. One ~m-thick cross-sections were mounted on gelatin-coated slides, stained with safranin, and examined by light microscopy.
To obtain a CCaMK-specific probe, a 438-bp fragment (base pairs 2076 to 2514 in lily CCaMK; Patil et al., Proc. Natl. Acad. Sci. USA 92 :4897-4901, 1995) was cloned into the pSPT18 plasmid (Boehringer Mannheim, DIG RNA Labeling Kit, cat. no. 1175025). Antisense and sense digoxigenin-labeled RNA was synthesized according to standard protocols (Boehringer Mannheim). After transcription, the RNA was hydrolyzed to approximately AED:en 4630-45115.app 5/23/96 --48--150 bp by 0.2 M NaHCO3/Na2CO3 (pH 10.2) at 60~C.
Sections were treated with 5 ~g/ml proteinase K
for 30 min at 37~C before hybridization. 15 ~l hybridization solution (Panoskaltsis-Mortair et al., 5 BioTechniques 18:300-307, 1995) containing the heat-denatured RNA probe were applied to each section, covered with a coverslip, sealed with rubber cement, and incubated in a humid chamber overnight at 48~C. The next day, the slides were washed in 2 x SSC at room temperature for 5 min, then incubated in RNase A (40 ~g/ml in STE) for 30 min at 37~C, washed with 2 x SSC, 50~ formamide at 50~C for 5 min, with 1 x SSC, 0.5 x SSC, and 0.2 x SSC at room temperature for 5 min each wash, then quickly rinsed with H2O. Signals were detected by immunolocalization (Li et al ., Cell 72: 869-879, 1993) using gold-conjugated sheep anti-digoxigenin, silver enhanced at room temperature for 18 min, then stained with safranin for 30 sec.
Protein Extraction. Anther tissue was frozen in liquid nitrogen and ground using a mortar and pestle, then taken into 4 to 5 volumes (w/v) of extraction buffer (40 mM Tris pH 7.6, 1 mM DTT, 1 mM EDTA, 0.1~
Triton X-100, 1 mM PMSF and 10 ~g/ml each of antipain, pepstatin, and leupeptin). Powdered tissue was vortexed in the extraction buffer for 1 min and centrifuged at 12,000 g for 10 min at 4~C. The supernatant was used for phosphorylation assays.
Inactivation of Endogenous Kinases. The total protein extract was heated at 60~C for 10 min, slowly cooled in water at room temperature, and centrifuged at 10,000 g for 10 min at 4~C. This method inactivates all endogenous kinases in plant extracts.
In vi tro Phosphorylation of Proteins. For in vitro phosphorylation, 25 ~g of heat-inactivated total proteins were used. The assay (50 ~l) was carried out for 10 min at 30~C in 50 mM Hepes (pH 7.6), 1 mM DTT, 10 mM Mg(Ac)2, 200 mM ~-32P-ATP (1500 - 2000 cpm/pmol), 0.5 mM CaCl2, l~M calmodulin and either with or without CCaMK

' ' AED:en 4630-45115.app 5/23/96 --49-(200 ng). The reaction was terminated by the addition of SDS-PAGE sample buffer and analyzed by SDS-PAGE using a 10~ gel. Proteins were visualized by staining with Coomassie Brilliant Blue. The gels were dried and subjected to autoradiography.
Autophosphorylation of CCaMK. Four ~g of CCaMK
were autophosphorylated at 30~C for 20 min in the presence of 50 mM Hepes (pH 7.5) containing 0.5 mM ~_32p_ ATP (8000 - 10,000 cpm/pmol), 10 mM Mg(Ac)2, 1 mM DTT.
Unincorporated ATP was removed by filtering the reaction mixture several times through a Microcon 10 filter (Amicon).
Gel Overlay Assay of CCaMK-binding Proteins.
Protein samples (100 ~g) from lily anthers at various stages of development were separated by SDS-PAGE using a 12~ gel and transferred to a PVDF membrane (Millipore) at room temperature at 150 V for 3 hrs in transfer buffer (39 mM glycine, 48 mM Tris base, 0.037~ SDS and 20~ methanol). The gel overlay assay was performed using the method of Carr and Scott ( Trends Biochem . Sci .
17:246-247, 1992). The membrane was blocked for 60 min in blocking buffer (50 mM Tris HCl/200 mM NaCl-TBS
containing Tween 20 (3% v/v) and non-fat powdered milk (5~ w/v)).
After washing the membrane with rinsing buffer (TBS containing Tween 20 (0.1% v/v) and non-fat powdered milk (5~ w/v)) for 30 min, the membrane was incubated with autophosphorylated CCaMK (32P-labeled) in rinse buffer for 2 hrs at room temperature with constant agitation. The membrane was washed extensively with a minimum of 4 to 5 changes of rinse buffer, dried, and subjected to autoradiography.
PCR cDNA Library Screening, and Sequencing. A
partial CCaMK cDNA clone (483 bp) was obtained from developing anthers of tobacco (Nicotiana tabacum SR1) by PCR using two degenerate oligonucleotide primers corresponding to two highly conserved regions of m~mm~l ian Ca2+/calmodulin-dependent protein kinases , AED:en 4630-45115.app 5/23/96 - 50--(DLKPEN and FNARRKL, Hanks et al., Science 241:42-52, 1988). A tobacco immature anther cDNA library was produced using the A ZAPII vector according to manufacturer's protocol (Stratagene) and screened using the PCR-amplified fragment as a probe. The sequencing of the cDNA was carried out by using the dideoxynucleotide chain-termination method.
RT-PCR Analysis. First-strand cDNA was synthesized from 5 ~g total RNA using a cDNA synthesis kit (Gibco BRL) and one out of 20 ~l was used as template. Two tobacco CCaMK gene-specific oligonucleotide primers (3'-coding region, amino acid residues 290-296 and 512-518) and two calmodulin degenerate primers (Takezawa et al ., Plant Mol. Biol.
27:693-703, 1995) were used in the same reaction. The cycling profile was 25 cycles of 94~C for 30 sec, 50~C
for 1 min, and 72~C for 1 min.
Southern Blot Analysis. Tobacco genomic DNA (5 ~g) was digested with various restriction enzymes and transferred to a nylon membrane. Hybridization was by standard protocols (Sambrook et al., 1989). The membrane was washed twice in 2 x SSC, 0.5% SDS at room temperature for 15 min and twice in 0.1 x SSC, 0.1% SDS
at 65~C, 20 min.
Plant Transformation. Binary plasmid pGA748 (An, Meth. Enzymol . 153:292-305, 1987) was used in preparing the sense and antisense constructs. To produce a sense construct, full-length cDNA was used; to produce an antisense construct, a BamHI cut fragment (amino acid residues 110-517) was used. The two constructs were then transferred to A. tumefaciens strain LBA4404 using a direct DNA transfer method (An, Meth. Enzymol. 153:292-305, 1987). Leaf discs of N.
tabacum xanthi were transformed according to the method of Horsch (Science 227:1229-1231, 1985). Transformants were selected on media containing kanamycin (100 mg/l).
Slot-Blot Analysis. Different tobacco parts and tobacco anthers from different stages were collected and -AED:en 4630-45115.app s/i3/96 --51--the RNA was isolated as described by Verwoerd (Nucl.
Acids Res. 17 :2362, 1989). Slot blot analysis was performed by using 5 ~g total RNA from different tobacco parts or anthers from different stages. A 330-bp fragment (5' coding region, amino acid residues 1-109, FIG. 11) was used as a probe. Hybridization was performed at 42~C overnight in a solution containing 50%
formamide, 6 x SSPE, 5 x Denhardt's solution, 0.5% SDS, 100 ~g/ml denatured herring sperm DNA and >109 cpm/~g of 32P-labeled cDNA probe. The membrane was washed once in 2 x SSC, 0.5% SDS at room temperature for 15 min and two times for 10 min each in 0.2 x SSC, 0.1% SDS at 55~C, and exposed to film. After the film was exposed, the membrane was washed in 0.1 x SSC, 0.1% SDS at 90~C for 5 min to remove the probe, then re-hybridized to 32p_ labeled PCM6 calmodulin cDNA, which shows the least changes during development (Takezawa et al., Plant Mol.
Biol . 27: 693-703, 1995) to confirm that the loaded RNA
was the same amount.
Scanning Electron Microscopy. Pollen grains from mature anthers were collected, freeze-dried, coated with gold and observed under a scanning electron microscope at 15 KV or 20 KV.
Histochemical Localization of Callose. Dehisced tobacco pollen grains were tapped directly into a 0.01%
aqueous solution of water-soluble aniline blue made up in 0.15 M K2HPO4. After 30 min, samples were observed using a microscope fitted with a fluorescence attachment as described by Worrall et al. (Plant Cell 4:759-771, 1992).
Pollen Germination. Dehisced tobacco pollen grains were collected and incubated at room temperature in a medium containing 10% sucrose, 0.0017 g/l KH2PO4, 0.025 g/l H3BO3, and 10 mM CaCl2. After 2 hrs incubation, the pollen suspension was mixed with diphenylboric acid 2-aminoethyl éster (0.5% in 50% MeOH, Sigma) and photographed.
Cross Pollination. Anthers in antisense plants -AED:en 4630-45115.app 5/23/96 --52--were removed before they matured, pollen grains from dehisced wild-type anthers were collected and applied to pistils of antisense plants.

Results CCaMK is Expressed in Anther in a Stage-specific Manner. Micrographs in FIGS. l9A-C show the progressive development of microspores in lily anther. FIG. l9A
shows the pollen mother cells (bud size 2.0 cm), FIG.
l9B shows stage I of meiosis (bud size 2.5 cm), and FIG.
l9C shows the uninucleate microspores (bud size 3.5 cm).
In order to investigate the cellular localization of CCaMK, in situ hybridization experiments were performed.
Thin sections of developing lily anthers were hybridized with an antisense or sense CCaMK RNA probe labeled with digoxigenin and detected by gold-conjugated sheep anti-digoxigenin. The antisense probe yielded hybridization signals mostly in tapetum and locules when the bud was 2.5 to 3.5 cm (FIGS. l9D and l9E). In contrast, epidermis, endothecium and middle layers showed little or no expression, while the sense control did not show hybridization in any tissues. Expression of CCaMK was specific to stages that coincided with meiosis and the uninucleate microspore stage (FIGS. l9B and l9C).
Identification of CCaMK Substrates in Lily Anthers. In order to identify substrates for CCaMK, total protein isolated from various stages of developing lily anthers was heated to inactivate endogenous kinases and subjected to Ca2+/calmodulin-dependent phosphorylation in the presence or absence of E. coli-expressed and purified CCaMK. The proteins were then separated on a 10~ SDS-polyacrylamide gel and the gel was dried was subjected to autoradiography. Several endogenous proteins were phosphorylated in a Ca2+/calmodulin-dependent manner by CCaMK (FIG. 20).
These proteins were shown to be present when the buds were 1.0 to 3.0 cm (fully opened lily flower is about 15 cm), coinciding with the pollen mother cell stage to the AE:D:en 4630-45115.app 5/23/96 --53-uninucleate microspore stages of lily anther development. The amount of these substrate proteins decreased in later stages of anther development and were absent in fully mature anthers and other parts of the plant, indicating the anther- and developmental stage-specificity of these substrates.
A polypeptide of approximately 24 kDa that was present at an extremely low level was phosphorylated to a very high level (FIG. 20), indicating its high specificity for CCaMK. The phosphorylation of this 24-kDa protein is Ca2+/calmodulin-dependent, since the addition of EGTA, a Ca2+ chelator or W-7 (Sigma), a calmodulin inhibitor, prevented its phosphorylation by CCaMK.
Binding Proteins of CCaMK. Autophosphorylated CCaMK (32P-labeled) was used in a gel overlay assay to identify the proteins interacting with CCaMK. Total protein (100 ~g) from various stages of lily anthers and other parts of the plant were separated on SDS-PAGE, transferred to a PVDF membrane, and probed with autophosphorylated CCaMK (32P-labeled) CCaMK. The dried membrane was then subjected to autoradiography. As shown in FIG. 21, several proteins that bind specifically to CCaMK were present only when the bud size was between 0.5 cm to 3.0 cm, coinciding with the pollen mother cell stage to the uninucleate microspore stage. There was a progressive decrease in binding proteins at later stages of anther development and a total absence in fully mature anthers. These proteins were not detected in other parts of the plant. These results indicate that CCaMK binds various proteins in developing anthers in an anther- and stage-specific manner.
Cloninq and Sequence Analysis of a Tobacco CCaMK
cDNA. A full-length (1776 bp) cDNA clone from tobacco (Nicotiana tabacum SR1) was obtained by screening an immature tobacco anther cDNA library using a PCR-amplified fragment (483 bp, corresponding to amino acid , AED:en 4630-45115.app 5/23/96 --54--residues 164-325) as a probe. The nucleotide sequence of tobacco CCaMK cDNA is shown in FIG. 22.
The coding region of the tobacco CCaMK cDNA
encodes a 517 amino acid polypeptide and is flanked by a 5 19 bp 5'-untranslated region and a 203 bp 3'-untranslated region. FIG. 23A shows the comparison of amino acid sequences of tobacco and lily CCaMKs. Both tobacco and lily CCaMKs contain all 11 major conserved subdomains of serine/threonine protein kinases (Hanks et 0 al., Science 241:42-52, 1988), the calmodulin-binding domain and the visinin-like Ca2'-binding domain (FIGS.
-23A and 23B). Tobacco CCaMK and lily CCaMK share 71%
identity and 82% similarity, with 66% identity and 79%
similarity in the kinase domain (amino acid residues 1-307). The 3' visinin-like domain (amino acid residues 339-517) is highly conserved, sharing 79% identity and 87% similarity, suggesting that the visinin-like domain is functionally conserved and plays an important role in regulating CCaMK activity. The calmodulin-binding domain (amino acid residues 320 to 335) is also conserved.
The helical wheel projection of the calmodulin-binding domain of tobacco and lily CCaMKs formed a basic amphipathic ~-helix (O'Neill et al., Trends Biochem Sci.
15:59-64, 1990), a characteristic feature of calmodulin-binding sites.
ExPression Pattern of CCaMK in Tobacco. As shown in FIG. 24, CCaMK mRNA was detected during meiosis (bud size 0.5 - 0.8 cm), and peaked following meiosis (bud size about 1.0 cm; when fully opened, the tobacco flower is approximately 4.5 cm). The message became undetectable at later stages of development. The CCaMK
gene was preferentially expressed during flower stage 3 to stage 2 of anther development (Koltunow et al., Plant Cell 2:1201-1224, 1990). RNA slot-blot analysis confirmed this result. No expression was detected in other tissues (including leaf, stem, root, pistil, ovary, and petal), indicating that CCaMK is expressed in AED:en 4630-45115.app 5/23/96 --55--an anther-specific and stage-specific manner during microsporogenesis.
Genomic Organization of CCaMK in Tobacco. To determine the approximate copy number of tobacco CCaMK, Southern blot analysis was carried out using a 600-bp fragment (amino acid residues 123-325) as a probe (FIG.
25). Two hybridization bands were observed in tobacco genomic DNA digested with EcoRI, EcoRV and HindIII.
Because the tobacco CCaMK cDNA sequence has one internal site for all three restriction enzymes, it is likely that tobacco CCaMK is encoded by a single copy gene.
In order to obtain a tobacco CCaMK genomic clone, a tobacco genomic library (Clontech) was screened with a probe consisting of the 5'-untranslated region of the tobacco CCaMK cDNA. A clone showing positive hybridization under stringent conditions was subcloned and sequenced.
FIG. 26 shows the nucleotide sequence of the promoter region of the tobacco CCaMK genomic sequence.
Plants Carrying CCaMK Antisense Construct Are Male Sterile. To study the function of CCaMK in vivo, sense and antisense constructs of tobacco CCaMK were fused to the CaMV 35S promoter and transgenic tobacco plants were produced. Seventeen antisense and 59 sense kanamycin-resistant transgenic plants were produced.
Four of the 17 antisense plants (A3, A4, A14, A17) showed extreme abnormality in anther development.
Control plants (transformed with vector pGA748 alone) showed normal development as compared to wild-type untransformed plants. Southern-blot analysis, using a 687-bp fragment of tobacco CCaMK (3' coding region, amino acid residues 289-517) as a probe, revealed that all of the antisense plants were transgenic. The anthers in transgenic plants looked normal until anthesis (FIGS. 27A and 27B). At stage 12 (Koltunow et al ., Plant Cell 2:1201-1224, 1990), when the anthers dehisced, wild-type anthers were fluffy with pollen, anthers in A3, A4, A14, and A17 plants were mostly bare , _ .
AED:en 4630-45115.app 5/23/96 - 56-(FIGS. 27D, 27F and 27H).
Morphological differences between transgenic and wild-type pollen grains were determined using a scanning electron microscope (FIG. 28). In general, the antisense pollen grains were much smaller and malformed as compared to wild-type pollen grains.
Histochemical localization of callose in the outer walls of pollen grains was compared between antisense and wild-type plants. The amount of callose in the outer walls of pollen grains was higher in the antisense plants. Callose distribution was uneven on the surface of the antisense pollen grains (FIG. 29H).
Bright field micrographs revealed that wild-type pollen had a very smooth ring-like wall (FIG. 29E); antisense pollen were misshapen and lacked uniformity (FIG. 29G).
A drastically reduced number of pollen grains were observed in the four antisense plants. Although remnants of pollen-like structures were observed in antisense plants (FIG. 29B and 29D), they were not viable. Germination tests demonstrated that the frequency of pollen germination of wild-type and transgenic plants transformed with vector alone was more than 95~. The frequency of germination of antisense pollen was around 1-7~ (Table 1 and FIG. 30).
Furthermore, pollen from antisense plants that did germinate showed drastically retarded pollen tube growth as compared to wild-type pollen (FIG. 30).
The four antisense plants failed to produce fruit capsules and seeds in self crosses, but when these plants were cross-pollinated with wild-type pollen, normal fruit capsules and seeds developed, indicating that these transgenic plants were male sterile. Pistils in these antisense plants recognize and transmit pollen normally.

AED:en 4630-45115.app 5/23/96 --57--Table 1. Germination of pollen grains from antisense and control plants Type of Plant Germination (~) Wild Type ~ 95.0 Control* , 95.0 A3 ~ 1.0 A4 ~ 1.5 A14 ~ 7.0 A17 ~ 5.0 *Vector DNA Alone RNA slot-blot hybridization was conducted to test the expression of CCaMK in antisense plants. RNA
from 1.0 cm bud size anthers was hybridized with a 32p_ labeled 330bp probe (5' coding region, amino acid residues 1-109). Antisense plants (A3, A4, A14, and A17) showed high levels of CCaMK antisense RNA compared to wild-type plants and plants transformed with vector alone. Anthers from 1.0 cm buds (the stage at which CCaMK had the highest expression level, FIG. 24) were collected from antisense plants as well as control plants. Endogenous CCaMK mRNA isolated from A3, A4, A14, and A17 anthers was suppressed as compared to wild-type plants (FIG. 31). RT-PCR analysis using tobacco CCaMK gene-specific primers confirmed these results.
Anther development of antisense plants was compared to about 50 other transgenic tobacco plants as well as to transgenic plants with vector DNA alone.
None of the antisense plants showed similar changes during anther development, suggesting that the observed male sterility is the result of suppression of CCaMK
mRNA and not an artifact of the point of insertion of the CCaMK transgene or tissue culture manipulation.

Discussion Although several anther-specific genes have been cloned, their role in microsporogenesis is not _ AED:en 4630-45115.app 5/23/96 -58--completely understood (Mascarenhas, Annu. Rev. Plant Physiol. Plant Mol. Biol. 41: 317-338, 1990; McCormick, Plant Cell 5:1265-1275, 1993). CCaMK is a novel Ca2+/calmodulin-dependent protein kinase expressed in an anther- and stage-specific manner during microsporogenesis.
The CCaMK gene of tobacco, a dicot, is similar in structure to the CCaMK gene of lily, a monocot, including the kinase catalytic domain, calmodulin-binding domain, and the visinin-like Ca2+-binding domain.
Tobacco and lily CCaMK share 71% identity and 82%
similarity. High homology in the visinin-like domain (79% identity, 87% similarity) indicates that the visinin-like domain is conserved and controls CCaMK
activity. The plant visinin-like domain contains three EF-hand motifs (FIGS. 22A and 22B), similar to animal visinin-like proteins such as frequenin, recoverin, and neurocalcin. These visinin-like proteins are members of Ca2+-sensitive guanylyl cyclase activators involved in cation channel regulation in neuronal tissues (Palczewski et al., Neuron 13:395-404 1994). In animals, visinin-like proteins are restricted to specialized tissues such as neurons. CCaMK, which has a visinin-like domain, is also expressed in an anther- and stage-specific manner during microsporogenesis. The CCaMK mutant lacking the visinin-like domain did not show Ca2+-dependent autophosphorylation. However, this mutant retained reduced activity as compared to the wild-type enzyme, suggesting that the visinin-like domain is crucial for maximal activation of CCaMK
(Takezawa et al., ~. Biol. Chem. 271:8126-8132, 1996).
Transient signal-induced changes in free Ca2+
concentration are known to switch on a series of biochemical changes, ultimately leading to a physiological response (Poovaiah and Reddy, CRC Crit.
Rev. Plant Sci. 12:185-211, 1993). Ca2+-induced conformational change in the visinin-like domain is believed to be critical for regulation of CCaMK

-AED:en 4630-45115.app 5/23/96 - 59--activity. Furthermore, biochemical characterization has revealed that autophosphorylation is Ca2+-dependent; a CCaMK mutant lacking this visinin-like domain did not show Ca2+-dependent autophosphorylation. In contrast, substrate phosphorylation requires both Ca2+ and calmodulin, suggesting a dual mode of regulation by Ca2+
and calmodulin.
Plants are known to have multiple isoforms of calmodulin, some of which are signal-responsive and developmentally regulated (Jena, et al., Proc. Natl.
Acad . Sci . USA 86:3644-3648, 1989; Braam and Davis, Cel l 60:357-364, 1990; Ling et al ., Plant Physiol . 96:1196-1202, 1991; Botella and Arteca, Plant Mol . Biol . 24: 757-766, 1994; Takezawa, et al., Plant Mol. Biol. 27:693-703, 1995). Plant calmodulin mRNA and protein are known to have a relatively rapid turnover rate in the cell (Perera and Zelinski, Plant Mol. Biol. 19:649-664, 1992). Signal-induced changes in the calmodulin level and a rapid turnover rate in plants suggests that there is a dynamic regulation of calmodulin in vivo. Hence, it is likely that CCaMK activity is differentially controlled by signal-induced transient changes in free Ca2+ concentration and calmodulin. The Ca2+-dependent autophoshorylation of CCaMK is suppressed by calmodulin, indicating that both the messenger (Ca2+) and the primary transducer of the Ca2+ signal (calmodulin) control the function of CCaMK, which in turn regulates the function of key anther proteins such as the 24-kDa protein (FIG.
20).
In plants, protein phosphorylation has been implicated in signal transduction (Poovaiah and Reddy, CRC Crit. Rev. Plant Sci . 12:185-211, 1993; Stone et al ., Plant Physiol . 108:451-457, 1995). Calcium controls CCaMK activity directly or indirectly through the action of calmodulin. The Ca2+ signal is amplified through Ca2+/calmodulin-dependent protein phosphorylation mediated by CCaMK. The coordinated regulation of CCaMK, its substrates, and binding proteins suggest that there AED:en 4630-45115.app 5/23/96 - 60--is a cascade of events that are switched on by changes in the Ca2+ level within the target cells. This transient change in Ca2+ and possibly calmodulin leads to the dual regulation of CCaMK either through the autophosphorylation of CCaMK or through the phosphorylation of substrate(s) in a Ca2+/calmodulin-dependent manner. Together, this coordinated regulation shows that CCaMK has a role in controlling the Ca2+-mediated signaling cascade during microsporogenesis.
The developmental events leading to pollen development and release are precisely timed and regulated. Events that occur in the tapetum profoundly affect microspore development. Cell differentiation and dehiscence events occur in an exact chronological order that correlates with floral bud size in tobacco (Koltunow et al., Plant Cell 2:1201-1224, 1990). The CCaMK gene is expressed in anther in a stage-specific manner, being detectable during meiosis, reaching highest levels following meiosis, then becoming undetectable in later stages of development (FIG. 24).
This programmed regulation, anther-specific expression, and induction of male sterility upon suppression of CCaMK message together indicate that CCaMK plays a role in microsporogenesis, affecting the deposition or degradation of callose in the outer wall of pollen.
Our attempts to suppress the CCaMK message using the CaMV 35S promoter have resulted in the production of male sterile plants. Plegt and Bino (Mol. Gen. Genet.
216:321-327, 1989) have shown that during premeiosis and meiosis, the 35S promoter is not active in the tapetum.
Because there is high endogenous GUS activity at later stages of anther development, it is uncertain whether the 35S promoter is active in the tapetum at later stages of anther development. To maximize the effect of CCaMK suppression, the CCaMK promoter or other well-known plant promoters can be used.

AED:en 4630-45115.app 5/23/96 -61--EXAMPLE 4: Trans~enic Plants As mentioned above, the tobacco CCaMK genomic clone has been obtained. FIG. 32 shows various constructs that have been introduced into tobacco to produce transgenic tobacco plants as described above.
In order to construct a transcriptional fusion, a tobacco CCaMK promoter fragment of either 1.7 kb (FIG.
32, constructs I, III, and IV) or 0.6 kb (FIG. 32, construct II) was fused to the ~-glucuronidase (GUS) reporter gene (FIG. 32, constructs I and II) or the tobacco CCaMK coding region in the antisense orientation (FIG. 32, construct IV). A perfect translational fusion was created between the tobacco CCaMK promoter (1.7 kb) and the tobacco CCaMK cDNA coding region in the sense orientation, to the 3'-end of which was fused the GUS
reporter gene. For comparison purposes, transcriptional fusions were produced between the tobacco TA29 promoter (FIG. 32, construct V) or the CaMV 35S promoter (FIG.
32, construct VI) and the tobacco CCaMK coding region in the antisense orientation. All constructs included the Agrobacterium tumefaciens nopaline synthase terminator sequence (Nos-ter). Transgenic plants including each of the constructs shown in FIG. 32 have been produced.
Having illustrated and described the principles of the present invention, it should be apparent to persons skilled in the art that the invention can be modified in arrangement and detail without departing from such principles. We claim all modifications that are within the spirit and scope of the appended claims.
All publications and published patent documents cited in this specification are incorporated herein by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

Claims (29)

1. An isolated nucleic acid comprising a sequence selected from the group consisting of:
(a) at least 15 contiguous nucleotides of a native CCaMK nucleic acid sequence that hybridizes specifically to a CCaMK sequence under stringent conditions; and (b) a polypeptide-coding sequence encoding a polypeptide having at least about 70% homology to a native lily or tobacco CCaMK polypeptide and having calcium/calmodulin-dependent protein kinase activity.

2. The nucleic acid of claim 1 comprising at least 15 contiguous nucleotides of a native CCaMK
nucleic acid sequence that hybridizes specifically to a CCaMK sequence under stringent conditions.

3. The nucleic acid of claim 1 comprising at least 15 contiguous nucleotides of a native lily or tobacco CCaMK nucleic acid sequence that hybridizes specifically to a CCaMK sequence under stringent conditions.

4. The nucleic acid of claim 2 comprising a native CCaMK promoter sequence.

5. The nucleic acid of claim 4 comprising a native tobacco promoter sequence.

6. The nucleic acid of claim 2 comprising a native CCaMK protein-coding sequence.

7. The nucleic acid of claim 6 wherein the sequence is a native lily or tobacco CCaMK gene.

8. The nucleic acid of claim 1 comprising a polypeptide-coding sequence encoding a polypeptide having at least about 70% homology to a native lily or tobacco CCaMK polypeptide and having calcium/calmodulin-dependent protein kinase activity.

9. The nucleic acid of claim 8 wherein the sequence encodes a polypeptide that comprises a native CCaMK sequence selected from the group consisting of a calmodulin-binding region, biotin-binding site, visinin-like domain, and autoinhibitory domain.

10. The nucleic acid of claim 8 encoding a polypeptide that has at least about 80% homology to a native lily or tobacco CCaMK polypeptide and has calcium/calmodulin-dependent protein kinase activity.

11. The nucleic acid of claim 10 encoding a polypeptide that has at least about 90% homology to a native lily or tobacco CCaMK polypeptide and has calcium/calmodulin-dependent protein kinase activity.

12. A nucleic acid of claim 8 that encodes a polypeptide sequence having only silent or conservative substitutions to a native CCaMK polypeptide sequence and having calcium/calmodulin-dependent protein kinase activity.

13. The nucleic acid of claim 8 encoding a polypeptide that has 100% homology to a native lily or tobacco CCaMK polypeptide.

14. A vector comprising the nucleic acid of claim 1.

15. A host cell comprising the nucleic acid of claim 1.

16. A transgenic plant comprising the nucleic acid of claim 1.

17. A male-sterile transgenic plant according to claim 16.

18. A purified polypeptide encoded by the nucleic acid of claim 1.

19. A method of producing a CCaMK polypeptide, the method comprising the steps of:
culturing the host cell of claim 12 under conditions suitable for expression of the nucleic acid to produce a CCaMK polypeptide; and purifying the CCaMK polypeptide.

20. A method of producing a male-sterile plant comprising the steps of:
introducing into a plant cell an isolated nucleic acid comprising a sequence selected from the group consisting of (a) at least 15 contiguous nucleotides of a native CCaMK nucleic acid sequence that hybridizes specifically to a CCaMK sequence under stringent conditions, or (b) a polypeptide-coding sequence encoding a polypeptide having at least about 70% homology to a native lily or tobacco CCaMK
polypeptide and having calcium/calmodulin-dependent protein kinase activity, thereby producing a transformed plant cell; and growing a male-sterile plant from the transformed plant cell.

21. A method of obtaining a CCaMK cDNA or genomic DNA of a plant comprising the steps of:
contacting cDNA or genomic DNA of the plant with a probe or primer under stringent hybridization conditions, wherein the probe comprises a detectable marker and at least 15 contiguous nucleotides of a native CCaMK nucleic acid sequence that hybridizes specifically to a CCaMK sequence under the stringent conditions to produce a hybridization product; and identifying the cDNA or genomic DNA of the plant to which the probe hybridizes.

22. A plant CCaMK gene produced by the method of claim 21.

23. A method of obtaining a CCaMK gene comprising the steps of:
contacting nucleic acids of a plant with a primer, wherein the primer comprises a detectable marker and at least 15 contiguous nucleotides of a native CCaMK
nucleic acid sequence that hybridizes specifically to a CCaMK gene under stringent conditions; and amplifying the CCaMK gene of the plant.

24. A method of producing a male-sterile plant comprising the steps of:
(a) introducing into a cell of the plant a nucleic acid comprising a sequence selected from the group consisting of: (i) at least 15 contiguous nucleotides of a native CCaMK nucleic acid sequence that hybridizes specifically to a CCaMK sequence under stringent conditions, and (ii) a polypeptide-coding sequence encoding a polypeptide having at least about 70% homology to a native lily or tobacco CCaMK
polypeptide and having calcium/calmodulin-dependent protein kinase activity, thereby producing a transformed cell; and (b) growing a plant having the altered phenotype from the transformed cell.

25. A method of expressing a nucleic acid sequence in an anther of a plant, the method comprising the steps of:
fusing a CCaMK promoter sequence to the nucleic acid sequence to produce a transgene;

introducing the transgene into a plant cell; and growing the plant cell to produce a transgenic plant having anthers.

26. An isolated plant CCaMK polypeptide.

27. A polypeptide of claim 26 selected from the group consisting of lily CCaMK and tobacco CCaMK.

28. An antibody specific for a plant CCaMK
polypeptide.

29. The antibody of claim 28 wherein the plant is lily or tobacco.