(Received for publication, August 13, 1996, and in revised form, January 13, 1997)
From the Division of Biomedical Polymer Science, Institute for Comprehensive Medical Science, Fujita Health University, Toyoake, Aichi 470-11, Japan
Growth-associated protein-43 (GAP-43) is believed
to be palmitoylated near the N terminus and the modification is assumed to be involved in the membrane anchoring of the protein. However, GAP-43 isolated from bovine brain is not palmitoylated as shown by mass
spectrometric analysis, but still retains the ability to bind
phospholipids, suggesting that other parts of the molecule are involved
in the interaction. Upon addition of acidic phospholipids, purified
GAP-43 showed a conformational change from random coil to -helix as
indicated by a change in CD spectra. A synthetic peptide
corresponding to the calmodulin-binding domain showed a similar
conformational change from random coil to
-helix in the presence of
various acidic phospholipids. These results suggest that the
calmodulin-binding domain of GAP-43 is directly involved in the
GAP-43-membrane interaction and undergoes a conformational change
upon binding to phospholipid membranes. After phosphorylation by
protein kinase C, the phospholipid-induced conformational changes were
no longer observed. Structural characteristics of the
calmodulin-binding domain peptide in aqueous and hydrophobic
solvents were further studied in detail by two-dimensional
1H nuclear magnetic resonance. The results obtained
suggest that the domain assumes a nascent
-helical structure in
aqueous solution, which is stabilized under hydrophobic
environments.
GAP-43 (growth-associated protein-43, also known as B50, F1, P56, or neuromodulin)1 has been characterized as one of the major phosphoproteins in neuronal growth cone as well as in adult nerve terminals, and is thought to be involved in neurite extension as well as in neuronal plasticity and regulation of neurotransmitter release (for reviews, see Refs. 1 and 2). Although the protein has been believed to be a specific in vivo substrate of protein kinase C (PKC), recent studies on the in vivo phosphorylation sites revealed that the protein is a substrate of so-called proline-directed protein kinases as well (3, 4). Since casein kinase II has also been shown to phosphorylate GAP-43 in vitro (5, 6), physiological functions of GAP-43 such as sequestering of calmodulin (7) and signal transduction through the binding to GO (8) may be regulated through various protein kinases in a very complex manner.
GAP-43 is found associated with membrane fractions or with so-called membrane-cytoskeleton fractions (9). Since the protein is very hydrophilic and lacks any apparent membrane-binding domain (10), the interaction of the protein with membranes is thought to be effected by palmitoylation at two cysteine residues near the N terminus (11, 12). However, our recent mass spectrometric study on the in vivo post-translational modifications of GAP-43 revealed that the protein isolated from the membrane fractions of adult brain lacks the palmitoyl group (3), suggesting that part(s) of the molecule other than the palmitoyl moiety is involved in the interaction of the protein with the membrane. In fact, the involvement of the calmodulin-binding domain, which is at the same time the phosphorylation domain by PKC, in the membrane association has been suggested (13, 14).
We have previously shown that the PKC phosphorylation and
calmodulin-binding domain of myristoylated alanine-rich protein kinase
C substrate, another in vivo major substrate of PKC
belonging to the same family of acidic hydrophilic membrane-associated
proteins (15), is directly involved in the interaction of the protein with membrane phospholipids (16). Furthermore, the phosphorylation of
the domain regulates the reversible membrane association of myristoylated alanine-rich protein kinase C substrate (14, 16). Thus,
the calmodulin-binding domain of basic amphiphilic -helical nature
seems to serve as a phosphorylation- and
calmodulin-dependent membrane binding motif as well (16,
17). A similar polybasic motif found in the Src family proteins has
been implicated in the membrane association of these proteins (18,
19).
In the present study, structural characteristics of the
calmodulin-binding domain of GAP-43 were studied by circular
dichroism (CD) spectroscopy and two-dimensional 1H nuclear
magnetic resonance (NMR). Purified non-palmitoylated GAP-43 and a
synthetic peptide corresponding to the domain showed similar
conformational changes from random coil to -helix upon binding to
acidic phospholipids. The calmodulin-binding domain of GAP-43 seems to
assume a "nascent"
-helical structure in aqueous solution, which
is further stabilized under hydrophobic environments.
GAP-43 was purified from membrane fractions of bovine brain after detergent extraction as described (3). PKC was purified from bovine brain as described previously (20). A peptide (QASFRGHITRKKLKGEK) corresponding to the calmodulin-binding domain of GAP-43, named GAP peptide, was synthesized using conventional t-butoxycarbonyl chemistry using an ABI 430A peptide synthesizer (Applied Biosystems), and purified over a C18 reversed-phase column (Vydac 218TP1010, The Separations Group, Hesperia, CA) using H2O-acetonitrile gradient in the presence of 0.1% trifluoroacetic acid. Phospholipids (Avanti Polar Lipids) were suspended in 5 mM phosphate buffer (pH 7.5) and sonicated in a sonicator (Branson Sonifier 250) for 30 min. After centrifugation in a tabletop centrifuge for 20 min, the supernatant was used as unilamellar liposomes.
Preparation of Phosphorylated GAP-43 and GAP PeptidePhosphorylation of the intact GAP-43 and GAP peptide by PKC was carried out in 25 mM Tris-HCl buffer (pH 7.5) containing 10 mM MgCl2, 100 µM CaCl2, 80 µg/ml phosphatidylserine (PS), 8 µg/ml dioleoyl glycerol, and 1 mM ATP at 35 °C for 90 min. The reaction was stopped by adding 0.1% final concentration of trifluoroacetic acid. The extent of the phosphorylation was analyzed by mass spectrometry as described previously (3, 21). The phosphorylated GAP peptide was purified by reversed-phase HPLC using a Vydac C18 column (218TP52, 0.46 × 25 cm), while the phosphorylated GAP-43 protein was purified by chromatography on a Mono Q column (HR 5/5) using a linear gradient of NaCl (0-0.5 M) in 20 mM Tris-HCl buffer (pH 7.5) containing 1 mM EDTA and 1 mM dithiothreitol. The purified phosphorylated protein was concentrated by ultrafiltration using a Centricon 10 cartridge (Amicon).
Mass Spectrometric AnalysisGAP-43 purified from the membrane fractions of bovine brain by the detergent extraction method (3, 20) was digested with L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin (Worthington), and the peptides were separated over a reversed-phase HPLC column as described previously (3, 21). Fractions containing the N-terminal peptide were combined and dried with a Speed-Vac concentrator. The peptide was dissolved in 5 ml of 50% acetonitrile containing 0.1% trifluoroacetic acid. Electrospray tandem mass spectra were recorded in a PE Sciex API-III mass spectrometer in tandem mass mode. A singly-charged precursor ion was selected with Q1 for collision-induced decomposition with argon in Q2. Fragment ion spectra were obtained by scanning Q3 of the triple quadrupole system.
Circular Dichroism (CD) SpectrometryCD spectra were
recorded at 25 °C in a JASCO J-720 CD spectropolarimeter using a
0.1-cm cell. The concentration of the peptide was 20 µM
in 5 mM phosphate buffer (pH 7.3) unless otherwise
indicated. The contents of -helix,
-sheet, and
-turn
structures were estimated from the CD spectra on an IRIS Indigo XS24
workstation (Silicon Graphics, Inc.) using a CONTIN program (22)
modified by Dr. F. Arisaka, Tokyo Institute of Technology.
1H NMR spectra were recorded on a Bruker DMX-500 spectrometer operating at 500 MHz. Chemical shifts were measured relative to the methyl resonance of 4,4-dimethyl-4-silapentane-1-sulfonate, used as an internal reference. GAP peptide (5 mM) was dissolved in H2O/D2O (9/1,v/v), D2O (99.98%), H2O/D2O/trifluoroethanol (TFE)-d3 (5/1/4), or D2O/TFE-d3 (6/4). The pH of the samples was 4.0 (direct meter reading). The sequence-specific assignment of resonances was obtained from two-dimensional TOCSY (23), NOESY (24, 25), DQF-COSY with phase cycling (26) or with pulsed field gradient (27, 28), and TQF-COSY with pulsed field gradient (27, 28) spectra, which were acquired at 25 °C in the phase-sensitive mode using the time proportional phase increment technique. Various mixing times were used for NOESY (200, 100, 50, 25, and 10 ms) and TOCSY (75, 50, and 25 ms).
Water suppression was accomplished using WATERGATE (29, 30) or presaturation. A total of 512 measurements with increasing t1 values were made, and 64 transients were accumulated for each measurement. For t2, 2048 data points were taken, and the spectral widths along f2 were 5000 Hz. The data were zero filled once in the f1 dimension, and a cosine window function was used in f1 dimension before Fourier transformation, and a Gaussian function was used in f2 dimension, except for the NOESY spectra whose time domain data were multiplied by Gaussian functions in both dimensions. All spectra were processed using Bruker XWIN-NMR or MSI Felix 95.0 software packages.
Sedimentation AssayThe direct binding of GAP-43 and that of GAP peptide to phospholipid membranes were assessed by cosedimentation analysis. GAP-43 (1 µM) or GAP peptide (20 µM) was mixed with indicated amounts of multilamellar liposomes containing 20% PS and 80% phosphatidylcholine (PC) in 5 mM potassium phosphate buffer (pH 7.5) containing 150 mM KCl, and incubated at 25 °C for 30 min. After centrifugation at 200,000 × g for 60 min, protein or peptide remained in supernatants were analyzed by SDS-gel electrophoresis and Coomassie-stained bands were quantified by densitometry (Molecular Dynamics, PDSI densitometer).
Titration CalorimetryHeat production by peptide binding to unilamellar vesicles of PC containing 20% PS was measured with a MicroCal MC-2 high-sensitivity titration calorimeter (MicroCal, Northampton, MA) as described previously (31, 32). Five microliters of the lipid vesicles (13.9 mM) were injected into the calorimeter cell containing 1.3 ml of 30 µM GAP peptide in 5 mM phosphate buffer containing 150 mM KCl at 25 °C.
Other Analytical MethodsProtein concentration was determined by quantitative amino acid analysis. Purity and authenticity of the peptide during synthesis and purification, and those of the proteins were routinely monitored by mass spectrometry as described previously (3, 21).
GAP-43
purified from membrane fractions of bovine brain was digested with
trypsin and the resulting peptide mixture was analyzed by the liquid
chromatography-electrospray mass spectrometry (3, 21). One peptide with
a mass of 796.3 Da has been tentatively assigned as the N-terminal
peptide in the previous study (3). The N-terminal peptide should give a
theoretical mass of 798.3 Da, when the N-terminal acetylation is
assumed. The reversed-phase column fractions containing the peptide
were combined, and the sample was analyzed by the electrospray tandem
mass spectrometry (3). In this technique, a precursor ion corresponding
to the peptide is selected with the first quadrupole mass analyzer, and the fragments obtained by the collision with argon are analyzed using
the second mass analyzer. Most of the cleavage occur randomly in the
middle of the peptide bonds, resulting in the formation of "ladder"
mixtures (33). As shown in Fig. 1, the tandem mass spectrum of the peptide clearly indicated that the N terminus is
acetylated, and that the two Cys residues, which have been assumed to
be palmitoylated (11), are not palmitoylated at all, but form an
intramolecular disulfide bond. GAP-43 has only two Cys residues, and no
other modification has been detected by mass spectrometric analysis
(3). Since the mass spectrometric analysis of the whole protein showed
a single peak with a minor peak corresponding to a phosphorylated
species (3), we concluded that the GAP-43 protein as isolated does not
contain palmitoylated species in significant amounts.
To confirm that the non-palmitoylated GAP-43 still has the ability to
bind to phospholipid membranes, the binding was studied by
sedimentation analysis. When GAP-43 was mixed with multilamellar liposomes made of a mixture of PC (80%) and PS (20%), which mimics the compositions of biomembranes, at a physiological ionic strength, most of GAP-43 (89%) was found associated with the lipids (Fig. 2a). No binding was observed with pure PC
liposomes (data not shown). The non-palmitoylated GAP-43, therefore,
still retains the ability to bind to phospholipid membranes containing
physiologically relevant combinations of lipids. A synthetic peptide
corresponding to the calmodulin-binding domain of GAP-43 showed a
similar binding to the phospholipid membranes. With increasing
concentrations of vesicles containing PC (80%) and PS (20%), the
amounts of peptide remained in the supernatant decreased (Fig. 2,
b and c). On the other hand, no significant
binding was observed when vesicles containing pure PC was used (Fig.
2c). A quantitative analysis of the peptide-phospholipid
interaction was carried out further by high-sensitivity titration
calorimetry (31, 32). As shown in Fig. 3a,
each injection of lipid vesicles into a peptide solution in the
measuring cell produced a marked heat production. With increasing
numbers of injections the magnitude of the reaction enthalpy decreased,
since less and less peptide is available for binding. From the
cumulative heat of reaction plotted against the injection numbers which
is correlated to the lipid concentration, one can calculate
thermodynamic parameters (32). The partition constant thus obtained,
Kp = 3.35 × 103
M1, suggests that GAP peptide has an affinity
comparable to other membrane-binding peptides (14, 32). A similar study
with the whole GAP-43 protein was not feasible due to the amounts of
samples needed and its tendency to aggregate at high
concentrations.
Conformational Change of GAP-43 Induced by Phospholipid Binding
To get more insights into the mode of the
GAP-43-phospholipid interactions, effects of phospholipids on the
conformation of GAP-43 were studied by CD spectroscopy. Purified GAP-43
in aqueous buffer showed a CD spectrum with a single large negative
peak at around 200 nm (Fig. 4a), suggesting
that the protein has a random structure. The addition of an acidic
phospholipid, phosphatidylglycerol (PG), caused a small but significant
change in the CD spectrum, which was dependent on the concentration of
PG added. There was a decrease in the intensity and a shift of the
largest negative peak around 200 nm with a concomitant increase in the
negative ellipticity between 210 and 230 nm. The characteristics of the change can be more clearly observed in the difference spectrum (Fig.
4a, inset). A broad negative peak between 210 and 230 nm with a positive peak below 200 nm suggests that a part of the protein
molecule underwent a conformational change from random coil to
-helix.
Interaction of Calmodulin-binding Domain Peptide with Phospholipids
The CD spectrum of the GAP peptide showed again a
typical CD spectrum of random coil with a single large negative peak at around 198 nm (Fig. 4b). When increasing amounts of PG were
added to the peptide, a broad negative peak between 210 and 230 nm due to -helix increased accordingly (Fig. 4b). The maximal
extent of the change in the CD spectra per mole of molecule was
comparable for the intact protein (0.98 × 105
degree·cm2·dmol
1) and the peptide
(1.22 × 105
degree·cm2·dmol
1), suggesting that only
the calmodulin-binding domain interacts with the lipids and undergoes
the conformational change.
Various phospholipids at a fixed concentration were mixed
with GAP peptide, and CD spectra were recorded to study the
phospholipid specificity. Although neutral phospholipids such as PC did
not affect the spectrum significantly, all acidic phospholipids
examined caused similar changes in the CD spectra (Fig.
5). Although the final spectral change observed at
saturating concentrations were very similar regardless of the
phospholipid used (data not shown), the affinities of various acidic
phospholipids varied appreciably, suggesting that the binding of the
peptide to the phospholipids involves not only ionic but also specific
structural interactions. Among various acidic phospholipids,
phosphatidic acid showed the highest affinity to GAP peptide. When
ionic strength of the buffer was increased, the apparent affinity
between the peptide and the phospholipids decreased, suggesting that
the electrostatic interaction plays an important role in the binding
(14, 34).
Effects of TFE on the Conformation of GAP Peptide
To study
the structural characteristics of the calmodulin-binding domain of
GAP-43 under hydrophobic environments in detail, effects of TFE, a
membrane mimicking reagent, on the conformation of GAP peptide were
studied. As shown in Fig. 6, the addition of TFE caused
a concentration dependent induction of a CD spectrum typical for
-helix with two negative peaks at 222 and 208 nm, a cross-over near
200 nm, and a maximum near 192 nm (35). The
-helical content reached
almost 100% in 40% TFE (Fig. 6, inset). Therefore the
secondary structure of GAP peptide is converted from random coli to
almost pure
-helix in the presence of TFE. Structural
characteristics of GAP peptide were further analyzed by two-dimensional
1H NMR. To reduce the amide exchange rate and
simultaneously increase the correlation time of the peptide, most of
the spectra were obtained at acidic pH (Fig.
7a). The assignment of the NMR resonances was
obtained by following standard procedures for two-dimensional 1H NMR of proteins (36). Fig. 8a
summarizes the short and medium range NOE cross-peaks observed with GAP
peptide. A substantial number of
(i, i+3) and
N (i, i+3) NOE
cross-peaks were observed with GAP peptide in 50% H2O,
10% D2O, 40% TFE-d3 solution,
suggesting that GAP peptide assumes an
-helical structure in the
presence of TFE. Furthermore, comparison of the chemical shifts of the
protons observed with GAP peptide to those obtained with peptides in random coil (37) shows upfield shifts characteristic for
-helix
(38, 39) as shown in Fig. 8b. Altogether, these results established that GAP peptide, especially about two-thirds of the N-terminal side of the peptide, assumes a regular
-helix in the presence of TFE.
Conformation of GAP Peptide in Aqueous Solution
In the
absence of TFE, many cross-peaks in the NMR spectra could not be
assigned due to resonance overlaps. However, it was possible to assign
several residues from their unique spin systems. Since
Ala2, Ile8, Thr9, and
Leu13 occurs only once in the peptide, their well resolved
methyl group signals in higher magnetic field region were used to
assign peaks belonging to these residues. Interestingly, the chemical
sifts of all the assigned residues showed intermediate values between those typical for random coil (37) and those for -helix.
Furthermore, the overlapped signals of cross-peaks between
protons
and amide protons as a whole showed a tendency of the upfield shift
(Fig. 7b). Again, most of the signals showed intermediate
chemical shifts between those observed with
-helical and random coil
peptides. These results suggest that the GAP peptide in aqueous
solution is not in a "pure" random coil, but assumes an
intermediate state between the random coil and the
-helix.
To examine effects of PKC-dependent phosphorylation of GAP-43 on the GAP-43-membrane interaction, GAP-43 and GAP peptide were phosphorylated by PKC and purified as described under "Experimental Procedures." The basic domain of GAP-43 contains only one phosphorylatable residue (Ser41) and the phosphorylation of the protein prevents its binding to calmodulin (7). Phosphorylated GAP-43 and phosphorylated GAP peptide showed CD spectra similar to those of non-phosphorylated species. When acidic phospholipids such as PG was added, no significant change in the CD spectra was observed with both the phosphorylated protein and the peptide (data not shown), suggesting that the phosphorylation by PKC reduced their affinities to the phospholipids. Phosphorylation of GAP peptide has already been shown to affect the interaction (14). It is easily conceivable that the incorporation of negative charges reduces the net positive charge of the basic domain, which should affect the electrostatic interaction involved in the binding.
Although the CD spectrum of the phosphorylated peptide was very similar
to that of the non-phosphorylated peptide in aqueous solution, those
obtained in the presence of TFE differed significantly. Even in the
presence of 50% TFE, the phosphorylated peptide showed an intermediate
state between the random and -helical structures (Fig.
9), similar to that observed with the non-phosphorylated peptide in 20% TFE (Fig. 6). These results suggest that
phosphorylation of the Ser41, which is located near the N
terminus of the peptide, not only changes the net charge of the
peptide, but also impairs significantly the ability of the peptide to
form the
-helical structure.
Effects of Calmodulin on GAP-43-membrane Interaction
Since
the same domain shows the ability to bind both acidic phospholipids and
calmodulin, it is of interest to examine the effects of calmodulin on
the GAP-43-membrane interactions. As shown in Fig. 10,
the addition of calmodulin to the membrane-bound GAP-43 reversed the
binding, and a part of the protein was found in the supernatant. In the
presence of 0.15 M KCl, the amounts of GAP-43 remained in
the supernatants were similar regardless of Ca2+,
reflecting the fact that GAP-43 has similar binding constants to both
Ca2+-free apo-calmodulin and Ca2+-calmodulin at
this ionic strength (40). However, at a low ionic strength, where
GAP-43 shows higher affinity to apo-calmodulin than to
Ca2+-calmodulin (40), the amounts of GAP-43 remained in
supernatant were higher in the absence of Ca2+ than in the
presence of Ca2+ (Fig. 10b). Therefore, the
cellular concentration of calmodulin available for binding can regulate
the GAP-43-membrane interactions. It should be noted that the addition
of Ca2+ also affected the GAP-43-phospholipid interaction,
although the effects seems to be less significant compared with those
observed with calmodulin under the experimental conditions employed.
This is probably due to the direct binding of Ca2+ to PS,
which has been well documented. These results suggest that the
GAP-43-membrane interaction is regulated in a very complex manner
including the PKC-dependent and the
calmodulin-dependent pathways.
Although GAP-43 has been reported to be palmitoylated and the
acylation has been believed to be responsible for the membrane localization of the protein, our previous (3) and present studies established that GAP-43 purified from membrane fractions is not palmitoylated but still retains the ability to bind to phospholipid membranes. Since GAP-43 is a very hydrophilic protein without any clear
membrane-binding domain, this seems to be puzzling. However, we have
already shown that myristoylated alanine-rich protein kinase C
substrate, which itself is a major PKC substrate protein and shows
stimulation-dependent reversible membrane binding, interacts with phospholipid membranes through the PKC phosphorylation domain of basic amphiphilic nature (16). A similar observation has been
reported with the GAP-43 calmodulin-binding domain (14). The
calmodulin-binding domain of nitric oxide synthase, too, seems to serve
as a membrane-binding domain (41).2
Although the calmodulin-binding motif has no clear conserved amino acid
sequence, basic hydrophilic and hydrophobic amino acids appear
alternately at certain intervals. When it assumes an -helical structure, the two groups of amino acids segregate on opposite sides of
the helices (42, 43). The calmodulin-binding domain of GAP-43, which is
at the same time the phosphorylation domain by PKC, belongs to the same
phosphorylation-dependent membrane binding motif.
As shown in the present study, both the intact protein and the
phosphorylation domain of GAP-43 bind to phospholipid membranes and
adopt helical structures under hydrophobic environments. The degree
of change in the CD spectra of the whole protein induced by the
phospholipid binding is similar to that observed with GAP peptide,
suggesting that upon phospholipid binding only the phosphorylation domain undergoes a conformational change while the conformation of the
rest of the molecule remains unchanged. It seems reasonable to assume
that only the phosphorylation domain of GAP-43 interacts with
phospholipid liposomes, since the whole protein except for the domain
is very hydrophilic and acidic.
Compared with conventional CD measurements, two-dimensional
1H NMR studies gave more accurate and residue-specific
information on the conformation. Most parts of the peptide, especially
the N-terminal two-thirds, formed a regular -helix in the presence of TFE, as was evidenced by the consecutive NOE connectivities and the
characteristic upfield shift of
protons. The C-terminal side of the
peptide, on the other hand, showed less tendency to form an
-helix.
The advantage to use the NMR technique was more pronounced in the
analysis of the conformation of the peptide in aqueous solution. The CD
spectrum of GAP peptide in aqueous solution showed no indication of the
presence of any regular structure. On the other hand, chemical shifts
of most of the
protons obtained in aqueous solution showed
intermediate values between those observable with
-helix and those
seen with "true" random coil. So-called random coil can be
considered to be a set of many different conformations, within which
rapid exchanges occur. While the CD spectroscopy does not indicate the
nature of the conformations included in such sets of the conformations,
the NMR studies suggest that a significant portion of the peptide
molecules adopts an
-helical conformation. Such a nascent helical
structure may deviate from ideal geometry, and/or the ends of the
-helix can fray (44, 45). The interaction of GAP peptide with
phospholipids seems to stabilize the conformation to induce an
-helical structure. Several calmodulin-binding peptides have been
reported to form such a nascent
-helical structure in aqueous
solution. Such a structure is usually further stabilized either by
addition of TFE (46, 47) or by binding to calmodulin (48, 49). It should be noted that the shape of the CD spectrum obtained in the
presence of phospholipids was not the same as that observed in TFE,
suggesting that the conformation of GAP peptide bound to phospholipids
is not a simple
-helix. Since the extents of the change observed
upon phospholipid binding were also smaller than those observed in the
presence of TFE, it is possible that only a part of the domain
undergoes such a conformational change. A more detailed study of the
interaction between phospholipid liposomes and the GAP-43
phosphorylation domain by NMR is now in progress.
The calmodulin-binding domain of GAP-43 interacts with membrane
phospholipids, and the phosphorylation by PKC reduces the interaction.
The phosphorylation seems to affect the interaction not only by
reducing the positive charge of the domain but also by impairing the
ability of the domain to form an -helical structure (Fig. 9). It is
reasonable to assume that the same mechanism is operational in the
regulation of the binding of the domain to calmodulin by
PKC-dependent phosphorylation (7). Since the binding of
calmodulin, in turn, affects the interaction of the domain with PKC and
phospholipids, physiological functions of GAP-43, whatever it is, are
regulated in a very complex manner which involves various components of
the signal transduction pathways (Fig. 11). The domain
of basic amphiphilic
-helical nature may function as one of the
cross-talk points in the calcium-dependent signal
transduction.
We thank M. Suzuki for technical assistance.