Roles of Arabidopsis cyclin-dependent kinase C complexes in cauliflower mosaic virus infection, plant growth, and development

Abstract

The C-terminal domain (CTD) of RNA polymerase II is phosphorylated during the transcription cycle by three cyclin-dependent kinases (CDKs): CDK7, CDK8, and CDK9. CDK9 and its interacting cyclin T partners belong to the positive transcription elongation factor b (P-TEFb) complexes, which phosphorylate the CTD to promote transcription elongation. We report that Arabidopsis thaliana CDK9-like proteins, CDKC;1 and CDKC;2, and their interacting cyclin T partners, CYCT1;4 and CYCT1;5, play important roles in infection with Cauliflower mosaic virus (CaMV). cdkc;2 and cyct1;5 knockout mutants are highly resistant and cdkc;2 cyct1;5 double mutants are extremely resistant to CaMV. The mutants respond normally to other types of plant viruses that do not replicate by reverse transcription. Expression of a reporter gene driven by the CaMV 35S promoter is markedly reduced in the cdkc;2 and cyct1;5 mutants, indicating that the kinase complexes are important for transcription from the viral promoter. Loss of function of CDKC;1/CDKC;2 or CYCT1;4/CYCT1;5 results in complete resistance to CaMV as well as altered leaf and flower growth, trichome development, and delayed flowering. These results establish Arabidopsis CDKC kinase complexes as important host targets of CaMV for transcriptional activation of viral genes and critical regulators of plant growth and development.

Figure 1.

Enhanced Resistance of cdkc;2 and cyct1;5 Mutants to CaMV. (A) Disease symptom development in mechanically inoculated plants. Forty wild-type, cdkc;2-1, and cyct1;5-1 single or double mutant plants were mechanically inoculated with partially purified CaMV virons (0.5 mg protein/mL), and plants with CaMV mosaic symptoms were scored at the indicated DAI. (B) Disease symptom development in plants inoculated by bombardment of the infectious CaMV DNA clone pCa122. Photographs of representative plants were taken at the indicated DAI. Disease symptom development in the cdkc;2-2 mutant was the same as in the cdkc;2-1 mutant. (C) Viral DNA accumulation. Wild-type and mutant plants were inoculated by bombardment of pCa122, and total DNA was isolated from leaf tissues harvested at the indicated DAI and probed with the P6 open reading frame of CaMV. Ethidium bromide staining of genomic DNA is shown as a loading control. (D) Viral RNA accumulation. Total RNA was isolated from inoculated plants and first probed with the P3 open reading frame of CaMV for detection of the 35S RNA. The blot was stripped and reprobed with the P6 open reading frame of CaMV for detection of the 19S RNA. Ethidium bromide staining of rRNA is shown for the assessment of equal loading.

Figure 2.

Response of Arabidopsis cdkc;2 and cyct1;5 Mutants to TMV-cg and CaLCuV. (A) Accumulation of TMV-cg viral RNAs. Three leaves of 4-week-old Arabidopsis plants were inoculated with TMV-cg (5 μg/mL). The lower, inoculated leaves and upper, systemic infected leaves of five plants were collected at the indicated DAI for RNA isolation. RNA gel blot analysis was performed with ³²P-labeled TMV-cg coat protein gene as a probe. (B) Accumulation of CaLCuV DNA-A and DNA-B. Four-week-old Arabidopsis plants were bombarded with the infectious DNA clones of CaLCuV. Total DNA was isolated from infected leaves at the indicated DAI. DNA gel blot analysis was performed with ³²P-labeled DNA fragments specific to the A or B components of CaLCuV.

Figure 3.

CDKC;2 and CYCT1;5 Dependence of the CaMV 35S Promoter Activity. (A) GUS activities in the wild type, cdkc;2, and cyct1;5 mutants transformed with a GUS transgene driven by the CaMV 35S promoter. Average GUS activities were calculated from 15 to 20 independent T1 transformants. GUS activities are expressed in units (nanomoles of 4-methylumbelliferone per minute per milligram of total soluble protein). (B) Accumulation of GUS transcripts. Total RNA was pooled from 15 to 20 T1 transformants and probed with the full-length GUS gene fragment.

Figure 4.

CaMV Resistance of the cdkc;2/CDKC;1 RNAi and cyct1;5/CYCT1;4 RNAi Mutants. (A) Viral DNA accumulation. Wild-type (lane 1), cdkc;2 cyct1;5 (lane 2), cdkc;2/CDKC;1 RNAi (lane 3), and cyct1;5/CYCT1;4 RNAi (lane 4) plants were inoculated by bombardment of pCa122, and total DNA was isolated from leaf tissues harvested at the indicated DAI and probed with the P6 open reading frame of CaMV. Ethidium bromide staining of genomic DNA is shown as a loading control. (B) Viral RNA accumulation. Total RNA was isolated from inoculated wild-type (lane 1), cdkc;2 cyct1;5 (lane 2), cdkc;2/CDKC;1 RNAi (lane 3), and cyct1;5/CYCT1;4 RNAi (lane 4) plants. The RNA gel blot was first probed with the P3 open reading frame of CaMV for detection of the 35S RNA. The blot was stripped and reprobed with the P6 open reading frame of CaMV for detection of the 19S RNA. Ethidium bromide staining of rRNA is shown for the assessment of equal loading.

Figure 5.

CDKC;1 and CDKC;2 Interact with CYCT1 Proteins and Phosphorylate the CTD Peptide. (A) CDKC;1 and CDKC;2 interact with CYCT1;3 but not CYCT1;4 or CYCT1;5 in yeast cells. The Gal4 DNA binding domain–CDKC;1 or –CDKC;2 fusion vector was cotransformed with various activation domain–CYCT1 fusion vectors into yeast cells and grown in nonselective (+His) or selective (−His) medium. The empty DNA binding (pBD) and activation (pAD) vectors were included as negative controls. Growth without added His (−His) demonstrates interaction between protein pairs. (B) Recombinant glutathione S-transferase (GST), GST-tagged CDKC;1 and CDKC;2, and His-tagged CYCT1;3 proteins were expressed and purified from E. coli. Phosphorylation of the GST-tagged CTD of RNAP II by the indicated proteins was assayed in the presence of [γ-³²P]ATP. (C) and (D) BiFC analysis of CDKC;1 (C) and CDKC;2 (D) interactions in planta with CYCT1 proteins. Yellow fluorescence was observed in the nuclear compartment of N. benthamiana leaf epidermal cells, which results from complementation of the N-terminal part of the YFP fused with CDKC;2 (CDKC;2-N-YFP) with the C-terminal part of the YFP fused with CYCT1;3 (CYCT1;3-C-YFP) or CYCT;5 (CYCT1;5-YFP). No fluorescence was observed when CDKC;1-N-YFP or CDKC;2-N-YFP was coexpressed with unfused C-YFP. Bright-field images (left), YFP epifluorescence images (middle), and overlay images of plasmid autofluorescence and YFP epifluorescence (right) of the same cells are shown.

Figure 6.

Induced Expression of CDKC and CYCT1 Genes by CaMV. Top, 4-week-old Arabidopsis plants were mechanically inoculated with TMV-cg or partially purified CaMV virions, with mock-inoculated (buffer only) plants used as controls. Bottom, plants were bombarded with the infectious DNA clones of CaLCuV or with gold particles (mock) as a control. Total RNA was isolated at the indicated DAI and probed with ³²P-labeled CDKC and CYCT1 gene-specific DNA fragments.

Figure 7.

Growth Phenotypes of Double Mutants. (A) Morphology of seeds from wild-type, cyct1;4^−/− cyct1;5^−/+, and cyct1;4^−/+ cyct1;5^−/− plants. The aborted seeds from siliques of mutant plants are either pale or brown and shrunken. (B) Morphology of 6-week-old wild-type and transgenic cdkc;2-1 mutant plants expressing a RNAi construct for CDKC;1. The extent of reduction in the size of the RNAi mutants was correlated with the extent of reduction in the level of CDKC;1 transcripts, as determined by RT-PCR (data not shown). (C) Morphology of wild-type and transgenic cyct1;5 mutant plants expressing a RNAi construct for CYCT1;4. The altered leaf growth in the RNAi mutants was observed only in mutant plants with reduced levels of CYCT1;4 transcripts, as determined by RT-PCR (data not shown). (D) Scanning electron microscopy images of rosette leaf surfaces of wild-type and cyct1;5/CYCT1;4 RNAi mutant plants. Bars = 50 μm.

Figure 8.

Phenotypes of Flowering and Flower Development of Double Mutants. (A) Plants of the wild type (plant 1), cdkc;2 (plant 2), cyct1;5 (plant 3), cdkc;2 cyct1;5 (plant 4), cdkc;2/CDKC;1 RNAi (plant 5), and cyct1;5/CYCT1;4 RNAi (plant 6) mutants grown under a 12-h-light/12-h-dark photoperiod. The photograph was taken 8 weeks after germination. (B) FLC transcripts in 5-week-old plants of the wild type (lane 1), cdkc;2 (lane 2), cyct1;5 (lane 3), cdkc;2 cyct1;5 (lane 4), cdkc;2/CDKC;1 RNAi (lane 5), and cyct1;5/CYCT1;4 RNAi (lane 6) mutants grown under a 12-h-light/12-h-dark photoperiod. (C) Siliques (top panel) and flower buds (bottom panel) of wild-type, rdr6-11, cdkc;2-1, and rdr6 cdkc;2 mutant plants. (D) GUS activities in flower buds of stably transgenic Arabidopsis plants expressing CDKC;1:GUS (left) or CDKC;2:GUS (right). These images are representative results of multiple flowers buds examined from five independent transgenic lines for each construct.