Autophosphorylation-dependent inactivation of plant chimeric
calcium/calmodulin-dependent protein kinase
P. V. Sathyanarayanan and B. W. Poovaiah
1
Laboratory of Plant Molecular Biology and Physiology, Department of Horticulture, Washington State University,
Pullman, WA, USA
Chimeric calcium/calmodulin dependent protein kinase
(CCaMK) is characterized by the presence of a visinin-like
Ca
2+
-binding domain unlike other known calmodulin-
dependent kinases. Ca
2+
-Binding to the visinin-like domain
leads to autophosphorylation and changes in the affinity
for calmodulin [Sathyanarayanan P.V., Cremo C.R. &
Poovaiah B.W. (2000) J. Biol. Chem. 275, 30417–30422].
Here, we report that the Ca
2+
-stimulated autophosphory-
lation of CCaMK results in time-dependent loss of enzyme
activity. This time-dependent loss of activity or self-inacti-
vation due to autophosphorylation is also dependent on
reaction pH and ATP concentration. Inactivation of the
enzyme resulted in the formation of a sedimentable enzyme
due to self-association. Specifically, autophosphorylation in
thepresenceof200l
M
ATP at pH 7.5 resulted in the for-
mation of a sedimentable enzyme with a 33% loss in enzyme
activity. Under similar conditions at pH 6.5, the enzyme lost

orchestrated through several calcium binding proteins such
as calmodulin, ion channels, Ca
2+
-dependent protein
kinases and Ca
2+
/calmodulin dependent protein kinases
[2,3]. A large number of plant Ca
2+
-dependent protein
kinases (CDPK) have been reported [4–7]. These kinases
require Ca
2+
for autophosphorylation and substrate phos-
phorylation [4–7]. However, there is only limited informa-
tion available about the Ca
2+
/CaM-dependent protein
kinases in plants [7,8].
Chimeric calcium calmodulin dependent protein kinase
(CCaMK) has been cloned from lily anthers [9]. CCaMK is
stage-specifically expressed in tapetal cells and pollen
mother cells of anthers during male gametophyte develop-
ment [10]. CCaMK is characterized by a serine-threonine
kinase domain, an autoinhibitory domain overlapping with
calmodulin binding domain and a C-terminal visinin-like
domain with three calcium-binding sites [9]. Visinin-like
proteins are high affinity Ca
2+
-binding proteins and

described as self-inactivation, is sensitive to reaction pH and
ATP concentration. Furthermore, the autophosphorylation-
dependent inactivation leads to the formation of a sediment-
able enzyme. When observed under transmission electron
microscope, the autophosphorylated kinase appeared as
particles that are clustered into branched complexes.
EXPERIMENTAL PROCEDURES
Materials
AMP-PNP and ATP were purchased from Sigma Chemical
Co. and [c-
32
P]ATP (3000 CiÆmmol
)1
) from Dupont Corp.
Correspondence to B. W. Poovaiah, Laboratory of Plant Molecular
Biology and Physiology, Department of Horticulture,
Washington State University, Pullman, WA 99164-6414, USA.
Fax: + 1 509 335 8690, Tel.: + 1 509 335 2487,
E-mail:
Abbreviations:CaM,calmodulin;CCaMK,chimericcalcium/
calmodulin-dependent protein kinase; TEM, transmission electron
microscopy; AMP-PNP, adenosine 5¢-(b,-imido)triphosphate;
CaMK II, calcium/calmodulin-dependent protein kinase II.
(Received 24 October 2001, revised 20 March 2002,
accepted 22 March 2002)
Eur. J. Biochem. 269, 2457–2463 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02904.x
Tabbed-copper grids (400 mesh) and polystyrene sizing
beads (93 and 262 nm) were obtained from Ted Pella
(Redding, CA, USA)
3

Hepes, at pH 6.5 and 8.5, were used under
the same conditions as described above. Aliquots (10 lL,
200 ng enzyme) were collected at 0, 2, 6 min and diluted
to 100 lLoficecold50m
M
Hepes, pH 7.5 with 10 m
M
EDTA. This terminated further autophosphorylation of
the enzyme. Samples were then centrifuged for 30 min at
16 000 g at 4 °C. The supernatant and pellet were
separated and suspended in SDS loading buffer (10%
glycerol, 15 m
M
dithiothreitol, 2.3% SDS and 62.5 m
M
Tris, pH 6.8) for SDS/PAGE analysis [14]. [
32
P]PO
4
incorporation was measured by excising the protein
bands from the gel and counting using a scintillation
counter.
Kinase assays
The autophosphorylation-induced changes in the kinase
activity was studied using a second stage assay. An
aliquot of autophosphorylated enzyme (2.5 lL, 100 ng of
enzyme) from the first stage reaction was added to the
second stage reaction mix (final volume, 20 lL) consisting
of 50 m
M

Transmission electron microscopy
All the solutions used for transmission electron microscopy
were filtered using 0.2-lm filters to remove any impurities.
CCaMK was autophosphorylated as described above and
the reaction was terminated by the addition of EDTA (to a
final concentration of 50 m
M
) and the sample was kept in
ice. The autophosphorylated and unphosphorylated kinase
samples were deposited onto the carbon coated Formvar
grids by the floating method, as described previously [15].
Drops (50 lL) of the kinase sample and other solutions
were placed on Parafilm stretched over a top of a Petri
dish. A grid was placed on a drop of the autophospho-
rylated kinase for 45 s and then placed on a drop of uranyl
acetate stain (2% uranyl acetate in 25% methanol) for
1 min. The grid was subsequently placed on a drop of
distilled water for 45 s and air dried. The polystyrene sizing
beads (93 and 262 nm diameter) were applied to the grids
following the procedures outlined above for size reference.
Transmission electron microscopy was performed using
conventional procedures on a JOEL-100CX operating at
80 kV.
RESULTS
Autophosphorylation and kinase activity of CCaMK
Autophosphorylation of CCaMK was associated with a
time dependent loss of the Ca
2+
/CaM-dependent enzyme
activity at pH 6.5, 7.5 and 8.5 (Fig. 1). Activity meas-

under these conditions. Figure 3 shows the formation of the
sedimentable enzyme after 2 min of autophosphorylation.
Autophosphorylation at different ATP concentrations
(200 l
M
and 1 m
M
; Fig. 3) shows that at higher concentra-
tions of ATP, less sedimentable enzyme is formed.
Transmission electron microscopy studies
of autophosphorylated CCaMK
The formation of sedimentable enzyme was visualized
using TEM. CCaMK phosphorylated at 200 l
M
ATP for
5 min produced distinct uranyl acetate staining structures
(Fig. 4A,B). These particles appeared to interconnect and
associate to form branched structures. Figure 4C shows the
electron micrograph of an autophosphorylation deficient
site directed mutant (T267A) of CCaMK that did not form
branched complexes in the presence of 200 l
M
ATP for
5 min. In addition, autophosphorylation of T267A for
10 min did not show any branched complexes under TEM.
Autophosphorylation in the presence of adenosine 5¢-(b-
imido) triphosphate (AMP-PNP)
4
, an unhydrolyzable ana-
logue of ATP also did not produce uranyl acetate staining

decrease in activity at 2 min [24]. CCaMK lost about 33%
of enzyme activity in 2 min, indicating that it is more
sensitive to autophosphorylation-dependent inactivation.
At pH 6.5, there was about 40% loss of enzyme activity in
2 min for CaMK II [24] and at this pH, CCaMK lost 67%
of enzyme activity. These results suggest that the enzyme
inactivation is dependent on the duration of autophospho-
rylation and reaction pH.
The fit to the time-dependent kinetics showed (Fig. 2)
that the inactivation followed exponential decay [R values:
Table 1. Time-dependent inactivation of kinase activity of CCaMK due to autophosphorylation. CCaMK was autophosphorylated as described in
Experimental procedures, and an aliquot (200 ng of CCaMK) was added at indicated time points to a second stage reaction mixture containing
200 l
M
Histone II AS, as substrate. The phosphorylation was allowed for 10 min and [
32
P]PO
4
incorporation into the Histone II AS was measured
by excising protein bands from the gel and counting using a scintillation counter. The phosphorylation (c.p.m.) represents mean of three
measurements.
Time
(min)
pH 6.5 pH 7.5 pH 8.5
Phosphorylation
(c.p.m.)
% Initial
activity
Phosphorylation
(c.p.m.)

inactivation), the enzyme loses activity very rapidly and in
the second phase (slow inactivation), loss of inactivation is
significantly slower. Loss of enzyme activity showed these
two phases of inactivation at all the pH conditions tried.
5
The autophosphorylation resulted in the formation of a
sedimentable enzyme (Fig. 3). After 2 min of autophospho-
rylation, CCaMK was detected in both the pellet and
supernatant fractions. However, at time zero, all the kinase
enzymes were seen in the supernatant, indicating that
autophosphorylation leads to the formation of sedimentable
CCaMK. The formation of a sedimentable enzyme was
observed after 2 min of autophosphorylation at all the
different reaction pH tested (Fig. 3). The role of the ATP
concentration in sedimentablity of the enzyme was tested by
conducting autophosphorylation at different ATP concen-
trations (200 l
M
and 1 m
M
). Figure 3 shows that a
sedimentable enzyme was formed at pH 6.5 and 8.5 at
both the ATP concentrations used for autophoshorylation.
At pH 7.5, a higher ATP concentration (1 m
M
)prevented
the formation of a sedimentable enzyme (Fig. 3). However,
under all the other conditions of pH and ATP concentra-
tions, kinase enzymes existed as both the sedimentable and
soluble form. This suggests that both pH and ATP

autophosphorylated kinase after denaturation (by boiling
in the presence of detergents) was observed under TEM.
After denaturation, the autophosphorylated kinase did not
show the formation of the network-like structures (data not
shown) indicating that these complex structures are formed
due to self-association caused by autophosphorylation.
The role of ATP binding and ATP hydrolysis in the
formation of the complexes was investigated by replacing
the ATP with an unhydrolyzable ATP analogue, AMP-
PNP, in the autophosphorylation reaction mix. Figure 4C
shows that the network-like complexes are not formed in the
presence of AMP-PNP. This suggests that ATP hydrolysis
is required for the complex formation. The autophospho-
rylation mutant T267A [12] did not show the complex
formation (Fig. 4C) under similar conditions that produced
complex structures of wild-type CCaMK. These results
further suggest that the formation of the complexes seen
under TEM is phosphorylation-dependent.
Self-inactivation may be a mechanism of regulating
enzyme activity as a means of modulating metabolic
processes or signal transduction pathway. Self-inactivation
imposes an upper limit on bioactive prostanoid synthesis by
prostaglandin H synthase (PGHS) [25]. The cytochrome
Fig. 3. Formation of sedimentable enzyme
during autophosphorylation of CCaMK.
CCaMK was autophosphorylated at different
pH (6.5, 7.5 and 8.5) and at different ATP
concentrations (200 l
M
and 1 m

phosphorylation, reaction pH, and ATP concentration. To
our knowledge, no other protein kinase reported from
plants shows such a phosphorylation-dependent loss of
activity. However, CCaMK-mediated protein phosphory-
lation is implicated in male gametophyte development in
plants [10]. Self-inactivation of CCaMK may be a mechan-
ism of modulating the Ca
2+
/CaM mediated signal trans-
duction pathway in anther. The elucidation of the molecular
mechanisms leading to the self-inactivation of CCaMK will
broaden our understanding of the regulation of Ca
2+
/
CaM-mediated signaling in plants.
ACKNOWLEDGEMENTS
We thank Dr Chris Davitt and Professor Vincent Franceschi of the
Electron Microscopy Center, WSU for their valuable help and
suggestions with electron microscopy, and Shima Nakanishi for help
with inactivation kinetics experiments. The support of the National
Science Foundation (Grant MCB 0082256) and the National Aero-
nautics and Space Administration (Grant NAG-10-0061) is gratefully
acknowledged.
REFERENCES
1. Poovaiah, B.W. & Reddy, A.S.N. (1993) Calcium and signal
transduction in plants. CRC Crit. Rev. Plant. Sci. 12, 185–211.
2. Clapham, D.E. (1995) Calcium signalling. Cell. 80, 259–268.
3. Berridge, M.J., Lipp, P. & Bootman, M.D. (2000) The versatility
and universality of calcium signaling. Nat Rev. Mol Cell Biol. 1,
11–21.

B.W. (1996) Dual regulation of a chimeric plant serine/threonine
kinase by calcium and calcium/calmodulin. J.Biol.Chem.271,
8126–8132.
12. Sathyanarayanan, P.V., Siems, W.F., Jones, J.P. & Poovaiah,
B.W. (2001) Calcium-stimulated autophosphorylation site of plant
chimeric calcium/calmodulin-dependent protein kinase. J.Biol.
Chem. 276, 32940–32947.
13. Colbran, R.J. & Soderling, T.R. (1990) Calcium/calmodulin-
dependent protein kinase II. Curr. Top Cell Regul. 31, 181–221.
14. Laemmli, U.K. (1970) Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680–685.
15. Sommerville, J. & Scheer, U. (1987) Electron microscopy in
molecular biology. A Practical Approach. IRL Press, Washington
DC.
6
16. Kuret, J. & Schulman, H. (1985) Mechanism of autopho-
sphorylation of the multifunctional Ca
2+
/calmodulin-dependent
protein kinase. J.Biol.Chem.260, 6427–6433.
17. Yamauchi, T. & Fujisawa, H. (1985) Self-regulation of calmodu-
lin-dependent protein kinase II and glycogen synthase kinase by
autophosphorylation. Biochem. Biophys. Res. Commun. 129,213–
219.
18. Bronstein, J.M., Farber, D.B. & Wasterlain, C.G. (1986) Auto-
phosphorylation of calmodulin kinase II: functional aspects.
FEBS Lett. 196, 135–138.
19. Lai, Y., Nairn, A.C. & Greengard, P. (1986) Autophosphorylation
reversibly regulates the Ca
2+

1364–1375.
24. Hudmon, A., Aronowski, J., Kolb, S.J. & Waxham, M.N. (1996)
Inactivation and self association of Ca
2+
/calmodulin-dependent
protein kinase II during autophosphorylation. J. Biol. Chem. 271,
8800–8808.
25. Zgoda, V.G., Karuzina, I.I., Nikitiuk, O.V. & Archakov, A.I.
(1996) Modification of cytochrome P-450 apoenzyme during its
oxidative self-inactivation in a reconstituted mono-oxygenase
system. Vopr. Med. Khim. 42, 203–210.
26. Wu, G., Vuletich, J.L., Kulmacz, R.J., Osawa, Y. & Tsai, A.L.
(2001) Peroxidase self-inactivation in prostaglandin H synthase-1
pretreated with cyclooxygenase inhibitors or substituted with
mangano protoporphyrin IX. J. Biol. Chem. 276, 19879–19888.
27. Shackelford, D.A., Yeh, R.Y. & Zivin, J.A. (1993) Inactivation
and subcellular redistribution of Ca
2+
/calmodulin-dependent
protein kinase II following spinal cord ischemia. J. Neurochem. 61,
738–747.
28. Aronowski,J.,Grotta,J.C.&Waxham,M.N.(1992)Ischemia-
induced translocation of Ca
2+
/calmodulin-dependent protein
2462 P. V. Sathyanarayanan and B. W. Poovaiah (Eur. J. Biochem. 269) Ó FEBS 2002
kinase II: potential role in neuronal damage. J. Neurochem. 58,
1743–1753.
29. Kolb, S.J., Hudmon, A. & Waxham, M.N. (1995) Ca
2+