Roles of OsCKI1, a rice casein kinase I, in root development and plant hormone sensitivity

Authors

  • Wei Liu,

    1. National Laboratory of Plant Molecular Genetics, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences (SiBS), Chinese Academy of Sciences, China,
    2. Partner Group of Max-Planck-Institute of Molecular Plant Physiology (MPI-MP) on Plant Molecular Physiology and Signal Transduction, 300 Fenglin Road, 200032 Shanghai, China, and
    3. Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany, Chinese Academy of Sciences, 20 Nanxincun Road, 100093 Beijing, China
    Search for more papers by this author
  • Zhi-Hong Xu,

    1. National Laboratory of Plant Molecular Genetics, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences (SiBS), Chinese Academy of Sciences, China,
    2. Partner Group of Max-Planck-Institute of Molecular Plant Physiology (MPI-MP) on Plant Molecular Physiology and Signal Transduction, 300 Fenglin Road, 200032 Shanghai, China, and
    Search for more papers by this author
  • Da Luo,

    1. National Laboratory of Plant Molecular Genetics, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences (SiBS), Chinese Academy of Sciences, China,
    Search for more papers by this author
  • Hong-Wei Xue

    Corresponding author
    1. National Laboratory of Plant Molecular Genetics, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences (SiBS), Chinese Academy of Sciences, China,
    2. Partner Group of Max-Planck-Institute of Molecular Plant Physiology (MPI-MP) on Plant Molecular Physiology and Signal Transduction, 300 Fenglin Road, 200032 Shanghai, China, and
      For correspondence (fax +86 21 64042385; e-mail hwxue@sibs.ac.cn).
    Search for more papers by this author

For correspondence (fax +86 21 64042385; e-mail hwxue@sibs.ac.cn).

Summary

Casein kinases are critical in cell division and differentiation across species. A rice cDNA fragment encoding a putative casein kinase I (CKI) was identified via cDNA macroarray under brassinosteroid (BR) treatment, and a 1939-bp full-length cDNA, OsCKI1, was isolated and found to encode a putative 463-aa protein. RT-PCR and Northern blot analysis indicated that OsCKI1 was constitutively expressed in various rice tissues and upregulated by treatments with BR and abscisic acid (ABA). Enzymatic assay of recombinant OsCKI1 proteins expressed in Escherichia coli showed that the protein was capable of phosphorylating casein. The physiological roles of OsCKI1 were studied through antisense transgenic approaches, and homozygous transgenic plants showed abnormal root development, including fewer lateral and adventitious roots, and shortened primary roots as a result of reduced cell elongation. Treatment of wild-type plants with CKI-7, a specific inhibitor of CKI, also confirmed these functions of OsCKI1. Interestingly, in transgenic and CKI-7-treated plants, exogenously supplied IAA could restore normal root development, and measurement of free IAA content in CKI-deficient primary and adventitious roots revealed altered auxin content, indicating that OsCKI1 is involved in auxin metabolism or that it may affect auxin levels. Transgenic plants were less sensitive than control plants to ABA or BR treatment during germination, suggesting that OsCKI1 may be involved in various hormone-signaling pathways. OsCKI1-GFP fusion studies revealed the localization of OsCKI1 to the nucleus, suggesting a possible involvement in regulation of gene expression. In OsCKI1-deficient plants, differential gene expression was investigated using cDNA chip technology, and results indicated that genes related to signal transduction and hormone metabolism were indeed with altered expression.

Introduction

Casein kinase I (CKI), a serine/threonine protein kinase, is highly conserved from yeast to mammals (reviewed by Gross and Anderson, 1998). Isolation and characterization of over 20 distinct mammalian isoforms, which included splice variants (Fish et al., 1995), indicated various developmental roles for the CKI isoforms, such as vesicular trafficking (Murakami et al., 1999; Panek et al., 1997), growth and morphogenesis (human casein kinase1 gamma1, CSNK1G1; Kusuda et al., 2000; Robinson et al., 1993), circadian rhythm (Kloss et al., 1998; Kusuda et al., 2000; Peters et al., 1999; Sakanaka et al., 1999), DNA-repair pathways (Dhillon and Hoekstra, 1994), and cell cycle progression and cytokinesis (Behrend et al., 2000).

Structural analysis of the CKI proteins identified a highly conserved kinase domain responsible for substrate recognition located at the N-terminus. The C-terminus, which is large and highly diverse between animal and yeast enzymes (Graves et al., 1993), is responsible for determination of substrate specificity (Cegielska et al., 1998; Graves and Roach, 1995). In vitro, CKI has been shown to accept many proteins as substrates, including cytoskeletal proteins, enzymes involved in DNA metabolism, and other metabolic enzymes. Cytological investigations have raised the possibility that CKIdelta interacts with the trans-Golgi network and cytoplasmic granular particles associated with microtubules in interphase cells of various mammalian species. In addition, CKIdelta has been shown to play regulatory roles in mitosis through association with the mitotic spindle (Behrend et al., 2000; Brockman et al., 1992).

CKIs are widely distributed in various tissues and subcellular compartments, and they act as ubiquitous cyclic nucleotide- and calcium-independent protein kinases. They preferentially phosphorylate acidic proteins such as casein and phosvitin, are insensitive to polyanic effectors such as heparin and polyglutamate, and are inhibited by polyamines. CKIs exclusively use ATP as a phosphate donor to phosphorylate serine and threonine residues, although some isoforms from yeast and Xenopus can phosphorylate tyrosine as well (Klimczak et al., 1995). Thus, CKIs can be regarded as dual-specificity kinases at a structural level. The consensus phosphorylation sequence for CKI is defined as S/T/Y(P) X1–2S/T/Y (Flotow et al., 1990; Meggio et al., 1992).

The wide in vitro substrate specificity of CKIs makes it likely that they are involved in a variety of important biological events, although their exact in vivo function is unknown.

The phosphate-presentation requirement in CKI substrate-recognition specificity suggests that CKIs could take part in hierarchical protein phosphorylation (Flotow and Roach, 1989; Hidalgo et al., 2001; Robinson et al., 1992) and modulate signal transduction operated by second-messenger-responsive protein kinases. However, to date, there is little evidence establishing a direct link between CKI phosphorylation and physiological events.

The large number of distinct CKI isoforms and splice variants afford considerable potential for modulation of distinct cellular processes. In plants, six cDNA clones encoding CKI isoforms have been isolated from Arabidopsis thaliana (Klimczak et al., 1995 and references therein), and at least 21 putative isoform sequences exist in the A. thaliana genome. Additional CKI genes have been isolated from a variety of plant species, including Glycine max (Murray et al., 1978a), Brassica cauliflora (Murray et al., 1978b), wheat germ (Rychlik and Zagorski, 1980), Nicotiana tabacum (Erdmann et al., 1982), maize seedlings (Dobrowolska et al., 1987), Brassica oleracea (Klimczak and Cashmore, 1993), Psophocarpus tetragonolobus (Mukhopadhyay, 1997), and developing maize endosperm (Babatsikos and Yupsanis, 2000). Again, the nature of the physiological substrates and the physiological functions remain unknown.

This study investigated both the regulation and function of CKI in rice. We first cloned the OsCKI1 gene using cDNA microarray and cDNA library screening techniques. Following characterization of the gene and protein, we transformed rice with an antisense OsCKI1 construct and treated control plants with a specific inhibitor of CKI, both of which allowed physiological and phenotypic comparisons to control plants. In summary, our studies provide evidence for the involvement of OsCKI1 in the regulation of plant root development and hormone-related functions.

Results

Isolation and structural analysis of a rice cDNA, OsCKI1, encoding a putative CKI

Brassinosteroid (BR) stimulates cell elongation, and thus, the isolation of responsive genes with its treatment will provide clues for the functional mechanisms. A BR-regulated rice cDNA fragment encoding a putative CKI was identified through cDNA macroarray analysis (Chu et al., submitted). A cDNA fragment showing 2.57-fold induction under BR treatment (Figure 1a, right) was isolated; sequence analysis indicated that the encoded peptide shared high homology with the CKI isoforms. Specific primers were then designed for PCR-based cDNA library screening that yielded a 1939-bp full-length cDNA, OsCKI1, which encodes a putative 463-aa polypeptide with a calculated molecular mass of 50 kDa. OsCKI1 shared high protein sequence homology with other CKIs, specifically, identities of 73 and 70% to AtCKI1 and AtCKI2, respectively (X78818, X78819; A. thaliana), 53% to yeast CKI1 (A53581), and 67% to human Hhp2 (S46358). The theoretical pI of OsCKI1 was calculated to be 9.55, similar to that of the above-mentioned CKIs. Searching the rice genome database with the cDNA sequence identified a corresponding contig clone (7136; http://210.83.138.51/rice), which allowed identification of a genomic structure comprising 14 exons (48–569 bp) and 13 introns (74–1091 bp).

Figure 1.

Isolation and structural organization of OsCKI1.

(a) Isolation of OsCKI1 via cDNA macroarray; the arrow indicates induction of OsCKI1 following BR treatment (left: hybridized with RNA probes from untreated control plants; right: hybridized with RNA probes from 1 µm BR-treated rice seedlings).

(b) Sequence alignment of the protein kinase domain of OsCKI1 with various Arabidopsis CKIs. Protein kinase motifs are indicated by roman numerals I–XI. CKI signature sequences are underlined with dashed lines. The near-consensus SV40 T antigen putative NLS is boxed, and the kinesin homology domain is solidly underlined. Accession numbers: AtCKI1 and AtCKI2 (Arabidopsis, X78818 and X78819), CKI1 (yeast, A53581), and Hhp2 (human, S46358).

The N-terminal protein kinase domain of OsCKI1, highly similar to those of other CKIs, contains domains I through XI (Hanks et al., 1988; Klimczak et al., 1995), and four short sequences characteristic of the CKI subfamily (Graves et al., 1993; Figure 1b). The presence of a putative nuclear localization site (NLS; Rowles et al., 1991; Tuazon and Traugh, 1991; Zhai et al., 1995) indicates that OsCKI1 may localize in the nucleus, and a kinesin homologous domain (Roof et al., 1992; Xu et al., 1995), which is necessary for interactions with microtubules, was also detected in OsCKI1 (Figure 1b). The absence of transmembrane regions, as predicted by the soui program, indicates that OsCKI1 is most likely a soluble protein.

OsCKI1 is expressed in various tissues and is regulated by plant hormones BR and abscisic acid (ABA)

To analyze the expression patterns of OsCKI1, particularly under plant hormone treatments, RT-PCR and Northern blot analysis were performed. As shown in Figure 2(a), OsCKI1 transcripts were constitutively detected in seed, seedling, root, stem, leaf, leaf sheath, and panicle. Northern blot analysis revealed that OsCKI1 was induced by BR (1 µm 24-epi-brassinolide (24-eBL)) within 6 h and declined to untreated levels after prolonged treatment (12 h; Figure 2b). Addition of ABA initially suppressed OsCKI1 expression after 2 h and then induced it to reach its maximum at 8 h (Figure 2c). Auxin did not affect OsCKI1 mRNA levels (data not shown).

Figure 2.

Expression pattern analysis of OsCKI1.

(a) RT-PCR analysis identified OsCKI1 transcripts in various rice tissues; the rice actin gene was used as an internal control.

(b) Northern blotting of OsCKI1 levels following 1 µm BR treatment for 0, 2, 6, 12, and 24 h.

(c) Northern blotting of OsCKI1 levels following 100 µm ABA treatment for 0, 2, 4, 8, and 12 h.

OsCKI1 could phosphorylate partially dephosphorylated casein

To determine whether OsCKI1 encoded an active CKI, recombinant OsCKI1 was in vitro expressed in Escherichia coli and assayed for enzyme activity. OsCKI1 was successfully produced following α-isopropyl-D-thio-galactoside (IPTG) induction, as shown by SDS–PAGE analysis (Figure 3a). The recombinant OsCKI1 had a molecular mass of approximately 64 kDa, which is in good accordance with the calculated molecular mass plus the 12.1-kDa fusion tag. The recombinant protein was able to phosphorylate a CKI-specific substrate (partially phosphorylated casein) with fivefold higher enzymatic activity than the control that harbored empty vector (Figure 3b). The phosphorylation activity of OsCKI1 was inhibited by the CKI-specific inhibitor CKI-7 (data not shown), providing further evidence that OsCKI1 is a functional CKI.

Figure 3.

In vitro expression of OsCKI1 and activity assay.

(a) Crude extracts of recombinant OsCKI1 induced with 1 mm IPTG were resolved on a 10% SDS–PAGE gel. Equal amounts of total protein were loaded, representing induction for 0, 1, 2, and 4 h. The arrow indicates the position of recombinant OsCKI1.

(b) Recombinant OsCKI1 shows CKI activity against a partially phosphorylated casein substrate. Equal amounts of crude extract were used in the measurements, and the experiment was performed three times on independent preparations of recombinant protein. The data presented are the mean ± SD.

Transgenic rice plants harboring antisense OsCKI1 showed abnormal root development because of reduced cell elongation

The binary vector p35S-1301-ACKI1 was constructed and used for rice transformation. Thirteen independent resistant T0 plants were regenerated after transformation, 10 of which had integrated a single copy of the T-DNA. Harvested T1 seeds were grown on a hygromycin-supplemented medium to test for resistance, i.e. single-copy integration of T-DNA, and histochemical analysis of GUS activity was performed to confirm T-DNA integration. As shown in Figure 4(a), PCR amplification using primers from the CaMV 35S promoter and OsCKI1 gene confirmed the integration of T-DNA into the rice genome (Figure 4a-1). RT-PCR analysis indicated the presence of antisense OsCKI1 transcripts (Figure 4a-2), which suppressed OsCKI1 mRNA levels (Figure 4a-3). Detailed analysis of mRNA transcripts in the transgenic plants revealed that OsCKI1 mRNA levels were suppressed to varying degrees (15–94%), which were in good agreement with the antisense OsCKI1 transcript levels in these transgenic lines. Further screening of T1 and T2 seeds identified five independent homozygous lines that were used for further study.

Figure 4.

Figure 4.

Abnormal root development in OsCKI1-deficient plants.

(a) RT-PCR analysis of transgenic rice plants. From top to bottom: 1, primers for the CaMV 35S promoter and OsCKI1 gene were used to test the integration of T-DNA in resistant plants (templates P.K.: transformed vector, C.K.: genomic DNA of untransformed plants, and L1, L2, L5: genomic DNA from different resistant lines); 2, primers OsCKI1-1 and OsCKI1-2 were used to test the amounts of antisense OsCKI1 transcripts; 3, primers OsCKI1-1 and OsCKI1-RT were used to test the transcripts of OsCKI1 in whole plants; the bottom line shows OsCKI1 mRNA levels in transgenic plants compared to control plants. Total RNA was isolated from leaf tissue of both resistant and untransformed plants; 4, RT-PCR was performed for 36 cycles with rice actin amplified as an internal control.

(b) Effects of CKI deficiency in decreasing primary root length and reducing numbers of lateral and adventitious roots of rice shoots at 7 days after germination (from left to right: control plants, transgenic plant from L5, control plants treated with CKI-7 (45 and 10 µm)).

(c) Detailed analysis of abnormal root development, including decreased length of primary roots, reduced numbers of adventitious roots, and reduced numbers of lateral roots following treatment with various concentrations of CKI-7 (0–100 µm). Exogenous supplemental IAA (0.1 µm) could partially restore the normal phenotypes. Measurements were performed at 7 days after germination, and all experiments were performed in triplicate. The data presented are the mean ± SD. C.K.: wild-type plants.

Figure 4.

Figure 4.

Abnormal root development in OsCKI1-deficient plants.

(a) RT-PCR analysis of transgenic rice plants. From top to bottom: 1, primers for the CaMV 35S promoter and OsCKI1 gene were used to test the integration of T-DNA in resistant plants (templates P.K.: transformed vector, C.K.: genomic DNA of untransformed plants, and L1, L2, L5: genomic DNA from different resistant lines); 2, primers OsCKI1-1 and OsCKI1-2 were used to test the amounts of antisense OsCKI1 transcripts; 3, primers OsCKI1-1 and OsCKI1-RT were used to test the transcripts of OsCKI1 in whole plants; the bottom line shows OsCKI1 mRNA levels in transgenic plants compared to control plants. Total RNA was isolated from leaf tissue of both resistant and untransformed plants; 4, RT-PCR was performed for 36 cycles with rice actin amplified as an internal control.

(b) Effects of CKI deficiency in decreasing primary root length and reducing numbers of lateral and adventitious roots of rice shoots at 7 days after germination (from left to right: control plants, transgenic plant from L5, control plants treated with CKI-7 (45 and 10 µm)).

(c) Detailed analysis of abnormal root development, including decreased length of primary roots, reduced numbers of adventitious roots, and reduced numbers of lateral roots following treatment with various concentrations of CKI-7 (0–100 µm). Exogenous supplemental IAA (0.1 µm) could partially restore the normal phenotypes. Measurements were performed at 7 days after germination, and all experiments were performed in triplicate. The data presented are the mean ± SD. C.K.: wild-type plants.

As shown in Figure 4(b) and Table 1, the primary roots of transgenic plants were 64.7–72.9% shorter than those of control plants. In addition, transgenic plants produced 60.6–67.4% fewer adventitious roots and 93% fewer lateral roots than that produced by control plants. The observed phenotypes of shortened primary roots (L1, L2, and L5 being 35, 24, and 27% of control, respectively) and fewer adventitious roots (L1, L2, and L5 being 63, 45, and 19.6% of control, respectively) showed that L2 and L5 had more severe phenotypes than L1, which corresponded to higher OsCKI1-suppression ratios in L2 and L5 than in L1, although there was no linear correlation. We therefore propose that primary root lengths and adventitious root numbers reflect OsCKI1 mRNA levels, and that suppressed transcription of OsCKI1 resulted in abnormal rice root development.

Table 1.  Length of primary roots and numbers of lateral and adventitious roots in control and transgenic plants harboring antisense OsCKI1
 Length of primary roots (cm)Number of adventitious rootsNumber of lateral roots
7 days14 days7 days14 days7 days
  1. Measurements were performed 7 or 14 days after germination. The numbers represent the average of 15 plants tested in triplicate.

C.K.6 ± 0.268.13 ± 2.045.48 ± 0.256.2 ± 1.8946 ± 4.97
L12.1 ± 0.265.73 ± 0.873.47 ± 1.024.7 ± 0.445.2 ± 3.0
L21.44 ± 0.276.69 ± 1.532.47 ± 1.152.33 ± 0.475.4 ± 3.58
L51.64 ± 0.344.12 ± 1.781.078 ± 0.273.6 ± 0.863.2 ± 1.24

Abnormal root development associated with OsCKI1 deficiency was further investigated by treating control plants with CKI-7, a specific inhibitor of CKI (Chijiwa et al., 1989). As shown in Figure 4(c), non-transgenic plants treated with different concentrations of CKI-7 showed phenotypes similar to those observed in the transgenic plants. At media concentrations of 10, 45, and 100 µm, CKI-7 plants showed 20.8, 35.1, and 45% inhibition of primary root length, 14.6, 28.4, and 52.2% fewer adventitious roots, and 32.6, 59.1, and 88% fewer lateral roots, respectively. This dose–response relationship between CKI-7 concentration and root abnormality provides further evidence that the transgenic phenotypes are correlated with relative suppression of OsCKI1.

In light of the effects of OsCKI1, we next investigated whether the short primary root phenotype was a result of altered cell division, decreased elongation, or both. The cell morphology of longitudinal sections from various maturation, elongation, and meristem zones were examined. As shown in Figure 5(a), the cells at the maturation zone of control plants were severely disrupted to form spaces between the epidermis and pericycle, whereas those of transgenic plants kept their cell shape and displayed few spaces, suggesting that the development of transgenic plants may be delayed. In the elongation zone, similar longitudinal arrangements were observed in both control and transgenic roots, but control plants possessed more lateral root primordia than did transgenic plants. In the meristem zone, organized cell arrangements were present in transgenic roots, but the cells were much shorter than those of control roots. These observations suggest that the shortening observed in transgenic primary roots was because of reduced cell elongation rather than changes in cell division. Quantitative analysis revealed approximately 500 cells in the examined portions of three zones of the root of control plants as compared to approximately 700 cells in the corresponding sections of transgenic plants, indicating smaller cell sizes in general. Light microscopy also confirmed that transgenic root cells were shorter than those of control plants (Figure 5b), which again indicated that the abnormal root development under OsCKI1 deficiency was a result of reduced cell elongation.

Figure 5.

Longitudinal sections of primary root showing reduced cell elongation but no change in cell division.

(a) Cell shapes and analysis of cell numbers in 1-week-old control and transgenic (L5) plants at maturation, elongation, and meristem zones of the primary roots. Squares indicate the space between the epidermis and pericycle at maturation zones of the primary roots of control plants, where the cell shapes are kept in transgenic plants. Cell numbers are the average of three to five independent roots (in a length of 0.5 mm). Bars = 100 µm.

(b) Cell shape in 1-week-old control and transgenic (L5) plants under light and fluorescence microscopy with different magnifications. Bars = 30 µm.

Supplementation of media with IAA rescues abnormal root development

Addition of exogenous IAA was able to partially rescue the abnormal development of transgenic primary, lateral, and adventitious roots. As shown in Figure 4(c), supplementation of culture media with 0.1 µm IAA increased the length of primary roots and numbers of adventitious and lateral roots from 27.1, 32.6, and 7% to 61.2, 52.6, and 39.5% of control plants, respectively. Examination of endogenous IAA levels in both primary and adventitious roots of transgenic plants, control plants, and CKI-7-treated plants showed that the levels of free IAA were significantly altered, i.e. decreased in adventitious roots but increased in primary roots in the downregulated transgenic lines and the CKI-7-treated plants (Table 2).

Table 2.  Auxin content in primary and adventitious roots of control plants, OsCKI1 transgenic plants, and control plants following treatment with CKI-7 (100 µm)
 Primary rootsAdventitious roots
  1. Each number indicates the auxin content (pmol g−1 FW) calculated as the average of two replicates each of 10 samples.

C.K.632529181
L15697818644
L22444917538
L55650519671
C.K. + CKI-7572244083

OsCKI1-deficient rice plants show reduced sensitivity to BR and ABA

As both BR and ABA regulated OsCKI1 expression, we next measured their effects on the growth and development of transgenic plants. Transgenic seedlings were less sensitive to 24-eBL (1 µm) than were control plants. Control plants grew normally in the absence of exogenously supplied 24-eBL, with coleoptile elongation stopping at an early stage of germination and the foliage leaves elongating to emerge from the coleoptile. When germinated in the presence of 24-eBL, the coleoptiles elongated in an abnormal pattern, resulting in a twisted shape, and the foliage leaves grew poorly and did not break through the coleoptile. Root elongation was inhibited, and the root grew in a wavy form (Figure 6a,b). In contrast, the seedlings of transgenic plants showed normal growth of foliage leaves and coleoptile elongation, but abnormal, wavy root growth (Figure 6c,d). Inhibition of root elongation by 24-eBL was moderate: 37% in transgenic plants compared to 49% in control plants (Figure 6e). Also, in transgenic plants, the angle between the leaf blade and sheath of seedlings, a measure that is used in the quantitative bioassay for BRs (Wada et al., 1981), was smaller than that of control plants (data not shown). Taken together, these results show that transgenic seedlings were less sensitive to exogenous BR.

Figure 6.

Transgenic plants show insensitivity to BR.

(a–d) Rice control (a, b) and transgenic plants (c, d) germinated on medium supplemented with (a, c) or without (b, d) 1 µm 24-eBL for 7 days. Control plants showed typical leaf sheath phenotypes when treated with 24-eBL (b), while transgenic plants (L2) were insensitive to the treatment (d).

(e) Transgenic plants (L2 and L5) showed less inhibition of root growth following treatment with 1 µm 24-eBL compared to control. Black bars show the root length of plants without 24-eBL treatment; white bars show that with 24-eBL treatment. Measurements were performed in triplicate and the data presented are the mean ± SD. C.K.: wild-type plants.

We also examined germination of ABA-treated transgenic seeds. As shown in Figure 7, the germination frequencies of transgenic seeds after 4, 7, and 9 days of ABA treatment (2, 4, 6, and 8 µm) were much higher than those of control plants. Similarly, inhibition of root growth was limited in transgenic plants, suggesting a lower sensitivity to ABA (data not shown).

Figure 7.

Insensitivity of transgenic plants to ABA.

Both control and transgenic (L5) plants were germinated on media supplemented with different concentrations of ABA; germination frequencies were calculated 4 (a), 7 (b), and 9 (c) days after sowing. Measurements were performed in triplicate and the data presented are the mean ± SD. C.K.: wild-type plants.

OsCKI1 deficiency results in altered expression of a variety of genes

Genes with altered transcription levels in OsCKI1-deficient plants were studied using cDNA chip technology. Approximately 300 differentially expressed clones were detected and analyzed by blast searches against the rice, Arabidopsis, and other databases. Thirty-two clones were not associated with known genes. The others were identified and assigned to different families according to their interpreted functions in metabolism, signal transduction, transcriptional regulation, and resistance (Table 3).

Table 3.  Differential expression of genes in OsCKI1-deficient plants revealed by hybridization to rice cDNA chips
 Genes under OsCKI1 deficiencyRatios
  1. The numbers indicate the ratio of Cy5/Cy3, i.e. the change in expression, and are the average of two independent hybridizations.

Upregulated
 Signal transductionMAP kinase homolog (MAPK2)2.414
Putative calcium-binding EF-hand protein2.231
Putative casein kinase3.492
Putative glycerophosphodiester phosphodiesterase2.561
Putative glycosylphosphatidylinositol-anchored protein (COB)3.697
MEK kinase (MAP3Ka)3.854
S-Adenosyl-l-methionine: JA carboxyl methyltransferase (JMT)4.737
Phosphatidic acid phosphatase5.427
 DevelopmentA. thaliana DNA for germin-like protein 2 precursor3.962
Root hair defective 3 (RHD3)2.228
Ethylene-responsive factor (OSERS)4.516
 ResistanceWound-induced protein homolog2.538
Heat shock protein 70 (Hsc70–5), 81–2 (HSP81-2), 82 (HSP82)4.662
Putative proline-rich protein4.860
Betaine aldehyde dehydrogenase2.500
 TranscriptionZinc-finger protein Lsd1 (LSD1)2.802
Putative DNA-binding protein4.080
Putative bHLH transcription factor3.919
KNAT1 knotted-like homeobox protein6.562
RING-H2 finger protein RHF2a7.473
 CytoskeletonProfilin 2 and 33.536
Actin-24.100
 MetabolismPutative protein synthesis initiation factor2.162
Sucrose transporter2.337
Pyruvate kinase4.478
Downregulated
 ResistanceHeat shock factor 50.345
Putative disease resistance protein0.448
Viral resistance protein (HRT)0.394
 Signal transductionReceptor-like protein kinase0.445
Kinase-like protein0.372
Putative protein kinase0.341
Calmodulin-domain protein kinase CDPK isoform 6 (CPK6)0.377
Putative casein kinase I0.443
Developmentally regulated GTP-binding protein (AtDRG1)0.452
Protein tyrosine phosphatase 1 (PTP1)0.497
 TranscriptionZinc finger and C2 domain protein0.403
Putative MYB47 transcription factor0.449
E2F transcription factor-1 E2F10.42
A. thaliana MYB37 homolog0.392
 CytoskeletonAtnack1 kinesin-like protein0.416
 IAA degradationPutative protease0.451
 MetabolismPutative sucrose synthase0.237
Glutathione S-transferase (GST)0.225
Cellulose synthase mRNA0.265
Catalytic subunit of cellulose synthase (Ath-A)0.397
Catalytic subunit of cellulose synthase (Ath-B)0.433

Discussion

Isolation of a cDNA encoding CKI, which is constitutively expressed in rice tissues

A rice cDNA, OsCKI1, was isolated through cDNA macroarray followed by library screening. Comparative analysis revealed that OsCKI1 had high identity with a variety of plant CKIs, contained conserved kinase domains, and possessed a high pI value, which is characteristic of all CKI homologs (pI 9.0–9.6) isolated to date (reviewed by Tuazon and Traugh, 1991). Recombinant OsCKI1 showed phosphorylation activity against a casein substrate, indicating that it is a functional protein kinase. Interestingly, OsCKI1 was constitutively expressed in most tissues, in contrast with the majority of kinases, which exhibit specific tissue expression patterns (Gross and Anderson, 1998). This, however, suggests that OsCKI1 plays general roles during plant growth and development.

OsCKI1 regulates rice root development

As a multifunctional protein kinase in animal cells, CKI is involved in various developmental events. Through hierarchical protein phosphorylation, CKI plays important roles in cellular functions including determination of cell state and growth rate, metabolism, and differentiation (Hunter, 1995; Krebs and Beavo, 1979; Stone and Walker, 1995; Trewavas and Gilroy, 1991). However, few studies have been undertaken to elucidate the physiological functions of CKI in plant cells. Here, using antisense and CKI-specific inhibitor strategies in rice, we showed that CKI deficiency results in shorter primary roots and fewer lateral and adventitious roots. This indicates a critical role for CKI in rice root development, although it is possible that CKI-7 additionally inhibited other homologous proteins to exert its effects. The reduced number of lateral root primordia in transgenics (Figure 5a) correlates to reduced numbers of lateral roots later in development, suggesting the reduced ability of initiation, which may be a result of less amounts of the active auxin at the region. That cells in the elongation zone are less active in cell division and differentiation in a CKI-deficient environment further indicated that CKI may interact with other factors to regulate the initiation of lateral roots.

Previous studies have shown that root length is decided by both cell division and ensuing cell elongation in the elongation zone (Hoshikawa, 1989). Our observation of the various root zones revealed similar arrangements of cells in these zones, but reduced cell sizes in the transgenic or inhibitor-treated roots, suggesting that reduced root length in CKI-deficient plants is because of reduced cell elongation and not because of reduced cell division.

Involvement of OsCKI1 in plant-hormone functions

Plant hormones such as auxins, ethylene, ABA, and BR affect root development in different ways. Auxins affect the elongation of primary roots and the initiation and elongation of lateral roots (Kim et al., 2001), and ABA and ethylene inhibit root elongation (Whalen and Feldman, 1988), while BR interacts with auxins to inhibit root elongation (Ephritikhine et al., 1999; Kim et al., 2000). The partial restoration of normal root development following supplementation of CKI-deficient plants with exogenous IAA suggests that in the context of root development, CKI may lie upstream of auxin-related signaling pathways. Indeed, endogenous auxin contents of primary and adventitious roots were altered in CKI-deficient plants, although there was no simple correlation between phenotype and auxin content. This may suggest that the effects of OsCKI1 on root development and auxin metabolism are not directly linked, although restoration of root development by exogenous IAA suggests that they are related. Further studies will be required, especially regarding possible interactions with cytokinin, which affect the initiation of adventitious roots.

OsCKI1 was upregulated by plant hormones ABA and BR, and CKI-deficiency in rice caused insensitivity to both BR and ABA, suggesting that CKI may be involved in the root development signaling pathways that are regulated by these hormones. It is possible that proteins or enzymes that mediate the function of ABA and BR may serve as substrates for CKI, allowing integration of the various hormone responses.

Downstream genes regulated by OsCKI1

Genes that were differentially expressed in OsCKI1-deficient rice were examined using cDNA chips, and the results identified changes in a variety of gene families including those associated with plant signal transduction, transcriptional regulation, and hormone metabolism (Table 3). Members of all these families are potentially involved in the regulation of root development through direct or indirect interactions, such as profilin 2 and 3, phosphatidic acid phosphatase and so on. Suppression of protease-encoding genes, which are involved in IAA degradation, fits the observation of increased IAA content in primary roots. Induction or suppression of genes encoding transcription factors and proteins related to seed development and hormone functions may influence the cross-talk between different signaling and metabolic pathways. Upregulation of genes related to resistance suggests that OsCKI1 may mediate the plant's responses to biotic and abiotic environmental factors. In addition, OsCKI1 may influence many aspects of metabolism, as expression of key genes of important metabolic pathways were altered in OsCKI1-deficient plants. Interestingly, OsCKI1 contains an NLS, and the recombinant OsCKI1 protein localizes to the nucleus (Figure 8), perhaps suggesting OsCKI1 involvement in the regulation of gene transcription. Basing on these hypotheses, we are currently testing the interactions between OsCKI1 and relevant signaling pathways.

Figure 8.

OsCKI1 localizes to the nucleus.

OsCKI1-GFP fusion studies via transient transformation of onion epidermis cells revealed the localization of OsCKI1 in nucleus (a). Empty vector (pCambia1302) was used as control (b). Bar = 50 µm (a) and 20 µm (b).

In summary, we have identified and cloned OsCKI1, which encodes a rice CKI that is critical in root development, especially root initiation and cell elongation. It seems likely that in this context, CKI functions, at least in part, through regulation of auxin metabolism, or alternatively, by mediating the interplay of auxin, BR, and ABA signaling through hierarchical phosphorylation.

Experimental procedures

Enzymes and chemicals

Enzymes used for DNA restriction and modifications were from Roche (Boehringer Mannheim, Germany). DNA primers for PCR were from GENECORE Biotech (Shanghai, China). [alpha-32P]dCTP and [gamma-32P]ATP were obtained from Yahui Company (Beijing, China). CKI-7 (product code no. 120837) was from Seikagaku Corp. (Tokyo, Japan) 24-eBL was obtained from Sigma Company (product code E-1641). All other reagents used were standard analytical or electrophoresis grade.

Bacteria and plants

Escherichia coli strain XL-1 Blue (Stratagene, La Jolla, CA, USA) was used for DNA cloning and cDNA library screening. Rice seeds (Oryza sativa L. cv. Zhonghua 11) were surface-sterilized, germinated on agar-solid 0.5× MS medium, and then grown in pots in a phytotron with a 12-h light (26°C) and 12-h dark (18°C) cycle. For hormone-based experiments, 1-week-old water-grown plants were treated with plant hormones at a final water concentration of 100 µm (IAA and ABA, treated for 0, 2, 4, 8, or 12 h), and 1 µm 24-eBL (0, 2, 6, 12, or 24 h). Treated rice materials were harvested and used for RNA extraction.

Isolation of the OsCKI1 cDNA

A rice cDNA clone encoding a putative CKI was identified through cDNA macroarray approaches using 24-eBL (the most active BR)-treated rice RNA as hybridization probes. Two-week-old rice shoots (grown in water in a phytotron) were treated with 1 µm 24-eBL for 4, 6, 12, and 24 h, respectively, before being prepared for macroarray probing.

Primers OsCKI1-1 (5′-TACGGAGAGCACACAAAGCACAG-3′) and OsCKI1-2 (5′-TCCTCAAAGTGGGATCATCATGG-3′) were used to isolate the positive clones through PCR-based cDNA library screening (Alfandari and Darribere, 1994). Positive clones were converted into pBluescript derivatives using helper phage ExAssist according to the supplier's (Stratagene) instructions. The clone with the longest insert (designated pOsCKI1) was used for further analysis.

DNA sequencing was performed by GENECORE Biotech (Shanghai, China). Computational analysis, including sequence comparisons with DNA and amino acid sequences in GenBank™, EMBL, dbEST, and SwissProt databases with ‘fasta’ and ‘blast’ search programs, sequence alignments with ‘pileup’, and exon/intron structure analysis with ‘bestfit’ were all performed using the gcg seqweb program (Version 2.0.2; Accelrys, Inc., Burlington, MA, USA). Protein domain analysis was performed with the soui program (http://www.expasy.org/soui).

RT-PCR and Northern blot analysis

Northern blot and RT-PCR analyses were employed to study OsCKI1 transcription levels in various rice tissues or leaves following different treatments with hormones and chemicals. Total RNA was isolated from seeds, seedlings, roots, stems, leaves, leaf sheaths, and immature spikes using the Trizol reagent (Huashun Company, Shanghai, China), and reverse transcribed according to the manufacturer's instructions (SuperScript Pre-amplification System, Promega, Madison, WI, USA). Reverse transcription was performed in a total volume of 40 µl using 4 µg total RNA as template, and incubated at 42°C for 60 min. RT-PCR was performed using primers OsCKI1-1 and OsCKI1-2, and amplified DNA products were collected and analyzed following different numbers of (30, 35, and 40) amplification cycles. The rice actin gene was amplified using primers OsActin1 (5′-GAACTGGTATGGTCAAGGCTG-3′) and OsActin2 (5′-ACACGGAGCTCGTTGTAGAAG-3′) to quantify the relative amounts of cDNA. Identical quantities of reverse-transcribed first-strand cDNAs were used as templates for the test PCR.

For Northern blot analysis, total RNA isolated from 2-week-old rice shoots treated with IAA, ABA, or 24-eBL was quantified at 260 nm. Denatured RNA (30 µg) was loaded onto a 1.5% agarose-formaldehyde gel (Logemann et al., 1987) and transferred to Hybond-N+ membranes (Amersham Pharmacia Biotech, Piscataway, NJ, USA) after electrophoresis. RNA was fixed to the membrane by incubation at 80°C for 2 h. A 1546-bp DNA fragment (corresponding to the SalI to HhaI fragment) of the OsCKI1 cDNA was used as a [α-32P]dCTP-labeled hybridization probe. Membranes were hybridized and washed at 65°C according to Xue et al. (1999). Autoradiographs were exposed for 2–3 days at −70°C with intensifying screens.

In vitro expression of recombinant OsCKI1 and enzymatic activity measurement

To study the biochemical characteristics of OsCKI1 in vitro, we expressed recombinant OsCKI1 in E. coli. The OsCKI1-coding region, amplified via PCR using primers CKI1-3 (5′-CATGCCATGGCTATCGAATTCCCGGGGAAG-3′, with added NcoI site underlined) and CKI1-4 (5′-CGGGATCCTTATTTCCTTCTGTCAGCACT-3′, with added BamHI site underlined), was subcloned into a pET-32a (Novagen, Madison, WI, USA) vector pre-cut with NcoI and BamHI. The resulting construct, pET-OsCKI1, was confirmed by direct sequencing, and was transformed into fresh colonies of E. coli strain BL21(DE3). Transformed E. coli cells were grown at 37°C in liquid LB medium containing the appropriate antibiotics to a density of OD600 of 0.3, and then induced with IPTG at a final concentration of 1 mm for 4 h. Cells were collected by centrifugation and washed once with ice-cold phosphate-buffered saline (PBS). After lysis by sonication, the lysate was centrifuged and the supernatant was used for SDS–PAGE analysis and enzymatic activity assay. Proteins were separated on 10% gels and stained with Coomassie Blue 250. Protein concentrations were determined by the method of Bradford (1976) using bovine serum albumin (BSA) as standard.

The activity assay was carried out according to the method of Klimczak and Cashmore (1993) with a few modifications. The assay was initiated by adding 6 µl of crude extract (5 µg protein) containing recombinant OsCKI1 in a total volume of 40 µl containing 50 mm Tris–HCl (pH 8.0), 10 mm MgCl2, 10 mm DTT, 10 mm ATP, 1 µCi of [gamma-32P]ATP, and 6 µg of partially de-phosphorylated casein (Sigma, C4032). Reactions were incubated at 37°C for 20 min and terminated by icing for 2–3 min. The samples were spotted on 2 cm × 2 cm squares of Whatmann 3 mm filter paper, washed three times for 10 min each in 10% trichloroacetic acid/1% sodium pyrophosphate, rinsed in 95% ethanol, and dried. Radioactivity was determined by scintillation counting (LS650; Beckman, USA). One unit of activity was defined as the amount of enzyme able to incorporate 1 pmol of phosphate into casein per 1 min. Identical quantities of crude extract from cells transformed with empty vector were used as controls.

Transgenic modification of OsCKI1 expression

An OsCKI1 cDNA fragment with SalI and blunted HhaI sites was subcloned into the p35S-1301 (modified from pCambia1301 by insertion of a CaMV 35S promoter and Nos terminator into the MCS) vector pre-cut with SalI and SmaI. The resulting fusion construct, p35S-1301-ACKI1, which contained antisense OsCKI1, was transformed into Agrobacterium tumefaciens strain EHA105 by electroporation, and positive clones were used for transformation of immature rice embryos. Resistant plants were confirmed as transgenic using the following strategies: (i) genomic DNA was extracted with the cetyltrimethyl ammonium bromide (CTAB) method (Sambrook et al., 1989) and used as template for PCR with primers from the transformed construct, i.e. a 35S sense primer (5′-ATGCCATCATTGCGATAAAGG-3′) and an antisense primer of OsCKI1-1 (5′-TACGGAGAGCACACAAAGCACAG-3′; Figure 4a); (ii) GUS activity was histochemically detected in the leaves of resistant plants according to the method described by Jefferson et al. (1987); (iii) Total RNA was isolated from leaf tissue of both resistant and untransformed plants, and RT-PCR was used to determine whether OsCKI1 expression had been altered. Primers OsCKI1-1 and OsCKI1-2 were used to test for transcripts of integrated antisense OsCKI1, and OsCKI1-1 and OsCKI1-RT (5′-GAGTATAGTGGCAGAGTGGC-3′) were used to test the OsCKI1 transcripts of whole plants, respectively.

T1 seeds from primary transformants were germinated in rice culture medium containing 40 µg ml−1 of hygromycin for selection of transgenic progeny. T2 seeds from individual T1 plants were germinated in the same medium to identify homozygous T2 lines, which were then used to generate the transgenic plants used for further analysis.

Phenotype observation of OsCKI1-deficient transgenic plants

The lengths of primary roots, and numbers of adventitious and lateral roots of homozygous transgenic plants were calculated at 7 days after germination from an average of at least 30 seeds. All experiments were performed in triplicate. The data presented is the mean ± SD.

To test the effects of CKI-7, a specific inhibitor of CKI, 10, 45, and 100 µm of CKI-7 was supplemented into the culture medium. To test the effects of exogenous auxin, IAA was added to the medium with a final concentration of 0.1 µm. Measurements of root phenotypes were performed as described above.

Transgenic growth characteristics and response to plant hormones

Seeds of wild-type and transgenic plants were germinated on 0.5× MS medium with or without 1 µm 24-eBL for 1 week. The length of primary roots, and the length and number of lateral and adventitious roots were determined. Seed germination frequencies were determined in the culture medium supplemented with or without ABA (final concentrations of 2, 4, 6, and 8 µm) on 4, 7, and 9 days. All measurements were performed in triplicate, each taking the average measurements of at least 15 seeds.

Preparation of root sections for microscopic observation

Seeds of wild-type and transgenic plants were germinated for 1 week, and 0.5 mm sections of the root tip (meristem zone), middle part (elongation zone), and upper part (maturation zone) were sampled, fixed in 1% amylaldehyde, and dehydrated in a graded ethanol series. The samples were embedded in Epon812 resin, polymerized at 60°C, cut into 3-µm-thick sections, and then stained with toluidine blue. Sections were microscopically examined (Nikon, 40×, Tokyo, Japan) and photographed.

Measurement of auxin content in root tissues

To measure the auxin content in root tissues, seeds of wild-type and transgenic plants were germinated in culture medium for 1 week, and the primary roots (cut into upper and lower regions) and adventitious roots were sampled. Similar amounts of root samples (100–400 mg FW) were used for measurement of free IAA content by ELISA as described elsewhere (Liang and Yin, 1994).

Analysis of genes with altered expression in OsCKI1-deficient plants

To study possible functional mechanisms of OsCKI1, we analyzed gene expression in OsCKI1-deficient plants. Analyses were performed on a P100S cDNA chip (Biostar Genechip, Inc., Shanghai, China), which contained around 10 000 independent rice clones. Root samples from 7-day-old shoots of control and transgenic plants were used for RNA extraction and RT preparation of fluorescent cDNA probes labeled with Cy3- or Cy5-deoxy CTP (Amersham Pharmacia Biotech, Piscataway, NJ, USA), respectively. Quantified probes were applied onto pre-hybridized P100S chips under a cover glass. After hybridization, the chips were scanned with a ScanArray 4000 (GSI Lumonics, Billerica, MA, USA) at two wavelengths (532 and 653 nm for Cy3 and Cy5, respectively) to detect emission from both Cy3 and Cy5. The resulting images were analyzed using genepix pro 3 software (Axon Instruments, Inc., USA). Overall intensities were normalized using the corresponding genepix default normalization factor. Regulated genes were identified by ratios more than 2.0 or less than 0.5, which indicates at least onefold up- or downregulation. Experiments were performed two times, and clones that were differentially expressed in both hybridizations were selected for further analysis.

Localization of OsCKI1-GFP fusion proteins

The whole coding sequence of OsCKI1 was amplified using primers OsCKI5 (5′-CATGCCATGGTAATGAAGATTGGGAGCGGA-3′, with an added NcoI site underlined, ‘ATG’ italicized) and OsCKI6 (5′-GACTAGTTCCTTCTGTCAGCACTGATTG-3′, with an added SpeI site underlined). The resulting PCR products were subcloned into the pCambia1302 vector to generate p1302-OsCKI1-gfp containing an OsCKI1/green fluorescent protein (GFP) fusion construct under the control of the CaMV 35S promoter. The resulting construct was confirmed by direct sequencing and used for transient transformation of onion epidermis via a gene gun (Bio-Rad, USA). Transformed onion cells were observed under a confocal microscope.

Acknowledgements

This study was supported by grants from the National Natural Science Foundation of China (Grant no. 30100101) and the State Key Project of Basic Research (Grant no. G1999011604). We would like to thank Dr Zhao-Qing Chu for assistance with the OsCKI1-GFP work and Dr Charles Brearley and Prof. Tian-Duo Wang for critical comments on the manuscript.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank™/EMBL Data Bank with accession number AJ487966.

Ancillary