Recombinant brassinosteroid insensitive 1 receptor-like kinase autophosphorylates on serine and threonine residues and phosphorylates a conserved peptide motif in vitro

Abstract

BRASSINOSTEROID-INSENSITIVE 1 (BRI1) encodes a putative Leucine-rich repeat receptor kinase in Arabidopsis that has been shown by genetic and molecular analysis to be a critical component of brassinosteroid signal transduction. In this study we examined some of the biochemical properties of the BRI1 kinase domain (BRI1-KD) in vitro, which might be important predictors of in vivo function. Recombinant BRI1-KD autophosphorylated on serine (Ser) and threonine (Thr) residues with p-Ser predominating. Matrix-assisted laser desorption/ionization mass spectrometry identified a minimum of 12 sites of autophosphorylation in the cytoplasmic domain of BRI1, including five in the juxtamembrane region (N-terminal to the catalytic KD), five in the KD (one each in sub-domains I and VIa and three in sub-domain VIII), and two in the carboxy terminal region. Five of the sites were uniquely identified (Ser-838, Thr-842, Thr-846, Ser-858, and Thr-872), whereas seven were localized on short peptides but remain ambiguous due to multiple Ser and/or Thr residues within these peptides. The inability of an active BRI1-KD to transphosphorylate an inactive mutant KD suggests that the mechanism of autophosphorylation is intramolecular. It is interesting that recombinant BRI1-KD was also found to phosphorylate certain synthetic peptides in vitro. To identify possible structural elements required for substrate recognition by BRI1-KD, a series of synthetic peptides were evaluated, indicating that optimum phosphorylation of the peptide required R or K residues at P - 3, P - 4, and P + 5 (relative to the phosphorylated Ser at P = 0).

Figure 1

Autophosphorylation and phosphoamino acid analysis of recombinant BRI1-KD. A, Affinity-purified FLAG-BRI1-KD (lane 1) or the mutant FLAG-BRI1-K911E (lane 2) was incubated with 20 μCi [γ-³²P]ATP in kinase buffer for 1 h at ambient temperature, followed by PAGE and visualization of incorporated isotopes with a phosphor imager. Lanes 3 and 4 represent the silver-stained gel corresponding to lanes 1 and 2. To confirm the identity of the labeled protein in lane 1, an SDS-PAGE gel was transferred to a polyvinylidene difluoride (PVDF) membrane, and the labeled band was digested with trypsin and subjected to MALDI-MS. Molecular mass was calculated from the predicted amino acid sequence of the recombinant protein. B, A similar analysis to that described above using calmodulin (CaM)-binding peptide (CBP)-BRI1-KD (lanes 1 and 3) and CBP-BRI1-K911E (lanes 2 and 4). The gel was stained with Coomassie Blue rather than silver. C, CBP-BRI1-KD was autophosphorylated and transferred to a PVDF membrane as described above. The membrane was digested with HCl and subjected to phosphoamino acid analysis by thin-layer electrophoresis (TLE) with p-Ser, p-Thr, and p-Tyr standards. CBP-BRI1-KD autophosphorylated primarily on Ser residues with a minor Thr component and no detectable phospho-Tyr residues.

Figure 2

Determination of autophosphorylation sites by MALDI-MS. A, A portion of the MALDI mass spectrum from one HPLC fraction of the tryptic digest of BRI1-KD. The numbers identify the proteolysis products and p represents a phosphate group. The peak at m/z 1,607.6 represents peptide 842 to 854, which contains two phosphate groups. The peaks marked with an asterisk result from non-specific cleavage products. B, A portion of the MALDI-PSD spectrum of peptide 842 to 854. The presence of two phosphate groups is confirmed by the sequential loss of 98 D H₃PO₄ and 80 D HPO₃ for each phosphate group. 2,5- Dihydroxybenzoic acid with diammonium citrate was used as the matrix in both spectra.

Figure 3

Occurrence of Ser or Thr in related kinases at sites corresponding to autophosphorylation in BRI1-KD. BLASTP analysis was performed with amino acids 815 through 1196 of BRI1, comprising the entire KD. The top 49 matches were examined for the number of times a Ser or Thr occurred at a position in the alignment partner equivalent to a confirmed or possible autophosphorylation site in BRI1, as determined by MALDI-MS.

Figure 4

Phosphorylation of synthetic peptides in vitro by FLAG-BRI1-KD. A typical 40-μL reaction mixture contained 1 μg of affinity-purified FLAG-BRI1-KD, 0.1 mm [γ-³²P]ATP (500 cpm/pmol), and 100 μg/mL synthetic peptide in kinase buffer. Reactions were incubated for 20 min at ambient temperature and incorporation of ³²P into the synthetic peptide was quantitated by binding to P81 phosphocellulose paper, followed by liquid scintillation spectrometry. Peptide BR1 corresponds to residues surrounding the phosphorylated Ser in the CBP vector tag of CBP-BRI1-KD. SP11, NR6, and hydroxymethyl glutaryl-coenzyme A reductase (HMR) are based on sequences surrounding the regulatory phosphorylation sites of spinach SPS, spinach nitrate reductase (NR), and Arabidopsis HMR, respectively. All other peptides are sequence variants of BR1, SP11 or NR6. J indicates nor-Leu, a nonoxidizing functional equivalent of Met. All velocities are normalized to SP11 = 1.0. Error bars are se, n = 3. A and B represent two separate experiments.

Figure 5

The synthetic peptide SP11 is a true substrate of BRI1-KD. To show that SP11 was in fact phosphorylated by BRI1-KD rather than stimulating BRI1-KD autophosphorylation, parallel replicate reactions were performed with BRI1-KD only, or BRI1-KD plus SP11, using conditions described in Figure 4. Reaction pairs were spotted on P81 paper, washed, and counted by liquid scintillation spectrometry to determine total radioactivity incorporated, or were separated by 10% (w/v) SDS-PAGE and the BRI1-KD band was excised and counted. SP11 had no detectable affect on incorporation of ³²P into BRI1-KD, showing that increased filter-bound radioactivity in BRI1-KD + SP11 was due to phosphorylation of the peptide by BRI1-KD. Error bars are se, n = 3

Figure 6

Dependence of BRI1-KD kinase activity in vitro on pH. Affinity-purified FLAG-BRI1-KD was assayed as described in Figure 4 except that the pH was varied from 5.5 to 8.5. A, Autophosphorylation (no synthetic peptide present). B, Phosphorylation of SP11. Error bars are se, n = 3

Figure 7

Effect of Mg²⁺ and Ca²⁺ on BRI1-KD in vitro kinase activity. Affinity-purified FLAG-BRI1-KD was assayed as described in Figure 4 in buffer containing 10 mm Mg₂Cl₂ plus 0.2 mm CaCl₂; 10 mm Mg₂Cl₂ only; 0.2 mm CaCl₂ only; 10 mm Mg₂Cl₂ plus 0.2 mm CaCl₂ with 6.0 mm EDTA; and 10 mm Mg₂Cl₂ plus 0.2 mm CaCl₂ with 1.0 mm EGTA. Both FLAG-BRI1-KD autophosphorylation and phosphorylation of SP11 peptide were dependent on Mg²⁺ but not CaCl_2. Error bars are se, n = 3.

Figure 8

Peptide substrate kinetics of BRI1-KD. Lineweaver-Burk double reciprocal plots for SP11 (A) and BR13 (B) were constructed using the indicated substrate concentrations, 1 μg of FLAG-BRI1-KD and 0.1 mm [γ-³²P]ATP (500 cpm/pmol), in kinase buffer. Reactions were incubated for 20 min at ambient temperature and processed as described in Figure 4. Linear regression lines, K_m and V_max were determined using Prism 2.0 graphics software (San Diego).

Figure 9

BRI1-KD autophosphorylation reaction mechanism. A, CBP-BRI1-KD (lane 1) or CBP-BRI1-KD + FLAG-BRI1-K911E (lane 2) was incubated with [γ-³²P]ATP as described in Figure 1, separated by SDS-PAGE, and the incorporated isotope was visualized with a phosphor imager. Lanes 3 and 4 represent the Coomassie Blue-stained gel corresponding to lanes 1 and 2. The band marked with an asterisk corresponds to the FLAG-BRI1-K911E protein. Lack of a labeled band of the same size in lane 2 shows that CBP-BRI1-KD cannot transphosphorylate the mutant kinase, suggesting an intramolecular autophosphorylation mechanism. B, A plot of relative phosphorylation rate versus FLAG-BRI1-KD concentration. Actual amounts of FLAG-BRI1-KD varied from 0.5 μg to 4.0 μg per 40-μL reaction. Reaction conditions were as described in Figure 4. The approximation of first-order kinetics again suggests an intramolecular autophosphorylation mechanism.