The variable domain of a plant calcium-dependent protein kinase (CDPK) confers subcellular localization and substrate recognition for NADPH oxidase

Abstract

Calcium-dependent protein kinases (CDPKs) are Ca(2+) sensors that regulate diverse biological processes in plants and apicomplexans. However, how CDPKs discriminate specific substrates in vivo is still largely unknown. Previously, we found that a potato StCDPK5 is dominantly localized to the plasma membrane and activates the plasma membrane NADPH oxidase (RBOH; for respiratory burst oxidase homolog) StRBOHB by direct phosphorylation of the N-terminal region. Here, we report the contribution of the StCDPK5 N-terminal variable (V) domain to activation of StRBOHB in vivo using heterologous expression system in Nicotiana benthamiana. Mutations of N-terminal myristoylation and palmitoylation sites in the V domain eliminated the predominantly plasma membrane localization and the capacity of StCDPK5 to activate StRBOHB in vivo. A tomato SlCDPK2, which also contains myristoylation and palmitoylation sites in its N terminus, phosphorylated StRBOHB in vitro but not in vivo. Functional domains responsible for activation and phosphorylation of StRBOHB were identified by swapping regions for each domain between StCDPK5 and SlCDPK2. The substitution of the V domain of StCDPK5 with that of SlCDPK2 abolished the activation and phosphorylation abilities of StRBOHB in vivo and relocalized the chimeric CDPK to the trans-Golgi network, as observed for SlCDPK2. Conversely, SlCDPK2 substituted with the V domain of StCDPK5 localized to the plasma membrane and activated StRBOHB. These results suggest that the V domains confer substrate specificity in vivo by dictating proper subcellular localization of CDPKs.

Keywords: Calcium-dependent Protein Kinase; NADPH Oxidase; Phosphorylation; Plant Defense; Plant Immunity; Reactive Oxygen Species (ROS); Signal Transduction.

FIGURE 1.

Ser-82 and Ser-97 in StRBOHB are required for its activation by StCDPK5. A, schematic structures of StCDPK5 variants. Diagonal boxes indicate a six-residue substitution in the J domain. K indicates a Lys residue for ATP binding in the K domain, and M indicates amino acid substitution in Lys to Met. Asterisks indicate point-mutated sites to Ala in predicted myristoylation and palmitoylation sites. B, kinase activity of StCDPK5 variants. Total proteins were prepared from N. benthamiana expressing HA-tagged StCDPK5 variants or GUS as a control. Immunoprecipitates using anti-HA antibody were incubated with histone IIIS and [γ-³²P]ATP with CaCl₂ (indicated as Ca²⁺ +) or EGTA (indicated as Ca²⁺ −). Phosphorylation of histone IIIS was detected by x-ray film (top two panels). Total protein extracts were used for immunoblot analysis with anti-HA antibody (middle panel). Protein loads were monitored by Coomassie Brilliant Blue (CBB) staining of the bands corresponding to ribulose-1,5-bisphosphate carboxylase (Rubisco) large subunit (bottom panel). C, ROS production in N. benthamiana leaves coexpressing HA-tagged StCDPK5CA or GUS as a control with StRBOHB or point-mutated StRBOHB (S82A/S97A). ROS production was measured 2 days after agroinfiltration by chemiluminescence mediated by l-012. Data are means ± S.D. from four experiments. CA, constitutively active.

FIGURE 2.

Myristoylation and palmitoylation of StCDPK5 are required for interaction with and activation of StRBOHB. A, requirement of myristoylation and palmitoylation in StCDPK5 for the localization at plasma membrane. GFP-tagged StCDPK5 variants were transiently coexpressed with RFP-StREM1.3 via agroinfiltration in N. benthamiana. The upper image is from the GFP channel, the middle image is from the RFP channel, and the lower image is the overlay of the GFP and RFP channels. Fluorescence intensity profile (GFP, green; RFP, red) across the pale blue line was performed using the analyzing software (Leica, bottom). An asterisk indicates the corresponding peak of fluorescence intensity. Scale bars, 10 μm. B, requirement of myristoylation and palmitoylation in StCDPK5 for interaction with StRBOHB in vivo. Co-immunoprecipitation was performed with extracts from N. benthamiana leaves coexpressing HA-tagged StCDPK5 variants with FLAG-tagged StRBOHB. MACS MicroBeads with monoclonal HA antibody was used for immunoprecipitation, and anti-HA and anti-FLAG antibodies were used to detect the related proteins in the immunoprecipitates. C, ROS production in N. benthamiana leaves coexpressing HA-tagged StCDPK5 variants or GUS as a control with StRBOHB. ROS production was measured as described in Fig. 1C. Data are means ± S.D. from four experiments. D, in vivo phosphorylation of Ser-82 in StRBOHB by StCDPK5 variants. StRBOHB was transiently coexpressed with HA-tagged StCDPK5 variants or GUS as a control via agroinfiltration in N. benthamiana leaves. Plasma membrane proteins were prepared at 2 days after agroinfiltration. The intensities of each band were quantified, and relative intensity in immunoblot analysis using anti-pSer-82 antibody to anti-StRBOHB antiserum was expressed (top). Plasma membrane proteins were used for immunoblot analysis with anti-HA antibody (bottom panel). CA, constitutively active.

FIGURE 3.

SlCDPK2CA phosphorylates StRBOHB in vitro but not in vivo. A, ROS production in N. benthamiana leaves coexpressing HA-tagged StCDPK5 variants, SlCDPK2CA, or GUS as a control with StRBOHB. ROS production was measured as described in Fig. 1C. Data are means ± S.D. from four experiments. B, in vivo phosphorylation of Ser-82 in StRBOHB by StCDPK5CA but not SlCDPK2CA. Immunoblot analysis was performed as described in Fig. 2D. C, phosphorylation of StRBOHB by recombinant StCDPK5CA and SlCDPK2CA. Purified N-terminal peptides of StRBOHB (left) or histone IIIS (right) were used as substrates for bacterially expressed NusA-fused StCDPK5CA and SlCDPK2CA. Phosphorylation of StRBOHB and histone IIIS was detected by x-ray film (top panel). Protein loads were monitored by Coomassie Brilliant Blue (CBB) staining (middle and bottom panels). Every image was taken from the same gel/autoradiogram and edited. D, phosphorylation of Ser-82 in StRBOHB by recombinant StCDPK5CA and SlCDPK2CA. N-terminal peptides of StRBOHB were used as substrates for NusA-fused StCDPK5CA and SlCDPK2CA. Immunoblot analysis was performed using anti-pSer82 antibody (top panel). Protein loads were monitored by Coomassie Brilliant Blue staining (middle and bottom panels).

FIGURE 4.

Requirement of the V domain of StCDPK5 for phosphorylation and activation of StRBOHB in vivo. A, schematic structures of StCDPK5, SlCDPK2, and chimeric constructs (5V2KJC and 2V5KJC). Diagonal boxes indicate a six-residue substitution in the J domain. B, kinase activity of chimeric CDPKs. The immunocomplex kinase assay and immunoblot analysis with anti-HA antibody in N. benthamiana leaves coexpressing HA-tagged StCDPK5CA, SlCDPK2CA, 5V2KJC, 2V5KJC, or GUS as a control with StRBOHB were done as described in Fig. 1B. Asterisks indicate the detected CDPK constructs. Every image was taken from the same gel/blot/autoradiogram and edited. C, ROS production in N. benthamiana leaves coexpressing HA-tagged StCDPK5CA, SlCDPK2CA, 5V2KJC, or 2V5KJC with StRBOHB. ROS production was measured as described in Fig. 1C. Data are means ± S.D. from three experiments. D, In vivo phosphorylation of Ser-82 in StRBOHB by StCDPK5CA and 5V2KJC but not SlCDPK2CA and 2V5KJC. Immunoblot analysis was performed as described in Fig. 2D.

FIGURE 5.

The V domain of StCDPK5 is required for plasma membrane localization and interaction with StRBOHB in vivo. A, subcellular localization of StCDPK5CA-GFP, SlCDPK2CA-GFP, 5V2KJC-GFP, and 2V5KJC-GFP. The indicated CDPK constructs were transiently expressed in N. benthamiana leaves. The upper image is from GFP channel, and the lower image is the overlay of differential interference contrast image and GFP channel. Images are single-plane confocal images. Scale bars, 10 μm. B, colocalization of SlCDPK2CA-GFP with Golgi markers. SlCDPK2CA-GFP was transiently coexpressed with an mCherry-tagged Golgi marker (top) and a RFP-tagged VHA-a1 (bottom) in N. benthamiana. The left image is from the GFP channel, the middle image is from mCherry/RFP channel, and the right image is the overlay of GFP and mCherry/RFP channels. Images are Z-stack confocal images. An inset indicates magnification of a selected single-plane confocal image. Scale bars, 10 μm. C, interaction of StCDPK5CA and 5V2KJC, but not SlCDPK2CA and 2V5KJC, with StRBOHB in vivo. Co-immunoprecipitation was performed with extracts from N. benthamiana leaves coexpressing the indicated HA-tagged CDPK constructs with FLAG-tagged StRBOHB. MACS MicroBeads with monoclonal HA antibody was used for immunoprecipitation and anti-HA, and anti-FLAG antibodies were used to detect the related proteins in the immunoprecipitates. The asterisks indicate the detected CDPK constructs.