Studies of the molecular mechanism of motile activity require the capacity to examine the properties of individual, isolated molecular components and the properties of these same molecular components in the organised system. Pressure perturbation is one method which can be applied to motile systems at different levels of organisation. We show here that pressure perturbs a specific interaction between actin and myosin in solution and also perturbs the cycling crossbridge in a contracting muscle.

Muscle contraction is the result of a dynamic interaction between several proteins within the muscle fibre. A complete molecular description of this process requires information on both the behaviour of individual isolated proteins and the behaviour of these same proteins when assembled into the contracting muscle. Few methods can be readily applied to proteins both in solution and in the cellular environment. Over the past 5 years we have been exploring the use of pressure perturbation methods for probing protein—protein interactions at different levels of organisation in muscle.

Before detailing the principal results we have obtained, a few words are appropriate about the background and methods of this novel approach.

The use of pressure to perturb a system at equilibrium (or in a steady state) depends upon there being a difference in the volume of the components of the equilibrium mixture.

where Keq is the equilibrium constant.

If the molar volume of A differs from that of B by ΔV then for a small pressure change ΔP:

where ΔK is the pressure-induced change in equilibrium constant, R is the gas constant and T is the absolute temperature. In solution, volume changes normally result from changes in the local water structure when the number of ionic and/or hydrophobic groups exposed to the solvent is changed. Protein-protein and protein-ligand interactions or the resulting protein conformational changes frequently involve changes in the side chains that are exposed to solvent and so can have significant volume changes.

The magnitudes of the perturbation induced in the concentrations of A and B depend not’only on ΔV but also on the magnitude of K. A 10 % change in K will result in a much greater change in [B]/([A]+[B]) when K=X than when K is much greater or much less than 1.

We have been using a pressure chamber for both solution and muscle fibre studies that is based on the equipment built by Davis and Gutfreund (1976). They originally used this apparatus to study the assembly of myosin into filaments (Davis, 1981a,b). The apparatus shown in Fig. 1A applies hydrostatic pressure via an oil line to a solution in the centre of a 3×3×3 in stainless-steel block. The oil is separated from the experimental solution by a deformable Teflon membrane. When the solution is stabilised at the elevated pressure, the pressure can be released via a specially constructed valve that returns the pressure to 101.3 kPa (=latm) within 200 ps.

Fig. 1.

Diagram of the pressure equipment. (A) A cross section of the pressure cell. The chamber is milled from a single piece of stainless steel (3×3×3 in). A, observation cell; B, hydraulic chamber; C, absorbancy photomultiplier; D, thermostated base; E, quartz fibre optic from light source; F, pressure transducer; G, pressure line; H and I, ports for filling sample chamber; J, fluorescence window; K, bursting disc release valve; M, trigger mechanism; N, reset mechanism; 0, valve seat. (B) Schematic drawing of the tension transducer assembly, a, strain gauge with hook enclosed in glass tube (d); b, stainless steel tube housing the insulated terminals of the transducer element; c, epoxy resin seal at the junction of the glass and steel tubes; e and f, Main body of the transducer plug comprising a stainless steel cap (e) and brass body (f); g, rubber ‘O’ rings to maintain pressure seal; h, transducer terminals.

Fig. 1.

Diagram of the pressure equipment. (A) A cross section of the pressure cell. The chamber is milled from a single piece of stainless steel (3×3×3 in). A, observation cell; B, hydraulic chamber; C, absorbancy photomultiplier; D, thermostated base; E, quartz fibre optic from light source; F, pressure transducer; G, pressure line; H and I, ports for filling sample chamber; J, fluorescence window; K, bursting disc release valve; M, trigger mechanism; N, reset mechanism; 0, valve seat. (B) Schematic drawing of the tension transducer assembly, a, strain gauge with hook enclosed in glass tube (d); b, stainless steel tube housing the insulated terminals of the transducer element; c, epoxy resin seal at the junction of the glass and steel tubes; e and f, Main body of the transducer plug comprising a stainless steel cap (e) and brass body (f); g, rubber ‘O’ rings to maintain pressure seal; h, transducer terminals.

For optical studies of protein solutions the cell has three sapphire windows, which allow both transmission and fluorescence measurements to be made. For mechanical studies on muscle fibres one of the windows is removed and is replaced by a tension transducer mounted on a pressure sealing plug (Fig. IB). The muscle fibre is glued between a hook attached to the transducer and a second hook fixed to the transducer mounting before the whole assembly is inserted into the pressure chamber.

The effect of pressure on actin—myosin subfragment 1 in solution

In 1982 we published experiments which showed that the interaction between actin and myosin subfragment 1 (actin.Sl) was pressure sensitive (Geeves and Gutfreund, 1982). Exposure of a solution of actin.Sl (or actin.Sl.ADP) to hydrostatic pressures of up to 20 MPa. (200 atm) resulted in the dissociation of a small fraction of the complex as judged by the change in light scattering. Rapid release of pressure back to 101 kPa resulted in an exponential increase in the concentration of the actin.Sl complex. A series of such relaxations induced by different pressure jumps is shown in Fig. 2. For a simple one step binding reaction the relaxation time is defined by:

Fig. 2.

The pressure-induced changes in light scattering of actin.Sl .ADP. The three traces represent the observed changes in light scattering (monitored as transmission changes at 360 nm) for pressure jumps of 17.5, 12.5, and 7.5 MPa. The arrow indicates the time at which pressure was released. Each trace is the average of five successive relaxations and the single exponential fit is superimposed. The reciprocal relaxation times were 0.56, 0.65, and 0.62 s-1 respectively. Conditions: pH 8, 21 °C and ionic strength of 0.135 M. Vertical bar, 0.5 % transmission. Horizontal bar, 2 s.

Fig. 2.

The pressure-induced changes in light scattering of actin.Sl .ADP. The three traces represent the observed changes in light scattering (monitored as transmission changes at 360 nm) for pressure jumps of 17.5, 12.5, and 7.5 MPa. The arrow indicates the time at which pressure was released. Each trace is the average of five successive relaxations and the single exponential fit is superimposed. The reciprocal relaxation times were 0.56, 0.65, and 0.62 s-1 respectively. Conditions: pH 8, 21 °C and ionic strength of 0.135 M. Vertical bar, 0.5 % transmission. Horizontal bar, 2 s.

where the bars over concentrations define equilibrium concentrations. The concentration dependence of τ-1 defines k+1 and k-i and the values obtained, over a range of reaction conditions, agreed with the results of earlier stopped-flow studies (White and Taylor, 1976; Trybus and Taylor, 1976; Marston, 1982).

The fluorescence of a pyrene group covalently attached to Cys 374 of actin is quenched by 70 % when SI binds to actin (Kouyama and Mihashi, 1981; Criddle et al. 1985) and provides a more sensitive monitor of the actin.SI interaction. Pressure jumps were performed on a solution of actin.Sl in the absence of nucleotide and show two clear relaxations (Fig. 3). No relaxation is observed in the absence of SI and, therefore, the two relaxations must represent two relaxations of the acto.SI complex. Monitoring the same reaction by light scattering shows only a single relaxation and the concentration dependences of the relaxation time and amplitude are identical to those of the slower relaxation observed by pyrene fluorescence. The fast relaxation was complete in the pressure release time (200,us) and the amplitude increased with protein concentrations. The amplitude of the slow relaxation decreased at higher protein concentration. The experimental data were interpreted in terms of the following model in which step 2 is fast, pressure-sensitive and results in the change in pyrene fluorescence. Step 1 is a second-order reaction and results in a change in light scattering (Coates et al. 1985):

Fig. 3.

The pressure-induced changes in fluorescence of pyrene-actin.Sl. The arrow indicates the time of pressure release from 10 MPa to 0.1 MPa. The traces are an average of five successive relaxations on the same solution with the best-fit single exponential to the slow phase superimposed. The [SI] and the reciprocal relaxation times in each case were: (A) 2.8μM, 1.24s-1; (B) 8.4μM, 7.32 s-1; (C) 19.5 pM, 24.3 s-1. The fluorescence is expressed relative to the signal at high pressure. Conditions: actin 2.5μM, pH 7.5, 20°C and ionic strength 0.13 M.

Fig. 3.

The pressure-induced changes in fluorescence of pyrene-actin.Sl. The arrow indicates the time of pressure release from 10 MPa to 0.1 MPa. The traces are an average of five successive relaxations on the same solution with the best-fit single exponential to the slow phase superimposed. The [SI] and the reciprocal relaxation times in each case were: (A) 2.8μM, 1.24s-1; (B) 8.4μM, 7.32 s-1; (C) 19.5 pM, 24.3 s-1. The fluorescence is expressed relative to the signal at high pressure. Conditions: actin 2.5μM, pH 7.5, 20°C and ionic strength 0.13 M.

Thus, an increase in pressure reduces K2 and will lead to an overall decrease in affinity of actin for SI. Only at low concentrations of protein will this result in dissociation of a significant fraction of the complex. Under these conditions, rapid release of pressure will result in rapid reequilibration of step 2, observed as a change in fluorescence. This will be followed by a slower rebinding of SI to actin which can be observed by both light scattering and fluorescence. However, at very high protein concentration, increases in pressure will cause no dissociation of the two proteins but will still perturb the first-order isomerization of step 2.

Geeves et al. (1984) had earlier proposed that such a two-step binding of actin to myosin and myosin-nucleotide complexes was a general property of the proteins and that the isomerization step may be associated with the force generating event of the crossbridge cycle. This view has been supported by a range of biochemical studies in solution, including both measurements of K2 for a range of nucleotides and nucleotide analogues and evidence that the rate of product release from actin.SI remains slow unless the isomerization takes place (Geeves et al. 1986; Geeves, 1989; Geeves and Jeffries, 1988; Woodward et al. 1991).

In order for the isomerization to be responsible for force generation it must involve a substantial structural change in the actin.SI complex. The pressure sensitivity of the isomerization step allows an estimation of a volume change of 100 cm3 mole-1 which suggests a significant rearrangement of the protein (Coates et al. 1985). The isomerization has also been shown to perturb the environment around Cys374 of actin and the 2’/3’ hydroxyl groups of the ADP ribose simultaneously, even though these groups are believed to be some distance apart within the acto.Sl.nucleotide complex (Woodward et al. 1991).

If the isomerization is coupled to the force-generating event then factors that affect the isomerization should also effect the generation of force. Both increases in ethylene glycol concentration and increases in ionic strength reduce the equilibrium constant for the isomerization step and reduce the level of isometric force generated in a muscle fibre (Coates et al. 1985; Clarke et al. 1980; Gulati and Podolsky, 1981). Increased pressure would also be expected to decrease isometric force. To test this prediction, we set out to measure the effect of hydrostatic pressure on isometric force in single muscle fibres.

The effect of pressure on muscle fibres

In the 1930s a series of studies reported the effects of changes in hydrostatic pressure on whole muscle from the frog and turtle (Brown, 1934a,b;Cattel and Edwards, 1928). Increases in pressure caused reversible changes in both tetanic and twitch tension; the size and magnitude of the effects depended upon the exact conditions of the experiment. To distinguish the effect of pressure on the crossbridge cycle from effects on excitation-contraction coupling we used chemically skinned rabbit psoas fibres. Fibres were fully calcium activated by exchanging the bathing solution around the fibre whilst it was mounted inside the pressure chamber. The results of a series of stepped increases in pressure on the measured tension of a single muscle fibre are shown in Fig. 4. The results show that in the relaxed state, changes in pressure cause no measurable change in the output of the tension transducer. The same result was observed if the relaxed fibre was stretched to hold a large passive tension before the pressure changes were applied. A similar negligible change in measured tension was observed for a stretched rubber filament (Geeves and Ranatunga, 1987; Fortune et al. 1989a,b).

Fig. 4.

The effect of changes in pressure on the steady-state tension of glycerinated muscle fibres from rabbit psoas. (A) A small fibre bundle under relaxing conditions; 10 mM ATP, low Ca2+, 20°C, pH7, ionic strength 200mM. (B) The same fibre bundle holding a rigor tension of 520 uN. Conditions as in (A) but with no ATP. (C) A single fibre holding a steady fully active isometric tension of 1980 kN m-2. Conditions as in (A) but with 30 μM free calcium.

Fig. 4.

The effect of changes in pressure on the steady-state tension of glycerinated muscle fibres from rabbit psoas. (A) A small fibre bundle under relaxing conditions; 10 mM ATP, low Ca2+, 20°C, pH7, ionic strength 200mM. (B) The same fibre bundle holding a rigor tension of 520 uN. Conditions as in (A) but with no ATP. (C) A single fibre holding a steady fully active isometric tension of 1980 kN m-2. Conditions as in (A) but with 30 μM free calcium.

A fibre holding rigor tension showed an increase in measured tension with increases in pressure. The increases in tension were approximately linearly related to pressure over the range 0.1–8 MPa, with some irreversible loss of tension occurring at higher pressures. Each increase in tension corresponded to that observed following a 0.3 % length increase. Increases in tension were also observed for stretched filaments of glass, copper, silk, collagen and human hair (Ranatunga et al. 1990). The magnitude of the increase was specific to each material and is believed to reflect the volume changes in the material in response to a length change (defined by the Poissons ratio for a pure material). The relaxed fibre is rubber-like in its response to a pressure change, i.e. there is no volume change of the system on stretching. The fibre in rigor has a different response and this must represent some additional load-bearing element present in the fibre in rigor. The only known difference is the presence of attached crossbridges and the pressure sensitivity is therefore most likely to reflect the mechanical response of the non-cycling crossbridge or a structural element in series with the crossbridge.

The steady isometric tension of a muscle fibre in a fully activated contraction decreases as the pressure is increased (by approximately 8% per 10 MPa pressure increase) and the tension is fully recovered on pressure release over the pressure range 0.1–15 MPa (Fig. 4C) (Geeves and Ranatunga, 1987; Fortune et al. 1989a). This decrease is specific to the active muscle fibre and must represent perturbation of some component which is not present in the rigor fibre, i.e it is likely to represent a perturbation of the dynamic crossbridge.

To investigate this possibility further we examined the rate of tension change following a rapid release of 10 MPa pressure. The results for rigor and active muscle fibres are shown in Fig. 5; the relaxed fibre showed no tension transient. The fibre in rigor showed a stepped decrease in pressure, occurring in phase with the pressure change (complete in approximately 2 ms) compatible with a mechanical relaxation of an elastic element. The active fibre, in contrast, showed a complex response. Initially, a rapid decrease in tension was seen in phase with the pressure release. This is similar to the response observed in the rigor fibre and may represent a perturbation of the same elastic element. If this rigor-like response were the only perturbation present in the active fibre, then a very rapid recovery of tension to the initial level might be expected, as is seen in response to a small length change. However, we observed a recovery of tension back to that originally observed at 0.1 MPa. The time course of the recovery can be fitted by two exponentials with l/ τ 1=60s-1 and 1/ τ 2=5.5S-1 (Fortune et al. 1989b; Geeves et al. 1990). These two additional relaxations are only observed in the active muscle and suggest that a crossbridge event is being perturbed. The solution studies show that the isomerization step of actin.Sl association is pressure sensitive and this isomerization remains the most likely candidate for the pressure response seen in the muscle fibre.

Fig. 5.

Pressure induced tension transients in single glycerinated rabbit psoas fibres. Pressure was released from 10 MPa to 0.1 MPa in about 1ms. Solution conditions were the same as in Fig. 4. (A) A single fibre holding a rigor tension of 180 μN. (B) and (C) A single fibre holding a steady active tension of 520 μN. The numbers in (B) identify the three principal phases observed. The same relaxation is shown on two time scales with a single exponential fit to the slowest phase (5.5 s-1) in (B) and a double exponential fit (5.6 and 60 s’1) in (C). The arrows indicate the time of pressure release, f.s., full scale.

Fig. 5.

Pressure induced tension transients in single glycerinated rabbit psoas fibres. Pressure was released from 10 MPa to 0.1 MPa in about 1ms. Solution conditions were the same as in Fig. 4. (A) A single fibre holding a rigor tension of 180 μN. (B) and (C) A single fibre holding a steady active tension of 520 μN. The numbers in (B) identify the three principal phases observed. The same relaxation is shown on two time scales with a single exponential fit to the slowest phase (5.5 s-1) in (B) and a double exponential fit (5.6 and 60 s’1) in (C). The arrows indicate the time of pressure release, f.s., full scale.

Two additional observations support this view. Increases in pressure in the range used here caused no measurable change in the equatorial X-ray scattering from relaxed, rigor or active muscle fibres, suggesting no change in lattice spacing or in the number of attached crossbridges (Knight et al. 1990). The presence of phosphate in the active solution increases the observed lμ2 in the active fibre, consistent with this process involving a reversible binding of phosphate (N. S. Fortune, M. A. Geeves and K. W. Ranatunga, unpublished data).

The series of studies outlined here has resulted in the identification of a pressure-sensitive transition occurring both in solutions of purified actin and myosin and in contracting muscle fibres. The results so far are consistent with a direct correlation between the specific isomerization of actin.Sl in solution, the phosphate release step and a force-generating transition in the crossbridge cycle.

This work has been supported by the Wellcome Trust and the European Economic Community. The author is a Royal Society University Research Fellow.

Brown
,
D. E. S.
(
1934
).
Effect of rapid changes in pressure upon the contraction of skeletal muscle
.
J. cell. comp. Physiol.
4
,
257
—281.
Brown
,
D. E. S.
(
1936
).
Effect of rapid compression on events in the isometric contraction of skeletal muscle
.
J. cell. comp. Physiol.
8
,
141
157
.
Cattel
,
McK.
and
Edwards
,
D. J.
(
1928
).
The energy changes of skeletal muscle accompanying contraction under high pressure
.
Am. J. Physiol.
86
,
370
382
.
Clarke
,
M. L.
,
Rodger
,
C. D.
,
Tregear
,
R. T.
,
Bordas
,
J.
and
Koch
,
M.
(
1980
).
Effect of ethylene glycol and low temperature on the structure and function of insect flight muscle
.
J. Mus. Res. Cell Motil.
1
,
195
196
.
Coates
,
J. H.
,
Criddle
,
A. H.
and
Geeves
,
M. A.
(
1985
).
Pressure relaxation studies of pyrene labelled actin and myosin subfragment 1 from rabbit muscle. Evidence for two states of acto-subfragment 1
.
Biochem. J.
232
,
351
—356.
Criddle
,
A. H.
,
Geeves
,
M. A.
and
Jeffries
,
T.
(
1985
).
The use of actin labelled with N-(l-pyrenyl)iodoacetamide to study the interaction of actin with myosin subfragments and troponin/tropomyosin
.
Biochem. J.
232
,
343
349
.
Davis
,
J. S.
(
1981a
).
The influence of pressure on the self assembly of the thick filament from myosin of vertebrate skeletal muscle
.
Biochem. J.
197
,
301
—308.
Davis
,
J. S.
(
1981b
).
Pressure jump studies of the length regulation kinetics of the self assembly of myosin from vertebrate skeletal muscle into thick filaments
.
Biochem. J.
197
,
309
314
.
Davis
,
J. S.
and
Gutfreund
,
H.
(
1976
).
The scope of moderate pressure changes for kinetic and equilibrium studies of biochemical systems
.
FEBS Lett.
72
,
199
207
.
Fortune
,
N. S.
,
Geeves
,
M. A.
and
Ranatunga
,
K. W.
(
1989a
).
Pressure sensitivity of active tension in glycerinated rabbit psoas muscle fibres; effects of ADP and phosphate
.
J. Muscle Res. Cell Motil.
10
,
113
123
.
Fortune
,
N. S.
,
Geeves
,
M. A.
and
Ranatunga
,
K. W.
(
1989b
).
Tension transients initiated by pressure perturbations in skinned rabbit muscle fibres
.
J. Physiol.
418
,
158P
.
Geeves
,
M. A.
(
1989
).
Dynamic interaction between actin and myosin subfragment 1 in the presence of ADP
.
Biochemistry
28
,
5864
5871
.
Geeves
,
M. A.
,
Fortune
,
N. S.
and
Ranatunga
,
K. W.
(
1990
).
Temperature dependence of pressure induced tension transients in skinned rabbit psoas fibres
.
Biophys. J.
57
,
409
a.
Geeves
,
M. A.
,
Goody
,
R. S.
and
Gutfreund
,
H.
(
1984
).
The kinetics of acto-Sl interaction as a guide to a model for the crossbridge cycle
.
J. Muscle Res. Cell. Motil.
5
,
351
361
.
Geeves
,
M. A.
and
Gutfreund
,
H.
(
1982
).
The use of pressure perturbations to investigate the interaction of rabbit muscle myosin subfragment 1 with actin in the presence of ADP
.
FEBS Lett.
140
,
11
15
.
Geeves
,
M. A.
and
Jeffries
,
T. E.
(
1988
).
Effect of nucleotide upon a specific isomerisation of acto.SI
.
Biochem. J.
256
,
41
46
.
Geeves
,
M. A.
,
Jeffries
,
T. E.
and
Millar
,
N. C.
(
1986
).
The ATP induced dissociation of rabbit skeletal actomyosin subfragment 1. Characterization of an isomerization of the ternary acto.SI.ATP complex
.
Biochemistry
25
,
8454
8458
.
Geeves
,
M. A.
and
Ranatunga
,
K. W.
(
1987
).
Tension responses to increased hydrostatic pressure in glycerinated rabbit psoas muscle fibres
.
Proc. R. Soc. B
232
,
217
226
.
Gulati
,
J.
and
Podolsky
,
R. J.
(
1981
).
Isotonic contraction of skinned muscle fibres on a slow time base
.
J. gen. Physiol.
78
,
233
257
.
Knight
,
P.
,
Fortune
,
N. S.
and
Geeves
,
M. A.
(
1990
).
Structural basis of reduction of active muscle tension under pressure
.
J. Muscle Res. Cell Motil.
11
,
73
.
Kouyama
,
T.
and
Mihashi
,
K.
(
1981
).
Fluorimetry study of 7V-(1- pyrenyl)iodoacetamide labelled F-actin
.
Eur. J. Biochem.
114
,
33
38
.
Marston
,
S. B.
(
1982
).
Formation and dissociation of actomyosin complexes
.
Biochem. J.
203
,
453
—460.
Ranatunga
,
K. W.
,
Fortune
,
N. S.
and
Geeves
,
M. A.
(
1990
).
Hydrostatic compression in glycerinated rabbit muscle fibres
.
Biophys. J. (in press)
.
Trybus
,
K. M.
and
Taylor
,
E. W.
(
1980
).
Kinetic studies of the cooperative binding of SI to regulated actin
.
Proc. natn. Acad. Sci. U.S.A.
77
,
7209
7213
.
White
,
H.
and
Taylor
,
E. W.
(
1976
).
Energetics and mechanism of action of actomyosin ATPase
.
Biochemistry
15
,
5818
5826
.
Woodward
,
S. K. A.
,
Eccleston
,
J. F.
and
Geeves
,
M. A.
(
1990
).
Kinetics of the interaction of mant.ATP with myosin Si and acto.Sl. Characterization of two acto.Sl ADP complexes
.
Biochemistry (in press)
.