Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Dec;25(12):5053-66.
doi: 10.1105/tpc.113.116921. Epub 2013 Dec 24.

Calcium/Calmodulin-dependent Protein Kinase Is Negatively and Positively Regulated by Calcium, Providing a Mechanism for Decoding Calcium Responses During Symbiosis Signaling

Affiliations
Free PMC article

Calcium/Calmodulin-dependent Protein Kinase Is Negatively and Positively Regulated by Calcium, Providing a Mechanism for Decoding Calcium Responses During Symbiosis Signaling

J Benjamin Miller et al. Plant Cell. .
Free PMC article

Abstract

The establishment of symbiotic associations in plants requires calcium oscillations that must be decoded to invoke downstream developmental programs. In animal systems, comparable calcium oscillations are decoded by calmodulin (CaM)-dependent protein kinases, but symbiotic signaling involves a calcium/CaM-dependent protein kinase (CCaMK) that is unique to plants. CCaMK differs from the animal CaM kinases by its dual ability to bind free calcium, via calcium binding EF-hand domains on the protein, or to bind calcium complexed with CaM, via a CaM binding domain. In this study, we dissect this dual regulation of CCaMK by calcium. We find that calcium binding to the EF-hand domains promotes autophosphorylation, which negatively regulates CCaMK by stabilizing the inactive state of the protein. By contrast, calcium-dependent CaM binding overrides the effects of autophosphorylation and activates the protein. The differential calcium binding affinities of the EF-hand domains compared with those of CaM suggest that CCaMK is maintained in the inactive state at basal calcium concentrations and is activated via CaM binding during calcium oscillations. This work provides a model for decoding calcium oscillations that uses differential calcium binding affinities to create a robust molecular switch that is responsive to calcium concentrations associated with both the basal state and with oscillations.

Figures

Figure 1.
Figure 1.
Mutations at the EF-Hand Domains Autoactivate CCaMK. (A) The kinase domain of CCaMK (1-311) gives rise to spontaneous nodulation in the absence of rhizobia, as do mutants lacking EF-hand 3 or EF-hands 2 and 3 (1-477 and 1-435, respectively). Bars = 0.5 mm. (B) Quantification of spontaneous nodulation in the absence of rhizobia with CCaMK truncation mutants and single, double, and triple EF-hand point mutants. Schematic representation of CCaMK shows the kinase domain (light gray), CaM binding domain (dark gray), EF-hand domains (black), and location of point mutations (red stars). Data represent means and se; all spontaneous nodulation phenotypes were significant relative to ccamk-1 mutants transformed with wild-type CCaMK (one-tailed binomial test; P < 0.005).
Figure 2.
Figure 2.
Ca2+-Dependent Autophosphorylation of CCaMK Occurs on Thr-271. (A) A state-specific antibody to phosphorylated Thr-271 (α-pT271) specifically recognizes Thr-271 phosphorylation of wild-type (WT) CCaMK but not the kinase-dead (K47E) or T271A mutants. (B) Treatment with λ protein phosphatase (λPPase) removes the phosphate from Thr-271 of CCaMK such that no phosphorylation can be detected with α-pT271. (C) Thr-271 autophosphorylation of CCaMK requires the divalent cations Mn2+ or Mg2+. Addition of Ca2+ strongly enhances Thr-271 autophosphorylation of CCaMK. (D) Immunoblotting of CCaMK mutants shows that gain-of-function activity is associated with a lack of phosphorylation of Thr-271. Wild-type CCaMK and the 1-346 deletion (neither of which gave rise to spontaneous nodulation in the absence of rhizobia; Figure 1B) showed strong autophosphorylation on Thr-271, while gain-of-function truncations and EF-hand point mutants all showed an absence of Thr-271 phosphorylation. Final concentrations of 10 mM Mg2+ and 0.2 mM Ca2+ were used in panels (A), (B), and (D). (A) and (B) used recombinant CCaMK with a GST affinity tag. (C) and (D) used recombinant CCaMK with an MBP affinity tag. K47E, kinase-dead mutant; EF1, D413A; EF2, D449A; EF3, D491A; EF1+2+3, triple EF-hand point mutant (D413A+D449A+D491A).
Figure 3.
Figure 3.
Phosphorylation of Thr-271 Stabilizes a Closed Conformation of CCaMK. (A) A predicted structural model of CCaMK with the kinase domain in red, the predicted CaM binding domain in blue, and the linker region in yellow. The location of the hydrogen bond between Thr-271 and Arg-323 is shown with further detail in the other panels. (B) Unphosphorylated Thr-271 forms one hydrogen bond (dashed line) to Arg-323. (C) Phosphorylated Thr-271 forms a total of three hydrogen bonds to neighboring residues Arg-323 (two bonds) and Ser-322. (D) The Asp substitution (T271D) forms one hydrogen bond to the backbone of Trp-272, which is present in the kinase domain. The homology model was generated using I-Tasser and DeepView and energy minimized using Gromac (see Methods).
Figure 4.
Figure 4.
CaM Binding Blocks Thr-271 Autophosphorylation and Is Required for Mycorrhization and Nodulation. (A) Total autophosphorylation of CCaMK is decreased by the presence of 0.5 μM CaM. CBB, Coomassie blue. (B) Ca2+-induced Thr-271 phosphorylation of CCaMK is decreased by the addition of CaM at the time of Ca2+ treatment. (C) Ala scanning of the CaM binding domain reveals that residues Glu-319, Leu-324, Leu-333, and Ser-343 are important during mycorrhization and nodulation. The ccamk-1 mutant was transformed with the different CCaMK genes, and lines were tested for levels of mycorrhizal colonization (8 weeks after inoculation; black) and numbers of nodules following S. meliloti inoculation (4 weeks after inoculation; gray). Spontaneous nodules only occur at later time points, and in this analysis the nodules observed will be only those induced by S. meliloti. Data represent means and se; asterisks denote mutants showing a significant nodulation and mycorrhization phenotype relative to ccamk-1 mutants transformed with wild-type CCaMK (two-tailed Student’s t test; P < 0.05). In total, an average of 23 independent transformed plants was scored per construct, per phenotype. These transformed plants were generated from at least three independent transformation experiments. EV, empty vector; WT, wild type. (D) Immunoblots assessing Thr-271 phosphorylation of CCaMK. Thr-271 phosphorylation is abolished in the L324A mutant, while the L333A mutant showed enhanced Thr-271 phosphorylation, suggesting a constitutively deactivated protein. The E319A and S343A mutants are able to phosphorylate Thr-271, but to lower levels than the wild-type protein. Final concentrations of 10 mM Mg2+ and 0.2 mM Ca2+ were used in (A), (B), and (D).
Figure 5.
Figure 5.
Activation of CCaMK during Ca2+ Spiking and Target Phosphorylation. (A) A schematic network of reactions regulating the activity of CCaMK showing all the chemical reactions that were taken into account in the mathematical model of the system. The complexes that are active for target phosphorylation are shown in the gray box, and only Ca2+-saturated CaM is assumed to bind to Ca2+-saturated CCaMK. (B) and (C) The simulation results of the system of ordinary differential equations of the reactions in (A) for different dissociation rate constants of CaM for CCaMKP3Ca2+ and CCaMK3Ca2+ at basal Ca2+ concentrations (0.1 s−1 in [B] and 0.01 s−1 in panel [C]). With the experimentally determined kinetic parameters, CCaMK complexes (green) and pT271 CCaMK (blue) rise and fall with oscillations in Ca2+ concentrations. The complexes assumed to be active for target phosphorylation (green) lead to an increase in the level of phosphorylated target during Ca2+ oscillations (provided that the target dephosphorylation rate does not reduce the phosphorylated target concentration to that prior to the previous Ca2+ spike). The kinase activity factor was set to 0.09 in (B) and 0.01 in (C) (see Methods). The equations and kinetic parameters are given in Methods and Supplemental Table 2 online, respectively.
Figure 6.
Figure 6.
Schematic Overview of CCaMK Activation. A hydrogen bond network between the kinase domain and the CaM binding domain is strengthened by Ca2+-induced Thr-271 phosphorylation at basal Ca2+ concentrations; in this state, CCaMK is inactive for target phosphorylation. With relatively low dephosphorylation rates, phosphorylated CCaMK is likely to be the dominant species at basal Ca2+ concentrations. At elevated Ca2+ concentrations during Ca2+ spiking, CaM binds to CCaMK, and this overrides the negative regulation caused by Thr-271 phosphorylation, making the kinase active for target phosphorylation. Phosphorylation of Ser-343/Ser-344 inhibits CaM binding, and this negatively regulates CCaMK activity (Liao et al., 2012; Routray et al., 2013). The hydrogen bond network in the gain-of-function T271A mutant (yellow star) is disrupted, rendering the protein active for target phosphorylation.

Similar articles

See all similar articles

Cited by 32 articles

See all "Cited by" articles

Publication types

MeSH terms

Substances

Associated data

LinkOut - more resources

Feedback