Evidence of a new
type of protein-protein interaction:
desensitized actomyosin blocks Ca2+-sensitivity
of the natural one.
A possible model for an intracellular signalling system
related to actin filaments
Physiological
Chemistry and Physics and Medical NMR, 2000, 32: 167-178
Vladimir
V. Matveev
Laboratory
of Cell Physiology, Institute of Cytology, Russian Academy of Sciences,
Tikhoretsky Ave 4, St. Petersburg 194064, Russia and
Abstract. Actin filaments are certainly believed to function as an intracellular signalling system; however, this is not confirmed by direct evidence. We used a two-layer actomyosin gel with a concen tration gradient of the troponin-tropomyosin complex (TT-complex, Ca2+-sensitive system) between the two layers. To prepare one layer of the system, natural actomyosin (nAM) rich in TT-complex was used. To prepare the second layer, we used desensitized actomyosin (dAM) without the complex. All experimental studies were made in medium with a low ionic strength. Two phenomena were ob served: (1) dAM blocks Ca2+-sensitivity of nAM when the dAM weight portion in the system (as well as in mixed nAM+dAM suspension) reaches 40% and more; further increase of the dAM portion does not affect the Ca2+-sensitivity; (2) it was electrophoretically shown that a rapid diffusion of the TT-complex from nAM gel into the dAM gel took place. The apparent diffusion coefficient for the TT-complex in dAM gel is about (1-4)·10-4 cm2/sec, i.e. three orders higher than the same values for protein diffusion in water.
The
cytoskeleton and, particularly, actin filaments are believed to
function as an intracellular signalling system (by signal we mean any cellular
event which has a regulatory significance). Edelman (1) was the first to
suggest participation of actin filaments in the transfer of signals from the
plasma membrane to the cytoplasm and nucleus.
To elucidate the mechanism of transfer of a signal related to elements of the
cytoskeleton, experiments with protein model systems reproducing some important
features of the cytoplasm can be useful. Such a convenient model system can be
actomyosin. Natural actomyosin (nAM)
contains a regulatory troponin-tropomyosin complex (TT-complex), whose content
may be easily changed by washing out actomyosin with a solution of a low ionic
strength and alkaline pH. After removal of the TT-complex, the actomyosin
superprecipitation by addition of MgATP becomes Ca2+-independent,
i.e., actomyosin becomes desensi tized (dAM).
Addition of the TT-complex to dAM
results in recovery of Ca2+-sensitivity (2). Therefore, after
removal of the TT-complex from actomyosin, its binding sites still are able to
bind again the regulatory complex. Thus, the use of actomyosin preparations
with different contents of TT-complex can produce gels with a concentration
gradient of the actin-binding proteins.
It should be emphasized that the TT-complex is strongly bound to actin
filaments at physiological ionic strength and pH and serves as a Ca2+
trigger of muscle contraction (3,4). Hence, the function and distribution of
this regulatory complex can be evaluated not only by analysis of protein
content in various areas of the gel, but also by measurement of its functional
parameter, Ca2+-sensitivity of the reaction of actomyosin
superprecipitation.
The goal of the present work was to study the interaction of actomyosin gels
with different contents of TT-complex to detect possible redistribution of the
TT-complex as a result of its diffusion, translocation, transport, etc.
Materials and Methods
Isolation
of natural actomyosin (nAM) from rabbit skeletal muscles
was performed at 4-6°C, as described by Haga et al. (5), with minor
modifications. The time of the extraction was 15-20 h. The actomyosin gel
(about 30 mg/ml of protein) with 50 mM KC1 was diluted by the same volume of
cold glycerol and stored at -20°C (6). Prior to the experiments, the
glycerinated actomyosin was washed free of glycerol with a solution containing
90 mM KCl and 14 mM Na-K-phosphate buffer, pH 6.7. After two washings by
tenfold volumes of this phosphate buffer (PB), nAM was dissolved in 0.5 M KC1 and
centrifuged for 20 min at 10,000 g. The supernatant was diluted with 9 volumes
of cold deionized water, and the collected precipitate was washed twice with PB
and then resuspended in PB.
Preparation of desensitized
actomyosin (dAM) was done by washing nAM at 4-6°C in pure water
containing a minimal amount of Tris-buffer, pH 8.5-9.0 (2). 15-20 ml aliquots
of the stored glycerolized actomyosin preparation were placed into every 100-ml
centrifuge test tube, and water was added up to the volume of 70 ml. After the
second or third washing, with centrifugation for 20 min at 10,000 g, the
actomyosin gel swelled (as a result of desalination) to such a degree that it
could no longer be precipitated. After this step, prior to every
centrifugation, 1-2 drops of 3 M KCl were added to every tube so that the
actomyosin gel volume was about 50% of the suspension volume after
centrifugation. After each centrifugation, the supernatant was replaced by
fresh pure water. This procedure was repeated 10 times. The above procedure was
confirmed electrophoretically to result in the nearly complete removal of the
TT-complex and in the complete elimination of the Ca2+-sensitivity.
The sensitivity was effectively restored by addition of the TT-complex isolated
as described by Spudich and Watt (7) to such dAM
preparations. After washing, dAM
was centrifuged in 0.5 M KCl and precipitated by dilution with 10 volumes of
water. The precipitate was washed twice with PB and then resuspended in the third
portion of the buffer.
Superprecipitation
of nAM, dAM, and their mixtures
was carried out at 20°C in PB con taining 0.5 mM Ca2+ or 0.5 mM EGTA (ethylene
glycol-bis(ß-aminoethyl ether) N,N,N',N'-tetraacetic acid) and 0.25 mM
MgATP at a protein concentration of 0.1 mg/ml. The reaction of
superprecipitation was recorded by the method of 90°-angle light scattering at
450 nm, using a Specol spectrometer (Carl Zeiss, Jena, Germany) with a built-in
magnetic mixer for continuous stirring of the samples in the course of
superprecipitation. For the superprecipitation, 30-ml cells were used; the
volume of samples was 20 ml.
After addition of MgATP to the actomyosin suspension, the light scattering
began to fall, reaching the minimum steady-state level after 1-2 min (for nAM) or after 5-7 min (for
dAM). For nAM preparations, the
decreasing value of the light scattering varied depending on the presence of Ca2+
or EGTA in the solution. This sensitivity to Ca2+ (Sca)
was ex pressed quantitatively as a light scattering decrease (LSD): LSD =10-1,
in the presence both of 0.5 mM Ca2+ (LSDca) and of 0.5 mM
EGTA (LSDegta), where 10 is the initial and 1, the final level of
the light scattering (in arbitrary units). So Sca = LSDca/LSDegta.
As a rule, three measurements of Sca were done for each actomyosin
sample.
The Sca values for dAM
preparations were close to 1, which indicates that the superprecipitation of
these preparations did not depend on Ca2+ and, accordingly, that no
significant amount of the regulatory TT-complex remained in dAM. The Sca
values for nAM
prepara tions were more than 1. This method of measurement of Ca2+-sensitivity
is described in de tail in our earlier work (6).
Electrophoresis
in polyacrylamide gels in the presence of sodium dodecyi sulphate (disc-SDS-electrophoresis)
was performed by the method of Sheludko (8), with some modifica tions for
block-electrophoresis. The gel components were from Reanal (Hungary). Their
composition in the separation gel, unless otherwise specified, was the
following: 0.43 M Tris-HCl (tris(oxymethyl)aminomethane), pH 8.9, 12%
acrylamide (AA), and 0.3% N,N'-methylene-bis-acrylamide (MBA), in the
concentration gel: 0.05 M Tris-HCl, pH 6.8, 2.5% AA, 0.6% MBA, 20% sucrose, and
2% sodium dodecyl sulphate. The electrode solutions contained 0.6 M glycine and
0.033 M Tris-buffer, pH 8.3; sodium dodecyl sulphate was added to the upper
electrode solution up to a concentration of 0.1%. Actomyosin samples were
solubilized for 5 min in a boiling water bath in a solution containing 50% glycerol,
5% sodium dodecyl sulphate, 50 mM dithiothreitol (Serva, Germany), and 25 mM
EDTA (ethyienediaminetetraacetic acid), pH 7.0. The gels were fixed by a
mixture of ethanol-CH3COOH-H2O (1:3:6) and then stained
with 0.125% Coomassie brilliant blue G-250 (Serva, Germany) dissolved in the
fixation mixture. After the dye excess had been removed from the gels with 7%
acetic acid, densities of protein bands were measured three times, us ing an
Elscript-3 densitometer (Hirschmann, Germany). The relative contents of
TT-complex proteins were expressed as percentage to actin.
Protein concentration
was determined by the micro-biuret method (9), using bovine se rum albumin as a
standard.
Formation of two-layer
gels. To form an actomyosin gel with non-uniform distribution
of the TT-complex, a two-layer system was prepared in 10-ml centrifuge tubes
(Figure 1), using the following steps. First, 30 mg of the dAM in suspension were
placed in each tube and precipitated by centrifugation for 20 min at 10,000 g
(with the final protein concentration of 30-40 mg/ml in the precipitate). Then
the supernatant was removed; different amounts of the nAM suspension were added
very carefully, and the centrifugation was repeated. The second centrifugation
produced a two-layer system (Figure 1) with the overlying supernatant that
remained in the tubes during the entire experiment. In order to test
deformability of the interlayer surface during the second layer formation, the
surface of the first layer was marked with the dye neutral red or with
colloidal carbon. No deformation was detected after the second centrifugation.
So the possibility of a mixture of different actomyosin preparations can be
ruled out.
FIGURE 1. A diagrammatic sketch of the two-layer actomyosin systems used, each with a different ratio of natural (nAM) and desensitized (dAM) actomyosin. The sizes of the layers are overesti mated to better demonstrate the arrangement of the experiments.
In separate experiments, it was found that
the nAM layer
(in the two-layer system with the nAM/dAM ratio 60:40,
respectively) lost its Ca2+-sensitivity as early as after 1-3 min.
But to insure the steady-state conditions, we used an overnight incubation both
for the two-layer systems and for two-component (n- and dAM)
suspensions. However, the apparent diffusion coefficient was determined during
much shorter time periods.
After formation of the two-layer systems, the tubes were chilled for 15-20 h at
4-6°C be fore the electrophoresis and superprecipitation. After incubation,
minimal, about 2 mg, samples of actomyosin were carefully taken exactly from
the surface of the upper layer of the two-layer system for measurements of the
Ca2+-sensitivity and electrophoretic analysis. The stock nAM and dAM preparations used to
form the two-layer system served as control.
To
determine the apparent diffusion coefficient (ADC) for the
TT-complex (tropomyosin was used as a marker of the complex), the nAM-dAM interaction was performed
in thin polyvinyl chloride tubes that were chosen so that the outside diameter
of one tube corre sponded to the inside diameter of the other. This made it
possible to form a very tight, mechanically strong connection. Tubes with the
internal diameters of 3 and 6 mm were used; they formed 2-3-cm-long gels; one
end of each tube was sealed.
The thin tubes were filled with nAM,
the thick ones, with dAM.
The filling of the tubes with actomyosin gel was carried out by the same
centrifugation procedure, as in the preparation of the two-layer systems
(10,000 g, 20 min). After formation of the precipitate, the tubes were cut at
the level of the precipitate/supernatant line and connected to each other very
carefully. The protein contamination from the surface of the joined tubes was
thoroughly removed.
Several joined tubes were then incubated at 4-6°C. These two-tube systems were
quickly frozen on ice at -20°C after different incubation time (from 3 min to 1
h). Then the thin and thick tubes were separated into two sections, left and
right, respectively, 1 mm from the point of gel contact. The segment isolated
after these first two cuts was discarded.
The separated tubes were cut into 5 mm slices (resulting in 4-6 segments).
After each cut, the cutting tool was cleaned of protein contamination. Protein
from each segment was solubilized and subjected to electrophoresis. This
allowed for the detection of the time-dependent tropomyosin decrease in the nAM
gel and the tropomyosin increase in dAM.
The TT-complex diffusion was studied at a nAM/dAM weight ratio 1:4. The
ADC was calculated as described by Chang (10).
Statistics.
Results of the experiments were expressed as means ± SEM. The differences were
considered statistically significant at P < 0.05, using Student's t-test.
Results
Figure 2 (curve A) illustrates typical results of the experiments with the two-layer systems. The control nAM preparation definitely demonstrated a high sensitivity to Ca2+ (curve A, the left end, 0% of dAM), while the other control preparation, dAM, was completely desensitized (curve A, the right end, 100% of dAM). In the two-layer system, an increase in the proportion of the dAM weight within narrow limits from 20 to 40% (20% range) resulted in the complete loss of Ca2+-sensitivity in nAM (the top layer).
FIGURE 2. The Ca2+-sensitivity of the natural actomyosin (nAM), A, arranged as the upper part of the two-layer system (See Figure 1), and the Ca2+-sensitivity of the two-component suspension mixtures (1 mg of protein per ml) of nAM with desensitized actomyosin (dAM.), B, following the 15-20 h incubation (at 4-6°C) plotted as a function of the weight portion of the dAM. The value "0" on the X-axis means nAM preparation alone. The value "100" on the X-axis means dAM preparation alone. The value "1" on the Y-axis means the absence of the Ca2+-sensitivity. The data points are the means ± SEM from triplicate assays.
As seen from the shape of the curve A, the
degree of the TT-complex redistribution depends on the weight ratio of the
layers, therefore, diffusion of the minor components is triggered by the weight
ratio, rather than by the existence of the gradient alone. This two-layer
system as a whole initiates the protein diffusion at a definite nAM/dAM threshold (critical)
ratio.
This "threshold" hypothesis is confirmed by the data shown in Figure
3. As seen from Figure 2, the curve describing the dependence of Ca2+-sensitivity
on the weight portion of dAM
(without the points for 0 and 100% dAM
because these points do not reflect interactions in two-layer systems) looks
like a hyperbola. After transformation of the data of Figure 2 into the
coordinates of Figure 3, this dependence becomes linear. The linear character
of the relationships demonstrated by Figure 3 is statistically significant for
the two-layer system at the 95% confidence level (correlation coefficient =
0.989, P < 0.05) and for the two-component suspension, at the 99% confidence
level (correlation coefficient = 0.994, P < 0.01). This linearity proves the
hyperbolic character of the relationships seen in Figure 2 for interacting
actomyosin preparations. Therefore, we can regard the X-axis asymptote (see the
legend for Figure 3) for each curve as a corresponding nAM/dAM threshold ratio. The
threshold value for dAM
content to initiate blockage of Ca2+-sensitivity of nAM was calculated to be
18 ± 5% for the two-layer system and 9 ± 2%, for the two-component suspension.
FIGURE 3. The linear representation of the data from Figure 2 for natural (nAM) + desensitized actomyosin (dAM) preparations in a form of two-layer systems, A, and as two-component suspension, B. The experimental results (except for the data for individual nAM and dAM preparations, in which the studied interaction does not occur) are well fitted by the linear model: W/S = a + b·W, where W is the weight portion (in %) of dAM; S is the Ca2+-sensitivity (in arbitrary units) of nAM (as a part of the two-layer system) or of the nAM + dAM suspension mixtures. The parameters of the equation for the suspensions are: W/S = -10.4(±2.7) + 1.18(±0.05) ·W, correlation coefficient = 0.994 (P-value < 0.01); for the two-layer systems: W/S = -26.4(±6.6) + 1.50(±0.16) ·W, correlation coefficient = 0.989 (P-value < 0.05). The statistically significant linearity of the relationships indicates that the data both for the two-layer systems and for the nAM + dAM suspension mixtures demonstrate a hyperbolic relation between S and W values. If so, it is possible to calculate the X-axis asymptote for each curve in Figure 2 as a value for the threshold nAM/dAM ratio in accordance with the relation: Threshold = a/b, where a and b are parameters of the linear model mentioned above (the absolute value of a is calculated). The following values for the thresholds were obtained: 9±2% for the suspensions, and 18±5% for the two-layer systems (means ± SEM). These thresholds are the levels of dAM (in %, w/w) when the redistribution of TT-complex begins in the systems.
As seen from the shape of the curve A, the
degree of the TT-complex redistribution depends on the weight ratio of the
layers, therefore, diffusion of the minor components is triggered by the weight
ratio, rather than by the existence of the gradient alone. This two-layer
system as a whole initiates the protein diffusion at a definite nAM/dAM threshold (critical)
ratio.
This "threshold" hypothesis is confirmed by the data shown in Figure
3. As seen from Figure 2, the curve describing the dependence of Ca2+-sensitivity
on the weight portion of dAM
(without the points for 0 and 100% dAM
because these points do not reflect interactions in two-layer systems) looks
like a hyperbola. After transformation of the data of Figure 2 into the
coordinates of Figure 3, this dependence becomes linear. The linear char acter
of the relationships demonstrated by Figure 3 is statistically significant for
the two-layer system at the 95% confidence level (correlation coefficient =
0.989, P < 0.05) and for the two-component suspension, at the 99% confidence
level (correlation coefficient = 0.994, P < 0.01). This linearity proves the
hyperbolic character of the relationships seen in Figure 2 for interacting
actomyosin preparations. Therefore, we can regard the X-axis asymptote (see the
legend for Figure 3) for each curve as a corresponding nAM/dAM thresh old ratio. The
threshold value for dAM
content to initiate blockage of Ca2+-sensitivity of nAM was calculated to be
18 ± 5% for the two-layer system and 9 ± 2%, for the two-component suspension.
Since nAM (the
top layer) lost its Ca2+-sensitivity, the question arises as to
whether the change in the content of the TT-complex actually takes place in
this layer of the two-layer system under the conditions of our experiment. To
answer this question, several two-layer systems with different nAM/dAM ratios were formed
(Figure 1), and the content of the regulatory proteins in the top layer of the
systems was determined (after 10-15 h of incubation) by the method of
electrophoresis. Figure 4 shows that the content of all the studied components of
the regulatory complex fall sharply when the weight proportion of dAM exceeded a certain
threshold.
FIGURE 4. The tropomyosin, troponin T, and troponin I contents in the natural actomyosin (nAM) arranged as the top layer of the two-layer systems (see Figure 1), plotted as a function of the weight portion of the desensitized actomyosin (dAM) in the lower layer after 15-20 h of incubation (at 4-6°C). These data indicate that the TT-complex diffuses from the nAM to the dAM layer in the two-layer systems. The value "0" on the X-axis means nAM preparation alone. The value "100" on the X-axis means dAM preparation alone. The data points are the means ± SEM from triplicate assays.
Figure 2 into the coordinates of Figure 3,
this dependence becomes linear. The linear character of the relationships
demonstrated by Figure 3 is statistically significant for the two-layer system
at the 95% confidence level (correlation coefficient = 0.989, P < 0.05) and
for the two-component suspension, at the 99% confidence level (correlation
coefficient = 0.994, P < 0.01). This linearity proves the hyperbolic
character of the relationships seen in Figure 2 for interacting actomyosin
preparations. Therefore, we can regard the X-axis asymptote (see the legend for
Figure 3) for each curve as a corresponding nAM/dAM thresh old ratio. The
threshold value for dAM
content to initiate blockage of Ca2+-sensitivity of nAM was calculated to be
18 ± 5% for the two-layer system and 9 ± 2%, for the two-component suspension.
Since nAM (the top layer) lost its Ca2+-sensitivity, the question arises as to
whether the change in the content of the TT-complex actually takes place in
this layer of the two-layer system under the conditions of our experiment. To
answer this question, several two-layer systems with different nAM/dAM ratios were formed
(Figure 1), and the content of the regulatory proteins in the top layer of the
systems was determined (after 10-15 h of incubation) by the method of
electrophoresis. Figure 4 shows that the content of all the studied components
of the regulatory complex fall sharply when the weight proportion of dAM exceeded a certain
threshold.
With respect to the Ca2+-sensitivity changes, similar results were
also obtained for mixed nAM
and dAM suspensions
containing different proportions of dAM
(Figure 2, curve B) (e.g., with a final protein concentration of 1 mg/ml, each
ml in the 40% dAM
mixture would contain 0.4 mg of dAM
and 0.6 mg of nAM).
Like in the case when the dAM
weight proportion in the two-layer systems reached the 40% level, the
two-component mixture completely lost Ca2+-sensitivity at the same
level of dAM,
although it consisted of 60% of nAM
containing a significant amount of the native TT-complex. The cooperative
character of the B
curve (Figure 2) might indicate an interaction between the two types of
actomyosin preparations, otherwise the relationship between the Ca2+-sensitivity
and dAM content
would have had a linear character.
As the TT-complex is the receptor of Ca2+-, the Ca2+-dependence
of the nAM-dAM interaction was
studied. These sets of experiments were performed as were other experiments
with two-component suspensions but with the following differences: incubation
time was 1 h; incubation medium contained 0.5 mM EGTA or 0.5 mM Ca2+.
When Ca2+-sensitivity of the preparations with EGTA was measured,
excess Ca2+ was injected about 1-2 sec prior to the MgATP injection.
Data plotted in Figure 5 show that free Ca2+ enhances the
inactivation of the Ca2+-sensitive system (by dAM) but does not affect
the regression slope of the Ca2+-sensitivity versus dAM: the close-to-linear
parts of the curves are almost parallel.
Our measurements showed that the tropomyosin (in fact, the TT-complex) ADC
(apparent diffusion coefficient) varied within the limits of 4.2-4.4·10-4 cm2/sec
in the presence of Ca2+ (0.5 mM) and within the limits of
2.7-4.2·10-4 cm2/sec in the presence of EDTA (0.5 mM). Thus, it is
evident that divalent cations (Ca2+ and Mg2+) do not
significantly affect the rate of the tropomyosin/TT-complex diffusion under the
conditions of our experiments.
FIGURE 5. The Ca2+-sensitivity of the two-component suspension mixtures (1 mg of protein per ml) of natural actomyosin (nAM) with desensitized actomyosin (dAM) plotted as a function of the weight portion of the dAM in the mixture after one-hour incubation (at 4-6°C) in the presence of 0.5 mM EGTA or 0.5 mM Ca2+. When Ca2+-sensitivity of the preparations with EGTA was measured, excess of Ca2+ was injected about 1-2 sec prior to the MgATP injection. The value "0" on the X-axis means 100% nAM content in the suspension. The data points are the means ± SEM from triplicate assays.
Discussion
An understanding of the nature of the Ca2+-sensitivity
of actomyosin superprecipitation is essential to interpreting the results
obtained in this work. The TT-complex has been shown to be the final component
of the excitation-contraction coupling system (11). This complex also
determines Ca2+-sensitivity of the natural actomyosin superprecipitation (2-4).
Its removal results in the loss of Ca2+-sensitivity in actomyosin,
but addition of the TT-complex to a reconstituted actomyosin that contains no
other proteins but actin and myosin produces Ca2+-sensitivity of its
superprecipitation.
In this connection, of interest are our data using the two-layer gel (Figure
1). We found a new type of interaction of the contractile proteins both in
two-layer systems and in suspen sion mixtures. The nAM preparations studied had a high Ca2+-sensitivity
(3.3 arbitrary units) before interaction with the dAM layer (Figure 2).
However, after the nAM gel con tacts with the dAM gel, the Ca2+-sensitivity of nAM (the top layer) begins
to depend on the dAM/nAM weight ratio in the
two-layer system; this dependence has a threshold and cooperative pattern: nAM completely loses Ca2+-sensitivity
within the narrow limits of this ratio changes, from 22/78 to 40/60. This
suggests that nAM,
the top layer of the two-layer gel, loses the regulatory TT-complex.
The amount of the Ca2+-sensitive complex in the top layer is
actually reduced with an in crease in the amount of dAM in the system (Figure 4). At the dAM/nAM ratio = 4/6, the
tropomyosin amount, for example, in the top layer decreases more than 3 times.
The amount of other regulatory proteins also falls markedly. The
above-mentioned ratio, 4/6, is the limit, after which the actomyosin
preparations lose completely their Ca2+-sensitivity.
The change in the Ca2+-sensitivity of nAM in the two-layer system has a definite
cooperative character that indicates that two actomyosin preparations somehow
interact with each other and this interaction spreads far from the line of
contact of the two gels.
It is clear that the nAM-dAM interaction begins in
a zone of contact of the two gels, but then this process involves the entire
system. Thus, the molecular events in the protein sys tem studied appear
spatially interconnected.
The mechanism of such changes requires the ability of the minor components of
actomyosin gel to move inside the gel. This conclusion is based on the data
presented in Figure 4. These data indicate that in the bulk of the two-layer
gel, after the start of contact was made, the spatial redistribution
(diffusion) of the TT-complex molecules begins to be visible as evidenced by
the minor components leaving the top layer of the system (nAM) to the point where nAM loses its Ca2+-sensitivity.
It is highly probable that the structural basis for the diffusion of the
TT-complex is F-actin, because tropomyosin and the complex strongly bind to it
(4), and the sigmoid character of the interaction is caused by the cooperative
properties of actin filaments and proteins associated with them (12,13).
The nAM-dAM interaction also
occurs in the two-component suspensions (Figure 2, curve B). It is seen that
with a decrease in the Ca2+-sensitive nAM fraction, the decrease in the Ca2+-sensitivity
of the whole mixture is not linear; this relationship might indicate a more
complex, unexpected interaction. In fact, as soon as the nAM content falls below
60%, the Ca2+-sensitivity of the mixture disappears completely.
The interaction of nAM
with dAM in the
presence of free Ca2+ causes a greater decrease in the Ca2+-sensitivity
(Figure 5). This is definitely reasonable, considering multiple Ca2+
effects on the properties and structure of contractile and regulatory proteins
and on their complexes (4).
It is well known that Ca2+, in addition to its effect on the
structure and properties of the TT-complex (14), enhances the cooperativity and
changes some other properties of actin filaments (13,15) as well as reduces by
twice or more the actin affinity to the TT-complex (16). The latter effect is
of interest, as the weakening of the F-actin-TT-complex interaction may
facilitate diffusion of the complex. In the presence of EGTA, actin filaments
are more rigid than in the presence of Ca2+ (17); but in the
presence of Ca2+, the regulated actin (F-actin + TT-complex) and
F-actin alone become very flexible (18,19). From our point of view, the
flexibility of actin filaments is one of the factors modulating the rate of the
de crease or loss of the Ca2+-sensitivity after the beginning of the
nAM-dAM interaction: the
flexible filament facilitates the diffusion process and thus accelerates
redistribution of the TT-complex.
It is interesting to compare the TT-complex apparent diffusion coefficient
(ADC) obtained in this work for the condensed actomyosin gel (at 4-6°C) with
that for diffusion of other substances in water (at 25°C) (10): 1.10·10-5,
ethanol; 1.18·10-5, urea; 4.6·10-6, sucrose; 6.9·10-7,
hemoglobin; 13.3·l0-7 cm2/sec, myoglobin. It should be
noted again that in our experiments, tropomyosin was used as a protein marker
for the TT-complex. It might well be that the ADC obtained for
"tropomyosin" is in fact the coefficient for TT-complex.
Thus, the TT-complex ADC measured at a lower temperature and in much more
viscous, almost solid, medium than water has an ADC value that is one order of
magnitude higher than that for small molecules of ethanol and urea and three
orders of magnitude higher than that of hemoglobin and myoglobin. The diffusion
coefficients for other proteins (lysozyme, ovalbumin, bovine serum albumin), and
tobacco virus, as well as for DNA, also appeared to be three orders of
magnitude lower (20) than that for TT-complex. Hence, if the TT-complex
diffused in the water inside the actomyosin gel, its diffusion coefficient
would be much lower. In other words, the mechanism of the fast diffusion does
not allow TT-complex to dissociate from actomyosin.
On the other hand, protein diffusion in the cytoplasm (condensed protein
system) is much slower than in aqueous solutions. Thus, the ADCs for proteins
in cytoplasm are: 1.0·10-8 cm2/sec, bovine serum albumin (human fibroblasts, at
22°C) (21); 5.7·10-8 cm2/sec, aldolase (Swiss 3T3 cells, at 36°C)
(22); 2-3·10-9 cm2/sec, G-actin (embryonic chicken
gizzard cells, at 36°C) (23); 1.2·10-7 cm2/sec, myoglobin
(muscle fibers, at 22°C) (24). As a rule, the protein mobility in the cytoplasm
is much slower than in aqueous solutions. This indicates that the mobility
inside the cell is limited by interaction of diffusible proteins with the
cytoplasmic matrix (25).
It is interesting to discuss which properties of F-actin can promote the
"sliding" diffusion of regulatory proteins bound to it: (1)
tropomyosin bound to actin has an ability for lateral movement (26-30); (2)
F-actin itself can facilitate diffusion of the TT-complex by sponta neous bends
(18,19), a "twist" of monomeric units in coordination with each other
(31), and rotational movements of actin globules (32).
Our findings seem to indicate that the actomyosin gel can be a convenient model
for studies on processes of intracellular signalling systems related to actin
filaments. This model (a protein gel based on F-actin) is most likely to
provide promising opportunities for studies of conditions required for protein
diffusion in condensed protein systems like the cytoplasm. The phenomenon of
the fast diffusion of TT-complex recalls findings of a higher mobility of
cations bound to polyanions in comparison with the their lower motility in
water (for review see 33). These data indicate that the association does not
necessarily lead to a pronounced reduction in mobility.
So the data obtained can be summarized in the following words. Natural
actomyosin enriched in the troponin-tropomyosin complex interacts specifically
(by a mechanism still un known) with desensitized actomyosin free of the
regulatory complex. This results in (i) loss of Ca2+-sensitivity in
natural actomyosin placed both as a layer of the two-layer system and as a
component of its mixture with desensitized actomyosin, and (ii) redistribution
of the TT-complex in the actomyosin gel. The unknown mechanisms of these
effects may play an important role in the living cell.
The author wishes to thank sincerely Dr. A.A. Vereninov for helpful support of
this study. The author is also grateful to Dr. L.Z. Pevzner for criticism and
to Mr. J. Picheny for his assistance.
REFERENCES
1. Edelman, G.M. (1976) J. Biochem. (Tokyo)
79:1P.
2. Schaub, M.C. and Perry, S.V. (1969) Biochem. J. 115:993.
3. Farah, C.S. and Reinach, F.C. (1995) FASEB J. 9:755.
4. Marston, S. (1995) Int. J. Biochem. Cell Biol. 27:97.
5. Haga, T., Maruyama, K. and Noda, H. (1965) Biochim. Biophys. Acta 94:226.
6. Matveev, V.V. and Suzdalskaya, I.P. (1980) Tsitologiia 22:676.
7. Spudich, J.A. and Watt, S. (1971) J. Biol. Chem. 246:4866.
8. Sheludko, N.S.(1975) Tsitologiia 17:1148.
9. Itzaki, R.F. and Gill, D.M. (1964) Analyt. Biochem. 9:401.
10. Chang, R. (1977) Physical Chemistry with Application to Biological Systems.
Macmillan Publishing Co. Inc., New York.
11. Katz, A.M. (1995) Cardioscience 6:1.
12. Brandt, P.W., Diamond, M.S., Gluck, B., Kawai, M. and Schachat, F. (1984)
Carlsberg Res. Commun. 49:155.
13. Praser, I.D. and Marston, S.B. (1995) J. Biol. Chem. 270:7836.
14. Leavis, P.C. and Gergely, J. (1984) CRC Crit. Rev. Biochem. 16:235.
15. Popp.D. and Maeda.Y. (1993) J. Mol. Biol. 229:279.
16. Wegner, A. and Walsh, T.P. (1981) Biochemistry 20:5633.
17. Schwienbacher, C., Magri, E., Trombetta, G. and Grazi, E. (1995)
Biochemistry 34:1090.
18. Fujime, S. and Ishiwata, S. (1972) J. Mol. Biol. 62:251.
19. Ishiwata, S. and Fujime, S. (1972) J. Mol. Biol. 68:511.
20. Dubin, S.B., Lunacek, J.H. and Benedek, G.B. (1967) Proc. Natl. Acad. Sci.
USA. 57:1164.
21. Wojcieszyn, J.W., Schlegel, R.A., Wu, E.S. and Jacobson, K.A. (1981) Proc.
Natl. Acad. Sci. USA. 78:4407.
22. Pagliaro, L. and Taylor, D.L. (1988) J. Cell Biol. 107:981.
23. Kreis, T.E., Geiger, B. and Schlessinger, J. (1982) Cell 29:835.
24. Jurgens, K.D., Peters, T. and Gros, G. (1994) Proc. Natl. Acad. Sci. USA.
91:3829.
25. Gershon, N.D., Porter, K.R. and Trus, B.L. (1985) Proc. Natl. Acad. Sci.
USA. 82:5030.
26. Hanson, J., Lednev, V., O'Brien, E.J. and Bennet, P.M. (1972) Cold Sring
Harbor Symp. Quant. Biol. 37:311.
27. Haselgrove, J.C. (1972) Cold Sring Harbor Symp. Quant. Biol. 37:341.
28. Huxley, H.E. (1972) Cold Sring Harbor Symp. Quant. Biol. 37:361.
29. Parry, D.A.D. and Squire, J.M. (1973) J. Mol. Biol. 75:33.
30. Lehman, W., Craig, R. and Vibert, P. (1994) Nature 368:65.
31. Stokes, D.L. and DeRosier, D.J. (1987) J. Cell Biol. 104:1005.
32. Yanagida, T. and Oosawa, F. (1980) J. Mol. Biol. 140:313.
33. Ling, G.N. In: Search of the Physical Basis of Life. Plenum Press, New York
and London, 1984.
Full text [PDF]
Author’s remarks made after publication of this article
1. The main phenomenon of this work is
illustrated by Fig.2. It is needed to demonstrate the protein-protein interaction
on Fig.2 by some direct method. If so, it will open a way for further
experiments.
2. In our conditions, diffusion of tropomyosin-troponin complex began after 6
hour’s period after creation of the two-tube systems and came to an end after
12 hours. So, the diffusion takes up about 6 hours.
3. The value of tropomyosin/actin ratio in dAM
has important role for nAM-dAM interaction.
4. In some cases (in 20% of all experiments) a freshly extracted natural
actomyosin has no Ca-sensitivity. All experiments described above were carried
out with natural actomyosin Ca-sensitivity of about 3 a.u.