Báo cáo khoa học: A hydrophilic cation-binding protein of Arabidopsis thaliana, AtPCaP1, is localized to plasma membrane via N-myristoylation and interacts with calmodulin and the phosphatidylinositol phosphates PtdIns(3,4,5)P3 and PtdIns(3,5)P2 pptx

A hydrophilic cation-binding protein of
Arabidopsis thaliana, AtPCaP1, is localized to plasma
membrane via N-myristoylation and interacts with
calmodulin and the phosphatidylinositol phosphates
PtdIns(3,4,5)P
3
and PtdIns(3,5)P
2
Nahoko Nagasaki, Rie Tomioka and Masayoshi Maeshima
Laboratory of Cell Dynamics, Graduate School of Bioagricultural Sciences, Nagoya University, Japan
The intracellular localization of proteins is critical for
expression of their cellular function, and is determined
by several mechanisms, including their primary
sequences, post-translational processing, covalent mod-
ifications and affinity to other elements. Most soluble
proteins are localized to the cytoplasm, intra-organelle
spaces, cytoskeletons or secreted out of the cells. How-
ever, some parts of hydrophilic proteins in cells can be
Keywords
Arabidopsis; calcium; myristoylation;
phosphatidylinositol phosphate; plasma
membrane
Correspondence
M. Maeshima, Laboratory of Cell Dynamics,
Graduate School of Bioagricultural Sciences,
Nagoya University, Nagoya 464-8601, Japan
Fax: +81 52 789 4096
Tel: +81 52 789 4096
E-mail:
(Received 19 October 2007, revised 5
February 2008, accepted 5 March 2008)

doi:10.1111/j.1742-4658.2008.06379.x
A hydrophilic cation-binding protein, PCaP1, was found to be stably
bound to the plasma membrane in Arabidopsis thaliana. PCaP1 was quanti-
fied to account for 0.03–0.08% of the crude membrane fractions from roots
and shoots. Its homologous protein was detected in several plant species.
We investigated the mechanism of membrane association of PCaP1 by
transient expression of fusion protein with green fluorescent protein. The
amino-terminal sequence of 27 residues of PCaP1 had a potential to local-
ize the fusion protein with green fluorescent protein to the plasma mem-
brane, and the substitution of Gly at position 2 with Ala resulted in the
cytoplasmic localization of PCaP1. When PCaP1 was expressed in the
in vitro transcription ⁄ translation system with [
3
H]myristic acid, the label
was incorporated into PCaP1, but not into a mutant PCaP1 with Gly2
replaced by Ala. These results indicate that PCaP1 tightly binds to the
plasma membrane via N-myristoylation at Gly2. We examined the binding
capacity with phosphatidylinositol phosphates (PtdInsPs), and found that
PCaP1 selectively interacts with phosphatidylinositol 3,5-bisphosphate and
phosphatidylinositol 3,4,5-triphosphate. Competition assay with the N-ter-
minal peptide and mutational analysis revealed that PCaP1 interacts with
these two PtdInsPs at the N-terminal part. Interaction of PCaP1 with the
membrane and PtdInsPs was not altered in the presence of Ca
2+
at physio-
logical concentrations. Furthermore, calmodulin associated with PCaP1 in
aCa
2+
-dependent manner, and its association weakened the interaction of
PCaP1 with PtdInsPs. These results indicate that the N-terminal part is

essential for both N-myristoylation and interaction with PtdInsPs, and that
PCaP1 may be involved in intracellular signalling through interaction with
PtdInsPs and calmodulin.
Abbreviations
CaM, calmodulin; GFP, green fluorescent protein; GPI, glycosylphosphatidylinositol; MAP, methionine aminopeptidase; NMT, myristoyl-
CoA:protein N-myristoyltransferase; PCaP1, plasma membrane-associated cation-binding protein; PtdIns(3,4,5)P
3
, phosphatidylinositol
3,4,5-triphosphate; PtdIns(3,5)P
2
, phosphatidylinositol 3,5-bisphosphate; PtdInsP, phosphatidylinositol phosphate.
FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS 2267
associated with the plasma and organelle membranes
via covalent modification with lipids, formation of
complexes with membrane integral proteins and associ-
ation with membrane components such as membrane
lipids. The strength and reversibility of the association
with membranes depends on the biochemical proper-
ties of the proteins.
Covalent modifications with lipids, in particular,
are of interest in relation to the cell signalling and
regulatory functions through these proteins [1–5].
Lipid modifications, in combination with other post-
translational changes, some reversible, often cause
proteins to undergo extensive intracellular transloca-
tion. Four types of lipid modification are known:
N-myristoylation, prenylation, palmitoylation and
modification with glycosylphosphatidylinositol (GPI)
anchor proteins [5]. Palmitoylation is the process of
introduction of palmitic acid into protein by substitu-

tion for a hydrogen atom of a Cys residue (S-acyla-
tion). Typical proteins with palmitoylation are
a-subunits of heterotrimeric G-proteins [6]. Palmitoy-
lation of proteins is a reversible process in living cells.
Therefore, the intracellular localization and physiolog-
ical functions can be regulated in cells. N-myristoyla-
tion is the covalent attachment of a myristoyl group
via an amide bond to the N-terminal Gly residue of
a nascent polypeptide. For example, some a-subunits
of G-protein heterotrimers, some small G-proteins
and several non-receptor-type tyrosine kinases are
N-myristoylated proteins. Proteins with lipid modifica-
tions come in many shapes, sizes and functions, even
in plants [7]. Specific primary sequences, such as a
myristoylation signal motif, determine the type of
lipid modification.
In addition to covalent lipid modifications, the spe-
cific interaction with phosphatidylinositol phosphates
(PtdInsPs) in the membrane plays a critical role in the
regulation of the function and intracellular localization
of proteins [8–11].
Very recently, a novel hydrophilic cation-binding
protein was identified in Arabidopsis thaliana [12]. The
protein is composed of 225 amino acid residues and is
rich in Glu and Lys. The protein has no transmem-
brane domain, but is associated with the plasma mem-
brane, and was tentatively named AtPCaP1 (hereafter
referred to as PCaP1). The gene coding for PCaP1 was
constitutively expressed in most organs, and the
mRNA level was enhanced by the treatment with a

pathological elicitor, sorbitol, and copper [12]. How-
ever, the physiological function of PCaP1 is unclear.
In this study, we focused our attention on the bio-
chemical mechanism of the association of PCaP1 with
the plasma membrane. We found that the protein con-
tains a candidate for the myristoylation signal at the
N-terminal region, and investigated this. Biochemical
analyses, including in vitro myristoylation, demon-
strated the N -myristoylation of PCaP1. In addition,
PCaP1 has a candidate for association with PtdInsPs.
We examined this possibility and determined quantita-
tively the specificity of the PtdInsP species. Further-
more, we observed that PCaP1 associated with
calmodulin (CaM) in the presence of calcium. These
observations are essential for understanding the bio-
chemical roles of the novel cation-binding protein and
its related proteins in various organisms. The present
study revealed that PCaP1 is a unique protein, which
is N-myristoylated and associated with specific
PtdInsPs. The biochemical meaning of these properties
is discussed.
Results
Immunochemical detection of PCaP1 orthologues
in several plant species
PCaP1 is composed of 225 amino acids and is rich
in Glu (44 residues), Lys (35 residues) and Val (25
residues). The protein has characteristic repeats
(IEEKK, VEEKK and VEETKK) (Fig. 1A). To
date, no motif has been found for enzymatic
function. A possible candidate for N-myristoylation

exists at the N-terminal region, as described later.
PCaP1 has many Ser and Thr residues, and some
residues have been estimated to be phosphorylation
sites. A homologous protein with high identity with
PCaP1 was found in Nicotiana tabacum by blast
search ( (Fig. 1A).
This protein was named DREPP1 (developmentally
regulated plasma membrane protein) [13]. Although
the protein was detected in the plasma membrane
and endomembrane fractions, its physiological and
biochemical properties are unknown. The N-terminal
halves are highly conserved between the two
sequences, suggesting that PCaP1 and its orthologues
are not unique to A. thaliana.
The calculated molecular mass of PCaP1 is 24 584;
however, the protein was detected with a molecular
mass of 36 kDa in an immunoblot with anti-PCaP1
IgG (Fig. 1B), which was raised against the peptide
with internal sequence of PCaP1 (positions 152–166).
The difference between the calculated and apparent
size may be caused by the amount of dodecyl-sulfate
bound to PCaP1 and ⁄ or the structure in SDS.
Immunoblotting showed bands in Raphanus sativus
(radish, 41 kDa), Brassica rapa (turnip, 42 kDa),
Brassica rapa var. glabra Regel (Chinese cabbage,
A novel cation-binding myristoylated protein N. Nagasaki et al.
2268 FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS
43 kDa) and Brassica oleracea var. italica (broccoli,
41 kDa) (Fig. 1B). The immunostained bands disap-
peared when the corresponding peptide was added to

the reaction medium. Thus, these bands were ortho-
logues of PCaP1. The low intensity of immunostain-
ing, except for A. thaliana, may be caused by the
partial difference in the sequence corresponding to
the epitope. We did not examine the membrane
preparation from N. tabacum, because the corre-
sponding sequence is not a match with that of
PCaP1 (Fig. 1A).
Quantification of PCaP1 in the membrane and
soluble fractions
To determine the amount of PCaP1 in tissues and the
distribution of PCaP1 in the membrane and soluble
fractions (by an immunochemical method), we pre-
pared the recombinant PCaP1 as the standard protein.
As shown in Fig. 2A, a highly purified preparation of
PCaP1 without any tag was obtained. The protein was
analysed by SDS-PAGE and immunoblotting with an
anti-PCaP1 IgG to obtain a calibration curve
A
B
Fig. 1. Detection of PCaP1 orthologues in plants. (A) Amino acid sequence alignment of A. thaliana PCaP1 and N. tabacum DREPP1.
Identical (*) and conserved (:) residues are marked. Gaps introduced to maximize alignment scores are denoted by hyphens. A putative
N-myristoylation site of PCaP1 is underlined. The overlined sequence was used for preparation of the anti-PCaP1 IgG. Characteristic VEEKK
motifs and variants are boxed. Possible phosphorylation sites were predicted using the N
ETPHOS 2.0 program ( />services/NetPhos/). Open circles indicate possible phosphorylation residues with a high score of more than 0.8, and filled circles indicate the
target residues of protein kinase-C-like enzymes with a high score of more than 0.7. (B) Immunoblot detection of PCaP1 orthologous protein
in crude membrane fractions with anti-PCaP1. Lanes 1 and 6, A. thaliana; lanes 2 and 7, Raphanus sativus; lanes 3 and 8, Brassica rapa;
lanes 4 and 9, B. rapa var. glabra; lanes 5 and 10, B. oleracea var. italica. The amount of protein applied was 4 lg for A. thaliana and 40 lg
for the other plants.
N. Nagasaki et al. A novel cation-binding myristoylated protein

FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS 2269
(Fig. 2B,C). The crude membrane fractions and soluble
fractions were prepared from shoots and roots and
subjected to immunoblotting (Fig. 2D). The absolute
amount of PCaP1 was calculated using a standard
curve. Most PCaP1 was recovered in the membrane
fractions, and the PCaP1 amounts in the shoot and
root fractions were 0.79 and 0.29 lgÆmg
)1
of total
membrane protein, respectively. There was only a trace
amount of PCaP1 in the soluble fractions. Thus,
PCaP1 was predominantly localized to the membrane
in the tissues, and was present at 0.079% and 0.029%
of total membrane proteins in the shoots and roots,
respectively.
The stability of the interaction of PCaP1 with the
plasma membrane was examined by treating the mem-
branes with several reagents (Fig. 3). PCaP1 was not
released from the plasma membrane by treatment with
0.1 m NaCl or 2 m urea. Even in 1 m NaCl, PCaP1
was stably associated with the membrane (data not
shown). PCaP1 was partially released from the mem-
brane by treatment with 0.1 m Na
2
CO
3
or 1% Tri-
ton X-100 (Fig. 3). In general, alkaline treatment with
Na

2
CO
3
removes peripheral membrane proteins, which
are associated with membrane intrinsic proteins, and a
mild detergent Triton X-100 is used to solubilize mem-
brane proteins, but not all membrane integral proteins.
Partial resistance to detergent and alkaline treatment
indicates that PCaP1 has properties similar to mem-
brane integral proteins.
Mode and sequence essential for membrane
association
The stable association of a protein without transmem-
brane domains with the plasma membrane led us to
determine the mode of interaction. The results shown
in Fig. 3 suggest that the interaction of PCaP1 with
the membrane does not occur electrostatically or by
association with transmembrane proteins. Indeed, we
failed to isolate a complex of PCaP1 with transmem-
brane protein(s). Therefore, we examined lipid modifi-
cation, especially N-myristoylation, as PCaP1 contains
a putative N-myristoylation consensus sequence,
Met-Gly-X-X-X-Ser-Lys, at the N-termini [4] (Fig. 1).
If the protein is N-myristoylated, Gly2 will be the
site of covalent modification. We prepared a PCaP1
mutant construct, whose Gly2 was replaced by Ala,
linked with green fluorescent protein (GFP) at the
C-terminus of PCaP1 (PCaP1
G2A
-GFP). We then

expressed the GFP fusion proteins in A. thaliana sus-
pension-cultured cells. We observed more than 25 cells
for each construct by confocal laser scanning micros-
copy. Green fluorescence of wild-type PCaP1 was
A
C
D
B
Fig. 2. Preparation of standard PCaP1 protein and immunochemical
quantification of PCaP1 in A. thaliana. (A) PCaP1 with (His)
6
tag
(His ⁄ PCaP1) was expressed in Escherichia coli cells and purified
from the soluble fraction. Purified His ⁄ PCaP1 was treated with
TAGZyme to remove the (His)
6
tag. Samples were subjected to
SDS-PAGE and stained with Coomassie brilliant blue. Lane 1, solu-
ble fraction (10 lg) prepared from E. coli cells; lane 2, preparation
(1.5 lg) after nickel nitrilotriacetic acid Superflow column chroma-
tography; lane 3, TAGZyme system-treated fraction (1.5 lg); lane 4,
peak fraction (1.5 lg) after Sephacryl S-300 HR column chromatog-
raphy. Black and white arrowheads indicate the position of His ⁄ P-
CaP1 (37 kDa) and PCaP1 (36 kDa), respectively. (B) Purified
PCaP1 (0, 5, 10, 15, 20, 30 and 40 ng) was subjected to SDS-
PAGE, followed by immunoblotting with anti-PCaP1 IgG. (C) Rela-
tive intensity of immunostained bands was plotted against the
amount of PCaP1 protein to prepare a calibration curve. (D) Crude
membrane (P100) and cytosol (S100) fractions were prepared from
shoots and roots of 2-week-old plants of A. thaliana by centrifuga-

tion at 100 000 g. The fractions (20 lg each) were subjected to
immunoblotting with anti-PCaP1 IgG (inset). The amount of PCaP1
protein on the basis of total protein in each fraction was calculated
using the standard curve.
A novel cation-binding myristoylated protein N. Nagasaki et al.
2270 FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS
clearly detected in the plasma membrane (Fig. 4A). In
contrast, the fluorescence of the PCaP1
G2A
mutant was
observed in the cytosol, but not in the plasma mem-
brane (Fig. 4B). This was not caused by the release of
the GFP moiety from the fusion protein by proteolytic
cleavage in the cells, because free GFP was always
localized to the cytosol and nucleus (data not shown).
To examine whether the N-terminal sequence included
a possible myristoylation signal, we prepared two addi-
tional GFP fusion proteins: one with the first
27-amino-acid sequence of PCaP1 (PCaP1
1)27
) and the
other with a modified N-terminal 27-residue sequence,
in which Gly2 was replaced by Ala (PCaP1
1)27 ⁄ G2A
).
Green fluorescence of PCaP1
1)27
-GFP and
PCaP1
1)27 ⁄ G2A

-GFP was detected in the plasma mem-
brane and cytosol, respectively (Fig. 4C,D). The results
indicate that the N-terminal part with 27 residues has
the ability to localize the protein to the plasma mem-
brane, and that Gly2 is essential for plasma membrane
localization.
In vitro myristoylation
To confirm the N-myristoylation of PCaP1, we carried
out an in vitro transcription ⁄ translation assay in the
presence of [
3
H]myristic acid, using rabbit reticulocyte
lysate, which contained N-myristoyltransferase activity
[14]. Because N-myristoylation occurs cotranslational-
ly, the experiments were carried out in a cell-free tran-
scription ⁄ translation system. CBL4 (also known as
A
B
C
D
Fig. 3. Tight association of PCaP1 with plasma membrane. (A) The
purified plasma membrane fraction was treated with 0.1
M NaCl,
2
M urea, 0.1 M Na
2
CO
3
or 1% Triton X-100 for 20 min, and then
centrifuged as described in Experimental procedures. The PCaP1

contents in the supernatant (S) and pellet (P) were determined by
immunoblotting with anti-PCaP1 IgG. (B) The relative content of
PCaP1 in the supernatant and pellet was expressed as the percent-
age of the total amount of PCaP1. The data are the averages from
two independent experiments. (C) The purified plasma membranes
were treated with 0.1
M Na
2
CO
3
to release peripheral membrane
proteins (left) and HCl was added to neutralize the suspension
(right). (D) The suspensions were centrifuged at 100 000 g, and the
supernatant (S) and pellet (P) were subjected to immunoblotting
with anti-PCaP1 IgG.
µ
A
B
C
D
Fig. 4. Plasma membrane localization of PCaP1 variants. (A–D)
Expression of PCaP1-GFP fusion proteins in suspension-cultured
cells of A. thaliana. Constructs of PCaP1-GFP (A), PCaP1
G2A
-GFP
(B), PCaP1
1)27
-GFP (C) and PCaP1
1)27 ⁄ G2A
-GFP (D) were transiently

expressed in the cells. Green fluorescence was viewed with a con-
focal laser scanning microscope (left panels). Nomarski images
were also recorded (right panels).
N. Nagasaki et al. A novel cation-binding myristoylated protein
FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS 2271
SOS3), a typical myristoylated protein, was synthesized
as a 28 kDa protein, as detected by immunoblotting in
this system, and was labelled with [
3
H]myristic acid as
shown by fluorography (Fig. 5). CBL4 is myristoylated
at Gly2 in the N-terminal sequence (MGCSVSKKK)
and functions as an EF-hand-type Ca
2+
-binding pro-
tein [14]. When the Gly residue was replaced by Ala,
the mutant CBL4 (CBL4
G2A
) did not incorporate
[
3
H]myristic acid. Both PCaP1 and its variant
(PCaP1
G2A
) were translated and detected as 36 kDa
proteins by immunoblotting. Radioactive [
3
H]myristic
acid was incorporated into PCaP1, but not into
PCaP1

G2A
(Fig. 5). These results indicate that PCaP1
is myristoylated at Gly2.
Specific interaction of PCaP1 with PtdInsPs
The N-terminal part of PCaP1 is rich in Lys and aro-
matic (Tyr, Trp and Phe) residues (Fig. 1A). These
characteristic sequences with clusters of basic ⁄ aromatic
residues have been found in domains for interaction
with PtdInsPs [8,15,16].
We determined the binding nature of PCaP1 using
PIP Strips
TM
, which were spotted with a series of
PtdInsPs (Fig. 6A). PCaP1 bound to phosphatidylino-
sitol 3,4-bisphosphate [PtdIns(3,4)P
2
], PtdIns(3,5)P
2
,
PtdIns(4,5)P
2
and phosphatidylinositol 3,4,5-triphos-
phate [PtdIns(3,4,5)P
3
], and weakly with PtdIns(3)P,
PtdIns(4)P and PtdIns(5)P. The protein did not associ-
ate with lysophosphatidic acid, lysophosphatidylcho-
line, phosphatidylinositol, phosphatidylethanolamine,
phosphatidylcholine, sphingosine 1-phosphate, phos-
phatidic acid or phosphatidylserine. Next, we exam-

ined the effect of calcium on these properties, because
PCaP1 has been demonstrated to bind calcium [12].
The selectivity of binding was not changed by the
addition of calcium to the reaction mixture at 4 lm
(Fig. 6B). Potassium did not affect the intensity or
selectivity, even at 10 mm, but magnesium weakened
the affinity but not the selectivity.
Further quantitative analysis was performed using a
PIP Array
TM
(Fig. 6C). PCaP1 bound PtdInsPsina
concentration-dependent manner (Fig. 6D). PCaP1
had a high affinity for PtdIns(3,5)P
2
and
PtdIns(3,4,5)P
3
, and bound even at 3.1 pmol on the
sheet. The affinity for PtdIns(3,4)P
2
and PtdIns(4,5)P
2
was relatively low. We examined the binding selectivity
of PCaP1 using an array sheet containing
PtdIns(3,5)P
2
, PtdIns(3,4)P
2
and PtdIns(4,5)P
2

(spot-
ted by ourselves; data not shown), because there was a
difference in the signal strength for the three PtdInsP
2
between the PIP Strips
TM
and PIP Array
TM
(Fig. 6A–C). This careful assay confirmed the high
affinity of PCaP1 for PtdIns(3,5)P
2
, but not for
PtdIns(3,4)P
2
or PtdIns(4,5)P
2
. The results indicated
that PCaP1 has an ability to bind selectively to
PtdIns(3,5)P
2
and PtdIns(3,4,5)P
3
amongst the various
PtdInsPs. It should be noted that a (His)
6
tag did not
affect the interaction of PCaP1 with PtdInsPs (data
not shown). At present, we cannot deny the weak
interaction with PtdIns(3,4)P
2

and PtdIns(4,5)P
2
in this
in vitro assay system.
Amino-terminal part of PCaP1 as the site of
binding to PtdInsPs
In general, the polybasic residue region is a good can-
didate for binding to PtdInsPs. The N-terminal part is
the most basic region, containing seven Lys residues.
Thus, we carried out a competition assay of PtdInsP
binding using a peptide that corresponds to the N-ter-
minal part (positions 2–24) (Fig. 7E). The purified
PCaP1 bound PtdInsP
2
and PtdIns(3,4,5)P
3
. The bind-
ing intensity was decreased markedly in the presence
of the PCaP1
2)24
peptide (Fig. 7A,B), suggesting that
PCaP1 binds PtdInsPs at the N-terminal region. This
possibility was confirmed by comparison of the
Fig. 5. Incorporation of [
3
H]myristic acid into PCaP1, but not into
PCaP1
G2A
, in a rabbit reticulocyte in vitro translation assay. In vitro
transcription and translation of wild-type PCaP1 (WT) and PCaP1

G2A
(G2A) were performed in the presence of [
3
H]myristic acid. The
translation products were subjected to SDS-PAGE, immunoblotting
(right and middle panels) and fluorography (left panel). Constructs
of CBL4 and its derivative with Gly2 replaced by Ala were exam-
ined as positive and negative controls, respectively. The same incu-
bation was performed without any template DNA (none).
Arrowheads indicate the positions of PCaP1 (36 kDa) and CBL4
(27 kDa). Translation products were detected by immunoblotting
with anti-PCaP1 (middle panel) and anti-(His)
6
(right panel) IgG.
Molecular masses (kDa) of the standard proteins are shown on the
right.
A novel cation-binding myristoylated protein N. Nagasaki et al.
2272 FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS
binding intensity to PtdInsPs of the wild-type and
N-terminal truncated protein D2-25PCaP1 ⁄ His (Fig. 7E).
The truncated form was slightly smaller than the wild-
type PCaP1 and highly purified, as shown in Fig. 7F.
This D2-25PCaP1 ⁄ His mutant protein gave no signal on
either the PIP Strips
TM
or PI P A rray
TM
(Fig. 7C,D).
Effect of calcium on the interaction of PCaP1 with
the plasma membrane and PtdInsPs

PCaP1 has been demonstrated to bind calcium by
the
45
Ca-overlay assay [12]. Thus, we examined the
effect of calcium on the association of PCaP1 with
the plasma membrane. The purified plasma mem-
branes did not release PCaP1, even in the presence
of 10 mm CaCl
2
(Fig. 8A). Thus, calcium cannot be
an effector or regulator for dissociation ⁄ association
of PCaP1 from the membrane. As shown in the PIP
Array
TM
test (Fig. 8B), calcium did not affect the
interaction with PtdInsPs. The selectivity to PtdInsPs
was unchanged and the interaction with
PtdIns(3,5)P
2
and PtdIns(3,4,5)P
3
was retained up to
100 lm CaCl
2
. The affinity was decreased strongly in
1mm Ca
2+
and lost at 5 mm, concentrations way
above the physiological concentration. Thus, binding
to the plasma membrane and PtdInsPs may be stable

in living cells at a physiological concentration of
0.1 lm [17,18].
Fig. 6. PCaP1 preferentially interacts with
phosphatidylinositol di- and triphosphates.
(A) Binding capacity of PCaP1 to PtdInsPs
was tested with PIP Strips
TM
on which 15
kinds of lipid were immobilized (left). The
strips were incubated in a solution contain-
ing the purified recombinant PCaP1
(50 ngÆmL
)1
) (+PCaP1) or the buffer without
the protein ()PCaP1) at 4 °C overnight
(right). The strips were stained with anti-
PCaP1 IgG, and the antigen PCaP1 bound
to the strips was visualized. LPA, lysophos-
phatidic acid; LPC, lysophosphocholine;
PtdIns, phosphatidylinositol; PtdIns(3)P,
phosphatidylinositol 3-monophosphate;
PtdIns(4)P, phosphatidylinositol 4-mono-
phosphate; PtdIns(5)P, phosphatidylinositol
5-monophosphate; PtdIns(3,4)P
2
, phosphati-
dylinositol 3,4-bisphosphate; PtdIns(3,5)P
2
,
phosphatidylinositol 3,5-bisphosphate;

PtdIns(4,5)P
2
, phosphatidylinositol
4,5-bisphosphate; PtdIns(3,4,5)P
3
, phos-
phatidylinositol 3,4,5-triphosphate, PE,
phosphatidylethanolamine; PC, phosphatidyl-
choline; S1P, sphingosine 1-phosphate; PA,
phosphatidic acid; PS, phosphatidylserine.
(B) Incubation of PIP Strips
TM
with PCaP1
was carried out in the presence of KCl
(b, c), MgCl
2
(d) or CaCl
2
(e). (C) Affinity of
PCaP1 for individual lipids was determined
using a PIP Array
TM
sheet (new version
after 2004), on which lipids were immobi-
lized at the indicated amount. (D) The signal
intensities of a representative assay shown
in (C) are expressed as a percentage of
PtdIns(3,4,5)P
3
at 100 pmol.

N. Nagasaki et al. A novel cation-binding myristoylated protein
FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS 2273
Calcium-dependent interaction of PCaP1 with
CaM
There is no motif of enzymatic function in PCaP1. In
order to understand the physiological role of a non-
enzymatic protein, it is worth surveying the partner of
the interaction protein. We examined the interaction of
PCaP1 with CaM. When purified recombinant PCaP1
was incubated with CaM-agarose, PCaP1 bound to
CaM-agarose, especially in the presence of Ca
2+
, and
no PCaP1 was recovered in the unbound fraction or
wash fraction (Fig. 9A, )CaM). This interaction was
competitively inhibited by free CaM in the incubation
medium (Fig. 9A, +CaM). The bound PCaP1 was
released and eluted by an SDS solution (Fig. 9B).
When free calcium was removed from the incubation
medium by EGTA, no PCaP1 was bound to CaM-
agarose (Fig. 9B). The results indicate that PCaP1
Fig. 7. An amino-terminal polybasic region is necessary for specific binding of PCaP1 to the phosphatidylinositol moiety. (A) The capacity of
binding of PCaP1 to PtdInsP s (abbreviations as in Fig. 6) was tested with PIP Strips
TM
in the absence (left panel) or presence (right panel) of
PCaP1
2)24
peptide. PCaP1 bound to the sheets was detected by immunoblotting with anti-PCaP1 IgG. (B) The signal intensities of a repre-
sentative assay shown in (A) are expressed as a percentage of PtdIns(3,4,5)P
3

without the peptide. Wild-type (PCaP1) and N-terminal trun-
cated PCaP1 (D 2-25PCaP1 ⁄ His) were assayed for PtdInsP binding capacity using PIP Strips
TM
(C) and PIP Array
TM
sheets (D). (E) Schematic
diagram of PCaP1
2)24
peptide, PCaP1 and D2-25PCaP1 ⁄ His. Peptide sequence of PCaP1
2)24
is boxed. (F) SDS-PAGE profile of the purified
D2-25PCaP1 ⁄ His. The D2-25PCaP1 ⁄ His protein was expressed in E. coli cells and purified from the soluble fraction. Samples were subjected
to SDS-PAGE and stained with Coomassie brilliant blue. Lane 1, soluble fraction (10 lg) prepared from the E. coli lysate; lane 2, preparation
(1.5 lg) after nickel nitrilotriacetic acid Superflow column chromatography; lanes 3 and 5, peak fractions (1.5 lg) after HiTrap Phenyl HP col-
umn chromatography (1.5 lg); lane 4, purified PCaP1 ⁄ His (1.5 lg). PCaP1 and D2-25PCaP1 ⁄ His were used in this assay. Black and white
arrowheads indicate the position of PCaP1 ⁄ His and D2-25PCaP1 ⁄ His, respectively.
A novel cation-binding myristoylated protein N. Nagasaki et al.
2274 FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS
associates with CaM in a calcium-dependent manner.
The membrane association of PCaP1 was not affected
by CaM, even in the presence of Ca
2+
(Fig. 9C). How-
ever, CaM suppressed the interaction of PCaP1 with
PtdInsPs in the presence of Ca
2+
(Fig. 9D); the
amount of PCaP1 bound to PtdInsPs was decreased to
25% as shown by the intensity of the PCaP1 signal in
the PIP Array

TM
test. These results indicate that CaM
binds with PCaP1 and affects the association of PCaP1
with PtdInsPs in a calcium-dependent manner.
Discussion
PCaP1 is a novel hydrophilic protein without a pre-
dicted transmembrane domain in nature [12]. PCaP1 is
a minor membrane component and accounts for
0.079% and 0.029% of the total membrane protein
from shoots and roots, respectively, of A. thaliana
seedlings (Fig. 2). The aim of this study was to clarify
the mechanism of the specific tight association of
PCaP1 with the plasma membrane in vivo and in vitro.
Almost all PCaP1 was associated with the membrane
and was not released by treatment with a high concen-
tration of salt or urea (Fig. 3). Alkaline treatment with
Na
2
CO
3
(pH 11.6) released PCaP1, but the released
PCaP1 was recovered in the membrane by neutraliza-
tion of the suspension with HCl (Fig. 3C,D), suggest-
ing the involvement of basic residues, such as Lys
(pK
a
for side-chain, 10.53), in the interaction with the
membrane.
The present study clearly reveals that N-myristoyla-
tion anchors PCaP1 to the plasma membrane. First,

when Gly2 of PCaP1 was replaced by Ala, the mutant
PCaP1 was localized to the cytoplasm (Fig. 4). A Gly
residue adjusted to the first Met is essential for N-myr-
istoylation [4,19]. Second, the first 27 residues of the
N-terminal sequence were sufficient for N-myristoyla-
tion, as GFP linked with the 27-residue peptide was
anchored to the membrane (Fig. 4). Third, [
3
H]myristic
acid was incorporated into PCaP1, but not into a
PCaP1
G2A
mutant (Fig. 5). Thus, we conclude that
PCaP1 is myristoylated at Gly2 and that cotranslation-
al myristoylation anchors the protein to the mem-
brane.
N-myristoylation is catalysed by two enzymes,
namely methionine aminopeptidase (MAP) and myri-
stoyl-CoA:protein N-myristoyltransferase (NMT).
Three MAP isoforms, MAP1A, MAP2A and MAP2B,
have been identified in A. thaliana as the cytoplasmic
forms [20,21]. These MAPs catalyse the excision of the
N-terminal Met residue from proteins. The subsequent
myristoylation reaction is catalysed by N-myristoyl-
transferase; for example, in A. thaliana, AtNMT1 has
been demonstrated to modify several known N-myri-
stoylated proteins in vitro [4]. A comprehensive study
of the substrate specificity of AtNMT1 has revealed
that the positive charge on residue 7 of the sub-
strate proteins is particularly important. The seventh

N-terminal residue of PCaP1 is Lys (Fig. 1). Thus,
PCaP1 may be cotranslationally N-myristoylated by a
A
B
Fig. 8. Association of PCaP1 with the plasma membrane and PtdInsPs (abbreviations as in Fig. 6) in the presence of calcium. (A) The puri-
fied plasma membranes were incubated with CaCl
2
at the indicated concentrations at room temperature for 20 min, and then centrifuged at
100 000 g for 15 min. Both the supernatant (S) and precipitate (P) fractions (20 lg of protein in each lane) were subjected to immunoblotting
with anti-PCaP1 IgG. (B) PIP Array
TM
sheets were incubated with the purified recombinant PCaP1 (50 ngÆmL
)1
) in the absence (a) or pres-
ence (b–f) of CaCl
2
. The bound PCaP1 was detected with anti-PCaP1 IgG.
N. Nagasaki et al. A novel cation-binding myristoylated protein
FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS 2275
cytoplasmic MAP and AtNMT1, and subsequently
anchored to the cytoplasmic face of the plasma mem-
brane. Furthermore, it is clear from the present study
that the N-terminal 27-residue sequence is necessary
and sufficient for the N-myristoylation of PCaP1.
PCaP1 is a novel protein with unique structural fea-
tures, namely an abundance of Glu and Lys residues
and a lack of common functional motifs. A previous
study has suggested the constitutive expression and
significant stimulation of gene expression by a patho-
logical elicitor (flagellin peptide) and copper [12]. In

general, N-myristoylation provides the primary mem-
brane-targeting signal for several plant protein kinases,
such as zucchini CpCDPK1, A. thaliana AtCPK2 and
tomato LeCPK1 [6,22–25]. A plant Rab GTPase,
Ara6, which plays a critical role in endosomal homo-
typic fusion, also requires N-myristoylation for its
endosomal localization [26]. In our preliminary experi-
ments, T-DNA insertion mutant lines of PCaP1
showed decreased tolerance to pathological infection
and heavy metal ion stresses. PCaP1 may be involved
in the intracellular response to some physiological
stresses. The biochemical role of PCaP1 remains to be
examined, with a consideration of the phenotypic
properties of the knockout mutant plants.
The elucidation of the specific interaction with
PtdInsPs provides essential information for an under-
standing of the physiological role of PCaP1 in plants.
A large number of proteins associate with PtdInsPsin
membranes with high or low specificity, and express
their own activities, such as intracellular signalling and
organization [8,15,16]. In eukaryotic cells, PtdInsPs
constitute a minor fraction of total membrane lipid,
but play many important roles [27,28]. We demon-
AB
C
D
Fig. 9. Interaction of PCaP1 with CaM and its effect on binding with PtdInsPs (abbreviations as in Fig. 6). (A) Purified PCaP1 was mixed
with CaM-agarose in the presence (lanes 5–8) or absence (lanes 1–4) of CaM (top panel). The same experiments were performed in the
presence of 0.5 m
M Ca

2+
(middle) or 1 mM EGTA (bottom). After centrifugation, the supernatant fractions (Ub, unbound fraction; lanes 1 and
5) were collected. The CaM-agarose beads were washed three times with the same buffer. The supernatants obtained (W1, W2, W3;
lanes 2–4 and 6–8) and the unbound fraction were subjected to SDS-PAGE and protein staining. (B) PCaP1 was incubated with CaM-agarose
in the presence of 0.5 m
M Ca
2+
(lanes 1 and 2) or 1 mM EGTA (lanes 3 and 4). CaM was added to the mixture (lanes 2 and 4). Proteins
bound to CaM-agarose were released with an SDS solution and subjected to SDS-PAGE. Lane 5, recombinant PCaP1 (0.0175 lg). (C) The
purified plasma membranes were incubated with or without 0.167 mgÆmL
)1
CaM, 0.1 m M CaCl
2
and 1 m M EGTA at room temperature for
20 min. After centrifugation at 100 000 g for 15 min, aliquots (20 lg) of the supernatant (S) and precipitated (P) fractions were subjected to
immunoblotting with anti-PCaP1 IgG. (D) PIP Array
TM
sheets were incubated with purified recombinant PCaP1 (50 ngÆmL
)1
) in the presence
of 0.167 mgÆmL
)1
CaM, 0.1 mM CaCl
2
and ⁄ or 1 mM EGTA. Bound PCaP1 was detected by immunoblotting.
A novel cation-binding myristoylated protein N. Nagasaki et al.
2276 FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS
strated that PCaP1 interacts with PtdIns(3,5)P
2
and

PtdIns(3,4,5)P
3
, and that the N-terminal region is nec-
essary and sufficient for specific binding with these two
PtdInsPs. The authentic peptide of the N-terminal
region (PCaP1
1)25
) competitively inhibited the binding
to PtdInsPs (Fig. 7). Furthermore, the N-terminal
truncated form of PCaP1 (D2-25PCaP1 ⁄ His) showed
no binding activity to PtdIns(3,5)P
2
and PtdIns(3,4,5)
P
3
. The polybasic region in the N- or C-terminal parts
of the proteins has been reported to be necessary for
binding to PtdInsPs [16]. PCaP1 contains seven Lys
residues in the N-terminal 25-residue sequence,
although PCaP1 is a totally acidic protein with a
pI value of 4.99. The polybasic region may bind to
phosphate moieties on the inositol ring by electrostatic
bonds. Several PtdInsP
2
-binding proteins contain clus-
ters of basic ⁄ aromatic residues of relatively undefined
structure that interact with PtdInsP
2
[8]. In the N-ter-
minal part of PCaP1, there are four aromatic residues

in addition to the seven Lys residues (Fig. 7). The
interaction with PtdInsPs may contribute to the stabil-
ity of the attachment of PCaP1 to the membrane.
Several consensus sequences have been reported as
PtdInsP-binding domains [9], namely pleckstrin homol-
ogy (PH) [29], phox homology (PX) [30], Fab1-
p ⁄ YOTB⁄ Vac1p ⁄ EEA1 (FYVE) [31–33] and Tubby
domains [34]. However, these domains were not found
in PCaP1. Thus, we conclude that the N-terminal part
of PCaP1 is the main binding site to PtdInsPs. It is
probable that the tertiary structure of a cluster of
basic ⁄ aromatic residues of the N-terminus fits appro-
priately with the structure of PtdIns(3,5)P
2
and
PtdIns(3,4,5)P
3
, and the Lys residues interact with
phosphate moieties at positions 3 and 5. A neuronal
calcium sensor protein hippocalcin has recently been
reported to be N-myristoylated, and also interacts with
PtdIns(4,5)P
2
at the N-terminal part [15]. However,
the N-terminal part of hippocalcin shows a low
sequence similarity to that of PCaP1. The molecular
characteristics of proteins binding PtdIns(4,5)P
2
have
been discussed in detail using a database of crystallo-

graphic structures [35]. To verify the specificity
of PCaP1 to PtdIns(3,5)P
2
and PtdIns(3,4,5)P
3
,a
high-resolution crystallographic study needs to be
performed to determine the binding site.
In plants, PtdIns(3)P, PtdIns(4)P, PtdIns(5)P,
PtdIns(3,4)P
2
, PtdIns(3,5)P
2
and PtdIns(4,5)P
2
have
been detected. In contrast, PtdIns(3,4,5)P
3
has not
been identified in plants [9,36]. Therefore,
PtdIns(3,5)P
2
is the most likely candidate for interac-
tion with PCaP1. Amongst PtdInsPs, PtdIns(4,5)P
2
is
the most abundant PtdInsP
2
, predominantly accumu-
lates in the plasma membrane and is well characterized

as a donor of the signaling compound, inositol 1,4,5-
phosphate [9]. Unfortunately, information concerning
PtdIns(3,5)P
2
in plants is limited. Recently, hyperos-
motic and salt stresses have been shown to induce an
increase in PtdIns(4,5)P
2
and PtdIns(3,5)P
2
in yeast,
Chlamydomonas and some plant cells [9,37,38]. In ani-
mal cells, although PtdInsPs are not exclusively local-
ized to certain organelle membranes, PtdIns(3,5)P
2
is
relatively concentrated on multivesicular bodies and
lysosomal membranes, and is required for cargo deliv-
ery to the former organelle [10]. At present, we cannot
conclude whether the selective interaction of PCaP1
with PtdIns(3,5)P
2
determines the plasma membrane
localization of PCaP1. In plant cells, PtdIns(3,5) P
2
is
accumulated under osmotic and salt stresses, and is
thought to be involved in responses to these specific
stresses [39].
The present study has revealed that CaM interacts

with PCaP1 in a calcium-dependent manner (Fig. 9).
CaM, an acidic protein with four high-affinity Ca
2+
-
binding sites, is well known as the protein mediator of
many Ca
2+
-stimulated enzymes, such as phosphoinosi-
tide 3-kinase and plasma membrane Ca
2+
-ATPase.
The presence of CaM and Ca
2+
suppressed the inter-
action of PCaP1 with PtdInsPs. The binding of PCaP1
to the plasma membrane was not affected by CaM,
even in the presence of Ca
2+
. Therefore, CaM and
Ca
2+
may regulate the unidentified function of PCaP1
localized in the plasma membrane. It has been
reported that a brain-specific protein CAP-23 ⁄ NAP22
is myristoylated and interacts with Ca
2+
⁄ CaM at the
myristoylated N-terminal domain [40]. If Ca
2+
⁄ CaM

binds to the N-terminal site of PCaP1 competitively
with PtdInsPs, the present observation that
Ca
2+
⁄ CaM suppresses the PCaP1–PtdInsP interaction
can be explained.
In conclusion, we have demonstrated that PCaP1
tightly binds to the plasma membrane via N-myristoy-
lation at Gly2 and specifically interacts with
PtdIns(3,5)P
2
and PtdIns(3,4,5)P
3
in the membrane.
N-myristoylation anchors PCaP1, and the interaction
with PtdInsPs may contribute to the stability of the
attachment of PCaP1 to the membrane. It has been
shown that the attachment to the membrane is stable
under physiological conditions (Fig. 8), and that
Ca
2+
⁄ CaM regulates the association of PCaP1 with
PtdInsPs (Fig. 9). Many proteins are N-myristoylated
or interact with PtdInsPs in various organisms. PCaP1
is a highly unique protein, because the N-terminal
domain is required for both N-myristoylation and the
specific interaction with PtdInsPs. Furthermore,
PCaP1 interacts with rare PtdInsPs, namely
PtdIns(3,5)P
2

and PtdIns(3,4,5)P
3
, and weakly with
N. Nagasaki et al. A novel cation-binding myristoylated protein
FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS 2277
PtdIns(3,4)P
2
and PtdIns(4,5)P
2
. Although PtdIns
(3,4,5)P
3
has not yet been detected in plant cells, fur-
ther studies should be performed using other methods,
such as detection with monoclonal antibody to
PtdIns(3,4,5)P
3
[41]. We estimate that PCaP1 is a key
factor for the transduction of signal(s) mediated by
PtdInsPs and Ca
2+
⁄ CaM. The quantification of
PtdIns(3,5)P
2
in the plasma membrane and further
investigation of the phenotypic properties of plants
overexpressing PCaP1 and of T-DNA inserted mutant
plants should provide further insight into the biochem-
ical function of this novel protein.
Experimental procedures

Plant materials
Seeds of A. thaliana (ecotype Col-0) were grown on sterile
gel plates for 2 weeks at 22 °C under continuous light. The
gel plates contained Murashige and Skoog (MS) salt,
2.5 mm 2-(N-morpholino) ethanesulfonate ⁄ KOH, pH 5.7,
2% (w ⁄ v) sucrose and 0.25% Gellangum (Wako Pure
Chemicals, Osaka, Japan) (1· MS-sucrose). A. thaliana sus-
pension-cultured cells (also known as ‘Deep’ cells) were a
kind gift from Masaaki Umeda (University of Tokyo,
Japan). The cells were cultured in MS medium at 22 °Cin
the dark. Other plants [Raphanus sativus (radish), Bras-
sica rapa (turnip), Brassica rapa L. var. glabra Regel (Chi-
nese cabbage) and Brassica oleracea var. italica (broccoli)]
were purchased from a market.
Subcellular fractionation
Two-week-old A. thaliana plants were used for membrane
fractionation [42,43]. The tissue homogenate was filtered
through two layers of Miracloth (EMD Bioscience, Darms-
tadt, Germany) and centrifuged at 10 000 g for 10 min.
The supernatant was centrifuged at 100 000 g for 30 min.
The pellet obtained was suspended in 50 mm Tris ⁄ acetate
buffer, pH 7.5, containing 5% glycerol and 1 mm dith-
iothreitol, and used as a crude membrane fraction. Plasma
membranes were isolated from the crude membranes with
an aqueous two-phase partitioning system [42,44]. Crude
membranes were prepared from taproots of R. sativus and
B. rapa, petioles of B. rapa var. glabra Regel and shoots of
B. oleracea var. italica by the same methods, and used for
immunoblotting.
Preparation of recombinant PCaP1

Three types of recombinant PCaP1 protein were prepared,
namely the full-length PCaP1 (PCaP1), PCaP1 with a
(His)
6
tag at the C-terminus (PCaP1 ⁄ His) and PCaP1 that
lacked the 24 N-terminal residues, except for the first Met,
and was tagged with (His)
6
at the C-terminus (D2-25PCa-
P1 ⁄ His). A construct for PCaP1 ⁄ His was prepared and
introduced into E. coli BL21(DE3) using the pET23b
expression vector (Novagen, Madison, WI, USA), as
described previously [12]. The expression vector, pET ⁄ PCa-
P1 ⁄ His, was directly amplified by PCR with a pair of
PCaP1-specific primers (forward, 5¢-TAAGCGGCCGC
ACTCGAG-3¢; reverse, 5¢-AGGCTTTGGTGGTTCAG
CC-3¢). PCR was performed using KOD-Plus DNA poly-
merase (TOYOBO, Osaka, Japan). The amplified DNA
fragment (pET ⁄ PCaP1) was self-ligated and subjected to
nucleotide sequencing. The vector, pET ⁄ PCaP1, was then
directly amplified by PCR with a pair of primers (forward,
5¢-CACCACCACCACCAGATGGGTTACTGGAATTCCA
AG-3¢; reverse, 5¢-GTGGTGTTTCATATGTATATCTCCT
TCTTAAAGTTAAAC-3¢; italic type shows the His-tag
adaptor sites). After confirmation of the nucleotide
sequences of the pET ⁄ His ⁄ PCaP1 obtained, the expression
vector was introduced into E. coli BL21(DE3) (Novagen).
Transformants were grown in Luria–Bertani (LB) broth for
2 h at 25 °C after induction with 1 mm isopropyl
thio-b-d-galactoside. For D2-25PCaP1, the DNA construct

was prepared from the pET vector possessing pET ⁄
PCaP1 ⁄ His by PCR with a pair of specific primers
(forward, 5¢-GCTGCTGAAGCTACCAAGAC-3¢; reverse,
5¢-CATATGTATATCTCCTTCTTAAAG-3¢). Transfor-
mants of E. coli BL21(DE3) were grown in LB broth at
37 °C and treated with 1 mm isopropyl thio-b-d-galactoside
for 3 h. The cells were then harvested and used for protein
purification.
Purification of recombinant proteins
E. coli cells expressing PCaP1 ⁄ His, D2-25PCaP1 ⁄ His and
His ⁄ PCaP1 were harvested by centrifugation and resus-
pended in 20 mm Tris ⁄ acetate, pH 7.5, containing 20%
(v ⁄ v) glycerol, 0.1 mgÆmL
)1
DNase I, 0.2 mgÆmL
)1
lyso-
zyme, 10 mm 2-mercaptoethanol and protein inhibitor
cocktail (0.5· Complete, EDTA-free) (Roche Applied Sci-
ence, Mannheim, Germany). The cells were disrupted by
sonication for 12.5 min on ice. After removal of cell debris
by centrifugation at 104 000 g for 30 min, the supernatant
was applied to a nickel nitrilotriacetic acid Superflow col-
umn (Qiagen, Valencia, CA, USA) equilibrated with 20 mm
imidazole containing 20 mm Tris ⁄ acetate, pH 7.5, 20%
(v ⁄ v) glycerol and 2 m NaCl. Recombinant PCaP1 proteins
tagged with (His)
6
were eluted with 300 mm imidazole con-
taining 20 m m Tris-acetate, pH 7.5, and 2 m NaCl. The

content of recombinant protein was measured by immuno-
blotting with anti-PCaP1 IgG. The peak content fractions
were collected and applied to a HiTrap Phenyl HP column
(GE Healthcare, Piscataway, NJ, USA) equilibrated with
20 mm Tris ⁄ acetate, pH 7.5, and 2 m NaCl. The His-tagged
PCaP1 proteins were recovered in the flow-through frac-
A novel cation-binding myristoylated protein N. Nagasaki et al.
2278 FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS
tion, desalted and used as the purified proteins. To prepare
PCaP1, we removed the (His)
6
tag from His ⁄ PCaP1 by
treatment with TAGZyme (Qiagen), according to the man-
ufacturer’s instructions. Then, PCaP1 and PCaP1 ⁄ His were
applied to a column (2.6 · 60 cm) of Sephacryl S-300 HR
(GE Healthcare). The peak content fractions were collected
and used as the purified preparations.
Protein quantification, SDS-PAGE and
immunoblotting
The protein concentration was determined using a bicinch-
oninic acid protein assay reagent kit (Pierce Biotechnology,
Rockford, CA, USA) [45]. Protein samples were subjected
to SDS-PAGE on a 12.5% (w ⁄ v) polyacrylamide gel and
immunoblotting. The blots were visualized with horseradish
peroxidase-coupled protein A and western blotting detec-
tion reagents (GE Healthcare).
Dissociation of PCaP1 from membranes
The purified plasma membranes were incubated with 0.1 m
NaCl, 2 m urea, 0.1 m Na
2

CO
3
or 1% Triton X-100 at
room temperature for 20 min to release PCaP1 protein
from the membrane. The suspensions were centrifuged at
100 000 g for 15 min to separate the soluble fraction from
the membrane fraction. Both fractions were subjected to
immunoblotting with polyclonal antibody to PCaP1, which
was prepared previously [12].
Membrane association assay
The purified plasma membranes were incubated with CaCl
2
at room temperature for 20 min, and then centrifuged at
100 000 g for 15 min to determine the effect of calcium on
the association of PCaP1 with the plasma membrane.
In some cases, the plasma membranes were incubated
with 0.1 mm CaCl
2
and 0.167 mgÆmL
)1
CaM (bovine
brain; Wako Pure Chemicals) at room temperature for
20 min, followed by centrifugation at 100 000 g for 15 min,
to determine the effect of calcium on the association of
PCaP1 with the plasma membrane. As a control experi-
ment, EGTA was added to the suspension to remove free
Ca
2+
.
Expression of PCaP1 linked to GFP in Arabidopsis

cells
PCaP1 linked to GFP was prepared, and the fusion protein
was transiently expressed using pENTR ⁄ D-TOPO (Invitro-
gen, Carlsbad, CA, USA) and pUGWB5 (developed by
T. Nakagawa, Shimane University, Japan), as described
previously [12]. GFP protein was linked to the C-terminus
of PCaP1 (PCaP1-GFP). Three other GFP fusion con-
structs were prepared: with the PCaP1 construct with Gly2
replaced by Ala (PCaP1
G2A
-GFP), with the N-terminal
27-residue sequence (PCaP1
1)27
-GFP) and with the
PCaP1
1)27
peptide with Gly2 replaced by Ala
(PCaP1
1)27 ⁄ G2A
-GFP). The PCR primers for pUGWB5-
PCaP1
1)27
-GFP were 5¢-ATGGTGAGCAAGGGCGAG-3¢
(forward) and 5¢-AGCAGCAGCAGCCTTCTTAG-3¢
(reverse). pUGWB5-PCaP1
G2A
-GFP and pUGWB5-
PCaP1
1)27 ⁄ G2A
-GFP were generated from pUGWB5-

PCaP1-GFP and pUGWB5-PCaP1
1)27
-GFP, respectively,
by site-directed mutagenesis [46], with a primer set of
5¢-ATGGCTTACTGGAATTCCAAG-3¢ (forward; italic
type, site-directed mutation site) and 5¢-AGGCTTTGGT
GGTTCAGCC-3¢ (reverse).
The four constructs of GFP fusion proteins (PCaP1-GFP,
PCaP1
G2A
-GFP, PCaP1
1)27
-GFP and PCaP1
1)27 ⁄ G2A
-GFP)
were introduced and expressed in A. thaliana suspension-
cultured cells, as described previously [43]. GFP fluores-
cence was visualized with a Fluoview FV500 confocal laser
scanning microscope (Olympus, Tokyo, Japan) using a
set of BA465–495 (excitation) and BA505–550 (emission)
filters.
In vitro myristoylation
In vitro transcription and translation were performed using
the TNT T7 Quick coupled transcription ⁄ translation system
(Promega, Madison, WI, USA), according to the manufac-
turer’s instructions. Constructs for PCaP1, PCaP1
G2A
,
CBL4 and CBL4
G2A

were prepared using the vector pBS
(Promega). The constructs obtained were transcribed and
translated in a TNT Quick master mix containing rabbit
reticulocyte lysate in a single test tube. A. thaliana CBL4
and its modified protein with the substitution of Gly2 with
Ala (SOS
G2A
) were used as controls. [
3
H]Myristic acid
(1.96 · 10
3
TBqÆmol
)1
) (GE Healthcare) was dried by heat-
ing at 65 °C for 40 min, suspended in distilled water and
then added to the reaction medium at a final concentration
of 37 kBqÆlL
)1
. After reaction at 30 °C for 60 min, aliqu-
ots were subjected to SDS-PAGE. The gel was treated with
Amplify (GE Healthcare). The dried gel was contacted with
an X-ray film (RX-U type; Fuji Film, Tokyo, Japan) and
packed in a TranScreen (Kodak, Rochester, NY, USA).
After incubation for 3 weeks at )80 °C, the X-ray film was
developed.
Lipid-binding assay
Sheets of PIP Strips
TM
(Echelon Biosciences, Salt Lake

City, UT, USA) and PIP Array
TM
(new version after 2004)
(Echelon Biosciences) were treated with 10 mm Tris ⁄ HCl,
pH 8.0, 150 mm NaCl and 0.1% (w ⁄ v) Tween 20 (TBST),
supplemented with 3% fatty acid-free BSA, for 1 h at
25 °C to block the sheets. The sheets were incubated for
N. Nagasaki et al. A novel cation-binding myristoylated protein
FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS 2279
16 h with 50 ngÆmL
)1
recombinant PCaP1 in TBST at
4 °C, and then washed with TBST and immunoblotted with
anti-PCaP1 IgG. The blots were then visualized with horse-
radish peroxidase-coupled protein A and western blotting
detection reagents (GE Healthcare, Nacalai Tesque, Kyoto,
Japan). For some experiments with PIP Strips
TM
,Ca
2+
,
Mg
2+
or K
+
was added to the reaction medium at a
concentration of 4 lm. To evaluate the role of the N-ter-
minal sequence on the lipid-binding capacity, we prepared
N-terminal truncated PCaP1 (D2-25PCaP1) and subjected it
to the lipid-binding assay as described above. The effect of

CaM on the interaction of PCaP1 with PtdInsPs was also
examined. Recombinant PCaP1 (50 ngÆmL
)1
) was incu-
bated with 2.4 lgÆmL
)1
CaM in the presence of 0.1 mm
CaCl
2
for the PIP Strips
TM
assay.
CaM-binding assay
The binding property of CaM was assessed using CaM-aga-
rose (Sigma-Aldrich, St Louis, MO, USA). Recombinant
PCaP1 (50 ngÆmL
)1
) was incubated with 30 lL of CaM-
agarose suspended in 20 mm Tris ⁄ HCl, pH 7.5, 150 mm
NaCl and 0.5 mm CaCl
2
[47]. After centrifugation at
1600 g for 2 min, the supernatant was removed as an
unbound fraction. CaM-agarose was washed three times
with the same buffer, and the proteins bound to CaM-aga-
rose were eluted with SDS sample buffer containing 2%
SDS. Both bound and unbound fractions were analysed by
SDS-PAGE and immunoblotting.
Acknowledgements
We are grateful to Dr Masashi Miyano for advice on

the importance of the N-terminal sequence for mem-
brane interaction, Mr Yuki Ide for contribution to the
initial stage of this study and Dr Sumiko Kaihara for
critical reading of the manuscript. This work was sup-
ported by Grants-in-Aid for Scientific Research
(18380064, 16085204) from the Ministry of Education,
Sports, Culture, Science, PROBRAIN, and Technol-
ogy of Japan, and a grant from the Global Research
Programme of the Ministry of Science and Technol-
ogy, South Korea (to M. M.). N. N. is a recipient of
the Ajinomoto Foundation for Young Scientists
Research Fellowship.
References
1 Resh MD (1994) Myristoylation and palmitoylation of
Src family members: the fats of the matter. Cell 76,
411–413.
2 Cadwallader KA, Paterson H, Macdonald SG & Han-
cock JF (1994) N-terminally myristoylated Ras proteins
require palmitoylation or a polybasic domain for
plasma membrane localization. Mol Cell Biol 14, 4722–
4730.
3 Boutin JA (1997) Myristoylation Cell Signal 9, 15–35.
4 Boisson B, Giglione C & Meinnel T (2003) Unexpected
protein families including cell defense components fea-
ture in the N-myristoylome of a higher eukaryote.
J Biol Chem 278, 43418–43429.
5 Petsko GA & Ringe D (2004) In Protein Structure and
Function. 3–19 Protein targeting by lipid modification,
pp. 124–125. New Science Press Limited, London.
6 Murase K, Shiba S, Iwano M, Che FS, Watanabe M,

Isogai A & Takayama S (2004) A membrane-anchored
protein kinase involved in Brassica self-incompatibility
signaling. Science 303, 1516–1519.
7 Thompson GA Jr & Okuyama H (2000) Lipid-linked
proteins of plants. Prog Lipid Res 39, 19–39.
8 McLaughlin S, Wang J, Gambhir A & Murray D
(2002) PIP
2
and proteins: interactions, organization,
and information flow. Annu Rev Biophys Biomol Struct
31, 151–175.
9 Meijer HJG & Munnik T (2003) Phospholipid-based
signaling in plants. Annu Rev Plant Biol 54, 265–306.
10 Shaw JD, Hama H, Sohrabi F, DeWald DB & Wendland
B (2003) PtdIns(3,5)P
2
is required for delivery of endocy-
tic cargo into the multivesicular body. Traffic 4, 479–490.
11 Roth MG (2004) Phosphoinositides in constitutive
membrane traffic. Physiol Rev 84, 699–730.
12 Ide Y, Nagasaki N, Tomioka R, Suito M, Kamiya T &
Maeshima M (2007) Molecular properties of a novel,
hydrophilic cation-binding protein associated with the
plasma membrane. J Exp Bot 58, 1173–1183.
13 Logan DC, Domergue O, Teyssendier de la Serve B &
Rossignol M (1997) A new family of plasma membrane
polypeptides differentially regulated during plant devel-
opment. Biochem Mol Biol Int 43, 1051–1062.
14 Ishitani M, Liua J, Halftera U, Kima C-S, Shia W &
Zhua JK (2000) SOS3 function in plant salt tolerance

requires N-myristoylation and calcium binding. Plant
Cell 12 , 1667–1678.
15 O’Callaghan DW, Haynes LP & Burgoyne RD (2005)
High-affinity interaction of the N-terminal myristoyla-
tion motif of the neuronal calcium sensor protein hipp-
ocalcin with phosphatidylinositol 4,5-bisphosphate.
Biochem J 391, 231–238.
16 Kaadige MR & Ayer DE (2006) The polybasic region
that follows the plant homeodomain zinc finger 1 of Pf1
is necessary and sufficient for specific phosphoinositide
binding. J Biol Chem 281, 28831–28836.
17 Sanders D, Pelloux J, Brownlee C & Harper JF (2002)
Calcium at the crossroads of signaling. Plant Cell 14,
S401–S417.
18 Sze H, Liang F, Hwang I, Curran AC & Harper JF
(2000) Diversity and regulation of plant Ca21 pumps:
insights from expression in yeast. Annu Rev Plant Phys
Plant Mol Biol 51, 433–462.
A novel cation-binding myristoylated protein N. Nagasaki et al.
2280 FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS
19 Farazi TA, Waksman G & Gordon JI (2001) The biol-
ogy and enzymology of protein N-myristoylation. J Biol
Chem 276, 39501–39504.
20 Giglione C, Serero A, Pierre M, Boisson B & Meinnel
T (2000) Identification of eukaryotic peptide deformy-
lases reveals universality of N-terminal protein process-
ing mechanisms. EMBO J 19, 5916–5929.
21 Ross S, Giglione G, Pierre M, Espagne C & Meinnel T
(2005) Functional and developmental impact of cyto-
solic protein N-terminal methionine excision in Arabid-

opsis. Plant Physiol 137, 623–637.
22 Ellard-Ivey M, Hopkins RB, White TJ & Lomax TL
(1999) Cloning, expression and N-terminal myristoyla-
tion of CpCPK1, a calcium-dependent protein kinase
from zucchini (Cucurbita prop L.). Plant Mol Biol 39,
199–208.
23 Lu SX & Hrabak E (2002) An Arabidopsis calcium-
dependent protein kinase is associated with the endo-
plasmic reticulum. Plant Physiol 128, 1008–1021.
24 Rutschmann F, Stalder U, Piotrowski M, Oecking C &
Schaller A (2002) LeCPK1, a calcium-dependent protein
kinase from tomato. Plasma membrane targeting and
biochemical characterization. Plant Physiol 129, 156–
168.
25 Martin ML & Busconi L (2000) Membrane localization
of a rice calcium-dependent protein kinase (CDPK) is
mediated by myristoylation and palmitoylation. Plant J
24, 429–435.
26 Ueda T, Yamaguchi M, Uchimiya H & Nakano A
(2001) Ara6, a plant-unique novel type Rab GTPase,
functions in the endocytic pathway of Arabidopsis thali-
ana. EMBO J 20, 4730–4741.
27 Stevenson JM, Perera IY, Heilmann I, Persson S &
Boss WF (2000) Inositol signaling and plant growth.
Trends Plant Sci 5, 252–258.
28 Jung J-Y, Kim Y-W, Kwak JM, Hwang J-U, Young J,
Schroeder JI, Hwang I & Lee Y (2002) Phosphatidyl-
inositol 3- and 4-phosphate are required for normal sto-
matal movements. Plant Cell 14, 2399–2414.
29 Varnai P, Lin X, Lee SB, Tuymetova G, Bondeva T,

Spat A, Rhee SG, Hajnoczky G & Balla T (2002) Inosi-
tol lipid binding and membrane localization of isolated
pleckstrin homology (PH) domains. Studies on the PH
domains of phospholipase C delta 1 and p130. J Biol
Chem 277, 27412–27422.
30 Yu JW & Lemmon MA (2001) All phox homology
(PX) domains from Saccharomyces cerevisiae specifically
recognize phosphatidylinositol 3-phosphate. J Biol
Chem 276, 44179–44184.
31 Misra S, Miller GJ & Hurley JH (2001) Recognizing
phosphatidylinositol 3-phosphate. Cell 107, 559–562.
32 Sbrissa D, Ikonomov OC & Shisheva A (2002) Phos-
phatidylinositol 3-phosphate-interacting domains in
PIKfyve: binding specificity and role in PIKfyve endo-
membrane localization. J Biol Chem 277, 6073–6079.
33 Simonsen A, Birkeland HCG, Gillooly DJ, Mizushima
N, Kuma A, Yoshimori T, Slagsvold T, Brech A &
Stenmark H (2004) Alfy, a novel FYVE-domain-
containing protein associated with protein granules
and autophagic membranes. J Cell Sci 117, 4239–
4251.
34 Santagata S, Boggon TJ, Baird CL, Gomez CA, Zhao
J, Shan WS, Myszka DG & Shapiro L (2001) G-protein
signaling through tubby proteins. Science 292
, 2041–
2050.
35 Rosenhouse-Dantsker A & Logothetis DE (2007)
Molecular characteristics of phosphoinositide binding.
Pflugers Arch – Eur J Physiol 455, 45–53.
36 Mueller-Roeber B & Pical C (2002) Inositol phospholip-

ids metabolism in Arabidopsis . Characterized and puta-
tive isoforms of inositol phospholipids kinase and
phosphoinositide-specific phospholipase C. Plant Phys-
iol 130, 22–46.
37 Dove SK, Cooke FT, Douglas MR, Sayers LG, Parker
PJ & Michell RH (1997) Osmotic stress activates phos-
phatidylinositol-3,5-bisphosphate synthesis. Nature 390,
187–192.
38 Meijer HJG, Berrie CP, Iurisci C, Divecha N, Musgrave
A & Munnik T (2001) Identification of a new polyphos-
phoinositide in plants, phosphatidylinositol 5-monopos-
phate (PtdIns5P), and its accumulation upon osmotic
stress. Biochem J 360, 491–498.
39 Meijer HJG, Divecha N, van den Ende H, Musgrave A
& Munnik T (1999) Hyperosmotic stress induces rapid
synthesis of phosphatidyl-d-inositol 3,5-bisphosphate in
plant cells. Planta 208, 294–298.
40 Matsubara M, Nakatsu T, Kato H & Taniguchi H
(2004) Crystal structure of a myristoylated CAP-
23 ⁄ NAP22 N-terminal domain complexed with
Ca
2+
⁄ calmodulin. EMBO J 23, 712–718.
41 Chen R, Kang VH, Chen J, Shope JC, Torabinejad J,
DeWald DB & Prestwich GD (2002) A monoclonal
antibody to visualize PtdIns(3,4,5)P
3
in cells. J Histo-
chem Cytochem 50, 697–708.
42 Kobae Y, Sekino T, Yoshioka H, Nakagawa T, Marti-

noia E & Maeshima M (2006) Loss of AtPDR8, a
plasma membrane ABC transporter of Arabidopsis thali-
ana, causes hypersensitive cell death upon pathogen
infection. Plant Cell Physiol 47, 309–318.
43 Ishikawa F, Suga S, Uemura T, Sato MH & Maeshi-
ma M (2005) Novel type aquaporin SIPs are mainly
localized to the ER membrane and show cell-specific
expression in Arabidopsis thaliana. FEBS Lett 579,
5814–5820.
44 Alsterfjord M, Sehnke PC, Arkell A, Larsson H, Sven-
nelid F, Rosenquist M, Ferl RJ, Sommarin M & Lars-
son C (2004) Plasma membrane H
+
-ATPase and 14-3-3
isoforms of Arabidopsis leaves: evidence for isoform
specificity in the 14-3-3 ⁄ H
+
-ATPase interaction. Plant
Cell Physiol 45, 1202–1210.
N. Nagasaki et al. A novel cation-binding myristoylated protein
FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS 2281
45 Smith PK, Krohn RI, Hermanson GT, Mallia AK,
Gartner FH, Provenzano MD, Fujimoto EK, Goeke
NM, Olson BJ & Klenk DC (1985) Measurement of
protein using bicinchoninic acid. Anal Biochem 150,
76–85.
46 Kirsch RD & Joly E (1998) An improved PCR-muta-
genesis strategy for two-site mutagenesis or sequence
swapping between related genes. Nucleic Acids Res 26,
1848–1850.

47 Matsubara M, Titani K, Taniguchi H & Hayashi N
(2003) Direct involvement of protein myristoylation in
myristoylated alanine-rich C kinase substrate
(MARCKS)–calmodulin interaction. J Biol Chem 278,
48898–48902.
A novel cation-binding myristoylated protein N. Nagasaki et al.
2282 FEBS Journal 275 (2008) 2267–2282 ª 2008 The Authors Journal compilation ª 2008 FEBS