- Email: [email protected]
The surface active properties of myosin and its proteolytic fragments
Meat Science 26 (1989) 131-140
The Surface Active Properties of Myosin and its Proteolytic Fragments E. O'Neill, a D. M. MulvihilP & P. A. Morrissey b a Department of Food Chemistry, b Department of Nutrition, UniversityCollege, Cork, Republic of Ireland (Received 1 November 1988; revised version receivedand accepted 6 March 1989) ABSTRACT The surface active properties of myosin and its proteolytic fragments, light meromyosin ( LMM), heavy meromyosin ( HMM), subfragment-I (S-l) and myosin rod, at initial bulk phase concentrations in the range of lO- 4% to 10-2% w/v were determined by the drop volume method. Overall, S-1 was the most effective surface tension depressor, whereas the tail portions of myosin, i.e. L M M and myosin rod were less surface active than the parent myosin molecule. The surface pressures attained after 40 min, at an initial bulk phase concentration of 10-2% (w/v), were 22"00, 21.77, 21"02, 16"77 and 16.77 mNm- 1for S-1, HMM, myosin, L M M and myosin rod, respectively. Furthermore, S-1 effected the most rapM change in surface pressure during the initial 5 rain period.
INTRODUCTION The interfacial protein film which surrounds lipid droplets or particles in comminuted sausage-type products is thought to play an important role in the stability of the systems (Swasdee et al., 1982, Jones & Mandigo, 1982). Galluzzo and Regenstein (1978) reported that myosin is rapidly taken up at the fat-aqueous interface during emulsification and recent studies in this laboratory have shown that myosin is more surface active than actin or actomyosin at the air-water interface (O'Neill et al., 1989a). Myosini is an unusual protein in that it contains structural characteristics of both globular and fibrous proteins in one covalently linked functional unit. The globular head region of the molecule hydrolyses ATP and binds strongly to F-actin, 131 Meat Science 030%1740/89/$03.50 © 1989ElsevierSciencePublishersLtd, England.Printed in Great Britain
132
E. O'Neill, D. M. Mulvihill, P. A. Morrissey
while the tail region is responsible for the formation of filaments at low ionic strength. Borejdo (1983) observed that the surface hydrophobicity of myosin is confined almost exclusively to the globular head region. However, the contribution of the different structural domains to the interfacial film forming properties of myosin is not known. In this study, using chymotrypsin digestion to cleave myosin at specific sites, we examined the surface active properties of the major proteolytic fragments of myosin in order to identify the main surface active portion of the molecule. The subfragments examined were as follows: subfragment-1 (S-l), which represents the individual myosin heads; heavy meromyosin (HMM), which contains both heads and a portion of the tail region of myosin; the myosin rod, which comprises almost the complete tail; and light meromyosin (LMM), which consists of a shortened fragment of the myosin tail. The possible structure of the interfacial film formed by myosin at an oil/water interface is also discussed.
MATERIALS AND METHODS All chemicals used were of reagent grade. Distilled water, glassware and dialysis tubing were prepared, as described by O'Neill et al. (1989a).
Determination of protein content Protein contents of the solutions were determined by the Biuret method.
Preparation of protein solutions Rabbit skeletal muscle was used in all studies. Myosin was prepared by the method outlined by Margossian and Lowey (1982), using ion exchange chromatography on DEAE-Sephadex to remove C-protein and traces of contaminating actin. S-I, HMM, LMM and myosin rod were prepared by chymotryptic digestion of myosin according to the procedure of Margossian and Lowey (1982), with HMM and S-1 being purified by ion-exchange chromatography on DEAE-cellulose followed by ammonium sulphate precipitation. All preparations of myosin and its fragments were checked for purity by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (Pollard, 1982), using the staining procedure of Greaser et al. (1983). All the protein solutions were exhaustively dialysed at 4°C against 0.6r~ KCI, pH 7.0. The protein solutions were centrifuged at 20000g for 30 min and the supernatants were filtered through Whatman No. 1 filter paper to remove
Surface active properties of myosin
133
any protein particles that could sediment during surface tension measurements. The protein contents were determined and appropriate concentrations (10-2-10-4% w/v) prepared by dilution with 0-6M KC1, pH 7"0. Surface tension measurements were made immediately after protein preparation. Surface tension measurements
The surface activities of the protein solutions at the air-solvent interface at 25°C were determined by the drop volume principle, using the apparatus described by Tornberg (1977). Calibration of the drop volume apparatus and calculation of surface tension were carried out, as described by Tornberg (1978) and Arnebrant and Nylander (1985). The protein solutions were prepared in triplicate. The surface activity of each protein preparation was determined in quadruplicate for each drop size, and the mean surface tensions were plotted as a function of time. A surface tension decay curve was then plotted for each protein from the mean of the three curves. The standard deviations of these curves were determined using the Minitab Statistical Package (Ryan et aL, 1976) on a DEC Vax 11/780 computer and were within the range of ___0.27mNm-
RESULTS AND DISCUSSION The time dependence of the surface tension, ~, of solutions of myosin and of its proteolytic fragments, light meromyosin (LMM), heavy meromysin (HMM), subfragment-1 (S-l) and myosin rod, in 0"6M KC1 at pH 7.0, at different bulk phase concentrations are shown in Figs 1-5. The initial surface tension of the solvent was found to be 73.27 mNm-1. An initial induction period, where no discernible change in surface tension occurred, was observed for all the proteins at a bulk phase concentration of 10-3% (w/v). DeFeijter and Benjamins (1987) suggested that the induction period is the time taken for the adsorbing protein molecules to attain a sufficient surface coverage to decrease 7. Thus, the difference observed in induction times may reflect the relative rates of protein accumulation at the air-solvent interface. Since dynamic surface properties are considered to be important in the stabilisation oflipid droplets during emulsion formation (Hailing, 1981), the rate of change of surface pressure exerted by a protein, particularly during the first few minutes, is important. The data show that the rate of surface tension decay was dependent on the identity of the adsorbing protein molecules and decreased as the bulk phase concentration of protein was reduced (Figs 1-5). In order to compare the surface behaviours of the
E. O'Neill, D. M. Mulvihill, P. A. Morrissey
134
75 li-
7O
55 A-O--
50
I
I
I
I
10
20
30
40
TIME
(rain)
Fig. 1. Time dependence of surface tension at the air-solvent interface, at 25°C, of myosin at initial bulk phase concentrations of 1 x 10 -4 (m); i x 10 -3 ( o ) ; 5 × 10 -3 (~.) and 1 x 10 -2 (I--q) % (w/v) protein.
various proteins the rates of change of surface pressure (H), AH/At, during the first 30s and 5min, at 1 × 10 - 2 , 5 × l 0 - 3 and 1 × 10-3% (w/v), were calculated from the data shown in Figs 1-5, and are reported in Table 1. Since the rate of change of H is considered to be a function of the physicochemical properties of proteins (Graham & Phillips, 1979a-c; Waniska & Kinsella, 1985; Shimuzu et al., 1985; Song & Damodaran, 1987), the different behaviours of the proteins evident from the data probably reflects the
75 m--
7O
o--
6O
&--
545
I
I
I
I
10
20
30
40
TIME
(min)
Fig. 2. Time dependenceof surface tension at the air-solvent interface, at 25°C, of light meromyosin at initial bulk phase concentrations of I × 10-4 (m); 1 x 10- 3 (O): 5 x 10- 3 (A) and 1 × 10 2 C]J % (w/v) protein.
Surface active properties of myosin
135
75 aDD
7O
0--
55 A-D B
UJ
U)
I
I
I
I
10
20
30
40
TIME
(rain)
Fig. 3. Time dependence of surface tension at the air-solvent interface, at 25°C, of heavy meromyosin at initial bulk phase concentrations of 1 x 10 -4 (ml); l × 10- 3 (O); 5 × 10- 3 (A) and l × l0 -2 (1-]) % (w/v) protein.
chemical and structural properties of the individual proteins. At 1 0 - 2 % (w/v) protein the rate of change of H was fastest for S-1 followed by HMM, myosin, LMM and myosin rod. We consider that AH/At during the initial 30s reflects the relative rates at which the protein molecules migrate and adsorb at the interface and is influenced by such factors as molecular size and surface hydrophobicity (MacRitchie, 1978; Kato & Nakai, 1980). Thus, it is not surprising that the S-1 fragment (mw ~ 115 000 daltons) accumulated
75 7O ] "
!0-A-DI
I 50 I
10
I
I
20
30
TIME
I
40
(min)
Fig. 4. Time dependence of surface tension at the air-solvent interface, at 25°C, of subfragment-l at initial bulk phase concentrations of 1 x l0 -4 (Hal); 1 x i0 -3 (O); 5 x 10 -3 (&) and l x 10 -2 ([]) % (w/v) protein.
E. O'Neill, D. M. Mulvihill, P. A. Morrissey
[36
75 70
O--
D--
ss
10
20 TIME
30
40
(rain)
Fig. 5. Time dependence of surface tension at the air-solvent interface, at 25°C, of myosin rod at initial bulk phase concentrations of 1 x 10 -4 (R); 1 x 10 -3 (O); 5 x 10 -3 (,L) and 1 x 10 -z (1:]) % (w/v) protein.
more readily at the interface than the larger myosin rod (mw ~ 220 000 daltons) and caused a more rapid change in I1 during the early stages. The faster All/At exerted by H M M and myosin compared with LMM and myosin rod, in spite of their large molecular weights (340 000 and 500 000, respectively) is probably due to the greater surface hydrophobicity of these fragments (Borejdo, 1983). The overall rate of change of surface pressure, All/At, during the initial 5 min followed the same order as that for the first 30s, i.e. S-1 exerted the fastest change in II, followed by HMM, myosin, LMM and myosin rod. Between 20 and 40min changes in II were very small for all the proteins (Figs 1-5) and probably reflect further minor rearrangements of the adsorbed molecules rather than adsorption of new protein molecules from the bulk phase. In addition to the rate of change of surface pressure exerted by proteins, the equilibrium surface pressure attained is also an important index of surface activity. The concentration dependence of the surface pressure attained after 40 min, 1-I4o,i.e. the reduction in surface tension after 40 rain, is clearly illustrated in Fig. 6 where 174o is plotted as a function of initial bulk phase concentration. Overall, S-1 was the most surface active protein, whereas the tail portions of myosin, i.e. LMM and myosin rod, were less surface active than the parent myosin molecule. H M M was almost as surface active as S- 1 at concentrations of 5 × 10- 3 and 1 × 10- 2% (w/v) and Flao was more or less independent of bulk phase concentration in this range for these two myosin fragments. However, when the bulk phase concentration was reduced to 10-3% (W/V), differences in surface activity between these fragments were evident, with S-1 being much more efficient at reducing
Surface active properties of myosin
TABLE
137
1
P a r a m e t e r s D e s c r i b i n g the R a t e s o f S u r f a c e T e n s i o n D e c a y at the A i r - S o l v e n t I n t e r f a c e for M y o s i n , S u b f r a g m e n t - 1 , H e a v y M e r o m y o s i n , L i g h t M e r o m y o s i n and Myosin Rod
Protein conc. (% w/v) Myosin I x 10 - z 5 × 10 -3 1 × 10 -3
Induction period (min)
All~At 0"5 rain (mNm- l rnin- l)
A FI/ At 5"0 rain (mNm- 1rain- 1)
---
18.04 10.54
3'25 3"00
4-67
---
0"35
Subfragment- 1 1 × 10 -2
22.00
3.90
-1"67
15.54 ---
3.54 2"01
1 x 10 -2 5 × 10 -3
--
18.54 13.00
3-40 3-10
1 × 10 -3
3'33
5 × 10 -3
1
×
10 - 3
Heavy meromyosin
Light m e r o m y o s i n 1 x 10 -2
1"09
--
14'94
2"85
5 × 10 -3
0.17
6'54
2-17
1 × 10 3
5"50
---
--
--
13"54
2"70
0.27 6.00
5"54 --
2-30 --
Myosin rod 1 x 10 -2 5 × 10 -3 1 x 10 -3
surface tension than HMM. At very low protein concentration, 10- 4% (w/v) only S-1 reduced surface tension substantially. In an earlier study we found that myosin was more surface active than actin or actomyosin (O'Neill et al., 1989a). We suggested that the superior surface active properties of myosin may be attributed to its greater surface hydrophobicity (Borejdo, 1983) and that the elongated nature of the myosin molecule (Lowey et al., 1969; Squire, 1981) may allow a larger surface available for interaction with interface, even when myosin is in its native configuration. However, the present results suggest that the globular heads of myosin play a critical role in the surface activity of myosin. Borejdo (1983) concluded that myosin is a unique protein in that its surface hydrophobicity is confined almost exclusively to the globular head region of the molecule and that the myosin tail is practically devoid of surface hydrophobic sites. Surface hydrophobicity has been reported to correlate significantly with
138
E. O'Neill, D. M. Mulvihill, P. A. Morrissey 25
20
o/ / IS °
10
I
10-4
I
10 "a
t
10-2
BULK PHASE CONCENTRATION (~WlV)
Fig. 6. The surface pressure attained after 40 min, n4o, as a function of initial bulk phase protein concentration for myosin (11); light meromyosin (A);heavy meromyosin (0); subfragment-1 (IS]) and myosin rod ((3).
surface activity of proteins (Kato & Nakai, 1980; Kato et al., 1981); thus, it is not surprising that the present findings show that the fragments which include the myosin head region, i.e. HMM and S-l, were more surface active than those which contain only the tail portion of the molecule. Differences in the interfacial properties of proteins are, in part, attributed to compositional differences (Waniska & Kinsella, 1985; Song & Damodaran, 1987). Thus, it may be possible to explain the observed differences in surface properties between the myosin subfragments on the basis of amino acid composition and distribution. The myosin rod sequence is highly repetitive and has the characteristics of an s-helical coiled coil (McLachlan & Karn, 1982). According to these workers, the rod amino acid sequence has hydrophobic residues strongly concentrated in the centre core of the helix. The helix surface has a high density of charged groups and is essentially devoid of hydrophobic groups. Furthermore, the data of Borejdo (1983) also imply that the myosin rod is essentially devoid of surface hydrophobic groups. Thus, it is likely that the hydrophilic nature of the helix surface militates against the rod portion of the myosin molecule penetrating into the interfacial film, since it would involve the thermodynamically
Surface active properties of myosin
139
unfavourable process of dehydration. Dickinson et al. (1987) suggested that the hinge region may play a crucial role in the interfacial behaviour of myosin by unfolding at low temperatures and exposing the hydrophobic surfaces of the ~-helix coiled coil. However, we did not observe any significant differences in surface activity of LMM, which does not include the hinge region and the myosin rod, which does. The present results are consistent with earlier findings in this laboratory, which showed that thermal denaturation of myosin which causes unfolding of the ~-helical portion of the molecule, exposing the previously hidden hydrophobic core, enhances the surface activity of myosin (O'Neill et al., 1989b). The head region of myosin includes amino acids which are strongly hydrophobic (Elzinga & Collins, 1977). While some of these hydrophobic groups are buried in the interior of the native molecule, many remain exposed and contribute t o surface hydrophobicity. Borejdo (1983) concluded that the hydrophobic pocket present on the surface of myosin is formed between the 41-residue region at the N-terminal end of the A-1 light chain and the heavy chain of S-1. Thus, the initial attraction of the myosin head to the interface and the subsequent rapid reduction in surface tension for S-1 (Table 1) is probably due to its native surface hydrophobicity. Surface denaturation and conformational rearrangement of the adsorbed protein molecules may also contribute to surface tension decay. Since the head region is more surface active than the tail region, as indicated by the present results, and hydrophobic groups are located almost exclusively in the head region (Borejdo, 1983), it is difficult to envisage how the interfacial membrane in meat emulsions could be orientated in the manner proposed by Schut (1976), i.e. with LMM orientated towards the fat phase and the head region attracted towards the aqueous phase. Jones (1984) and the present workers (Morrissey et al., 1987) proposed that the hydrophobic heads of native myosin molecules are orientated towards the surface of the fat globule. On the basis of the present findings we suggest that the proposed model is basically correct. Thus, the myosin heads are strongly adsorbed at the interface between oil and water. It is also likely that the myosin heads undergo a certain degree of rearrangement and surface denaturation following adsorption. The myosin tails projecting from the fat globule surface probably penetrate into the surrounding matrix and undergo protein-protein interaction to form a viscoelastic interfacial film.
REFERENCES Arnebrant, T. & Nylander, T. (1985). J. Dispersion Sci. Technol., 6, 209. Borejdo, J. (1983). Biochern., 22, 1182.
140
E. O'Neill, D. M. Mulvihill, P. A. Morrissey
DeFeijter, J. A. & Benjamins, J. (1987). In Food Emulsions and Foams, ed. E. Dickinson, Royal Society of Chemistry, London, p. 72. Dickinson, E., Murray, B. S., Stainsby, G. & Brock, C. J. (1987). Int. J. Biol. Macromol., 9, 302. Elzinga, M. & Collins, J. H. (1977). Proc. Natl. Acad. Sci. USA, 74, 4281. Galluzzo, S. J. & Regenstein, J. M. (1978). J. Food Sci., 43, 1761. Graham, D. E. & Phillips, M. C. (1979a). J. Colloid Interface Sci., 70, 403. Graham, D. E. & Phillips, M. C. (1979b). J. Colloid Interface Sci., 70, 415. Graham, D. E. & Phillips, M. C. (1979c). J. Colloid Interface Sci., 70, 427. Greaser, M. L., Yates, L. D., Krzywicki, K. & Roelke, D. L. (1983). In Proceedings of the 36th Annual Reciprocal Meat Conference. National Livestock and Meat Board, Chicago, IL, p. 87. Hailing, P. J. (1981). CRC Crit. Rev. Food Sci. Nutr., 15, 155. Jones, K. W. (1984). In Proceedings of the 37th Annual Reciprocal Meat Conference. National Livestock and Meat Board, Chicago, IL, p. 52. Jones, K. W. & Mandigo, R. W. (1982). J. Food Sci., 47, 1930. Kato, A. & Nakai, S. (1980). Biochim. Biophys. Acta, 624, 13. Kato, A., Tsutsui, N., Matsudomi, N., Kobayashi, K. & Nakai, S. (1981). Agric. Biol. Chem., 45, 2755. Lowey, S., Slayter, H. S., Weeds, A. G. & Baker, H. (1969). J. MoL Biol., 42, 1. Margossian, S. S. & Lowey, S. (1982). Method Enzymol., 85, 55. MacRitchie, F. (1978). Adv. Protein Chem., 32, 283. McLachlan, A. D. & Karn, J. (1982). Nature, 299, 226. Morrissey, P. A., Mulvihull, D. M. & O'Neill, E. M. (1987). In Developments in Food Proteins--& ed. B. J. F. Hudson, Elsevier Applied Science Publishers, London, p. 195. O'Neill, E., Morrissey, P. A. & Mulvihill, D. M. (1989a). Food Chem., 35, I. O'Neill, E., Morrissey, P. A. & Mulvihill, D. M. (1989b). Food Chem., 34, 295. Pollard, T. D. (1982). Method Enzymol., 85, 123. Ryan, T. A., Joiner, B. C. & Ryan, B. F. (1976). Minitab Student Handbook. Duxbury Press, MA. Schut, J. (1976). In Food Emulsions, ed. S. Friberg. Marcel Dekker, NY, p. 385. Shimuzu, M., Saito, M. & Yamauchi, K. (1985). Agr&. Biol. Chem., 49, 189. Song, K. B. & Damodaran, S. (1987). J. Agric. Food Chem., 35, 236. Squire, J. (1981). The Structural Basis of Muscular Contraction. Plenum Press, NY. Swasdee, R. L., Terrell, R. N., Dutson, T. R. & Lewis, R. E. (1982). J. Food Sci., 47, 1011. Tornberg, E. (1977). J. Colloid Interface Sci., f~, 50. Tornberg, E. (1978). J. Colloid Interface Sci., 64, 391. Waniska, R. D. & Kinsella, J. E. (1985). J. Agric. Food Chem., 23, 1143.
The Surface Active Properties of Myosin and its Proteolytic Fragments E. O'Neill, a D. M. MulvihilP & P. A. Morrissey b a Department of Food Chemistry, b Department of Nutrition, UniversityCollege, Cork, Republic of Ireland (Received 1 November 1988; revised version receivedand accepted 6 March 1989) ABSTRACT The surface active properties of myosin and its proteolytic fragments, light meromyosin ( LMM), heavy meromyosin ( HMM), subfragment-I (S-l) and myosin rod, at initial bulk phase concentrations in the range of lO- 4% to 10-2% w/v were determined by the drop volume method. Overall, S-1 was the most effective surface tension depressor, whereas the tail portions of myosin, i.e. L M M and myosin rod were less surface active than the parent myosin molecule. The surface pressures attained after 40 min, at an initial bulk phase concentration of 10-2% (w/v), were 22"00, 21.77, 21"02, 16"77 and 16.77 mNm- 1for S-1, HMM, myosin, L M M and myosin rod, respectively. Furthermore, S-1 effected the most rapM change in surface pressure during the initial 5 rain period.
INTRODUCTION The interfacial protein film which surrounds lipid droplets or particles in comminuted sausage-type products is thought to play an important role in the stability of the systems (Swasdee et al., 1982, Jones & Mandigo, 1982). Galluzzo and Regenstein (1978) reported that myosin is rapidly taken up at the fat-aqueous interface during emulsification and recent studies in this laboratory have shown that myosin is more surface active than actin or actomyosin at the air-water interface (O'Neill et al., 1989a). Myosini is an unusual protein in that it contains structural characteristics of both globular and fibrous proteins in one covalently linked functional unit. The globular head region of the molecule hydrolyses ATP and binds strongly to F-actin, 131 Meat Science 030%1740/89/$03.50 © 1989ElsevierSciencePublishersLtd, England.Printed in Great Britain
132
E. O'Neill, D. M. Mulvihill, P. A. Morrissey
while the tail region is responsible for the formation of filaments at low ionic strength. Borejdo (1983) observed that the surface hydrophobicity of myosin is confined almost exclusively to the globular head region. However, the contribution of the different structural domains to the interfacial film forming properties of myosin is not known. In this study, using chymotrypsin digestion to cleave myosin at specific sites, we examined the surface active properties of the major proteolytic fragments of myosin in order to identify the main surface active portion of the molecule. The subfragments examined were as follows: subfragment-1 (S-l), which represents the individual myosin heads; heavy meromyosin (HMM), which contains both heads and a portion of the tail region of myosin; the myosin rod, which comprises almost the complete tail; and light meromyosin (LMM), which consists of a shortened fragment of the myosin tail. The possible structure of the interfacial film formed by myosin at an oil/water interface is also discussed.
MATERIALS AND METHODS All chemicals used were of reagent grade. Distilled water, glassware and dialysis tubing were prepared, as described by O'Neill et al. (1989a).
Determination of protein content Protein contents of the solutions were determined by the Biuret method.
Preparation of protein solutions Rabbit skeletal muscle was used in all studies. Myosin was prepared by the method outlined by Margossian and Lowey (1982), using ion exchange chromatography on DEAE-Sephadex to remove C-protein and traces of contaminating actin. S-I, HMM, LMM and myosin rod were prepared by chymotryptic digestion of myosin according to the procedure of Margossian and Lowey (1982), with HMM and S-1 being purified by ion-exchange chromatography on DEAE-cellulose followed by ammonium sulphate precipitation. All preparations of myosin and its fragments were checked for purity by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (Pollard, 1982), using the staining procedure of Greaser et al. (1983). All the protein solutions were exhaustively dialysed at 4°C against 0.6r~ KCI, pH 7.0. The protein solutions were centrifuged at 20000g for 30 min and the supernatants were filtered through Whatman No. 1 filter paper to remove
Surface active properties of myosin
133
any protein particles that could sediment during surface tension measurements. The protein contents were determined and appropriate concentrations (10-2-10-4% w/v) prepared by dilution with 0-6M KC1, pH 7"0. Surface tension measurements were made immediately after protein preparation. Surface tension measurements
The surface activities of the protein solutions at the air-solvent interface at 25°C were determined by the drop volume principle, using the apparatus described by Tornberg (1977). Calibration of the drop volume apparatus and calculation of surface tension were carried out, as described by Tornberg (1978) and Arnebrant and Nylander (1985). The protein solutions were prepared in triplicate. The surface activity of each protein preparation was determined in quadruplicate for each drop size, and the mean surface tensions were plotted as a function of time. A surface tension decay curve was then plotted for each protein from the mean of the three curves. The standard deviations of these curves were determined using the Minitab Statistical Package (Ryan et aL, 1976) on a DEC Vax 11/780 computer and were within the range of ___0.27mNm-
RESULTS AND DISCUSSION The time dependence of the surface tension, ~, of solutions of myosin and of its proteolytic fragments, light meromyosin (LMM), heavy meromysin (HMM), subfragment-1 (S-l) and myosin rod, in 0"6M KC1 at pH 7.0, at different bulk phase concentrations are shown in Figs 1-5. The initial surface tension of the solvent was found to be 73.27 mNm-1. An initial induction period, where no discernible change in surface tension occurred, was observed for all the proteins at a bulk phase concentration of 10-3% (w/v). DeFeijter and Benjamins (1987) suggested that the induction period is the time taken for the adsorbing protein molecules to attain a sufficient surface coverage to decrease 7. Thus, the difference observed in induction times may reflect the relative rates of protein accumulation at the air-solvent interface. Since dynamic surface properties are considered to be important in the stabilisation oflipid droplets during emulsion formation (Hailing, 1981), the rate of change of surface pressure exerted by a protein, particularly during the first few minutes, is important. The data show that the rate of surface tension decay was dependent on the identity of the adsorbing protein molecules and decreased as the bulk phase concentration of protein was reduced (Figs 1-5). In order to compare the surface behaviours of the
E. O'Neill, D. M. Mulvihill, P. A. Morrissey
134
75 li-
7O
55 A-O--
50
I
I
I
I
10
20
30
40
TIME
(rain)
Fig. 1. Time dependence of surface tension at the air-solvent interface, at 25°C, of myosin at initial bulk phase concentrations of 1 x 10 -4 (m); i x 10 -3 ( o ) ; 5 × 10 -3 (~.) and 1 x 10 -2 (I--q) % (w/v) protein.
various proteins the rates of change of surface pressure (H), AH/At, during the first 30s and 5min, at 1 × 10 - 2 , 5 × l 0 - 3 and 1 × 10-3% (w/v), were calculated from the data shown in Figs 1-5, and are reported in Table 1. Since the rate of change of H is considered to be a function of the physicochemical properties of proteins (Graham & Phillips, 1979a-c; Waniska & Kinsella, 1985; Shimuzu et al., 1985; Song & Damodaran, 1987), the different behaviours of the proteins evident from the data probably reflects the
75 m--
7O
o--
6O
&--
545
I
I
I
I
10
20
30
40
TIME
(min)
Fig. 2. Time dependenceof surface tension at the air-solvent interface, at 25°C, of light meromyosin at initial bulk phase concentrations of I × 10-4 (m); 1 x 10- 3 (O): 5 x 10- 3 (A) and 1 × 10 2 C]J % (w/v) protein.
Surface active properties of myosin
135
75 aDD
7O
0--
55 A-D B
UJ
U)
I
I
I
I
10
20
30
40
TIME
(rain)
Fig. 3. Time dependence of surface tension at the air-solvent interface, at 25°C, of heavy meromyosin at initial bulk phase concentrations of 1 x 10 -4 (ml); l × 10- 3 (O); 5 × 10- 3 (A) and l × l0 -2 (1-]) % (w/v) protein.
chemical and structural properties of the individual proteins. At 1 0 - 2 % (w/v) protein the rate of change of H was fastest for S-1 followed by HMM, myosin, LMM and myosin rod. We consider that AH/At during the initial 30s reflects the relative rates at which the protein molecules migrate and adsorb at the interface and is influenced by such factors as molecular size and surface hydrophobicity (MacRitchie, 1978; Kato & Nakai, 1980). Thus, it is not surprising that the S-1 fragment (mw ~ 115 000 daltons) accumulated
75 7O ] "
!0-A-DI
I 50 I
10
I
I
20
30
TIME
I
40
(min)
Fig. 4. Time dependence of surface tension at the air-solvent interface, at 25°C, of subfragment-l at initial bulk phase concentrations of 1 x l0 -4 (Hal); 1 x i0 -3 (O); 5 x 10 -3 (&) and l x 10 -2 ([]) % (w/v) protein.
E. O'Neill, D. M. Mulvihill, P. A. Morrissey
[36
75 70
O--
D--
ss
10
20 TIME
30
40
(rain)
Fig. 5. Time dependence of surface tension at the air-solvent interface, at 25°C, of myosin rod at initial bulk phase concentrations of 1 x 10 -4 (R); 1 x 10 -3 (O); 5 x 10 -3 (,L) and 1 x 10 -z (1:]) % (w/v) protein.
more readily at the interface than the larger myosin rod (mw ~ 220 000 daltons) and caused a more rapid change in I1 during the early stages. The faster All/At exerted by H M M and myosin compared with LMM and myosin rod, in spite of their large molecular weights (340 000 and 500 000, respectively) is probably due to the greater surface hydrophobicity of these fragments (Borejdo, 1983). The overall rate of change of surface pressure, All/At, during the initial 5 min followed the same order as that for the first 30s, i.e. S-1 exerted the fastest change in II, followed by HMM, myosin, LMM and myosin rod. Between 20 and 40min changes in II were very small for all the proteins (Figs 1-5) and probably reflect further minor rearrangements of the adsorbed molecules rather than adsorption of new protein molecules from the bulk phase. In addition to the rate of change of surface pressure exerted by proteins, the equilibrium surface pressure attained is also an important index of surface activity. The concentration dependence of the surface pressure attained after 40 min, 1-I4o,i.e. the reduction in surface tension after 40 rain, is clearly illustrated in Fig. 6 where 174o is plotted as a function of initial bulk phase concentration. Overall, S-1 was the most surface active protein, whereas the tail portions of myosin, i.e. LMM and myosin rod, were less surface active than the parent myosin molecule. H M M was almost as surface active as S- 1 at concentrations of 5 × 10- 3 and 1 × 10- 2% (w/v) and Flao was more or less independent of bulk phase concentration in this range for these two myosin fragments. However, when the bulk phase concentration was reduced to 10-3% (W/V), differences in surface activity between these fragments were evident, with S-1 being much more efficient at reducing
Surface active properties of myosin
TABLE
137
1
P a r a m e t e r s D e s c r i b i n g the R a t e s o f S u r f a c e T e n s i o n D e c a y at the A i r - S o l v e n t I n t e r f a c e for M y o s i n , S u b f r a g m e n t - 1 , H e a v y M e r o m y o s i n , L i g h t M e r o m y o s i n and Myosin Rod
Protein conc. (% w/v) Myosin I x 10 - z 5 × 10 -3 1 × 10 -3
Induction period (min)
All~At 0"5 rain (mNm- l rnin- l)
A FI/ At 5"0 rain (mNm- 1rain- 1)
---
18.04 10.54
3'25 3"00
4-67
---
0"35
Subfragment- 1 1 × 10 -2
22.00
3.90
-1"67
15.54 ---
3.54 2"01
1 x 10 -2 5 × 10 -3
--
18.54 13.00
3-40 3-10
1 × 10 -3
3'33
5 × 10 -3
1
×
10 - 3
Heavy meromyosin
Light m e r o m y o s i n 1 x 10 -2
1"09
--
14'94
2"85
5 × 10 -3
0.17
6'54
2-17
1 × 10 3
5"50
---
--
--
13"54
2"70
0.27 6.00
5"54 --
2-30 --
Myosin rod 1 x 10 -2 5 × 10 -3 1 x 10 -3
surface tension than HMM. At very low protein concentration, 10- 4% (w/v) only S-1 reduced surface tension substantially. In an earlier study we found that myosin was more surface active than actin or actomyosin (O'Neill et al., 1989a). We suggested that the superior surface active properties of myosin may be attributed to its greater surface hydrophobicity (Borejdo, 1983) and that the elongated nature of the myosin molecule (Lowey et al., 1969; Squire, 1981) may allow a larger surface available for interaction with interface, even when myosin is in its native configuration. However, the present results suggest that the globular heads of myosin play a critical role in the surface activity of myosin. Borejdo (1983) concluded that myosin is a unique protein in that its surface hydrophobicity is confined almost exclusively to the globular head region of the molecule and that the myosin tail is practically devoid of surface hydrophobic sites. Surface hydrophobicity has been reported to correlate significantly with
138
E. O'Neill, D. M. Mulvihill, P. A. Morrissey 25
20
o/ / IS °
10
I
10-4
I
10 "a
t
10-2
BULK PHASE CONCENTRATION (~WlV)
Fig. 6. The surface pressure attained after 40 min, n4o, as a function of initial bulk phase protein concentration for myosin (11); light meromyosin (A);heavy meromyosin (0); subfragment-1 (IS]) and myosin rod ((3).
surface activity of proteins (Kato & Nakai, 1980; Kato et al., 1981); thus, it is not surprising that the present findings show that the fragments which include the myosin head region, i.e. HMM and S-l, were more surface active than those which contain only the tail portion of the molecule. Differences in the interfacial properties of proteins are, in part, attributed to compositional differences (Waniska & Kinsella, 1985; Song & Damodaran, 1987). Thus, it may be possible to explain the observed differences in surface properties between the myosin subfragments on the basis of amino acid composition and distribution. The myosin rod sequence is highly repetitive and has the characteristics of an s-helical coiled coil (McLachlan & Karn, 1982). According to these workers, the rod amino acid sequence has hydrophobic residues strongly concentrated in the centre core of the helix. The helix surface has a high density of charged groups and is essentially devoid of hydrophobic groups. Furthermore, the data of Borejdo (1983) also imply that the myosin rod is essentially devoid of surface hydrophobic groups. Thus, it is likely that the hydrophilic nature of the helix surface militates against the rod portion of the myosin molecule penetrating into the interfacial film, since it would involve the thermodynamically
Surface active properties of myosin
139
unfavourable process of dehydration. Dickinson et al. (1987) suggested that the hinge region may play a crucial role in the interfacial behaviour of myosin by unfolding at low temperatures and exposing the hydrophobic surfaces of the ~-helix coiled coil. However, we did not observe any significant differences in surface activity of LMM, which does not include the hinge region and the myosin rod, which does. The present results are consistent with earlier findings in this laboratory, which showed that thermal denaturation of myosin which causes unfolding of the ~-helical portion of the molecule, exposing the previously hidden hydrophobic core, enhances the surface activity of myosin (O'Neill et al., 1989b). The head region of myosin includes amino acids which are strongly hydrophobic (Elzinga & Collins, 1977). While some of these hydrophobic groups are buried in the interior of the native molecule, many remain exposed and contribute t o surface hydrophobicity. Borejdo (1983) concluded that the hydrophobic pocket present on the surface of myosin is formed between the 41-residue region at the N-terminal end of the A-1 light chain and the heavy chain of S-1. Thus, the initial attraction of the myosin head to the interface and the subsequent rapid reduction in surface tension for S-1 (Table 1) is probably due to its native surface hydrophobicity. Surface denaturation and conformational rearrangement of the adsorbed protein molecules may also contribute to surface tension decay. Since the head region is more surface active than the tail region, as indicated by the present results, and hydrophobic groups are located almost exclusively in the head region (Borejdo, 1983), it is difficult to envisage how the interfacial membrane in meat emulsions could be orientated in the manner proposed by Schut (1976), i.e. with LMM orientated towards the fat phase and the head region attracted towards the aqueous phase. Jones (1984) and the present workers (Morrissey et al., 1987) proposed that the hydrophobic heads of native myosin molecules are orientated towards the surface of the fat globule. On the basis of the present findings we suggest that the proposed model is basically correct. Thus, the myosin heads are strongly adsorbed at the interface between oil and water. It is also likely that the myosin heads undergo a certain degree of rearrangement and surface denaturation following adsorption. The myosin tails projecting from the fat globule surface probably penetrate into the surrounding matrix and undergo protein-protein interaction to form a viscoelastic interfacial film.
REFERENCES Arnebrant, T. & Nylander, T. (1985). J. Dispersion Sci. Technol., 6, 209. Borejdo, J. (1983). Biochern., 22, 1182.
140
E. O'Neill, D. M. Mulvihill, P. A. Morrissey
DeFeijter, J. A. & Benjamins, J. (1987). In Food Emulsions and Foams, ed. E. Dickinson, Royal Society of Chemistry, London, p. 72. Dickinson, E., Murray, B. S., Stainsby, G. & Brock, C. J. (1987). Int. J. Biol. Macromol., 9, 302. Elzinga, M. & Collins, J. H. (1977). Proc. Natl. Acad. Sci. USA, 74, 4281. Galluzzo, S. J. & Regenstein, J. M. (1978). J. Food Sci., 43, 1761. Graham, D. E. & Phillips, M. C. (1979a). J. Colloid Interface Sci., 70, 403. Graham, D. E. & Phillips, M. C. (1979b). J. Colloid Interface Sci., 70, 415. Graham, D. E. & Phillips, M. C. (1979c). J. Colloid Interface Sci., 70, 427. Greaser, M. L., Yates, L. D., Krzywicki, K. & Roelke, D. L. (1983). In Proceedings of the 36th Annual Reciprocal Meat Conference. National Livestock and Meat Board, Chicago, IL, p. 87. Hailing, P. J. (1981). CRC Crit. Rev. Food Sci. Nutr., 15, 155. Jones, K. W. (1984). In Proceedings of the 37th Annual Reciprocal Meat Conference. National Livestock and Meat Board, Chicago, IL, p. 52. Jones, K. W. & Mandigo, R. W. (1982). J. Food Sci., 47, 1930. Kato, A. & Nakai, S. (1980). Biochim. Biophys. Acta, 624, 13. Kato, A., Tsutsui, N., Matsudomi, N., Kobayashi, K. & Nakai, S. (1981). Agric. Biol. Chem., 45, 2755. Lowey, S., Slayter, H. S., Weeds, A. G. & Baker, H. (1969). J. MoL Biol., 42, 1. Margossian, S. S. & Lowey, S. (1982). Method Enzymol., 85, 55. MacRitchie, F. (1978). Adv. Protein Chem., 32, 283. McLachlan, A. D. & Karn, J. (1982). Nature, 299, 226. Morrissey, P. A., Mulvihull, D. M. & O'Neill, E. M. (1987). In Developments in Food Proteins--& ed. B. J. F. Hudson, Elsevier Applied Science Publishers, London, p. 195. O'Neill, E., Morrissey, P. A. & Mulvihill, D. M. (1989a). Food Chem., 35, I. O'Neill, E., Morrissey, P. A. & Mulvihill, D. M. (1989b). Food Chem., 34, 295. Pollard, T. D. (1982). Method Enzymol., 85, 123. Ryan, T. A., Joiner, B. C. & Ryan, B. F. (1976). Minitab Student Handbook. Duxbury Press, MA. Schut, J. (1976). In Food Emulsions, ed. S. Friberg. Marcel Dekker, NY, p. 385. Shimuzu, M., Saito, M. & Yamauchi, K. (1985). Agr&. Biol. Chem., 49, 189. Song, K. B. & Damodaran, S. (1987). J. Agric. Food Chem., 35, 236. Squire, J. (1981). The Structural Basis of Muscular Contraction. Plenum Press, NY. Swasdee, R. L., Terrell, R. N., Dutson, T. R. & Lewis, R. E. (1982). J. Food Sci., 47, 1011. Tornberg, E. (1977). J. Colloid Interface Sci., f~, 50. Tornberg, E. (1978). J. Colloid Interface Sci., 64, 391. Waniska, R. D. & Kinsella, J. E. (1985). J. Agric. Food Chem., 23, 1143.