A conserved cysteine motif is critical for rice ceramide kinase activity and function

Abstract

Background: Ceramide kinase (CERK) is a key regulator of cell survival in dicotyledonous plants and animals. Much less is known about the roles of CERK and ceramides in mediating cellular processes in monocot plants. Here, we report the characterization of a ceramide kinase, OsCERK, from rice (Oryza sativa spp. Japonica cv. Nipponbare) and investigate the effects of ceramides on rice cell viability.

Principal findings: OsCERK can complement the Arabidopsis CERK mutant acd5. Recombinant OsCERK has ceramide kinase activity with Michaelis-Menten kinetics and optimal activity at 7.0 pH and 40°C. Mg2+ activates OsCERK in a concentration-dependent manner. Importantly, a CXXXCXXC motif, conserved in all ceramide kinases and important for the activity of the human enzyme, is critical for OsCERK enzyme activity and in planta function. In a rice protoplast system, inhibition of CERK leads to cell death and the ratio of added ceramide and ceramide-1-phosphate, CERK's substrate and product, respectively, influences cell survival. Ceramide-induced rice cell death has apoptotic features and is an active process that requires both de novo protein synthesis and phosphorylation, respectively. Finally, mitochondria membrane potential loss previously associated with ceramide-induced cell death in Arabidopsis was also found in rice, but it occurred with different timing.

Conclusions: OsCERK is a bona fide ceramide kinase with a functionally and evolutionarily conserved Cys-rich motif that plays an important role in modulating cell fate in plants. The vital function of the conserved motif in both human and rice CERKs suggests that the biochemical mechanism of CERKs is similar in animals and plants. Furthermore, ceramides induce cell death with similar features in monocot and dicot plants.

Figure 1. Sequence analysis and expression of rice ceramide kinase (OsCERK) in planta.

(A) Schematic representation of the sequence organization of OsCERK. The sizes of the 5′-UTR (white), exons (black) and 3′-UTR (grey) are indicated by numbers below the boxes. The sizes of introns are shown above. (B) Phylogenetic analysis of CERKs and the related enzymes sphingosine kinase (SPK)s, and diacylglycerol kinase (DAGK)s. The architecture of each gene was obtained from SMART using the architecture analysis function, and conserved domains (letters a–d) are indicated as follows: a, Pleckstrin homology domain (PH domain; accession number SM00233); b, Diacylglycerol kinase catalytic domain (DAGKc; SMART accession number SM00046); c, segments of low compositional complexity determined by the program; d, diacylgycerol kinase catalytic domain (DAGK_cat; Pfam accession number PF00781). Detailed sequences information is described in Materials and methods. AtCERK is ACD5. (C) CLUSTAL W alignment of CERK from various organisms showing the highly conserved CXXXCXXC motif. (D) RT-PCR expression analysis of OsCERK in leaves, stem and roots of 2-week-old seedlings. β-actin was used as a control. (E) CERK activity of rice extracts. Crude plant extracts made from leaves and roots of rice, respectively, were incubated at 40°C for 30 min as described in Materials and methods. The reaction mixture (1 µl) was resolved on a TLC plate. Purified OsCERK recombinant protein was used as the positive control (+CK). For the negative control (−CK) no extract was added. (F) Comparison of recombinant OsCERK and rACD5. The recombinant protein (1.0 µg) was added in 100 µl reaction system, and incubated at 40°C for 30 min. –CK, no recombinant protein was added. Arrows in (E and F) indicated the formed product C6-NBD-ceramide-1-phosphate (C6-NBD-Cer-1P).

Figure 2. Biochemical characterization of the recombinant rice ceramide kinase.

(A) Effect of different temperatures on CERK activity. (B) Effect of different pH values on CERK activity. (C) and (D) Michaelis-Menten representation for OsCERK activity with increasing concentrations of D-erythro-C6-NBD-ceramide and ATP, respectively. (E) and (F) Lineweaver-Burk plot of OsCERK activity. (G) Effect of calcium on ceramide kinase activity; for this experiment, ceramide phosphorylation was determined in the absence of MgCl₂. (H) Effect of magnesium on ceramide kinase activity; for this experiment, the effect on activity by MgCl₂ was measured in the absence of CaCl₂. For (A)–(H), 1 µg purified recombinant OsCERK was used for OsCERK activity assay. The reaction time was 30 min. Values are means ± standard deviation (SD) of four replicates from two independent experiments.

Figure 3. Complementation of Arabidopsis acd5 mutant plants by OsCERK.

(A) Comparison of early development between acd5 mutant plants expressing OsCERK and nontransgenic plants. Note that no phenotype differences were observed in 25-day-old transgenic acd5 plants expressing OsCERK (acd5+OsCERK) and empty vector when compared with wild-type and acd5 plants. (B) Phenotypes of 48-day-old plant leaves (upper part) as shown in (A). Cell death was determined by trypan blue staining on the indicated days in leaf tissue (lower part). Note that acd5 and transgenic acd5 plants expressing EV showed obviously spontaneous cell death in both leaves and stained tissues; in contrast, acd5 plants carrying OsCERK showed a completely normal phenotype, just as wild-type plants. This experiment was repeated twice with similar results. Each time 9–12 leaves from at least 6 plants were used for trypan blue staining and over 85% showed full complementation. (C) Phenotypes of 55-day-old plants as shown in (A). Arrowheads indicate the cell death phenotype of acd5 and transformed plants carrying the empty vector in leaves and stems. Note the lack of cell death regions in the transformed acd5 plants carrying OsCERK. At least ten independent lines were obtained. More than 100 plants were observed and phenotypic complementation was 95% in all homozygous lines. (D) Steady-state levels of PR1 gene transcripts in wild-type and transgenic plants at different times. Tubulin indicates the loading control. This experiment was repeated twice with equivalent results. (E) Growth of P. syringae virulent strain DG3 in the indicated genotypes after infection at OD600 = 0.001. Lesion-free plants were inoculated with bacteria at 3 wk of age. The mean value of the growth of bacteria in six leaves is indicated in each case. Letters indicate that values are different using Student-Newman-Keuls test (P<0.05). Bars indicate standard deviations. This experiment was repeated three times with similar results. (F) CERK activity assay of plant extract in the indicated plants. Equal amounts of protein were extracted from 50-day-old leaves as described in Materials and Methods. Arrow indicates the formed product C6-1-³²P. Purified ACD5 recombinant protein was used as the positive control (+CK). For negative control (−CK) no isotope-labeled ATP was added. This experiment was repeated twice with equivalent results. (G) Amount of C6-1-³²P quantified from panel (F).

Figure 4. Effects of mutations in highly conserved motif on OsCERK activity and function.

(A) Effects of mutations in the highly conserved CXXXCXXC motif on OsCERK activity. Site directed mutagenesis of OsCERK was performed as described in Materials and Methods. Equal quantities of protein were added into the 100 µl reaction system, and 1 µl of reaction mix was resolved on a TLC plate after 30 min incubation at 40°C. Purified recombinant ACD5 protein (rACD5) was used as a positive control. Arrowhead indicates the product of C6-NBD-Cer-1P. W.T. means recombinant OsCERK. (B) Relative OsCERK activity quantified from same samples as in panel (A). The values on the vertical axis, were normalized to wild-type (W.T.) rOsCERK activity at 100%. Note that the three variant proteins showed decreased activity, with the C461A variant showing the least activity. Bars represent the mean ± SD of two independent experiments. (C) Western blots with Flag antibody show the stable production of OsCERK and its modified variants in yeast. This experiment was repeated twice with equivalent results. (D) Effects of OsCERK and modified versions of OsCERK on yeast growth. Note the dramatic growth inhibition in yeast expressing OsCERK. This experiment was repeated four times with equivalent results. (E) Expression level of OsCERK and PR1 in acd5 and acd5 transgenic plants. Two different lines in each site-mutated OsCERK were used. Note that the plant materials of line acd5+C458A2 were taken from 26-day-old seedlings as shown in (F), the other transgenic lines and acd5 were taken from 36-day-old plants as shown in (F). (F) Comparison of phenotype between acd5 mutant plants expressing mutated OsCERK and non-transgenic acd5 plants. At least 50 plants from 2 independent lines were used for phenotype identification in each site-mutated OsCERK plants. Arrows indicate cell death lesions.

Figure 5. Inhibition assay and effects of site-mutation on OsCERK activity.

(A) Effect of ceramide kinase inhibitors K1 and DMS on OsCERK activity. The activity of OsCERK was assayed as described in Materials and Methods using TLC and scanned for NBD fluorescence. Arrowhead indicates the formed product C6-NBD-Cer-1P. K1 and DMS were used at different concentrations. Purified OsCERK recombinant protein was used as a positive control (+CK). EtOH (0.5%) and MeOH (0.5%) indicated solvent controls for K1 and DMS, respectively. (B) and (C) Quantification of data from (A). Bars indicate means ± SD of two independent experiments. (D) Effect of the ceramide kinase inhibitor K1 on rice protoplasts. Protoplasts were isolated from 10-day-old rice seedling and treated with K1 (0.5 µM and 5 µM) for the indicated times under light. (E) Effect of ceramide kinase inhibitor K1 on Arabidopsis protoplasts. Protoplasts were isolated from 18-day-old wild-type and acd5 mutant plants and treated with 5 µM K1 for 24 h under light. The percentage of surviving cells in (D, E) was estimated by FDA staining. These experiments were repeated three times with similar results. Letters indicate that values differed in Fisher's protected least significant difference (PLSD) test, a post hoc multiple t test (P<0.05). Error bars indicate standard deviations. Control treatment was with 0.15% ethanol (the solvent in which K1 was dissolved).

Figure 6. Effects of ceramide related inhibitors on cell death induction.

(A) Effects of cycloheximide (CHX) on C6 ceramide-induced cell death. (B) Effects of K252a on C6 ceramide-induced cell death. (C) Effects of C2 ceramide-1-phosphate on C2 ceramide induced cell death. Protoplasts were treated with less than 0.15% solvent or the indicated reagent. Percentage of surviving cells was estimated by FDA staining. These assays were repeated three times with similar results. Letters indicate that values are different using Fisher's PLSD test (P<0.05). Bars show standard deviations.

Figure 7. Ceramide induced programmed cell death is associated with Δψ_m loss in rice protoplasts.

Mitochondrial membrane potential Δψ_m loss detected by flow cytometry. Protoplasts were cultured under light for the indicated times with 0.1% ethanol (solvent control), 100 µM C6, 100 µM C6-DHC and 100 µM protonophore CCCP (positive control), respectively. Δψ_m was analyzed by flow cytometry using the fluorescent probe DiOC₆(3). Results are from a single analysis, representative of three independent experiments that showed similar results. M1, region of high Δψ_m. The percentages in each panel indicate the proportion of the cells with high Δψ_m.