| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, July 26, 1994; and in revised form, December 27,
1994) From the
The binding of chicken gizzard caldesmon to actin was studied
both in the presence and the absence of caltropin using Airfuge
centrifugation experiments, disulfide cross-linking studies, and the
fluorescent probe acrylodan (6-acryloyl-2-(dimethylamino)napththalene).
In co-sedimentation studies most of the caldesmon pelleted along with
actin. However, when caldesmon in the presence of caltropin was mixed
with actin, caldesmon did not pellet along with actin following high
speed centrifugation, suggesting that caltropin has significantly
weakened its binding to actin. The caltropin effect was noticed even
when tropomyosin was included in the reaction mixture.
Acrylodan-labeled caldesmon, when excited at 375 nm, had an emission
maximum at 515 ± 2 nm. The addition of actin produced a nearly
70% increase in fluorescent intensity, accompanied by a blue shift in
the emission maximum (i.e.
In smooth muscle, calcium exerts its control via mechanisms
based on both the thick (myosin) filaments (Sobieszek and Bremel, 1975)
and the thin (actin-based) filaments (Chalovich, 1993; Sobue and
Sellers, 1991). The thin filament-linked pathway involves
actin-mediated regulation by caldesmon, a protein that is found bound
to the thin filaments (Furst et al., 1986; Lehman et
al., 1987; Walsh and Sutherland, 1989). When purified caldesmon
binds to actin-tropomyosin, it inhibits the superprecipitation of
actomyosin (Sobue et al., 1981) and the actin-activated ATPase
activity of both myosin (Dabrowska et al., 1985; Moody et
al., 1985; Sobue et al., 1985) and its subfragments (Lash et al., 1986; Chalovich et al., 1987). Recent studies
with caldesmon fragments and smooth muscle fibers (Pfitzer et
al., 1994; Katsuyama et al., 1992) demonstrate the
physiological feasibility of the regulation of smooth muscle
contraction by caldesmon. However, defining the precise role of
caldesmon in cells requires an understanding of its interactions with
several key proteins, namely actin, myosin, calmodulin, and caltropin. Caldesmon binds to actin and myosin with strong affinity (Velaz et al., 1989; Hemric and Chalovich, 1990). Recent studies
suggest that caldesmon inhibits the binding of myosin-ATP to actin,
thereby inhibiting ATP hydrolysis (Velaz et al., 1989). In an
earlier study (Mani and Kay, 1993), we have shown that caldesmon binds
to smooth muscle heavy meromyosin with high affinity and that this
interaction was weakened in the presence of caltropin
(Ca
Figure 1:
Interaction between caldesmon and
actin. The binding of caldesmon to actin was examined in the presence
and absence of Ca
Figure 2:
Fluorescence emission spectra of
acrylodan-caldesmon (solid line) and acrylodan-caldesmon-actin (dashed line) complexes in 25 mM Tris, pH 7.5, 42
mM NaCl, 2 mM MgCl
Figure 3:
Influence of actin on acrylodan-caldesmon
(
Figure 4:
Scatchard plot of actin binding to
acrylodan caldesmon (
The
change in fluorescence intensity as a function of added actin was also
analyzed using double-inverse plots. The affinities of actin for
caldesmon and the caldesmon-caltropin complex determined from the
abscissa intercepts of the double inverse plots were 1 The observed increase in fluorescence intensity
accompanied by a blue shift in the emission maximum upon actin binding
could be due to a decrease in the exposure of the probe to solvent
quenchers. Fig. 5shows the effect of the acrylamide quencher on
the fluorescence intensity of caldesmon, the caldesmon-actin complex,
the caldesmon-caltropin (Ca
Figure 5:
Steady-state acrylamide quenching plots
for the acrylodan-caldesmon (
Caldesmon binds tightly to myosin (K The
interaction between caldesmon and actin is important for several
reasons. First, the binding of caldesmon to the NH The addition of actin to labeled caldesmon produced an increase in
fluorescence intensity that leveled off when the mole ratio of actin to
caldesmon was nearly 7. There is uncertainty in the in vivo and in vitro stoichiometric ratio of caldesmon to actin,
raising questions as to the assembly and arrangement of caldesmon on
the actin filament. According to Marston et al.(1992), the in vivo ratio is about one caldesmon molecule/14 actin
monomers, whereas the in vitro values reported range from 3 to
7 actin monomers/caldesmon molecule for maximum inhibition of ATPase
activity. The length of caldesmon has been estimated to be 740 Å.
If caldesmon binds to both sides of an actin filament like tropomyosin,
then a maximum of 14 actin monomers could be covered by a single
caldesmon molecule, consistent with the in vivo determinations. Graceffa and Jancso(1991) studied cross-linking of
caldesmon to actin as a function of caldesmon concentration and
demonstrated that only 3-4 actin monomers could be cross-linked
to one caldesmon molecule, hence reinforcing the difference in the in vivo and in vitro determinations. Native thin
filaments also contain calponin (Nishida et al., 1990), which
also binds actin. The presence of calponin might limit the amount of
caldesmon bound to actin, and this could explain the observed
difference between the in vivo and the in vitro estimations. From this study we conclude that caltropin in the
presence of Ca A possible fine tuning mechanism for
cooperative caldesmon regulation of the thin filament is shown in Fig. 6. The regulatory properties of caldesmon are similar to
troponin in most details as indicated by Marston and Redwood(1993). The
COOH-terminal region of caldesmon named domain 4b comprising amino acid
residues 658-756 is closely analogous to the inhibitory peptide
of troponin (troponin-I). Domain 3 of caldesmon (i.e. residues
508-565) is homologous to the CB
Figure 6:
A troponin-like model for cooperative
caldesmon regulation of the thin filament caldesmon (CaD),
tropomyosin (TM), actin (A), and myosin (M).
Domain 1 corresponds to the NH
Caldesmon can bind tightly to myosin as well as to actin, and
caltropin has no effect on these interactions in the absence of calcium
as shown in our model (Fig. 6). Caldesmon is known to compete
with myosin and myosin fragments for the binding to actin in rigor
(Hemric and Chalovich, 1988), and this indicates the existence of at
least a partial overlap in their actin-binding sites and could form the
basis of the regulatory function of caldesmon. According to Chalovich et al.(1987), caldesmon acts as a competitive inhibitor of the
binding of myosin to actin, because the COOH-terminal end of caldesmon
that does not bind to myosin is able to inhibit ATP hydrolysis.
Calmodulin and caltropin in the presence of Ca Current studies have clearly demonstrated the importance of this
cooperative unit consisting of caldesmon-tropomyosinactin in regulating
the smooth muscle thin filament. Even though in vitro studies
involving the interaction of caldesmon with individual proteins, namely
tropomyosin, actin, and myosin, and the calcium-binding proteins have
been studied in detail, caution should be exercised in interpreting
these results, because most of these studies involve interaction of
caldesmon with either thick or thin filament proteins. It is necessary
to relate these findings with the individual proteins to the integrated
system consisting of both the thick and thin filament assembly, namely
the actomyosin system.
Volume 270,
Number 12,
Issue of March 24, 1995 pp. 6658-6663
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
=
505 ± 2 nm), suggesting that the probe now occupies a more
nonpolar environment. Titration of labeled caldesmon with actin
indicated a strong affinity (K
=
6 
10
M
). When
actin was titrated with labeled caldesmon in the presence of caltropin
in a 0.2 mM Ca
medium, its affinity for
caldesmon was lowered (K = 2
10M). Caltropin, which
is very effective in reversing caldesmon's inhibition of the
actin-activated myosin ATPase (Mani, R. S., McCubbin, W. D., and Kay,
C. M.(1992) Biochemistry 31, 11896-11901), is shown in
the present study to have a pronounced effect on its binding to actin,
suggesting a major role for caltropin in regulating caldesmon in smooth
muscle.
), a high affinity smooth muscle calcium-binding
protein. Caltropin, which is very potent in reversing the inhibitory
effect of caldesmon in the presence of calcium (Mani et al.,
1992), modulates the interaction between caldesmon and heavy meromyosin
in a calcium-dependent manner. The caltropin effect on caldesmon may be
summarized as follows. Caldesmon, by virtue of its ability to interact
with both myosin and actin, is able to inhibit the actin-activated
myosin ATPase activity by interfering in the actin-myosin interaction.
In the presence of Ca caltropin, the affinity of
caldesmon for myosin is lowered; as a consequence myosin is able to
interact effectively with actin, resulting in restoring the
actin-activated myosin ATPase activity. However, any proposed mechanism
to account for the effect of caltropin on caldesmon should include not
only the caldesmon-myosin interaction but also the caldesmon-actin
interaction. For this reason, in the present study the effect of
caltropin on the interaction between caldesmon and actin was
investigated by the co-sedimentation method and fluorescence
measurements using the fluorescent probe acrylodan (
)to
label both the sulfhydryl groups in caldesmon. Using these techniques
we have shown that caltropin in the presence of calcium was effective
in weakening the interaction between caldesmon and actin.
Protein Purification
Caltropin and caldesmon
were isolated from chicken gizzard using the procedure described in our
earlier papers (Mani and Kay, 1990; Mani et al., 1992).
F-actin was purified from rabbit skeletal muscle according to Pardee
and Spudich(1982).Airfuge Binding Experiments
Rabbit skeletal muscle
actin (10 µM) was polymerized using a solvent system
consisting of 2 mM MgCl
, 42 mM NaCl, 20
mM Tris, pH 7.5, 0.5 mM dithiothreitol, and 0.2
mM CaCl. Subsequent to polymerization, F-actin was
incubated with caldesmon (1.0-3.0 µM) resulting in a
range of caldesmon to actin ratios of 0.1-0.3. Following
incubation for 20 min, the protein mixtures were centrifuged at 100,000
g for 30 min in an air-driven ultracentrifuge (Beckman
Instruments). Under these conditions caldesmon was not pelleted when
centrifuged in the absence of actin. The pelleted samples were
solubilized in SDS-gel running buffers and were analyzed using 12%
SDS-polyacrylamide gels.Fluorescence Studies
Fluorescence spectra were
obtained with a Perkin-Elmer model MPF-44 spectrofluorometer, and all
measurements were made at 20 °C. Caldesmon was labeled with
acrylodan following the procedure of Prendergast et al. (1983)
as described in our earlier paper (Mani et al., 1992). The
ratio of label to protein was determined to be 1.9, suggesting that
both of the sulfhydryl groups in caldesmon were labeled under these
conditions. Labeled caldesmon was titrated with F-actin by following
the changes in fluorescence intensity at 470 nm.Acrylamide Fluorescence Quenching Studies
The
excitation wavelength used was 375 nm, at which the probe acrylodan has
maximum absorption. The initial A
of the labeled
caldesmon was 
0.05. The fluorescence quenching was measured at the
emission maximum (505 nm) for the labeled caldesmon-actin complex
following the addition of 8 M acrylamide in small aliquots.
The theory of acrylamide quenching and the mathematical treatment of
the data are described by Lehrer and Leavis(1978).
Binding Experiments
The binding of caldesmon to
actin was studied by a co-sedimentation method. When caldesmon/actin
mixtures of varying ratios (see ``Materials and Methods'')
were subjected to high speed centrifugation, most of the caldesmon
co-sedimented with F-actin whether Ca was present or
not (Fig. 1, lane b). The addition of
Ca/caltropin to actin prior to the addition of
caldesmon had a pronounced effect on the ability of caldesmon to
interact with actin. Under these conditions most of the caldesmon did
not co-sediment with actin (Fig. 1, lane a), i.e. caldesmon was now present in the supernatant along with caltropin.
However, when this experiment was repeated in the presence of 1 mM EGTA (i.e. in the absence of calcium) most of the
caldesmon could be pelleted along with actin (Fig. 1, lane
d), suggesting that caltropin has no significant effect on
caldesmon-actin interaction in the absence of calcium. This finding is
consistent with our earlier studies involving bioassays (Mani et
al., 1992) where caltropin was shown to be effective in releasing
the inhibition of caldesmon only in the presence of calcium. Caltropin
was also effective in releasing caldesmon from actin filaments even in
the presence of tropomyosin. Control experiments indicated that neither
caldesmon nor caltropin sedimented by itself, and we also found no
interaction between caltropin and actin.
and caltropin by the high speed
centrifugation method described under ``Materials and
Methods.'' The proteins that were pelleted were analyzed by 12%
SDS-gels. Lanea, 1.5 µM caldesmon (CaD), 1 µM caltropin, and 10 µM actin in 0.2 mM Ca; laneb, 1.5 µM caldesmon and 10 µM actin in 0.2 mM Ca; lanec, 1.5 µM caldesmon and 10 µM actin in 1 mM EGTA; laned, 1.5
µM caldesmon, 1 µM caltropin, and 10
µM actin in 1 mM EGTA.
Fluorescence Spectroscopy
Acrylodan, a
sulfhydryl-specific fluorescent probe, is very useful for studying
protein-protein interactions. Because sulfhydryl groups in chicken
gizzard caldesmon can be modified without the protein losing its
activity (Hemric and Chalovich, 1990), we decided to label both the
sulfhydryl groups in caldesmon with acrylodan. When the labeled
caldesmon was excited at 375 nm, the emission maximum occurred at 515
± 2 nm in both the presence and the absence of calcium. The
addition of actin resulted in a blue shift in the emission maximum,
which was then centered at 505 ± 2 nm. Even though there was no
significant change in fluorescence intensity around the emission
maximum, a significant increase in relative fluorescence intensity
(nearly 70%) was observed at 470 nm (Fig. 2). For this reason,
during the titration of labeled caldesmon with actin, the fluorescence
intensity at 470 nm was monitored. Analysis of the titration curve
indicated that the maximum increase in fluorescence intensity was
achieved by the time 7 mol of actin were added per mol of caldesmon (Fig. 3). Also shown in the figure is the titration of actin
with the caldesmon-caltropin complex. Addition of caltropin to labeled
caldesmon in the presence of calcium produced an increase in
fluorescence intensity. The observed fluorescence intensity in the
presence of caltropin was taken as the starting value for the titration
with actin. In this instance also the maximum change was reached when 7
mol of actin were added. However, as can be seen from Fig. 3,
actin binds to the caldesmon-caltropin complex with lower affinity. For
example, for a 50% increase in fluorescence intensity, the amount of
actin required for caldesmon and the caldesmon-caltropin complex
corresponded to 0.15 and 0.25 µM, respectively. The shape
of the titration curve suggests the existence of more than one class of
binding site. If all seven actins were to bind with the same affinity,
then the observed increase in fluorescence intensity would be linear
until 7 mol of actin were added, at which stage the observed increase
in fluorescence intensity would correspond to the maximum level (i.e. it will level off). If one were to draw tangents to the
initial slope of the titration curve and to the asymptotic part of the
curve (i.e. where the curve levels off), then the point of
intersection should correspond to the number of high affinity binding
sites. The stoichiometry obtained using this approach corresponds to a
value of 2.8 and 3.5 mol of actin for the caldesmon and the
caldesmon-caltropin complex, respectively. In other words, out of seven
actins that are required for saturation, only three are binding with
high affinity. If one assumes that the observed fluorescence change is
the result of actin binding to caldesmon, then one can calculate the
amount of actin bound at any given point in the titration curve.
Knowing the amount of actin added, one can deduce the amount of free
actin that is present. When Scatchard analysis was carried out, with
the amounts of the bound and free actin and the caldesmon concentration
used for the titration known, a best fit was also obtained when the
ratio of actin to caldesmon was 3:1 (Fig. 4). From the slope
term of the Scatchard plot, a K
value of (6.0
± 0.5) 10M was
calculated from three sets of titration data, suggesting that caldesmon
binds to actin with high affinity. Also shown in Fig. 3is the
titration of actin with the caldesmon-caltropin complex. Actin binds to
this binary complex with lower affinity, because the K value obtained was only (2.5 ± 0.5) 10M. The presence of caltropin has
lowered the affinity of caldesmon for actin more than 2-fold.
, 0.2 mM CaCl, and 1 mM dithiothreitol at 20 °C.
The excitation wavelength was 375 nm.
) and acrylodan caldesmon-caltropin fluorescence (
). The
initial caldesmon concentration was 1.0 10M. Measurements were carried out in
25 mM Tris, pH 7.5, 42 mM NaCl, 2 mM MgCl, 0.2 mM CaCl, and 1 mM dithiothreitol at 20 °C. Relative changes in fluorescence
intensities (
F/F
) at 470 nm
are plotted as functions of actin concentration. The excitation
wavelength was 375 nm. The data points shown on the titration curves
are based on the results obtained from three sets of titrations. The
size of the symbols on these curves takes into
account the observed experimental errors. The inset shows the
relative changes in fluorescence intensities
(F/F) at 470 nm as functions
of molar ratios of actin added to
caldesmon.
) and acrylodan-caldesmon-caltropin ()
with [actin] corrected to [free actin] using a 3:1
stoichiometry for actin to caldesmon. Bound actin was determined from
the change in fluorescence intensity, which is directly proportional to
binding. 
is the number of moles of ligand bound to 1 mol of
macromolecules; it is defined as =
[A]
/[P]
.
The conditions were the same as described in the legend of Fig. 3.
10
and 5 10
M,
respectively (data not shown), and this probably represents the average K values for the high and the low affinity binding
sites. In earlier studies involving the interaction of caldesmon and
actin, several other investigators have also suggested the existence of
high and low affinity binding sites (Moody et al., 1985; Lash et al., 1986; Horiuchi et al., 1986; Smith et
al., 1987). We also wanted to see if caltropin could influence
caldesmon-actin interaction in the presence of tropomyosin. For this
reason we carried out fluorescence titration experiments with labeled
caldesmon containing tropomyosin with actin in the presence and the
absence of caltropin. In this instance the affinity of caldesmon for
actin was also lowered by more than 2-fold when caltropin was present,
suggesting that caltropin could function even in the presence of
tropomyosin.)-actin ternary complex,
the caldesmon-tropomyosin-actin complex, and the
caldesmon-tropomyosin-caltropin (Ca)-actin complex.
The linear portion of the plots at low acrylamide concentration gave
Stern-Volmer constants (K
) of 2.1, 1.7, 1.2, 1.1,
and 0.6 for the caldesmon, caldesmon-actin, caldesmon-caltropin-actin,
caldesmon-tropomyosin-actin, and caldesmon-tropomyosin-caltropin-actin
complexes, respectively. The value of the Stern-Volmer quenching
constant obtained is higher for caldesmon than for the caldesmon-actin
complex. Thus actin does shield the label on caldesmon from the solvent
quencher. The addition of Ca caltropin to the
caldesmon-actin complex results in further lowering of the quenching
constant, suggesting that the label now moves to a more hydrophobic
environment (i.e. the label is now more protected from the
quencher). This effect of caltropin was also noticed for the
caldesmon-tropomyosin-actin complex. In other words, caltropin was
effective in shielding the label even when tropomyosin was present in
the caldesmon-actin complex. Even though both the sulfhydryl groups in
caldesmon (Cys-153 and Cys-580) are labeled, we believe the observed
effect of caltropin and actin on acrylamide quenching is due to the
binding of these proteins at or near the label Cys-580, because actin
and caltropin are known to bind at the COOH-terminal end of caldesmon
(Szpacenko and Dabrowska, 1986; Zhuang et al., 1994) and thus
shield the label from collisions with quenchers in the medium.
), acrylodan-caldesmon-actin
(), acrylodan-caldesmon-caltropin-actin (),
acrylodan-caldesmon-tropomyosin B-actin (
), and the
acrylodan-caldesmon-tropomyosin B-caltropin-actin (
) complexes.
Excitation and emission wavelengths were 375 and 505 nm, respectively.
The solvent system used was the same as described in the legend of Fig. 3.
= 10M),
actin (K =
10-10M), and tropomyosin (K = 10M) and
inhibits the actin-activated ATPase of smooth and skeletal myosins and
their subfragments (Lash et al., 1986; Hemric and Chalovich,
1988; Velaz et al., 1989; Horiuchi and Chacko, 1988). These
characteristics of caldesmon make it a potential candidate for the
regulation of actomyosin interaction. Simultaneous binding of caldesmon
to actin and myosin may provide an additional role for caldesmon in
stabilizing the structure of the contractile apparatus or may even
maintain tension in the latch state of smooth muscle.-terminal
end of actin results in the inhibition of ATP hydrolysis by actomyosin
(Crosbie et al., 1994), suggesting a regulatory role for
caldesmon in the filament-linked regulation of smooth muscle
contraction. Even though we know that the COOH-terminal region of
caldesmon binds to the NH-terminal end of actin, many
questions regarding the interaction remain unanswered. Caldesmon
appears to bind along the actin filament (Lehman et al., 1989)
with each caldesmon monomer covering several actin monomers (Velaz et al., 1989). Unlike tropomyosin, caldesmon does not have
repeating actin-binding structural units, and for this reason it is
unclear how the binding of the COOH-terminal region of caldesmon
stabilizes its binding along several actin monomers. In the presence of
calcium, calcium-binding proteins such as caltropin and calmodulin can
release caldesmon's inhibition of the actin-activated myosin
ATPase. In order to get an insight into the mechanism by which these
calcium-binding proteins release the inhibition of caldesmon, we
studied the influence of caltropin on the caldesmon-actin interaction
using co-sedimentation and fluorescence measurements using labeled
caldesmon. Caltropin in the presence of Ca was very
effective in weakening the interaction between caldesmon and actin. is effective in weakening the
interaction between caldesmon and actin. Caldesmon has been implicated
to act as a competitive inhibitor of myosin binding to actin. Caldesmon
can bind to actin with strong affinity at or near the myosin binding
region (i.e. at the NH-terminal end of actin),
thus blocking myosin from binding to actin and resulting in lowered
ATPase activity (Chalovich et al., 1987). Caldesmon, which
binds caltropin in the presence of Ca with high
affinity (Mani et al., 1992), undergoes a conformational
change; as a consequence we believe it binds to actin with lower
affinity and is no longer able to compete with myosin for binding to
actin, resulting in the recovery of the ATPase activity. In this
context Graceffa and Jansco(1991) have also evoked a similar mechanism
to explain the effect of Ca/calmodulin on
caldesmon-actin interaction. region of troponin-T (i.e. residues 71-151, which are implicated in
tropomyosin binding). For this reason we have indicated in our model
that region 4b and domain 3 of caldesmon bind to actin and tropomyosin,
respectively. Tropronin-C (the Ca-binding subunit) is
part of the troponin complex, whereas in smooth muscle the
Ca factor caltropin is not a part of caldesmon.
-terminal region of
caldesmon; domains 3 and 4b represent COOH-terminal
amino acid residues 508-565 and 658-756, respectively. CaT, caltropin.
can
release the inhibition of caldesmon without displacing caldesmon from
the actin filament (Smith et al., 1987; Marston and Smith,
1985; Mani et al., 1992). According to our proposed model, in
the presence of Ca, caltropin binds to caldesmon with
high affinity (Mani et al., 1992) and induces a conformational
change in caldesmon with the result that caldesmon binds to actin with
lower affinity (3-fold) and is no longer able to compete with
myosin for binding to actin resulting in the release of the inhibition.
)
We thank A. Keri, K. Oikawa, and L. Hicks for
excellent technical assistance.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  N. Courtois-Coutry, C. Le Moellic, S. Boulkroun, M. Fay, F. Cluzeaud, B. Escoubet, N. Farman, and M. Blot-Chabaud Calcyclin Is an Early Vasopressin-induced Gene in the Renal Collecting Duct. ROLE IN THE LONG TERM REGULATION OF ION TRANSPORT J. Biol. Chem., July 5, 2002; 277(28): 25728 - 25734. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Stradal and M. Gimona Ca2+-dependent Association of S100A6 (Calcyclin) with the Plasma Membrane and the Nuclear Envelope J. Biol. Chem., October 29, 1999; 274(44): 31593 - 31596. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Wang, Z.-Q. Yang, and S. Chacko Functional and Structural Relationship between the Calmodulin-binding, Actin-binding, and Actomyosin-ATPase Inhibitory Domains on the C Terminus of Smooth Muscle Caldesmon J. Biol. Chem., July 4, 1997; 272(27): 16896 - 16903. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
