Plant Cell Physiol. 48(11): 1558–1566 (2007)
doi:10.1093/pcp/pcm126, available online at www.pcp.oxfordjournals.org
ß The Author 2007. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists.
All rights reserved. For permissions, please email: [email protected]
Binding of Chara Myosin Globular Tail Domain to Phospholipid Vesicles
Shun-ya Nunokawa 1, Hiromi Anan 1, Kiyo Shimada 1, You Hachikubo 1, Taku Kashiyama
and Keiichi Yamamoto 1, *
1
2
1, 2
, Kohji Ito
1
Department of Biology, Chiba University, Yayoicho, Inage-ku, Chiba, 263-8522 Japan
Department of Pharmacology, Juntendo University, Medical School, Hongo, Bunkyo-ku, Tokyo, 113-8500 Japan
Binding of Chara myosin globular tail domain to
phospholipid vesicles was investigated quantitatively. It was
found that the globular tail domain binds to vesicles made
from acidic phospholipids but not to those made from neutral
phospholipids. This binding was weakened at high KCl
concentration, suggesting that the binding is electrostatic by
nature. The dissociation constant for the binding of the
globular tail domain to 20% phosphatidylserine vesicles
(similar to endoplasmic reticulum in acidic phospholipid
contents) at 150 mM KCl was 273 nM. The free energy
change due to this binding calculated from the dissociation
constant was 37.3 kJ mol1. Thus the bond between the
globular tail domain and membrane phospholipids would not
be broken when the motor domain of Chara myosin moves
along the actin filament using the energy of ATP hydrolysis
(G80 ¼ 30.5 kJ mol1). Our results suggested that direct
binding of Chara myosin to the endoplasmic reticulum
membrane through the globular tail domain could work
satisfactorily in Chara cytoplasmic streaming. We also
suggest a possible regulatory mechanism of cytoplasmic
streaming including phosphorylation-dependent dissociation
of the globular tail domain from the endoplasmic reticulum
membrane.
Keywords: Ca2þ regulation — Chara corallina —
Cytoplasmic streaming — Globular tail domain — Plant
myosin — Phospholipid binding.
Abbreviations: BSA, bovine serum albumin; DTT, dithiothreitol; ER, endoplasmic reticulum; GST, glutathione S-transferase; GTD, globular tail domain; PC, phosphatidylcholine; PE,
phosphatidylethanolamine; PI, phosphatidylinositol; PIP2, phosphatidylinositol 4,5-bisphosphate; PMSF, phenylmethylsulfonyl
fluoride; PS, phosphatidylserine; TAME, Na-p-tosyl-L-arginine
methyl ester hydrochloride; TPCK, N-p-tosyl-L-phenylalanine
chloromethylketone.
Introduction
Myosin is a molecular motor that moves along actin
filaments using ATP as the energy source. Phylogenetic
analysis of all the myosins sequenced so far revealed that
there are at least 24 classes of myosin (Foth et al. 2006),
though only three of them (classes VIII, XI and XIII) were
found in plants (Knight and Kendrick-Jones 1993,
Kinkema and Schiefelbein 1994, Vugrek et al. 2003).
Several lines of evidence suggested that plant myosin XI
plays important functional roles within cells in driving
actin-based motility, such as intracellular vesicle transport
and cytoplasmic streaming (Reddy 2001, Yamamoto 2007).
Association with the organelle membrane is essential for its
proper localization and function. However, little is known
about the manner of binding of plant myosin XI to the
membrane.
Detailed studies on this interaction with organelle
membrane have been carried out on animal myosin V. It is
known that each organelle has a specific receptor for the
globular tail domain (GTD) of certain myosin V molecules,
which allows it to be transported to its destination by the
movement of the motor domain along actin filaments
(Govindan et al. 1995, Hill et al. 1996, Catlett and Weisman
1998, Schott et al. 1999, Beach et al. 2000, Yin et al. 2000,
Hoepfner et al. 2001, Boldogh et al. 2004, Itoh et al. 2004,
Pashkova et al. 2006). Plant myosin XI has structural
similarity to myosin V and has a similar GTD (Yamamoto
et al. 1995, Kashiyama et al. 2000, Morimatsu et al. 2000,
Tominaga et al. 2003). Hashimoto et al. (2005) reported
that an antibody raised against a peptide with a sequence
specific for Arabidopsis myosin XI (MYA2) co-localized
with peroxisomes, suggesting specific interaction of MYA2
with this organelle. Reisen and Hanson (2007) recently
expressed six Arabidopsis myosin XI tails fused with yellow
fluorescent protein (YFP) in various plant cells and
observed their association with unidentified vesicles that
move at a velocity of about 1 mm s1.
On the other hand, it is known that animal myosin I
can bind directly to the phospholipid bilayer through the
interaction of its basic tail domain with the head group
of acidic phospholipids such as phosphatidylserine (PS;
Adams and Pollard 1989, Hayden et al. 1990, Doberstein
and Pollard 1992, Zot et al. 1992). It was shown that myosin
XI from the freshwater alga Chara corallina also bound to
PS vesicles (Yamamoto et al. 1995). This myosin is known
to generate very fast cytoplasmic streaming in Chara cells
(Yamamoto et al. 1994, Kashiyama et al. 2000, Ito et al.
2003). Hydrodynamic considerations suggested that
*Corresponding author: E-mail, [email protected]; Fax, þ81-43-290-2809.
1558
Phospholipid binding of Chara myosin
Chara myosin must move with bound organelles such as
the endoplasmic reticulum (ER) to generate bulk water flow
(Nothnagel and Webb 1982).
Specific association of myosin with a particular
organelle is necessary for the transport of organelles to
their destination. However, such an association with a
specific organelle might not be necessary for Chara myosin
if its function was only for cytoplasmic streaming. In this
study, we expressed the GTD of Chara myosin fused with
glutathione S-transferase (GST) and performed quantitative investigation of its binding to phospholipid vesicles to
assess the possibility of its binding to organelles through
phospholipids. We examined the GTD because it is the
organelle-binding domain in most of myosins known so far
(Kieke and Titus 2003).
Results
Globular tail domain construct
Chara myosin consists, from the N-terminus, of a
motor domain having an ATPase site and an actin-binding
site, a neck domain comprising six tandem repeats of IQ
motifs (light chain-binding sites), an a-helical coiled-coil
domain leading to dimer formation, and a globular tail
domain (AB007459; Kashiyama et al. 2000). Analysis with
COILS (Lupas 1996) revealed that the coiled-coil domain
begins at Lys875 and ends at Val1,635. We assumed,
therefore, that the GTD is from Val1,636 to Ala2,182.
The cDNA region encoding Gly1,639–Ala2,182 of Chara
myosin was fused with GST. The fused protein is composed
of residues 1–221 of GST (all the amino acids residues of
GST), a linker (EVLFQGPLGSELEICSWYR) and residues 1,639–2,182 of Chara myosin. This fusion protein
(calculated mol. wt. 89,392 Da) was expressed and purified
as described in Materials and Methods.
Binding of Chara myosin GTD to phospholipid vesicles
It was reported that native Chara myosin co-sedimented with PS vesicles but not with phosphatidylcholine (PC)
vesicles (Yamamoto et al. 1995). We first examined if
Chara myosin interacted with PS vesicles through the GTD.
A binding assay using the GTD fused with GST clearly
showed that the GTD interacts with PS but not with
PC (Fig. 1), suggesting that the phospholipid binding
characteristics of Chara myosin are conferred by the
GTD. Binding assays with other phospholipids revealed
that the GTD interacts with acidic phospholipids such as
PS and phosphatidylinositol (PI), but not with neutral
phospholipids such as PC and phosphatidylethanolamine
(PE) (Fig. 1). Since we used a fusion protein of GTD
and GST, we tested if GST co-sedimented with these
phospholipid vesicles. It was found that GST itself did
1559
PS
S
PI
P
S
PC
P
S
PE
P
S
P
220 K
97 K
GTD
67 K
BSA
43 K
30 K
Fig. 1 Binding of GTD to various phospholipids. GTD fused with
GST (0.7 mM) was incubated with vesicles made from phosphatidylserine (PS), phosphatidylinositol (PI), phosphatidylcholine (PC)
and phosphatidylethanolamine (PE) in 80 mM KCl, 0.5 mM
DTT, 0.5 mg ml1 BSA and 10 mM Tris–HCl pH 7.4 at 258C.
The concentration of phospholipids was 300 mM. BSA was
included to prevent non-specific adsorption of the GTD to the
centrifuge tubes. The mixture was centrifuged at 95,000g for
20 min at 258C to separate bound from unbound GTD. Both the
supernatant (S) and pellet (P) were analyzed by SDS–PAGE using
12% gels. The positions of molecular weight markers are shown on
the left.
not co-sediment with any of the phospholipids (data not
shown).
Effect of ionic strength on the interaction between the GTD
and PS vesicles
Because the charge of the phospholipid is an important
factor for the interaction between the GTD and phospholipid vesicles, the electrostatic interaction is a candidate for
the attractive force between them, as in the case of myosin I.
To test this possibility, we examined the effect of KCl
concentration on the interaction between the GTD (0.7 mM)
and PS vesicles (300 mM). The amount of the GTD which
co-sedimented with PS vesicles started to decrease at KCl
concentrations 4150 mM and only 10% of the GTD
co-sedimented at 500 mM KCl (Fig. 2).
Binding affinity of GTD for PS vesicles
As mentioned before, myosin I binds to PS vesicles
with very high affinity (Kd ¼ 300 nM; Hayden et al. 1990).
To estimate the binding affinity of the Chara myosin GTD
to PS vesicles, we measured the dissociation constant for the
binding of the GTD for PS vesicles under ionic conditions
close to those in Chara cells [150 mM KCl, 2 mM MgCl2,
1 mM dithiothreitol (DTT), 0.5 mg ml1 bovine serum
albumin (BSA) and 20 mM HEPES pH 7.4]. It was found
that the dissociation constant was 72 25 nM and the
1560
Phospholipid binding of Chara myosin
0.7
100
0.5
40
20
0.4
16
Bound / Free
60
Bound GTD (µM)
GTD bound (%)
0.6
80
0.3
0.2
0
100
200
300
400
500
maximum binding capacity was (0.120 0.017) 101 mol
GTD mol1 PS (Fig. 3). The dissociation constant for
Chara myosin GTD was much smaller than that for myosin
I even though the binding assay was done at a KCl
concentration much higher than that used for myosin I
(75 mM KCl, 2.5 mM MgCl2, 1 mM EGTA and 10 mM
imidazole pH 7.5). Considering the effect of ionic strength
on the interaction between GTD and PS vesicles as
mentioned above (Fig. 2), the dissociation constant would
be much smaller than 72 nM at 75 mM KCl.
Affinity for mixed PS vesicles
It is well known that the content of acidic phospholipids in natural organelle membranes is not 100%, and
the acidic phospholipid content in bean cotyledon membrane vesicles is reported to be 20% (Allen et al. 1971).
We therefore measured the binding of the GTD to vesicles
made from 20% PS and 80% PC. The GTD concentration
was varied from 0.05 to 10 mM at a mixed phospholipid
concentration of 300 mM. We raised the concentration of
mixed phospholipids to 300 mM to make the PS concentration the same as that used in the above-mentioned
experiment (60 mM) and to make a measurable amount
of the GTD co-sediment with the mixed phospholipid
vesicles. Analysis of the amount of co-sedimented GTD
showed that the dissociation constant for the binding to the
mixed phospholipid vesicles was 273 34 nM and the
maximum binding capacity was (0.930 0.079) 103 mol
GTD mol1 of the mixed phospholipids (Fig. 4).
4
0
0.2
0.4
0.6
0.8
Bound (µM)
0
0
0.1
0.2
0.3
0.4
0.5
0.6
Free GTD (µM)
Fig. 3 Binding affinity of the GTD for PS vesicles. The GTD
(0.05–1 mM) was incubated with fixed amount of PS vesicles
(60 mM) in 150 mM KCl, 2 mM MgCl2, 1 mM DTT, 0.5 mg ml1 BSA
and 20 mM HEPES pH 7.4, and the bound GTD was separated
from the unbound GTD by centrifugation. The amount of
co-sedimented GTD was determined by SDS–PAGE and the
densitometry of stained gels. Data were fit to hyperbolae by
using KALEIDAGRAPH (Synergy Software). Binding data are also
plotted according to Scatchard (inset). The dissociation constant
and maximum binding capacity were 72 25 nM and
(0.120 0.017) 101 mol GTD mol1 PS, respectively.
0.25
0.2
Bound GTD (µM)
Fig. 2 Effect of KCl concentration on the binding of the GTD to PS
vesicles. The GTD (0.7 mM) and PS vesicles (300 mM) were
incubated in varying concentrations of KCl, 0.5 mM DTT,
0.5 mg ml1 BSA and 10 mM Tris–HCl pH 7.4 at 258C, and the
bound GTD was separated from the unbound GTD by centrifugation. The amount of co-sedimented GTD was determined by
SDS–PAGE and the densitometry of stained gels.
8
0
0.1
KCl concentration (mM)
0.15
1.2
Bound / Free
0
12
0.1
0.8
0.4
0.05
0
0
0.1
0.2
0.3
Bound (µM)
0
0
0.2
0.4
0.6
0.8
1
Free GTD (µM)
Fig. 4 Binding affinity of GTD for 20% PS vesicles. The GTD
(0.05–1 mM) was incubated with vesicles made from 20% PS and
80% PC (total 300 mM) in 150 mM KCl, 2 mM MgCl2, 1 mM DTT,
0.5 mg ml1 BSA and 20 mM HEPES pH 7.4, and the bound GTD
was separated from the unbound GTD by centrifugation. The
amount of bound and unbound GTD was determined and
analyzed as in Fig. 3. The dissociation constant and maximum
binding capacity were 273 34 nM and (0.930 0.077) 103
mol GTD mol1 of mixed phospholipids, respectively.
Phospholipid binding of Chara myosin
1561
Bound construct (µM)
0.12
0.08
1.2
0.06
Bound \ Free
Bound GTD (µM)
0.1
0.04
0.02
0.1
0.4
AAAA
0.4
0
0
0
RAAR
0.2
0.8
0
0
Wild type
0.6
0.04
0.08
Bound (µM)
0.2
0.3
Free GTD (µM)
0.4
0.12
0.5
Fig. 5 Binding affinity of GTD for 2% PIP2 vesicles. The GTD
(0.05–0.5 mM) was incubated with vesicles made from 2% PIP2 and
98% PC (total 300 mM) in 150 mM KCl, 2 mM MgCl2, 1 mM DTT,
0.5 mg ml1 BSA and 20 mM HEPES pH 7.4, and the bound GTD
was separated from the unbound GTD by centrifugation. The
amount of bound and unbound GTD was determined and
analyzed as in Fig. 3. The dissociation constant and maximum
binding capacity were 157 17 nM and (0.493 0.013) 103
mol GTD mol1 of mixed phospholipid, respectively.
Affinity for phosphatidyl inositol 4,5 bisphosphate
It is reported that one subclass of myosin I, myo1c,
interacts specifically with phosphatidylinositol 4,5-bisphosphate (PIP2) and its binding to the plasma membrane is
regulated by Ca ions and phospholipase C (Hokanson and
Ostap 2006, Hokanson et al. 2006). We therefore studied
the binding of the GTD (0.05–0.5 mM) to vesicles made
from 2% PIP2 and 98% PC (total 300 mM). It was found
that the dissociation constant was 157 17 nM and the
maximum binding capacity was (0.493 0.013) 103 mol
GTD mol1 of the mixed phospholipids (Fig. 5).
Binding of mutant GTDs to PS vesicles
To investigate the site of electrostatic interaction, an
arginine cluster of the GTD from Arg1,870 to Arg1,873 was
altered to reduce its positive charge. Mutants RAAR and
AAAA had the sequences Arg-Ala-Ala-Arg and Ala-AlaAla-Ala, respectively, instead of Arg-Arg-Arg-Arg in the
wild-type GTD. The manner of binding of these mutant
GTDs to PS vesicles was examined and compared with that
of the wild-type GTD (Fig. 6). As seen in the figure, the
binding affinity decreased drastically as the charge
decreased, while the maximum binding capacity remained
almost unchanged. The dissociation constants for the wild
type, RAAR and AAAA were 72, 408 and 1,280 nM,
respectively. The maximum binding capacities for the wild
type, RAAR and AAAA were 0.012, 0.01 and 0.014 mol
protein mol1 PS, respectively.
0
0.2
0.4
0.6
0.8
1
Free construct (µM)
1.2
1.4
Fig. 6 Binding of mutant GTDs to PS vesicles. Mutant GTDs
(0.05–1 mM) were incubated with PS vesicles (60 mM) in 150 mM
KCl, 2 mM MgCl2, 1 mM DTT, 0.5 mg ml1 BSA and 20 mM HEPES
pH 7.4, and bound and unbound mutant GTDs were separated
by centrifugation. The amount of bound and unbound mutant
GTD was determined and analyzed as in Fig. 3. Data for the wildtype GTD were those shown in Fig. 3. The dissociation constants
for the wild type, RAAR and AAAA were 72, 408 and 1,280 nM,
respectively. The maximum binding capacities for the wild
type, RAAR and AAAA were 0. 12 101, 0. 12 101 and 0.
14 101 mol protein mol1 PS, respectively.
Discussion
Number of PS molecules interacting with the GTD
We showed in this study that the Chara myosin
GTD binds strongly to acidic phospholipids through an
electrostatic interaction (Figs. 1, 2). The affinity of the GTD
for PS vesicles was much higher than that of myosin I whose
direct interaction with plasma membrane phospholipid
is well documented. The dissociation constant for the
binding of brush border myosin I to PS vesicles was 300 nM
and this value was obtained at 75 mM KCl. Our assay
was done at a KCl concentration much closer to the
physiological level (150 mM) and the dissociation constant
was 72 nM (Fig. 3). The maximum binding capacity of
PS vesicles for the GTD was 12 103 mol mol1 PS, and
this value was of a similar order to that for brush border
myosin I (5.5 103 mol mol1 PS). The value suggests
that 1 mol of GTD occupies an area composed of about
40 molecules of PS if we consider the bilayer structure of
the phospholipid vesicle membrane. This number is, of
course, not the number of PS molecules interacting
electrostatically with the GTD. The free energy change
(G80 ) due to the binding of the GTD to PS vesicles can
be calculated from the binding constant K using the
following equation:
G ¼ RT ln K
1562
Phospholipid binding of Chara myosin
where R and T are the gas constant and absolute
temperature, respectively. Since the binding constant is the
inverse of the dissociation constant Kd, this equation can be
rewritten as follows:
G ¼ RT ln K d
The free energy change due to the binding is, therefore,
40.6 kJ mol1. The energy of electrostatic interaction in
water is not as high as in a vacuum and is 510 kJ mol1
(Lodish et al. 2004). Therefore, the interaction of 4–5
positively charged amino acids with the head groups of PS
would be sufficient for the binding of the GTD to PS
vesicles. The maximum binding capacity of 2% PIP2 vesicles
for the GTD is relevant to this question. Although one to
one binding of the myo1c tail to PIP2 was suggested
(Hokanson and Ostap 2006), the Chara myosin GTD does
not seem to be tethered by a single PIP2. If the binding of
the GTD to PIP2 is one to one, the maximum binding
capacity should be 0.02 mol GTD mol1 of mixed phospholipid containing 2% PIP2. The value of the maximum
binding capacity of 0.49 103 mol mol1 (Fig. 5) was
much closer to 0.02 0.02, suggesting that at least two
closely associated PIP2 molecules are required for the
binding. Since the effective charge of PIP2 is 4 at neutral
pH (McLaughlin et al. 2002), the number of molecules of PS
(with a charge of 1) interacting with the GTD should be
44. The number of 44 is supported by the maximum
binding capacity of 20% PS vesicles for the GTD (Fig. 4)
because the value 0.930 103 mol mol1 is between (0.2)4
and (0.2)5.
The calculated area on the PS vesicle surface probably
corresponds to a space physically occupied by the GTD,
which would interfere with the binding of other GTD
molecules. Our finding that the maximum binding capacity
of PS vesicles for RAAR and AAAA mutants was almost
the same as that for the wild type supports this idea. Since
the center-to-center distance of phospholipid molecules in a
bilayer is about 1 nm (Dufrene et al. 1997), the size of the
space is about 40anm2. It is reported that the GTD of
myosin V has a slightly elongated structure (10 4.5 nm;
Pashkova et al. 2006). The Chara myosin GTD together
with GST also has an elongated structure and gyrates by
Brownian motion around the binding site, which excludes
the binding of other GTD molecules from the space of
about 40 nm2.
The organelle to which Chara myosin GTD binds
It is generally assumed that the organelle being pulled
by Chara myosin is the ER because it exists most
abundantly in cells and its well-developed meshwork
structure is suitable for generating bulk flow of cytosol.
Electron microscopy revealed that the tips of the ER
membrane were attached to actin bundles in rapidly frozen
cytoplasm of Chara (Kachar and Reese 1988). However, if
the Chara myosin GTD binds to acidic phospholipids in the
ER membrane, it should also bind to the vacuolar
membrane and plasma membrane. Most of the Chara
myosin molecules were localized around actin bundles when
observed by immunofluorescence microscopy (Yamamoto
et al. 2006) and their association with the vacuolar
membrane or plasma membrane was scarcely observed
(unpublished observation). The reason may simply be that
the affinity for actin recruits the Chara myosin motor
domain around the actin bundles while the GTD interacts
with acidic phospholipids in the membrane of various
organelles. The amount of myosin on the vacuolar
membrane and plasma membrane will be small in the
final equilibrium state because they are far from the actin
bundles and because the head to tail length of the Chara
myosin molecule is only 0.1 mm (Yamamoto et al. 1995).
The Chara myosin GTD does not bind to chloroplasts,
though they are the closest to the actin bundles, probably
because the lipid composition of the outer membrane of
their envelope is very different from that of other internal
membranes. It is reported that the outer membrane of
spinach chloroplast envelope contains large amounts of
galactolipids (monogalactosyl diacylglycerol and digalactosyl diacylglycerol) and does not contain PE which exists
commonly in the cytosolic leaflet of the internal and plasma
membrane (Block et al. 1983). The polar lipid composition
of the chloroplast envelope is similar to that of blue-green
algae, except for PC (Murata et al. 1981), which probably
reflects the evolutionary origin of chloroplasts.
The target of myosin generating cytoplasmic streaming
in other plant cells may also be the ER because of the
reasons mentioned above. However, the manner of binding
to the ER membrane and the regulatory mechanism
of cytoplasmic streaming may not be the same as in
Chara. It is possible that other plant myosins bind to the
ER membrane through receptors for the GTD because
association of one subclass of myosin XI with a specific
organelle (peroxisome) is reported in higher plants
(Hashimoto et al. 2005).
As discussed above, the Chara myosin GTD can bind
to acidic phospholipids in the membrane of any organelles,
and it would bind many kinds of organelles when examined
in vitro. It is the organization of actin filaments in Chara
cells that makes the ER the most favorable organelle for the
GTD to bind. We discuss in the following sections the
binding between the GTD and ER membrane using data
obtained from the binding of the GTD to 20% PS vesicles.
Bond strength between GTD and the ER membrane
The Chara myosin GTD could also bind strongly
to 20% PS and 2% PIP2 vesicles under conditions close to
Phospholipid binding of Chara myosin
those in Chara cells (Figs. 4, 5). Since PIP2 is synthesized
from PI in the plasma membrane and the majority of PIP2
localizes in the plasma membrane (Roth 2004), we
estimated the bond strength between the GTD and ER
membrane from the free energy change (G80 ) due to the
binding of the GTD to 20% PS vesicles. The free energy
change calculated from the dissociation constant as mentioned above was 37.3 kJ mol1. Thus the energy required
to dissociate the GTD from the ER membrane is about
37.3 kJ mol1. This value is quite important in assessing
whether the direct binding of the GTD to phospholipids of
the ER membrane works functionally in Chara cells or not
because the binding force must be stronger than the force
generated by the interaction between the motor domain of
Chara myosin and actin. If this binding force was weak,
myosin could not pull the ER to produce bulk water flow.
Since the energy source for the motion of myosin is ATP,
it would not exceed the energy liberated by ATP hydrolysis
(30.5 kJ mol1). Our calculation suggests that the bond
between the GTD and ER membrane in Chara cells is
strong enough to allow the ER to be pulled by myosin for
cytoplasmic streaming.
Amount of Chara myosin molecules associated with the ER
Another important point to be considered is whether
the amount of ER-bound Chara myosin is enough to
maintain the flow of cytoplasm. Energy calculation
suggested that only 1% of existing myosin molecules is
sufficient to generate the force required for cytoplasmic
streaming in Chara cells even if the efficiency of transducing
chemical energy of ATP to mechanical force is as low as
10% (Yamamoto et al. 2006). The amount of ER-bound
myosin can be calculated if we can estimate the amount of
ER membrane in Chara cells because the myosin content
has already been determined (2 107 M; Yamamoto et al.
2006). We made two assumptions to estimate the amount of
ER membrane. The first is that the ER membrane is a
meshwork filling the cytoplasm of 10 mm thickness (Kachar
and Reese 1988) and, when being projected radially, its area
is about the same as that of the surface area of Chara cells.
The second is that the acidic phospholipid content of the
ER membrane of Chara is 20% and we can use values of the
Kd and the maximum binding capacity for the binding of
GTD to 20% PS vesicles (Fig. 4). An internodal cell of
Chara with a length of 10 cm and a diameter of 1 mm has a
surface area of 3.16 102 mm2. Considering the tubular
structure of the ER membrane and the area occupied by one
phospholipid being about 1 nm2 (Dufrene et al. 1997),
the number of phospholipid molecules in this imaginary
cell is 1.26 1015. The volume of cytoplasm in the
internodal cell of Chara is about 5% of the total cell
volume due to a large central vacuole, and is 3.93 106 l.
We can calculate the concentration of phospholipid in the
1563
cytoplasm by dividing the number of phospholipid molecules by both the Avogadro number and the volume of
cytoplasm. The phospholipid concentration thus calculated
is 5.35 104 M. Since the maximum binding capacity
of 20% PS vesicles is 0.93 103 mol mol1 (Fig. 4), the
concentration of binding site is 4.98 107 M. From
the dissociation constant of 2.73 107 M (Fig. 4) and the
myosin concentration of 2.0 107 M (Yamamoto et al.
2006), the concentration of bound myosin was calculated to
be 1.17 107 M. This value suggests that about 60% of
existing myosin molecules bind to the ER membrane
through electrostatic interaction. Even if the amount of
ER membrane is half of the assumed value, about 40% of
myosin molecules still bind to the ER membrane. Since
these values are for the monomeric GTD, native myosin
with two GTDs would bind to the ER membrane more
tightly and the amount of bound myosin molecules would
be much larger than these values. If myosin molecules are
recruited around actin bundles as we reasoned previously,
the myosin concentration around actin bundles would be
much higher than 2 107 M and the amount of ER-bound
myosin becomes much larger than that. These calculations
suggest that the amount of Chara myosin molecules bound
to the ER membrane through electrostatic interaction is
sufficient to maintain cytoplasmic streaming.
Sites of electrostatic interaction
Little is known about the phospholipid-binding sites on
the surface of the Chara myosin GTD. Since charge
reduction of the sequence from Arg1,870 to Arg1,873
drastically lowered its affinity for PS vesicles (Fig. 6), part
of the binding site may be this arginine-rich sequence.
However, there must be other binding sites because the
AAAA mutant could still bind, though weakly, to PS
vesicles (Fig. 6), and interaction with 44 PS molecules is
suggested, as discussed above. There are other sequences
rich in positively charged amino acids in the GTD
(AB007459), e.g. Lys-Lys-Ser-Lys (from 1,765 to 1,768)
and Arg-Leu-Lys-Lys (from 1,934 to 1,937). These
sequences may also serve as binding sites for acidic
phospholipids.
Regulatory mechanism of cytoplasmic streaming
If the membrane binding of the Chara myosin GTD is
solely electrostatic in nature, phosphorylation of amino acid
side chains near the basic amino acid cluster would weaken
the binding, as in the case of RAAR and AAAA mutants
(Fig. 6), and cause dissociation of moving myosin from the
ER membrane. Since movement of myosin alone cannot
generate bulk flow of water (Nothnagel and Webb 1982),
such phosphorylation would stop cytoplasmic streaming in
Chara cells. It is known that the cytoplasmic streaming
stops when the intracellular Ca2þ concentration rises above
1564
Phospholipid binding of Chara myosin
106 M (Williamson and Ashley 1982, Kikuyama and
Tazawa 1982). Tominaga et al. (1987) suggested that Ca2þ
regulation involves a phosphorylation–dephosphorylation
mechanism. In their scheme, streaming is inhibited when a
protein component is phosphorylated by Ca2þ-dependent
protein kinase, and streaming is activated when the same
component is dephosphorylated. There are sequences
around the arginine cluster of the Chara myosin GTD
that match the consensus sequences of the Ca2þ-dependent
protein kinase phosphorylation site (S/T X R/K) and
the Ca2þ/calmodulin-dependent protein kinase II phosphorylation site (R X X S/T) (Fig. 7). Regulation of
cytoplasmic streaming through the phosphorylation of
the GTD by these kinases fits well with the scheme
proposed by Tominaga et al. (1987), and a similar
phosphorylation-dependent dissociation of the myosin V
tail from organelles was reported (Karcher et al. 2001).
Movement of Chara myosin along the actin filament
without bound ER membrane may seem a waste of
energy, but Chara myosin will dissociate and diffuse away
from the actin filament immediately after one or two steps
since it is a non-processive motor protein (Awata et al.
2001, Ito et al. 2007). Dissociated myosin molecules will
rebind weakly to the ER membrane around them because
there is no force to pull them apart from the ER membrane,
or they may rebind strongly to the ER membrane after
being dephosphorylated.
If phosphorylation of the GTD is a regulatory
mechanism of cytoplasmic streaming in Chara cells, it will
work efficiently when only myosin molecules moving along
the actin filament are phosphorylated. Such selective
phosphorylation will be possible if kinase molecules are
tethered near actin filament bundles. McCurdy and
Harmon (1992) studied the localization of Ca2þ-dependent
protein kinase in Chara cells by immunofluorescence
microscopy and showed its localization on actin filament
bundles and on small organelles associated with the
bundles.
We, of course, noticed that this regulatory mechanism
is one of many possible mechanisms. Although the motor
activity of Chara myosin is known not to be inactivated
directly by Ca2þ (Awata et al. 2001), it can be inactivated by
Ca2þ-dependent phosphorylation of the motor domain.
Phosphorylation of the motor domain may allow it to
interact with the GTD, and this interaction may dissociate
the GTD from the ER. We previously reported that the
affinity of Chara myosin for PS vesicles becomes high (more
myosin molecules co-sedimented) in the presence of Ca2þ
(Yamamoto et al. 1995). However, we did not observe any
effect of Ca2þ on the affinity of the GTD for PS vesicles
when we used sucrose-loaded vesicles (data not shown).
Since divalent cations are known to catalyze lipid aggregation and fusion (Wilschut et al. 1985), our previous report
1855
1890
LQRTLKTGGGGGTTPRRRRQATLFGRMTQRFSSQQE
Ca2+ dependent protein kinase phosphorylation site
Ca2+ /calmodulin dependent protein kinase phosphorylation site
Fig. 7 Phosphorylation sites near the arginine cluster. Filled
triangles indicate threonine residues that can be phosphorylated by
Ca2þ-dependent protein kinase (S/T x R/K). Open triangles indicate
serine or threonine residues that can be phosphorylated by Ca2þ/
calmodulin-dependent protein kinase II (R x x S/T).
on the effect of Ca2þ may have been due to changes in
vesicle size and structure. Sucrose-loaded vesicles sedimented equally regardless of Ca2þ, but a larger amount of
unloaded vesicles probably sedimented with bound myosin
in the presence of Ca2þ in our previous study. A similar
observation was reported for myosin I by Hokanson and
Ostap (2006).
Materials and Methods
Reagents
L-a-phosphatidylserine (PS), L-a-phosphatidylcholine (PC),
L-a-phosphatidylethanolamine (PE) and PIP2 were purchased
from Avanti Polar Lipids (Alabaster, AL, USA). L-a-phosphatidylinositol (PI) was obtained from Sigma (St Louis, MO, USA).
Na-p-tosyl-L-arginine methyl ester hydrochloride (TAME),
N-p-tosyl-L-phenylalanine chloromethylketone (TPCK), leupeptin,
pepstatin A and phenylmethylsulfonyl fluoride (PMSF) were
purchased from Sigma. Glutathione–SepharoseÕ 4B is a product
of Amersham Biosciences (Buckinghamshire, UK). Malachite
green oxalate was purchased from Chroma-Gesellschaft
(Germany). SYPRO-red was obtained from Molecular Probes
Inc. (Eugene, OR, USA).
Expression and purification of the GTD of Chara myosin
The fusion gene of GST and the GTD was subcloned into the
baculovirus transfer vector pFastBac1 (Invitrogen, Carlsbad, CA,
USA) between the XbaI and XhoI sites. This plasmid (pFastBac1GST-GTD) was introduced into Escherichia coli DH10Bac cells to
produce bacmid DNA. Purified bacmid DNA encoding GST–
GTD was used to infect Sf-9 cells to produce baculovirus
containing the GST–GTD gene. The GST–GTD fusion protein
was expressed in High FiveÕ cells and purified as follows.
Cells were harvested and washed with 150 mM NaCl and 10 mM
Tris–HCl pH 7.4. Colleted cells were sonicated in a lysis buffer
containing 200 mM NaCl, 1 mM EDTA, 0.1 mM EGTA, 30 mM
Tris–HCl pH 7.4 and protease inhibitor mixture (50 mg ml1
TAME, 40 mg ml1 TPCK, 5 mg ml1 leupeptin, 3 mg ml1 pepstatin
A and 0.1 mM PMSF), and intact cells were removed by
centrifugation at 30,000g for 30 min. Supernatant was centrifuged further at 260,000g for 30 min and the resultant supernatant was mixed with glutathione–SepharoseÕ 4B resin
equilibrated with the lysis buffer and left for 4 h with gentle
agitation. The resin was packed in a column and washed first with
a 40-fold column volume of the lysis buffer and then with a
solution containing 150 mM KCl and 10 mM Tris–HCl pH 7.4.
The resin was suspended with one column volume of glutathione
Phospholipid binding of Chara myosin
solution (150 mM KCl, 20 mM glutathione, 100 mM HEPES pH
8.0), left for 30 min with gentle agitation, and the released protein
was collected. This procedure was repeated five times. Glycerol was
added to this protein solution to make 10% for preservation.
Mutant GTDs with altered electrostatic charge
The wild-type GTD has the sequence Arg-Arg-Arg-Arg from
1,870 to 1,873 (AB007459). Since this basic arginine cluster is a
candidate for the acidic phospholipid-binding site on the GTD as
shown in the Results, we altered this sequence to Arg-Ala-Ala-Arg
or Ala-Ala-Ala-Ala by PCR-based site-directed mutagenesis
(ExSite mutagenesis kit, Stratagene, La Jolla, CA, USA). The
primers used to generate the Arg-Ala-Ala-Arg mutation were 50 -G
CCGCCCAAGCAACGTTGTTCGGG-30 and 50 -CGCGGCTGG
AGTTGTCCCACCTC-30 , and those for the Ala-Ala-Ala-Ala
mutation were 50 -CGCCAAGCAACGTTGTTCGGGA-30 and
50 -CGCTGCTCGTGGAGTTGTCCCAC-30 . After verifying their
sequences, the mutant genes were subcloned into pFastBac1-GSTGTD and then expressed and purified as mentioned above for the
wild-type GST–GTD.
Preparation of phospholipid vesicles
Phospholipid vesicles were prepared according to Hayden
et al. (1990) with slight modification. Phospholipids dissolved in
chloroform–methanol were dried in vacuo and suspended in a
solution containing 176 mM sucrose and 20 mM HEPES pH 7.4 by
sonication. The large aggregate was removed by centrifugation and
the homogenous supernatant was dialyzed against 150 mM KCl
and 20 mM HEPES pH 7.4. This sucrose-loaded phospholipid
vesicle suspension was used within 3 d of preparation.
Measurement of phospholipid concentration
The concentration of phospholipid was determined by
measuring acid-labile phosphate. We placed the phospholipid
vesicle suspension (100 ml) into a test tube and evaporated the
solvent by heating with a gas burner. The test tube was cooled
and 1 ml of 9.24 M perchloric acid was added to it. Then the test
tube was sealed with a screw cap and heated at 1508C for 6 h.
The solution was diluted 30-fold with glass-distilled water. An
aliquot was withdrawn and mixed with an equal volume of a
malachite green solution containing 0.7 N HCl, 0.03% malachite
green oxalate, 0.2% sodium molybdate and 0.05% Triton X-100.
This mixture was kept at 308C for 25 min and the absorption at
650 nm was measured. The phosphate concentration was determined from the standard curve made by measuring the color
development of a known amount of phosphate.
Binding assay
Binding of the Chara myosin GTD to phospholipid vesicles
was determined by incubating vesicles made from various
phospholipids with the GTD (0.05–1 mM) in the presence of
80–500 mM KCl, 2 mM MgCl2, 0.5 mM DTT, 10 mg ml1 leupeptin, 0.7 mg ml1 pepstatin A, 0.5 mg ml1 BSA, 10 mM Tris–HCl
pH 7.4 at 258C for 20 min, followed by centrifugation at 95,000g
for 30 min at 258C to separate the vesicle-bound GTD from the
unbound form. BSA was included to prevent non-specific
adsorption of the GTD to the centrifuge tubes. Both the
supernatant and pellet were recovered and treated with SDS
sample buffer. Protein samples were resolved by SDS–PAGE and
gels were stained either with Coomassie brilliant blue or with
SYPRO-red. Gels stained with SYPRO-red were scanned by using
1565
a Molecular Imager FX (Bio-Rad, Hercules, CA, USA), and data
were analyzed by Quantity One (Bio-Rad).
Acknowledgments
This work was supported by a Grant-in-Aid for Scientific
Research and Grant-in-Aid for Scientific Research on Priority
Areas from The Ministry of Education, Culture, Sports, Science
and Technology of Japan.
References
Adams, R.J. and Pollard, T.D. (1989) Binding of myosin I to membrane
lipids. Nature 340: 565–568.
Allen, C.F., Good, P., Mollenhauer, H.H. and Totten, C. (1971) Studies on
seeds. IV. Lipid composition of bean cotyledon vesicles. J. Cell Biol. 48:
542–546.
Awata, J., Saitoh, K., Shimada, K., Kashiyama, T. and Yamamoto, K.
(2001) Effects of Ca2þ and calmodulin on the motile activity of characean
myosin in vitro. Plant Cell Physiol. 42: 828–834.
Beach, D.L., Thibodeaux, J., Maddox, P., Yeh, E. and Bloom, K. (2000)
The role of the proteins Kar9 and Myo2 in orienting the mitotic spindle
of budding yeast. Curr. Biol. 10: 1497–1506.
Block, M.A., Dorne, A.-J., Joyard, J. and Douce, R. (1983) Preparation and
characterization of membrane fractions enriched in outer and inner
envelope membranes from spinach chloroplasts. J. Biol. Chem. 258:
13281–13286.
Boldogh, I.R., Ramcharan, S.L., Yang, H.C. and Pon, L.A. (2004) A type V
myosin (Myo2p) and a Rab-like G-protein (Ypt11p) are required for
retention of newly inherited mitochondria in yeast cells during cell
division. Mol. Biol. Cell 15: 3994–4002.
Catlett, N.L. and Weisman, L.S. (1998) The terminal tail region of a yeast
myosin-V mediates its attachment to vacuole membranes and sites of
polarized growth. Proc. Natl Acad. Sci. USA 95: 14799–14804.
Doberstein, S.K. and Pollard, T.D. (1992) Localization and specificity of
the phospholipid and actin binding sites on the tail of Acanthamoeba
myosin IC. J. Cell Biol. 117: 1241–1249.
Dufrene, Y.F., Barger, W.R., Green, J.D. and Lee, G.U. (1997) Nanometerscale surface properties of mixed phospholipid monolayers and bilayers.
Langmuir 13: 4779–4784.
Foth, B.J., Goedecke, M.C. and Solsati, D. (2006) New insights into myosin
evolution and classification. Proc. Natl Acad. Sci. USA 103: 3681–3686.
Govindan, B., Bowser, R. and Novick, P. (1995) The role of Myo2, a yeast
class V myosin, in vesicular transport. J. Cell Biol. 128: 1055–1068.
Hashimoto, K., Igarashi, H., Nishimura, M., Shimmen, T. and Yokota, E.
(2005) Peroxisomal localization of a myosin XI isoform in Arabidopsis
thaliana. Plant Cell Physiol. 46: 782–789.
Hayden, S.M., Wolenski, J.S. and Mooseker, M.S. (1990) Binding of brush
border myosin I to phospholipid vesicles. J. Cell Biol. 111: 443–451.
Hill, K.L., Catlett, N.L. and Weisman, L.S. (1996) Actin and myosin
function in directed vacuole movement during cell division in
Saccharomyces cerevisiae. J. Cell Biol. 135: 1535–1549.
Hoepfner, D., van den Berg, M., Philippsen, P., Tabak, H.F. and
Hettema, E.H. (2001) A role for Vps1p, actin and the Myo2p motor in
peroxisome abundance and inheritance in Saccharomyces cerevisiae.
J. Cell Biol. 155: 979–990.
Hokanson, D.E., Laakso, J.M., Lin, T., Sept, D. and Ostap, E.M. (2006)
Myo1c binds phosphoinositides through a putative pleckstrin homology
domain. Mol. Biol. Cell 17: 4856–4865.
Hokanson, D.E. and Ostap, E.M. (2006) Myo1c binds tightly and
specifically to phosphatidylinositol 4,5-bisphosphate and inositol
1,4,5-trisphosphate. Proc. Natl Acad. Sci. USA 103: 3118–3123.
Ito, K., Ikebe, M., Kashiyama, T., Mogami, T., Kon, T. and
Yamamoto, K. (2007) Kinetic mechanism of the fastest motor protein,
Chara myosin. J. Biol. Chem. 282: 19534–19545.
Ito, K., Kashiyama, T., Shimada, K., Yamaguchi, A., Awata, J.,
Hachikubo, Y., Manstein, D.J. and Yamamoto, K. (2003)
1566
Phospholipid binding of Chara myosin
Recombinant motor domain constructs of Chara corallina myosin display
fast motility and high ATPase activity. Biochem. Biophys. Res. Commun.
312: 958–964.
Itoh, T., Toh, E.A. and Matsui, Y. (2004) Mmr1p is a mitochondrial factor
for Myo2p-dependent inheritance of mitochondria in the budding yeast.
EMBO J. 23: 2520–2530.
Kachar, B. and Reese, T.S. (1988) The mechanism of cytoplasmic streaming
in characean algal cells: sliding of endoplasmic reticulum along actin
filaments. J. Cell Biol. 106: 1545–1552.
Karcher, R.L., Roland, J.T., Zappocosta, F., Huddleston, M.J.,
Annan, R.S., Carr, S.A. and Gelfand, V.I. (2001) Cell cycle regulation
of myosin-v by calcium/calmodulin-dependent protein kinase II. Science
293: 1317–1320.
Kashiyama, T., Kimura, N., Mimura, T. and Yamamoto, K. (2000)
Cloning and characterization of a myosin from characean alga, the fastest
motor protein in the world. J. Biochem. 127: 1065–1070.
Kieke, M.C. and Titus, M.A. (2003) Myosin superamily: an overview.
In Molecular Motors. Edited by Schliwa, M. pp. 3–44. Wiley-VCH,
Weinheim, Germany.
Kikuyama, M. and Tazawa, M. (1982) Ca2þ ion reversibly inhibits the
cytoplasmic streaming of Nitella. Protoplasma 113: 241–243.
Kinkema, M. and Schiefelbein, J. (1994) A myosin from a higher plant has
structural similarities to class V myosins. J. Mol. Biol. 239: 591–597.
Knight, A.E. and Kendrick-Jones, J. (1993) A myosin-like protein from a
higher plant. J. Mol. Biol. 231: 148–154.
Lodish, H., Berk, A., Matsudaira, P., Kaiser, C.A., Krieger, M.,
Scott, M.P., Zipursky, S.L. and Darnell, J. (2004) Molecular Cell
Biology, 5th edn. Freeman & co.
Lupas, A. (1996) Prediction and analysis of coiled coil structures. Methods
Enzymol 266: 513–525.
McCurdy, D.W. and Harmon, A.C. (1992) Calcium-dependent protein
kinase in the green alga Chara. Planta 188: 54–61.
McLaughlin, S, Wang, J., Gambhir, A. and Murray, D. (2002) PIP(2) and
proteins: interactions, organization and information flow. Annu. Rev.
Biophys. Biomol. Struct. 31: 151–175.
Morimatsu, M., Nakamura, A., Sumiyoshi, H., Sakaba, N., Taniguchi, H.,
Kohama, K. and Higashi-Fujime, S. (2000) The molecular structure of
the fastest myosin from green algae, Chara. Biochem. Biophys. Res.
Commun. 270: 147–152.
Murata, N., Sato, N., Omata, T. and Kuwabara, T. (1981) Separation and
characterization of thylakoid and cell envelope of the blue-green alga
(cyanobacterium) Anacystis nidulans. Plant Cell Physiol. 22: 855–866.
Nothnagel, E.A. and Webb, W.W. (1982) Hydrodynamic models of viscous
coupling between motile myosin and endoplasm in characean algae.
J. Cell Biol. 94: 444–454.
Pashkova, N., Jin, Y., Ramaswamy, S. and Weisman, L.S. (2006) Structural
basis for myosin V discrimination between distinct cargoes. EMBO J. 25:
693–700.
Reddy, A.S.N. (2001) Molecular motors and their functions in plants.
Int. Rev. Cytol. 204: 97–178.
Reisen, D. and Hanson, M.R. (2007) Association of six YFP–myosin
XI-tail fusions with mobile plant cell organelles. BMC Plant Biol. 7:6
doi:10.1186/1471-2229-7-6
Roth, M.G. (2004) Phosphoinositides in constitutive membrane traffic.
Physiol. Rev. 84: 699–730.
Schott, D., Ho, J., Pruyne, D. and Bretscher, A. (1999) The COOH-terminal
domain of Myo2p, a yeast myosin V, has a direct role in secretory vesicle
targeting. J. Cell Biol. 147: 791–808.
Tominaga, M., Kojima, H., Yokota, E., Orii, H., Nakamori, R.,
Katayama, E., Anson, M., Shimmen, T. and Oiwa, K. (2003) Higher
plant myosin XI moves processively on actin with 35 nm steps at high
velocity. EMBO J. 22: 1263–1272.
Tominaga, Y., Wayne, R., Tung, H.Y.L. and Tazawa, M. (1987)
Phosphorylation–dephosphorylation is involved in Ca2þ-controlled
cytoplasmic streaming of characean cells. Protoplasma 136: 161–169.
Vugrek, O., Sawitzky, H. and Menzel, D. (2003) Class XIII myosins from
the green alga Acetabularia: driving force in organelle transport and tip
growth? J. Muscle Res. Cell Motil. 24: 87–97.
Williamson, R.E. and Ashley, C.C. (1982) Free Ca2þ and cytoplasmic
streaming in the alga Chara. Nature 296: 647–650.
Wilschut, J., Duzgunes, N., Hoekstra, D. and Papahadjopoulos, D (1985)
Modulation of membrane fusion by membrane fluidity: temperature
dependence of divalent cation induced fusion of phosphatidylserine
vesicles. Biochemistry 24: 8–14.
Yamamoto, K. (2007) Plant Myosins VIII, XI, and XIII. In Myosins.
Edited by Coluccio, L.M. pp. 375–390. Springer, Berlin.
Yamamoto, K., Kikuyama, M., Sutoh-Yamamoto, N. and Kamitsubo, E.
(1994) Purification of actin-based motor protein from Chara corallina.
Proc. Jpn. Acad. 70: 175–180.
Yamamoto, K., Kikuyama, M., Sutoh-Yamamoto, N., Kamitsubo, E. and
Katayama, E. (1995) Myosin from alga Chara: unique structure revealed
by electron microscopy. J. Mol. Biol. 254: 109–112.
Yamamoto, K., Shimada, K., Ito, K., Hamada, S., Ishijima, A.,
Tsuchiya, T. and Tazawa, M. (2006) Chara myosin and the energy of
cytoplasmic streaming. Plant Cell Physiol. 47: 1427–1431.
Yin, H., Pruyne, D., Huffaker, T.C. and Bretscher, A. (2000) Myosin V
orientates the mitotic spindle in yeast. Nature 406: 1013–1015.
Zot, H.G., Doberstein, S.K. and Pollard, T.D. (1992) Myosin-I moves actin
filaments on a phospholipid substrate: implications for membrane
targeting. J. Cell Biol. 116: 367–376.
(Received July 20, 2007; Accepted September 26, 2007)