- Email: [email protected]
Effect of trypsin on rabbit skeletal muscle α-actinin
Biochimica et Biophysica Acta, 446 (1976) 445-456
© Elsevier/North-Holland Biomedical Press BBA 37470 E F F E C T OF TRYPSIN ON RABBIT S K E L E T A L M U S C L E a-ACTININ*
GERALD R. HOLMES, DARREL E. GOLL**, A. SUZUKI***, R. M. ROBSON and M. H. STROMER Muscle Biology Group, Departments of Animal Science, Biochemistry and Biophysics, and Food Technology, Cooperating, Iowa State University, Ames, Iowa 50011 (U.S.A.)
(Received April 20th, 1976)
SUMMARY 5 min of tryptic digestion of purified rabbit skeletal a-actinin decreases by approximately 75 % the ability of a-actinin to cross-rink F-actin filaments as measured viscometrically at 27 °C, but has little effect on the sedimentation coefficient of a-actinin at 20 °C or on a-actinin's ability to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions. Twenty to sixty min of trypsin treatment reduces the sedimentation coefficient of a-actinin and destroys much of tt-actinin's ability to increase the Mg2+-modified ATPase and rate of turbidity increase of reconstituted actomyosin suspensions. Therefore, the ability of a-actinin to increase the rate of in vitro measures of muscle contraction may not result directly from a-actinin's ability to cross-link F-actin filaments. Trypsin does not split a-actinin into large fragments as it does myosin. Previous studies have shown that 35 to 65 ~o of total tryptic-susceptible peptide bonds i'n a-aetinin are split after 60 rain of incubation with t~ypsin and that 30 % of these bonds sprit in 60 min are cleaved during the first 5 rain in a rapid reaction. That splitting of this group of peptide bonds has little effect on the sedimentation coefficient of a-actinin indicates that these bonds are located in a region of the a-actinin molecule where noncovalent forces are strong enough to maintain conformation of the native a-actinin molecule even after these bonds have been split. This ostensible segregation of a-actinin's ability to cross-link F-actin filaments from its ability to increase rate of in vitro assays of contraction by tryptic digestion may suggest that a-actinin could have at least two different physiological roles: (1) to bind actin filaments to each other or to basal structures, and (2) to enhance the effectiveness of actin in supporting movement.
" Journal Paper No. J-8450 of the Iowa Agriculture and Home Economics Experiment Station, Projects 1795, 1796, and 2025. ** Address inquiries to Darrel E. Goll, Department of Nutrition and Food Science, University of Arizona, Tucson, Ariz. 85721, U.S.A. *'" Present address: Department of Animal Science, Faculty of Agriculture, the University of Niigata, Niigata, Japan.
446 INTRODUCTION a-Actinin was discovered as a new protein component of skeletal muscle myofibrils because of the ability of low ionic strength extracts of crude myofibrillar residues to accelerate the rate of turbidity development in reconstituted actomyosin suspensions [1-3]. This discovery was soon followed by the finding that a-actinin exerts its effects on actomyosin suspensions by combining with the actin part of the actin-myosin complex [2-4]. The early studies on a-actinin/F-actin mixtures [4, 5] indicated that a-actinin cross-linked F-actin filaments under certain in vitro conditions, and it was subsequently shown that a-actinin is locafized in Z-disks of skeletal myofibrils [6-11], which is the only place where cross-linking of actin filaments occurs in skeletal muscle cells. The available evidence, based on yields of purified a-actinin obtainable from skeletal muscle, indicates that a-actinin is a relatively minor component of skeletal myofibrils and makes up less than 3 ~ of total myofibrillar protein by weight [12-14]. That a-actinin has been found in every motile system that has been examined for its presence [9-11, 13, 15, 16], including systems such as smooth muscle [9, 15], brush borders of intestinal epithelial cells [16], acrosomal processes of Limulus sperm [16], and fibroblasts [16] that have no Z-disks, suggests, despite its relatively small content in skeletal myofibrils, that a-actinin may have a fundamental role in motility. Although it is now possible to obtain highly purified a-actinin in sizable quantities [14, 17-19], the nature of a-actinin's role in motility remains unclear. The ability of a-actinin to increase the Mg2+-modified ATPase activity [14, 17] and rate of turbidity development [2, 14, 17] of reconstituted actomyosin suspensions may indicate that a-actinin can modify the structure of actin monomers while these monomers are in the aggregated filamentous state and thereby enhance the effectiveness of actin in supporting contraction [20]. It is not immediately obvious, however, how a-actinin located in the Z-disk of skeletal myofibrils can affect the actin-myosin interaction that occurs in situ at the overlap of thick and thin filaments approximately 0.5 #m distant. On the other hand, the presence of a-actinin in Z-disks of skeletal myofibrils and its ability to cross-link actin filaments in vitro [21] may indicate that the role of a-actinin is to bind actin filaments to each other in skeletal myofibrils or to certain basal structures in some nonmuscle motile systems. Controlled digestion by proteolytic enzymes such as trypsin has been very useful in studying structure of fibrous proteins such as collagen [22-24], paramyosin [25], and myosin [26-29]. Indeed, carefully controlled proteolytic digestion of myosin used in conjunction with electron microscope examination of the proteolytic fragments [30] has localized many of the biological properties of myosin to selected parts of the myosin molecule. We have, therefore, examined the effects of trypsin on some physical and biological properties of purified skeletal muscle a-actinin to determine whether proteolytic digestion might enable us to segregate different biological properties of a-actinin to different parts of the molecule. The results indicate that trypsin does not break skeletal muscle a-actinin into several large fragments as it does myosin but that tryptic digestion seems to decrease the ability of a-actinin to cross-link actin filaments before affecting its ability to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions.
447 MATERIALS AND METHODS Back and leg muscles of rabbits were used for all experiments reported here. Rabbits were anesthesized with sodium pentobarbital and o-tubocurarine chloride [17], and muscleswere handled as described by Arakawa et al. [17]. Unless specified otherwise, all preparations were done at 0 to 4 °C with precooled solutions and with double-deionized, distilled water that had been redistiUed in glass and stored in polyethylene containers.
Protein preparations A crude P0-30 a-actinin extract was made according to Goll et al. [18] and was purified by successive chromatography on two DEAE-ceUulose columns [14]. Actin was made according to Arakawa et al. [17], and myosin was made as described by Seraydarian et al. [3]. All protein preparations were routinely monitored for purity by using polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate [31]. As discussed previously [19], our myosin preparations were not chromatographically purified and therefore contained the B- and C-proteins described by Offer and coworkers [32, 33]. Our actin preparations contained 1 to 2 ~o of their total protein as a-actinin; this small residue of a-actinin is detectable only at very heavy loads [35-45/tg] on sodium dodecyl sulfate-polyacrylamide gels and is also impossible to remove without using column chromatography [19]. The presence of these small amounts of contaminants in our actin and myosin preparations had no evident effects on sensitivity of our assays for a-actinin activity. Protein concentrations were determined by the biuret method [34] as modified by Robson et al. [35]. Trypsin digestion Salt-free, 2 times crystallized trypsin was purchased from Sigma Chemical Co. and possessed 16 000 to 20 000 p-toluenesulfonyl-L-arginine methyl ester units of activity per mg [36]. Trypsin was dissolved at 2.0 mg/ml in 0.001 M HC1 and then diluted to 1.0 mg/ml in 160 mM Tris/acetate buffer (pH 7.2) just before use. Stock trypsin solutions were stable for 3 to 4 months at 2 °C. Soybean trypsin inhibitor (purest salt-free form from Sigma Chemical Co.) was dissolved in water at 4 to 10 mg/ml. Trypsin treatment of a-actinin was done in 112 mM KCI, 56 mM Tris/ acetate, pH 8.0, at 25 °C. a-Actinin concentration was 5.97 mg/ml, and the trypsin to a-actinin ratio was 1:50 by weight. Trypsin digestion was stopped by adding 4 parts of soybean trypsin inhibitor to part of trypsin by weight. Controls were made by incubating a-actinin for 60 rain at 25 °C under ionic conditions identical to those used for trypsin treatment and then adding trypsin and soybean trypsin inhibitor that had been premixed in a ratio of 1:4 by weight. Details on the ionic conditions and protein concentrations used during the different assays of trypsin-treated a-actinin will be given with the individual experiments. ATPase and turbidity assays Reconstituted actomyosin was made by mixing 2 parts of myosin to 1 part actin, w/w, and then washing as described by Arakawa et al. [17]. Ability of a-actinin to increase the MgZ+-modified ATPase activity and rate of turbidity increases in reconstituted actomyosin suspensions was assayed according to our previously
448 described procedures [17]. Details on ionic conditions and experimental protocol will be given with the individual experiments. As we have discussed previously [7, 17], total final absorbance under the conditions we use for the turbidity assay of a-actinin is not related to a-actinin activity. Indeed, very active or high concentrations of a-actinin frequently produce a remarkable flocculation of actomyosin suspensions so these suspensions settle to the bottom of the curvette before a high final absorbance is obtained (for example, see Fig. 3 in the present paper). Tris/acetate buffers were prepared by dissolving an appropriate amount of Tris, free base, and then adding glacial acetic acid to obtain the desired final pH.
Viscosity measurements Viscosity measurements were done by using Ostwald viscometers with flow times of 65 to 70 sec for water at 20 °C. 5-ml samples were used in the viscometers. Specific viscosities were calculated by subtracting one from the ratio obtained by dividing flow times of protein solutions with flow times of the respective solvents. No kinetic energy corrections were made when calculating specific viscosities. Waterbath temperatures were controlled to within 4- 0.05 °C for all viscosity measurements. We have previously found [37] that G-actin containing very little a-actinin aggregates to F-actin very slowly at 0 °C, even in the presence of a-actinin; therefore, regardless of the temperature at which the experiment was done, G-actin was converted to F-actin by incubating at 27 °C for 15 min. We have previously shown [37] that the effect of a-actinin of F-actin viscosity can be altered reversibly by varying temperature between 0 and 37 °C so temperature of the conversion of G to F-actin had no irreversible effects on the results of our viscosity experiments. Two different experimental procedures were used to investigate the effect of tryptic digestion on the ability of a-actinin to increase F-actin viscosity. In the first procedure, purified a-actinin was treated with 1 part of trypsin to 50 parts a-actinin for different periods of time between 0 and 60 min at 25.0 °C. After a selected duration of trypsin treatment, digestion was stopped with soybean trypsin inhibitor, the treated a-actinin was mixed with G-actin, and the actin in this mixture was converted to F-actin by adding MgC12 to a final concentration of 2 mM at 27 °C. Viscosity of the a-actinin/F-actin mixture was then measured at 0 °C or 27 °C. Control (zero-time) samples were made by stirring purified a-actinin under the same ionic conditions as the treated samples at 25 °C for 60 min and then adding premixed trypsin-soybean trypsin inhibitor. The second procedure used to assay the effects of trypsin on a-actinin's ability to increase F-actin viscosity involved mixing G-actin and a-actinin in a viscometer at 27 °C, adding MgCI2 to a final concentration of 2 mM to convert the G-actin to F-actin, and then measuring viscosity of the resulting mixture. 18 min after conversion of G to F-actin, 1 part of trypsin in 0.1 ml to 50 parts of actin, by weight, was added directly to the viscometer, and viscosity was measured again as quickly after mixing as possible. The control in this procedure was done by adding 0.10 ml buffer solution containing no trypsin. Details on protein concentrations and ionic conditions will be given with the individual experiments.
Analytical ultracentrifugation Analytical ultracentrifuge studies were done with a Spinco Model E analytical
449 ultracentrifuge equipped with phase plate, rotor temperature indicator control unit, electronic speed control, and a high-temperature accessory. Kel-F centerpieces were used, and plates were measured with a Nikon 6C profile projector. Sedimentation velocity runs were done at 20.0 °C, but because all sedimentation runs were done in identical solvents at identical concentrations and were for comparative purposes only, no attempt was made to correct the observed sedimentation coefficients to water or zero protein concentration. RESULTS
Effect of trypsin on sedimentation diagram of a-actinin We have previously reported [12, 18] that skeletal muscle a-actinin is moderately susceptible to degradation by trypsin with 35 to 65% of tryptic-labile bonds (as calculated from the sum of the lysine and arginine contents of purified a-actinin) hydrolyzed after 60 rain of trypsin treatment at p H 8.5, 25.0 °C, and trypsin to purified a-actinin ratios o f 1:50, by weight. Fig. 1 shows that 60 rain of trypsin treatment of purified rabbit skeletal muscle a-actinin under these same conditions diminishes both the size and the sedimentation coefficient of the 6 S a-actinin peak in the analytical ultracentrifuge. It is clear from Fig. 1, however, that trypsin does not cleave the a-actinin molecule into two large fragments as it does myosin [26-30]. Instead, trypsin seems to "nibble" at the a-actinin molecule, releasing small fragments
Fig. 1. Sedimentation patterns of purified rabbit skeletal a-actinin after different times of tryptic digestion. Tryptic digestion was done as described in Materials and Methods. Ionic conditions during the analytical ultracentrifuge run and the observed sedimentation coefficient of the main boundary in each of the runs a r e g i v e n in the figure. Temperature of analytical ultracentrifuge run = 20.0 °C, and phase plate angle = 70 °.
450 that do not sediment as a distinct boundary and leaving a large fragment whose observed sedimentation coefficient gradually decreases from 6.03 S to 5.03 S after 60 min of digestion (Fig. 1). It therefore seems that use of trypsin to study substructure of the a-actinin molecule will require a careful and detailed analysis of the fragments released after different times of treatment and will not be as simple as the analysis initially afforded by the observation that trypsin splits myosin into two large flagments [38]. Such a detailed analysis of the products produced by tryptic digestion of purified a-actinin is currently in progress and will be reported in a subsequent paper (Arakawa, Robson, Huiatt, and Goll, manuscript in preparation).
Effect of trypsin on ability of a-actinin to increase in vitro contractile responses Although the effects of trypsin on the sedimentation pattern of skeletal muscle a-actinin do not immediately produce new insights into substructure of the a-actinin molecule, it was also of interest to learn whether trypsin would segregate some of the biological properties of the a-actinin molecule. Increasing time of trypsin treatment between 0 and 60 min gradually reduces ability of a-actinin to increase the Mg z+-
0,35
o. o
0,25 z ~- 0.20
~0,15 > 0.I0
0.05
ooo
~
CONTROL. ACTOMYOSIN ONLY
~o
/5
;o
~s
~o
~5
4'o
~5
;o
~5
6'o
TIME OF TRYPSINTREATMENT(MINUTES)
Fig. 2. Effect of trypsin treatment of rabbit skeletal a-actinin on its ability to increase the Mg2+modified ATPase activity of reconstituted actomyosin. Tryptic digestion of a-actinin was done as described in Materials and Methods. Conditions of ATPase assay were: 100 mM KCI, 20 mM Tris/acetate, pH 7.0, 1 mM MgC12,0.05 mM CaCI~, 1 mM ATP, 0.2 mg reconstituted rabbit skeletal actomyosin/ml, 0.04 nag a-actinin/ml when added, 25.0 °C. Activity of actomyosin in the absence of a-actinin is shown on the ordinate. modified ATPase activity (Fig. 2) or the rate of turbidity increase (Fig. 3) of reconstituted actomyosin suspensions. Very little of a-actinin's ability to increase these t w o in vitro measures of contractile activity remained after 60 rain of tryptic digestion at trypsin-to-a-actinin ratios of 1:50, by weight (Figs. 2 and 3). Ability of a-actinin to increase the MgZ+-modified ATPase activity and rate of turbidity response of reconstituted actomyosin suspensions, however, decreased only slightly during the first 5 min of trypsin digestion (Figs. 2 and 3), even though our earlier p H stat
451
04
/f 20mi=~ £0rain CONTROL
i0.3 Omin
0,I o,c
~
ib
~'5
2'o
2'5
3'o
3'5
4
4
55
TIME (MINUTES)
Fig. 3. Effectof trypsin treatment of rabbit skeletal a-aetinin on its ability to acceleraterate of turbidity increase in reconstituted actomyosin suspensions.Tryptic digestion of a-~tinin was done as de~ribed in Materials and Methods. Conditions of the turbidity assay were: 100 mM KCI, 20 mM Tris/aeetate, pH 7.0,1 mM MgCI2,0.05 mM CaCI2,1 mM ATP, 0.4 mg reconstitutedrabbit skeletalactomyosin/ml, 0.02 mg a-actinin/ml when added, 25.0 °C. studies [12, 18] showed that approximately 30 ~o of all peptide bonds split by trypsin during the first 60 min of incubation with a-actinin are split during the first 5 min in a rapid reaction that cleaves 1.4.10 -7 mol of bonds/min. The analytical ultracentrifugal and ATPase and turbidity results presented here suggest that cleavage of this group of peptide bonds has little effect on either the sedimentation coefficient of the ability of a-actinin to increase the Mg2+-modified ATPase activity or rate of turbidity increase of reconstituted actomyosin suspensions. Indeed, an average of two different experiments indicated that only 7~o of a-actinin's ability to increase the Mg 2+modified ATPase activity of reconstituted actomyosin suspensions was lost during the first 5 min of trypsin treatment under the conditions used in this study. Most of the trypsin-induced loss in ability of tt-actinin to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions occurs between 5 and 20 min of trypsin digestion (Figs. 2 and 3) when only approximately 24 ~o of total bonds split during 60 min of tryptic digestion are cleaved. An average of two different experiments showed that approximately 30 ~o of a-actinin's ability to increase the Mg2+-modified ATPase activity of reconstituted actomyosin suspensions was lost between 5 and 20 rain of trypsin treatment. A large decrease in sedimentation coefficient from 6.0 S to 5.5 S also occurs between 5 and 20 rain of tryptic digestion of a-actinin (Fig. 1). Hence, peptide bonds split by trypsin between 5 and 20 rain of digestion have a larger effect on a-actinin's structure as measured by sedimentation coefficient and ability to +increase in vitro measures of contraction than do those cleaved between 0 and 5 min of digestion
452
Effect of trypsin on ability of a-actinin to increase F-actin viscosity The ability of a-actinin to increase viscosity of F-actin is ostensibly due to its capacity to cross-link F-actin filaments. As described in Materials and Methods, the effect of tryptic digestion of a-actinin on its ability to cross-link actin filaments as measured viscometrically was assayed in two different ways. The results of the first method, in which a-actinin was treated with trypsin before being mixed with actin, are shown in Fig. 4. Control a-actinin samples, made simply by stirring at 25 °C for 60 rain before adding premixed trypsin and soybean trypsin inhibitor, had already lost some of their ability to increase the viscosity of F-actin at 27 °C (Fig. 4); this loss was especially noticeable at 0 °C when 0.02 part of a-actinin was added to 1 part actin by weight (Fig. 4). In contrast to the effects of trypsin treatment on ability of a-actinin to increase the Mg z+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions, the first 5 min tryptic digestion of a-actinin destroyed approximately 75 % of the ability of c~-actinin to increase F-actin viscosity at 27 °C (Fig. 4); longer tryptic treatment up to 60 min had little additional effect on c~-actinin's ability to increase F-actin viscosity at 27 °C. In contrast to its destruction of a-actinin's ability to increase F-actin viscosity at 27 °C, tryptic digestion up to 60 min had only a slight effect on a-actinin's ability to increase F-actin viscosity at 0 °C (Fig. 4). We [18, 37, 39] have previously shown that the a-actinin/F-actin interaction at 0 °C is completely different from the ~-actinin/F-actin interaction at higher temperatures, but the cause of this difference is still unknown. 3,0~
2. ~. W~¢-"-"
2% Untreated
a-Actinin,O °
2.0 10% Untreated Q-Actinin,O*
nsp
/ '~'~
I,C
•~
O,5X
2% Trypsin-treated
~-Actinin,
10% Trypsin-treated
a-Actinin,
s-Actinin,0 ~
o °
1o% Untreated ~ - A c t i n i n , 27°
} Control
0,1 . . . . . 0.
10% Trypsin-treated
27 °
- No Q-Actinin
5
I~
'
' ' 3'0 . . 20 TIME OF TRYPSINTREATMENT(MINUTES)
. . NO
.
.
50
60
Fig. 4. Effect of trypsin treatment of rabbit skeletal a-actinin on its ability to increase viscosity of actin polymerized by Mg 2+ at 27.0 °C. Tryptic digestion of (t-actinin was done as described in Materials and Methods. Rabbit skeletal G-actin was converted to F-actin at 27 °C in the absence or presence of trypsin-treated or untreated a-actinin by addition of Mg 2+, and viscosity of the resulting solution was measured. Final conditions of viscosity assay were: 1.0 rag actin/ml, a-actinin indicated as percentage of actin present by weight, 2 m M MgC12, 20 m M Tris/acetate, pH 7.5. All samples were polymerized at 27 °C but some were measured at 27 °C and some were measured at 0 °C. Untreated a-actinin samples were not subjected to any experimental handling before being mixed with actin in the viscometer; zero-time samples were treated as described in Materials and Methods.
453 1,50
1.25 Added Trypsin or B ff r
1
1.00
In
.
0 ° Bath
10~ a-Actinin
Plus
Buffer
I
nsp io~ ~-Actxnin Plus Trypsin
0,75 0% a-Actinin Plus Buffer
0,50 0% n-Actxnin Plus Trypsin
0.25
f}.0
I
I
I
I
I
rIME AFTERPOLYMERIZATION(MINUTES)
Fig. 5. Effect of trypsin on viscosity of the a-actinin-F-actin complex. Rabbit skeletal G-actin, in the presence or absenceof rabbit skeletal a-actinin in an Ostwald viscometer, was converted to F-actin
by addition of MgCI2 at 27 °C, and viscosity of the resulting solution was measured at the same temperature. Eighteen rain after the first viscosity measurement, 1 part of trypsin in 0.10 ml was added to 50 parts of actin by weight in the viscometer, and viscosity of the resulting mixture measured. 12 min later, the viscometer was placed in a 0 °C bath and viscosity measured again. Control samples had 0.10 ml buffer added in place of the trypsin. Final concentrations: 1.0 mg actin/ml, a-actinin indicated as percentage of actin present by weight, 2 mM MgCI2, 20 mM Tris/acetate, pH 7.5. The second method used to assay the effects of trypsin on a-actinin's ability to increase F-actin viscosity involved measuring the effects of trypsin on viscosity of a preformed a-actinin/F-actin complex (Fig. 5). The results show that, 5 to 10 min after addition of trypsin to the viscometer, viscosity of the a-actinin/F-actin mixture had decreased nearly to the viscosity of F-actin alone (Fig. 5). Addition of buffer instead of trypsin caused a small decrease in specific viscosity of a-actinin/F-actin mixtures at 27 °C (Fig. 5), but this effect was not nearly as large as the effect caused by addition of trypsin. Consequently, the results of this second method for assaying the effects of trypsin on a-actinin's ability to increase F-actin viscosity agree with the results of the first procedure used to assay this effect and show that trypsin almost completely eliminates the ability of a-actinin to increase F-actin viscosity at 27 °C within 5 to 10 min. Putting the viscometers containing the treated and control a-actinin/F-actin mixtures in a 0 °C bath 12 min after addition of trypsin or buffer resulted in an immediate increase in specific viscosity of both trypsin-treated and control a-actinin/F-actin mixtures, although viscosity of the trypsin-treated mixture did not increase to quite the same extent as viscosity of the control mixture (Fig. 5). Specific viscosity of trypsin-treated and control solutions of F-actin without a-aetinin was not affected by immersion in a 0 °C bath. These results, therefore, substantiate the findings obtained by the first procedure used to assay the effects of trypsin on a-actinin's ability to increase F-actin viscosity and show that trypsin treatment has less effect on a-actinin's ability to increase F-actin viscosity at 0 °C than it has on a-actinin's ability to increase F-actin viscosity at 27 °C.
454 DISCUSSION The results of this study show that 5 min of trypsin treatment decreases aactinin's ability to cross-link F-actin filaments as measured viscometrically at 27 °C by approximately 75 ~ , but has little effect on sedimentation pattern of a-actinin at 20 °C or on a-actinin's ability to increase the Mg2+-modified ATPase activity or rate of turbidity increase of reconstituted actomyosin suspensions at 25 °C. Tryptic treatment for 20 to 60 min eventually reduces the sedimentation coefficient of a-actinin and also decreases its ability to increase the rate of in vitro measures of contraction. These results suggest that tryptic digestion reduces a-actinin's ability to cross-link F-actin filaments, as measured by a-actinin's capacity to increase F-actin viscosity, more rapidly than it reduces a-actinin's ability to increase the Mg 2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions. Although it may be suggested that the rapid tryptic-induced loss of a-actinin's ability to cross-link F-actin filaments results in part from the lower a-actinin-to-actin ratios used in the viscosity experiments (viz., 0. I0 parts a-actinin to 1.0 part actin in viscosity experiments compared with 0.15 parts a-actinin to 1.0 part actin in the turbidity experiments), so that tryptic digestion rapidly reduces native a-actinin concentration below some threshold level required for activity, this possibility seems unlikely for several reasons. First, it seems very improbable that 5 min of tryptic treatment when the a-actinin to actin ratio is 0.10 to 1.0 would lower the proportion of undigested a-actinin to a level that gives no detectable response, but that 20 min of tryptic treatment when the a-actinin to actin ratio is 0.15 to 1.0 would not also lower the proportion of undigested a-actinin to a level that gives no detectable response. Yet, our results show that, even after 20 min of trypsin treatment, a-actinin retains considerable activity in the turbidity assay (Fig. 3), though 5 min of trypsin treatment causes almost complete loss of activity in the viscosity assay (Figs. 4 and 5). Second, a-actinin elicits a clear viscosity response in F-actin solutions even at a-actinin to actin ratios below 0.05 to 1.0 [18, 37]. Hence, over 50~o of total a-actinin molecules in the a-actinin added to the viscosity assays would have to be destroyed in the first 5 min of tryptic treatment to reduce undigested a-actinin concentrations below the level required to produce a detectable viscosity response. Destruction of 50 ~ of the a-actinin molecules in the first 5 min of tryptic digestion is not consistent with the sedimentation, pH stat, and turbidity results presented in this paper. Therefore, the results on tryptic digestion of --actinin indicate that the ability of a-actinin to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions may not be the direct result of a-actinin's ability to cross-link F-actin filaments as measured by a-actinin's capacity to increase F-actin viscosity. Although experimental evidence on the exact nature of the a-actinin- F-actin complex is still lacking, it may be suggested that simple monovalent binding of a-actinin to F-actin filaments is sufficient to cause the a-actinin-induced increase in Mgz+modified ATPase activity and rate of turbidity response of reconstituted actomyosin suspensions, whereas bivalent binding to different F-actin filaments would obviously be required for cross-linking of actin strands by a-actinin. Our previous pH stat studies [12, 18] show that approximately 30~o of all peptide bonds split by trypsin during 60 min of incubation with a-actinin are split during the first 5 rain in a rapid reaction that cleaves 1.4.10 -7 tool of bonds per rain.
455 Cleavage of this group of peptide bonds, however, has little or no effect on the sedimentation pattern of purified u-actinin. Evidently, this group of peptide bonds residues in regions of the u-actinin molecule where noncovalent forces are sufficiently strong to maintain the three-dimensional structure of the a-actinin, even after these bonds have been split. That much of the ability of a-actinin to increase F-actin viscosity is lost during this same period when no large changes evidently occur in conformation of the a-actinin molecule may be explained in several ways. For example, the first 5 min of tryptic digestion may remove a peptide that is essential in one of the two binding sites of u-actinin for actin but that is so small that its loss causes no large change in sedimentation coefficient of the 206 000 dalton, a-actinin molecule [19]. Alternatively, bivalent binding to two F-actin filaments may strain the a-actinin molecule sufficiently to disrupt the noncovalent forces holding the trypsin-treated molecule with its cleaved peptide bonds together, and this disruption may then result in loss of one of a-actinin's two binding sites for actin. That the ability of a-actinin to cross-link F-actin filaments seems to be segregated by trypsin treatment from its ability to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions may suggest that a-actinin, like myosin, could actually have several different physiological roles. Thus, a-actinin may act both to bind F-actin filaments to each other or to certain basal structures in some nonmuscle motile systems and to modify conformation of G-actin monomers in the F-actin filaments to which it binds in a way that enhances the effectiveness of these monomers in supporting movement. A dual physiological role for a-actinin may explain conservation of a-actinin's properties in widely divergent motile systems [40]. Although the ATPase and the filament-forming properties of myosin are clearly separated into different, large fragments by brief tryptic digestion, the F-actin cross-linking and enhancement of contraction activities of a-actinin are not so clearly separated into different fragments by tryptic digestion. Further studies of the tryptic fragments of a-actinin are necessary to determine whether trypsin releases a fragment that binds monovalently to F-actin and increases the effectiveness of F-actin in supporting movement. ACKNOWLEDGEMENTS We are grateful to Joan Andersen for skillful and devoted assistance with preparation of this manuscript. We also thank Joanne Temple, Jean Fatka, Jackie Harvey, Darlene Markley, Mary Bremner, and Diane Rath for indefatigable and expert technical assistance. This work was supported in part by Grants AM-12654 and HL-15679 from the National Institutes of Health, and by grants from the Muscular Dystrophy Association of America, from the American Heart Association (No. 71-679), and from the Iowa Heart Association. During part of this work, Dr. Atsushi Suzuki was recipient of a Visiting Professorship from Training Grant No. FD-00005-05, Food and Drug Association, U.S. Public Health Service. REFERENCES 1 Ebashi, S., Ebashi, F. and Maruyama, K. (1964) Nature 203, 645-646 2 Ebashi, S. and Ebashi, F. (1965) J. Biochem. 58, 7-12
456 3 Seraydarian, K., Briskey, E. J. and Mommaerts, W. F. H. M. (1967) Biochim. Biophys. Acta 133,399-411 4 Maruyama, K. and Ebashi, S. (1965) J. Biochem. 58, 13-19 5 Briskey, E. J., Seraydarian, K. and Mommaerts, W. F. H. M. (1967) Biochim. Biophys. Acta 133,424-434 6 Goll, D. E., Mommaerts, W. F. H. M. and Seraydarian, K. (1967) Federation Proc. 26, 499 7 Goll, D. E., Mommaerts, W. F. H. M., Reedy, M. K. and Seraydarian, K. (1969) Biochim. Biophys. Acta 175, 174-194 8 Masaki, T., Endo, M. and Ebashi, S. (1967) J. Biochem. 62, 630-632 9 Schollmeyer, J. E., Goll, D. E., Robson, R. M. and Stromer, M. H. (1973) J. Cell Biol. 59, 306a I0 Schollmeyer, J. E., Stromer, M. H. and Goll, D. E. (1972) Biophys. J. 12, 280a 11 Schollmeyer, J. V., Goll, D. E., Stromer, M. H., Dayton, W., Singh, I. and Robson, R. M. (1974) J. Cell Biol. 63,303a 12 Suzuki, A., Goll, D. E., Stromer, M. H., Singh, I. and Temple, J. (1973) Biochim. Biophys. Acta 295, 188-207 13 Robson, R. M. and Zeece, M. G. (1973) Biochim. Biophys. Acta 295, 208-224 14 Robson, R. M., Goll, D. E., Arakawa, N. and Stromer, M. H. (1970) Biochim. Biophys. Acta 2130, 296--318 15 Robson, R. M., Huiatt, T. W., Zeece, M. G., Goll, D. E. and Stromer, M. H. (1976) submitted 16 Schollmeyer, J. V., Goll, D. E., Tilney, L., Mooseker, M., Robson, R. and Stromer, M. (1974) J. Cell Biol. 63,304a 17 Arakawa, N., Robson, R. M. and Goll, D. E. (1970) Biochim. Biophys. Acta 200, 284-295 18 Goll, D. E., Suzuki, A., Temple, J. and Holmes, G. R. (1972) J. Mol. Biol. 67, 469-488 19 Suzuki, A., Goll, D. E., Singh, I., Allen, R. E., Robson, R. M. and Stromer, M. H. (1976) J. Biol. Chem., in the press 20 Maruyama, K. and Kimura, S. (1971) J. Biochem. 69, 983-985 21 Kawamura, M., Masaki, T., Nonomura, Y. and Maruyama, K. (1970) J. Biochem. 68, 577-580 22 von Hippel, P. H. and Harrington, W. F. (1959) Biochim. Biophys. Acta 36, 427-447 23 Harrington, W. F., von Hippel, P. H. and Mihalyi, E. (1959) Biochim. Biophys. Acta 32, 303-304 24 Halsey, J. F. and Harrington, W. F. (1973) Biochemistry 12, 693-701 25 Burke, M., Himmelfarb, S. and Harrington, W. F. (1973) Biochemistry 12, 701-710 26 Lowey, S., Slayter, H. S., Weeds, A. G. and Baker, H. (1969) J. Mol. Biol. 42, 1-29 27 Margossian, S. and Lowey, S. (1973) J. Mol. Biol. 74, 301-311 28 Margossian, S. and Lowey, S. (1973) J. Mol. Biol. 74, 313-330 29 Mihalyi, E. and Harrington, W. F. (1959) Biochim. Biophys. Acta 36, 447-466 30 Slayter, H. S. and Lowey, S. (1967) Proc. Natl. Acad. Sci. 58, 1611-1618 31 Weber, K. and Osborn, M. (1969) J. Biol. Chem. 244, 440~4412 32 Starr, R. and Offer, G. (1971) FEBS Lett. 15, 40-44 33 Offer, G., Moos, C. and Starr, R. (1973) J. Mol. Biol. 74, 653-676 34 Gornall, A. G., Bardawill, C. J. and David, M. M. (1949) J. Biol. Chem. 177, 751-766 35 Robson, R. M., Goll, D. E. and Temple, M. J. (1968) Anal. Biochem. 24, 339-341 36 Worthington Biochemical Corporation, Freehold, New Jersey, 1969 Catalog 37 Holmes, G. R., Goll, D. E. and Suzuki, A. (1971) Biochim. Biophys. Acta 253,240-253 38 Gergely, J. (1950) Fed. Proc. 9, 176 39 Stromer, M. H. and Goll, D. E. (1972) J. Mol. Biol. 67, 489-494 40 Singh, I., Goll, D. E., Robson, R. M. and Stromer, M. H. (1976) submitted
© Elsevier/North-Holland Biomedical Press BBA 37470 E F F E C T OF TRYPSIN ON RABBIT S K E L E T A L M U S C L E a-ACTININ*
GERALD R. HOLMES, DARREL E. GOLL**, A. SUZUKI***, R. M. ROBSON and M. H. STROMER Muscle Biology Group, Departments of Animal Science, Biochemistry and Biophysics, and Food Technology, Cooperating, Iowa State University, Ames, Iowa 50011 (U.S.A.)
(Received April 20th, 1976)
SUMMARY 5 min of tryptic digestion of purified rabbit skeletal a-actinin decreases by approximately 75 % the ability of a-actinin to cross-rink F-actin filaments as measured viscometrically at 27 °C, but has little effect on the sedimentation coefficient of a-actinin at 20 °C or on a-actinin's ability to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions. Twenty to sixty min of trypsin treatment reduces the sedimentation coefficient of a-actinin and destroys much of tt-actinin's ability to increase the Mg2+-modified ATPase and rate of turbidity increase of reconstituted actomyosin suspensions. Therefore, the ability of a-actinin to increase the rate of in vitro measures of muscle contraction may not result directly from a-actinin's ability to cross-link F-actin filaments. Trypsin does not split a-actinin into large fragments as it does myosin. Previous studies have shown that 35 to 65 ~o of total tryptic-susceptible peptide bonds i'n a-aetinin are split after 60 rain of incubation with t~ypsin and that 30 % of these bonds sprit in 60 min are cleaved during the first 5 rain in a rapid reaction. That splitting of this group of peptide bonds has little effect on the sedimentation coefficient of a-actinin indicates that these bonds are located in a region of the a-actinin molecule where noncovalent forces are strong enough to maintain conformation of the native a-actinin molecule even after these bonds have been split. This ostensible segregation of a-actinin's ability to cross-link F-actin filaments from its ability to increase rate of in vitro assays of contraction by tryptic digestion may suggest that a-actinin could have at least two different physiological roles: (1) to bind actin filaments to each other or to basal structures, and (2) to enhance the effectiveness of actin in supporting movement.
" Journal Paper No. J-8450 of the Iowa Agriculture and Home Economics Experiment Station, Projects 1795, 1796, and 2025. ** Address inquiries to Darrel E. Goll, Department of Nutrition and Food Science, University of Arizona, Tucson, Ariz. 85721, U.S.A. *'" Present address: Department of Animal Science, Faculty of Agriculture, the University of Niigata, Niigata, Japan.
446 INTRODUCTION a-Actinin was discovered as a new protein component of skeletal muscle myofibrils because of the ability of low ionic strength extracts of crude myofibrillar residues to accelerate the rate of turbidity development in reconstituted actomyosin suspensions [1-3]. This discovery was soon followed by the finding that a-actinin exerts its effects on actomyosin suspensions by combining with the actin part of the actin-myosin complex [2-4]. The early studies on a-actinin/F-actin mixtures [4, 5] indicated that a-actinin cross-linked F-actin filaments under certain in vitro conditions, and it was subsequently shown that a-actinin is locafized in Z-disks of skeletal myofibrils [6-11], which is the only place where cross-linking of actin filaments occurs in skeletal muscle cells. The available evidence, based on yields of purified a-actinin obtainable from skeletal muscle, indicates that a-actinin is a relatively minor component of skeletal myofibrils and makes up less than 3 ~ of total myofibrillar protein by weight [12-14]. That a-actinin has been found in every motile system that has been examined for its presence [9-11, 13, 15, 16], including systems such as smooth muscle [9, 15], brush borders of intestinal epithelial cells [16], acrosomal processes of Limulus sperm [16], and fibroblasts [16] that have no Z-disks, suggests, despite its relatively small content in skeletal myofibrils, that a-actinin may have a fundamental role in motility. Although it is now possible to obtain highly purified a-actinin in sizable quantities [14, 17-19], the nature of a-actinin's role in motility remains unclear. The ability of a-actinin to increase the Mg2+-modified ATPase activity [14, 17] and rate of turbidity development [2, 14, 17] of reconstituted actomyosin suspensions may indicate that a-actinin can modify the structure of actin monomers while these monomers are in the aggregated filamentous state and thereby enhance the effectiveness of actin in supporting contraction [20]. It is not immediately obvious, however, how a-actinin located in the Z-disk of skeletal myofibrils can affect the actin-myosin interaction that occurs in situ at the overlap of thick and thin filaments approximately 0.5 #m distant. On the other hand, the presence of a-actinin in Z-disks of skeletal myofibrils and its ability to cross-link actin filaments in vitro [21] may indicate that the role of a-actinin is to bind actin filaments to each other in skeletal myofibrils or to certain basal structures in some nonmuscle motile systems. Controlled digestion by proteolytic enzymes such as trypsin has been very useful in studying structure of fibrous proteins such as collagen [22-24], paramyosin [25], and myosin [26-29]. Indeed, carefully controlled proteolytic digestion of myosin used in conjunction with electron microscope examination of the proteolytic fragments [30] has localized many of the biological properties of myosin to selected parts of the myosin molecule. We have, therefore, examined the effects of trypsin on some physical and biological properties of purified skeletal muscle a-actinin to determine whether proteolytic digestion might enable us to segregate different biological properties of a-actinin to different parts of the molecule. The results indicate that trypsin does not break skeletal muscle a-actinin into several large fragments as it does myosin but that tryptic digestion seems to decrease the ability of a-actinin to cross-link actin filaments before affecting its ability to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions.
447 MATERIALS AND METHODS Back and leg muscles of rabbits were used for all experiments reported here. Rabbits were anesthesized with sodium pentobarbital and o-tubocurarine chloride [17], and muscleswere handled as described by Arakawa et al. [17]. Unless specified otherwise, all preparations were done at 0 to 4 °C with precooled solutions and with double-deionized, distilled water that had been redistiUed in glass and stored in polyethylene containers.
Protein preparations A crude P0-30 a-actinin extract was made according to Goll et al. [18] and was purified by successive chromatography on two DEAE-ceUulose columns [14]. Actin was made according to Arakawa et al. [17], and myosin was made as described by Seraydarian et al. [3]. All protein preparations were routinely monitored for purity by using polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate [31]. As discussed previously [19], our myosin preparations were not chromatographically purified and therefore contained the B- and C-proteins described by Offer and coworkers [32, 33]. Our actin preparations contained 1 to 2 ~o of their total protein as a-actinin; this small residue of a-actinin is detectable only at very heavy loads [35-45/tg] on sodium dodecyl sulfate-polyacrylamide gels and is also impossible to remove without using column chromatography [19]. The presence of these small amounts of contaminants in our actin and myosin preparations had no evident effects on sensitivity of our assays for a-actinin activity. Protein concentrations were determined by the biuret method [34] as modified by Robson et al. [35]. Trypsin digestion Salt-free, 2 times crystallized trypsin was purchased from Sigma Chemical Co. and possessed 16 000 to 20 000 p-toluenesulfonyl-L-arginine methyl ester units of activity per mg [36]. Trypsin was dissolved at 2.0 mg/ml in 0.001 M HC1 and then diluted to 1.0 mg/ml in 160 mM Tris/acetate buffer (pH 7.2) just before use. Stock trypsin solutions were stable for 3 to 4 months at 2 °C. Soybean trypsin inhibitor (purest salt-free form from Sigma Chemical Co.) was dissolved in water at 4 to 10 mg/ml. Trypsin treatment of a-actinin was done in 112 mM KCI, 56 mM Tris/ acetate, pH 8.0, at 25 °C. a-Actinin concentration was 5.97 mg/ml, and the trypsin to a-actinin ratio was 1:50 by weight. Trypsin digestion was stopped by adding 4 parts of soybean trypsin inhibitor to part of trypsin by weight. Controls were made by incubating a-actinin for 60 rain at 25 °C under ionic conditions identical to those used for trypsin treatment and then adding trypsin and soybean trypsin inhibitor that had been premixed in a ratio of 1:4 by weight. Details on the ionic conditions and protein concentrations used during the different assays of trypsin-treated a-actinin will be given with the individual experiments. ATPase and turbidity assays Reconstituted actomyosin was made by mixing 2 parts of myosin to 1 part actin, w/w, and then washing as described by Arakawa et al. [17]. Ability of a-actinin to increase the MgZ+-modified ATPase activity and rate of turbidity increases in reconstituted actomyosin suspensions was assayed according to our previously
448 described procedures [17]. Details on ionic conditions and experimental protocol will be given with the individual experiments. As we have discussed previously [7, 17], total final absorbance under the conditions we use for the turbidity assay of a-actinin is not related to a-actinin activity. Indeed, very active or high concentrations of a-actinin frequently produce a remarkable flocculation of actomyosin suspensions so these suspensions settle to the bottom of the curvette before a high final absorbance is obtained (for example, see Fig. 3 in the present paper). Tris/acetate buffers were prepared by dissolving an appropriate amount of Tris, free base, and then adding glacial acetic acid to obtain the desired final pH.
Viscosity measurements Viscosity measurements were done by using Ostwald viscometers with flow times of 65 to 70 sec for water at 20 °C. 5-ml samples were used in the viscometers. Specific viscosities were calculated by subtracting one from the ratio obtained by dividing flow times of protein solutions with flow times of the respective solvents. No kinetic energy corrections were made when calculating specific viscosities. Waterbath temperatures were controlled to within 4- 0.05 °C for all viscosity measurements. We have previously found [37] that G-actin containing very little a-actinin aggregates to F-actin very slowly at 0 °C, even in the presence of a-actinin; therefore, regardless of the temperature at which the experiment was done, G-actin was converted to F-actin by incubating at 27 °C for 15 min. We have previously shown [37] that the effect of a-actinin of F-actin viscosity can be altered reversibly by varying temperature between 0 and 37 °C so temperature of the conversion of G to F-actin had no irreversible effects on the results of our viscosity experiments. Two different experimental procedures were used to investigate the effect of tryptic digestion on the ability of a-actinin to increase F-actin viscosity. In the first procedure, purified a-actinin was treated with 1 part of trypsin to 50 parts a-actinin for different periods of time between 0 and 60 min at 25.0 °C. After a selected duration of trypsin treatment, digestion was stopped with soybean trypsin inhibitor, the treated a-actinin was mixed with G-actin, and the actin in this mixture was converted to F-actin by adding MgC12 to a final concentration of 2 mM at 27 °C. Viscosity of the a-actinin/F-actin mixture was then measured at 0 °C or 27 °C. Control (zero-time) samples were made by stirring purified a-actinin under the same ionic conditions as the treated samples at 25 °C for 60 min and then adding premixed trypsin-soybean trypsin inhibitor. The second procedure used to assay the effects of trypsin on a-actinin's ability to increase F-actin viscosity involved mixing G-actin and a-actinin in a viscometer at 27 °C, adding MgCI2 to a final concentration of 2 mM to convert the G-actin to F-actin, and then measuring viscosity of the resulting mixture. 18 min after conversion of G to F-actin, 1 part of trypsin in 0.1 ml to 50 parts of actin, by weight, was added directly to the viscometer, and viscosity was measured again as quickly after mixing as possible. The control in this procedure was done by adding 0.10 ml buffer solution containing no trypsin. Details on protein concentrations and ionic conditions will be given with the individual experiments.
Analytical ultracentrifugation Analytical ultracentrifuge studies were done with a Spinco Model E analytical
449 ultracentrifuge equipped with phase plate, rotor temperature indicator control unit, electronic speed control, and a high-temperature accessory. Kel-F centerpieces were used, and plates were measured with a Nikon 6C profile projector. Sedimentation velocity runs were done at 20.0 °C, but because all sedimentation runs were done in identical solvents at identical concentrations and were for comparative purposes only, no attempt was made to correct the observed sedimentation coefficients to water or zero protein concentration. RESULTS
Effect of trypsin on sedimentation diagram of a-actinin We have previously reported [12, 18] that skeletal muscle a-actinin is moderately susceptible to degradation by trypsin with 35 to 65% of tryptic-labile bonds (as calculated from the sum of the lysine and arginine contents of purified a-actinin) hydrolyzed after 60 rain of trypsin treatment at p H 8.5, 25.0 °C, and trypsin to purified a-actinin ratios o f 1:50, by weight. Fig. 1 shows that 60 rain of trypsin treatment of purified rabbit skeletal muscle a-actinin under these same conditions diminishes both the size and the sedimentation coefficient of the 6 S a-actinin peak in the analytical ultracentrifuge. It is clear from Fig. 1, however, that trypsin does not cleave the a-actinin molecule into two large fragments as it does myosin [26-30]. Instead, trypsin seems to "nibble" at the a-actinin molecule, releasing small fragments
Fig. 1. Sedimentation patterns of purified rabbit skeletal a-actinin after different times of tryptic digestion. Tryptic digestion was done as described in Materials and Methods. Ionic conditions during the analytical ultracentrifuge run and the observed sedimentation coefficient of the main boundary in each of the runs a r e g i v e n in the figure. Temperature of analytical ultracentrifuge run = 20.0 °C, and phase plate angle = 70 °.
450 that do not sediment as a distinct boundary and leaving a large fragment whose observed sedimentation coefficient gradually decreases from 6.03 S to 5.03 S after 60 min of digestion (Fig. 1). It therefore seems that use of trypsin to study substructure of the a-actinin molecule will require a careful and detailed analysis of the fragments released after different times of treatment and will not be as simple as the analysis initially afforded by the observation that trypsin splits myosin into two large flagments [38]. Such a detailed analysis of the products produced by tryptic digestion of purified a-actinin is currently in progress and will be reported in a subsequent paper (Arakawa, Robson, Huiatt, and Goll, manuscript in preparation).
Effect of trypsin on ability of a-actinin to increase in vitro contractile responses Although the effects of trypsin on the sedimentation pattern of skeletal muscle a-actinin do not immediately produce new insights into substructure of the a-actinin molecule, it was also of interest to learn whether trypsin would segregate some of the biological properties of the a-actinin molecule. Increasing time of trypsin treatment between 0 and 60 min gradually reduces ability of a-actinin to increase the Mg z+-
0,35
o. o
0,25 z ~- 0.20
~0,15 > 0.I0
0.05
ooo
~
CONTROL. ACTOMYOSIN ONLY
~o
/5
;o
~s
~o
~5
4'o
~5
;o
~5
6'o
TIME OF TRYPSINTREATMENT(MINUTES)
Fig. 2. Effect of trypsin treatment of rabbit skeletal a-actinin on its ability to increase the Mg2+modified ATPase activity of reconstituted actomyosin. Tryptic digestion of a-actinin was done as described in Materials and Methods. Conditions of ATPase assay were: 100 mM KCI, 20 mM Tris/acetate, pH 7.0, 1 mM MgC12,0.05 mM CaCI~, 1 mM ATP, 0.2 mg reconstituted rabbit skeletal actomyosin/ml, 0.04 nag a-actinin/ml when added, 25.0 °C. Activity of actomyosin in the absence of a-actinin is shown on the ordinate. modified ATPase activity (Fig. 2) or the rate of turbidity increase (Fig. 3) of reconstituted actomyosin suspensions. Very little of a-actinin's ability to increase these t w o in vitro measures of contractile activity remained after 60 rain of tryptic digestion at trypsin-to-a-actinin ratios of 1:50, by weight (Figs. 2 and 3). Ability of a-actinin to increase the MgZ+-modified ATPase activity and rate of turbidity response of reconstituted actomyosin suspensions, however, decreased only slightly during the first 5 min of trypsin digestion (Figs. 2 and 3), even though our earlier p H stat
451
04
/f 20mi=~ £0rain CONTROL
i0.3 Omin
0,I o,c
~
ib
~'5
2'o
2'5
3'o
3'5
4
4
55
TIME (MINUTES)
Fig. 3. Effectof trypsin treatment of rabbit skeletal a-aetinin on its ability to acceleraterate of turbidity increase in reconstituted actomyosin suspensions.Tryptic digestion of a-~tinin was done as de~ribed in Materials and Methods. Conditions of the turbidity assay were: 100 mM KCI, 20 mM Tris/aeetate, pH 7.0,1 mM MgCI2,0.05 mM CaCI2,1 mM ATP, 0.4 mg reconstitutedrabbit skeletalactomyosin/ml, 0.02 mg a-actinin/ml when added, 25.0 °C. studies [12, 18] showed that approximately 30 ~o of all peptide bonds split by trypsin during the first 60 min of incubation with a-actinin are split during the first 5 min in a rapid reaction that cleaves 1.4.10 -7 mol of bonds/min. The analytical ultracentrifugal and ATPase and turbidity results presented here suggest that cleavage of this group of peptide bonds has little effect on either the sedimentation coefficient of the ability of a-actinin to increase the Mg2+-modified ATPase activity or rate of turbidity increase of reconstituted actomyosin suspensions. Indeed, an average of two different experiments indicated that only 7~o of a-actinin's ability to increase the Mg 2+modified ATPase activity of reconstituted actomyosin suspensions was lost during the first 5 min of trypsin treatment under the conditions used in this study. Most of the trypsin-induced loss in ability of tt-actinin to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions occurs between 5 and 20 min of trypsin digestion (Figs. 2 and 3) when only approximately 24 ~o of total bonds split during 60 min of tryptic digestion are cleaved. An average of two different experiments showed that approximately 30 ~o of a-actinin's ability to increase the Mg2+-modified ATPase activity of reconstituted actomyosin suspensions was lost between 5 and 20 rain of trypsin treatment. A large decrease in sedimentation coefficient from 6.0 S to 5.5 S also occurs between 5 and 20 rain of tryptic digestion of a-actinin (Fig. 1). Hence, peptide bonds split by trypsin between 5 and 20 rain of digestion have a larger effect on a-actinin's structure as measured by sedimentation coefficient and ability to +increase in vitro measures of contraction than do those cleaved between 0 and 5 min of digestion
452
Effect of trypsin on ability of a-actinin to increase F-actin viscosity The ability of a-actinin to increase viscosity of F-actin is ostensibly due to its capacity to cross-link F-actin filaments. As described in Materials and Methods, the effect of tryptic digestion of a-actinin on its ability to cross-link actin filaments as measured viscometrically was assayed in two different ways. The results of the first method, in which a-actinin was treated with trypsin before being mixed with actin, are shown in Fig. 4. Control a-actinin samples, made simply by stirring at 25 °C for 60 rain before adding premixed trypsin and soybean trypsin inhibitor, had already lost some of their ability to increase the viscosity of F-actin at 27 °C (Fig. 4); this loss was especially noticeable at 0 °C when 0.02 part of a-actinin was added to 1 part actin by weight (Fig. 4). In contrast to the effects of trypsin treatment on ability of a-actinin to increase the Mg z+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions, the first 5 min tryptic digestion of a-actinin destroyed approximately 75 % of the ability of c~-actinin to increase F-actin viscosity at 27 °C (Fig. 4); longer tryptic treatment up to 60 min had little additional effect on c~-actinin's ability to increase F-actin viscosity at 27 °C. In contrast to its destruction of a-actinin's ability to increase F-actin viscosity at 27 °C, tryptic digestion up to 60 min had only a slight effect on a-actinin's ability to increase F-actin viscosity at 0 °C (Fig. 4). We [18, 37, 39] have previously shown that the a-actinin/F-actin interaction at 0 °C is completely different from the ~-actinin/F-actin interaction at higher temperatures, but the cause of this difference is still unknown. 3,0~
2. ~. W~¢-"-"
2% Untreated
a-Actinin,O °
2.0 10% Untreated Q-Actinin,O*
nsp
/ '~'~
I,C
•~
O,5X
2% Trypsin-treated
~-Actinin,
10% Trypsin-treated
a-Actinin,
s-Actinin,0 ~
o °
1o% Untreated ~ - A c t i n i n , 27°
} Control
0,1 . . . . . 0.
10% Trypsin-treated
27 °
- No Q-Actinin
5
I~
'
' ' 3'0 . . 20 TIME OF TRYPSINTREATMENT(MINUTES)
. . NO
.
.
50
60
Fig. 4. Effect of trypsin treatment of rabbit skeletal a-actinin on its ability to increase viscosity of actin polymerized by Mg 2+ at 27.0 °C. Tryptic digestion of (t-actinin was done as described in Materials and Methods. Rabbit skeletal G-actin was converted to F-actin at 27 °C in the absence or presence of trypsin-treated or untreated a-actinin by addition of Mg 2+, and viscosity of the resulting solution was measured. Final conditions of viscosity assay were: 1.0 rag actin/ml, a-actinin indicated as percentage of actin present by weight, 2 m M MgC12, 20 m M Tris/acetate, pH 7.5. All samples were polymerized at 27 °C but some were measured at 27 °C and some were measured at 0 °C. Untreated a-actinin samples were not subjected to any experimental handling before being mixed with actin in the viscometer; zero-time samples were treated as described in Materials and Methods.
453 1,50
1.25 Added Trypsin or B ff r
1
1.00
In
.
0 ° Bath
10~ a-Actinin
Plus
Buffer
I
nsp io~ ~-Actxnin Plus Trypsin
0,75 0% a-Actinin Plus Buffer
0,50 0% n-Actxnin Plus Trypsin
0.25
f}.0
I
I
I
I
I
rIME AFTERPOLYMERIZATION(MINUTES)
Fig. 5. Effect of trypsin on viscosity of the a-actinin-F-actin complex. Rabbit skeletal G-actin, in the presence or absenceof rabbit skeletal a-actinin in an Ostwald viscometer, was converted to F-actin
by addition of MgCI2 at 27 °C, and viscosity of the resulting solution was measured at the same temperature. Eighteen rain after the first viscosity measurement, 1 part of trypsin in 0.10 ml was added to 50 parts of actin by weight in the viscometer, and viscosity of the resulting mixture measured. 12 min later, the viscometer was placed in a 0 °C bath and viscosity measured again. Control samples had 0.10 ml buffer added in place of the trypsin. Final concentrations: 1.0 mg actin/ml, a-actinin indicated as percentage of actin present by weight, 2 mM MgCI2, 20 mM Tris/acetate, pH 7.5. The second method used to assay the effects of trypsin on a-actinin's ability to increase F-actin viscosity involved measuring the effects of trypsin on viscosity of a preformed a-actinin/F-actin complex (Fig. 5). The results show that, 5 to 10 min after addition of trypsin to the viscometer, viscosity of the a-actinin/F-actin mixture had decreased nearly to the viscosity of F-actin alone (Fig. 5). Addition of buffer instead of trypsin caused a small decrease in specific viscosity of a-actinin/F-actin mixtures at 27 °C (Fig. 5), but this effect was not nearly as large as the effect caused by addition of trypsin. Consequently, the results of this second method for assaying the effects of trypsin on a-actinin's ability to increase F-actin viscosity agree with the results of the first procedure used to assay this effect and show that trypsin almost completely eliminates the ability of a-actinin to increase F-actin viscosity at 27 °C within 5 to 10 min. Putting the viscometers containing the treated and control a-actinin/F-actin mixtures in a 0 °C bath 12 min after addition of trypsin or buffer resulted in an immediate increase in specific viscosity of both trypsin-treated and control a-actinin/F-actin mixtures, although viscosity of the trypsin-treated mixture did not increase to quite the same extent as viscosity of the control mixture (Fig. 5). Specific viscosity of trypsin-treated and control solutions of F-actin without a-aetinin was not affected by immersion in a 0 °C bath. These results, therefore, substantiate the findings obtained by the first procedure used to assay the effects of trypsin on a-actinin's ability to increase F-actin viscosity and show that trypsin treatment has less effect on a-actinin's ability to increase F-actin viscosity at 0 °C than it has on a-actinin's ability to increase F-actin viscosity at 27 °C.
454 DISCUSSION The results of this study show that 5 min of trypsin treatment decreases aactinin's ability to cross-link F-actin filaments as measured viscometrically at 27 °C by approximately 75 ~ , but has little effect on sedimentation pattern of a-actinin at 20 °C or on a-actinin's ability to increase the Mg2+-modified ATPase activity or rate of turbidity increase of reconstituted actomyosin suspensions at 25 °C. Tryptic treatment for 20 to 60 min eventually reduces the sedimentation coefficient of a-actinin and also decreases its ability to increase the rate of in vitro measures of contraction. These results suggest that tryptic digestion reduces a-actinin's ability to cross-link F-actin filaments, as measured by a-actinin's capacity to increase F-actin viscosity, more rapidly than it reduces a-actinin's ability to increase the Mg 2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions. Although it may be suggested that the rapid tryptic-induced loss of a-actinin's ability to cross-link F-actin filaments results in part from the lower a-actinin-to-actin ratios used in the viscosity experiments (viz., 0. I0 parts a-actinin to 1.0 part actin in viscosity experiments compared with 0.15 parts a-actinin to 1.0 part actin in the turbidity experiments), so that tryptic digestion rapidly reduces native a-actinin concentration below some threshold level required for activity, this possibility seems unlikely for several reasons. First, it seems very improbable that 5 min of tryptic treatment when the a-actinin to actin ratio is 0.10 to 1.0 would lower the proportion of undigested a-actinin to a level that gives no detectable response, but that 20 min of tryptic treatment when the a-actinin to actin ratio is 0.15 to 1.0 would not also lower the proportion of undigested a-actinin to a level that gives no detectable response. Yet, our results show that, even after 20 min of trypsin treatment, a-actinin retains considerable activity in the turbidity assay (Fig. 3), though 5 min of trypsin treatment causes almost complete loss of activity in the viscosity assay (Figs. 4 and 5). Second, a-actinin elicits a clear viscosity response in F-actin solutions even at a-actinin to actin ratios below 0.05 to 1.0 [18, 37]. Hence, over 50~o of total a-actinin molecules in the a-actinin added to the viscosity assays would have to be destroyed in the first 5 min of tryptic treatment to reduce undigested a-actinin concentrations below the level required to produce a detectable viscosity response. Destruction of 50 ~ of the a-actinin molecules in the first 5 min of tryptic digestion is not consistent with the sedimentation, pH stat, and turbidity results presented in this paper. Therefore, the results on tryptic digestion of --actinin indicate that the ability of a-actinin to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions may not be the direct result of a-actinin's ability to cross-link F-actin filaments as measured by a-actinin's capacity to increase F-actin viscosity. Although experimental evidence on the exact nature of the a-actinin- F-actin complex is still lacking, it may be suggested that simple monovalent binding of a-actinin to F-actin filaments is sufficient to cause the a-actinin-induced increase in Mgz+modified ATPase activity and rate of turbidity response of reconstituted actomyosin suspensions, whereas bivalent binding to different F-actin filaments would obviously be required for cross-linking of actin strands by a-actinin. Our previous pH stat studies [12, 18] show that approximately 30~o of all peptide bonds split by trypsin during 60 min of incubation with a-actinin are split during the first 5 rain in a rapid reaction that cleaves 1.4.10 -7 tool of bonds per rain.
455 Cleavage of this group of peptide bonds, however, has little or no effect on the sedimentation pattern of purified u-actinin. Evidently, this group of peptide bonds residues in regions of the u-actinin molecule where noncovalent forces are sufficiently strong to maintain the three-dimensional structure of the a-actinin, even after these bonds have been split. That much of the ability of a-actinin to increase F-actin viscosity is lost during this same period when no large changes evidently occur in conformation of the a-actinin molecule may be explained in several ways. For example, the first 5 min of tryptic digestion may remove a peptide that is essential in one of the two binding sites of u-actinin for actin but that is so small that its loss causes no large change in sedimentation coefficient of the 206 000 dalton, a-actinin molecule [19]. Alternatively, bivalent binding to two F-actin filaments may strain the a-actinin molecule sufficiently to disrupt the noncovalent forces holding the trypsin-treated molecule with its cleaved peptide bonds together, and this disruption may then result in loss of one of a-actinin's two binding sites for actin. That the ability of a-actinin to cross-link F-actin filaments seems to be segregated by trypsin treatment from its ability to increase the Mg2+-modified ATPase activity and rate of turbidity increase of reconstituted actomyosin suspensions may suggest that a-actinin, like myosin, could actually have several different physiological roles. Thus, a-actinin may act both to bind F-actin filaments to each other or to certain basal structures in some nonmuscle motile systems and to modify conformation of G-actin monomers in the F-actin filaments to which it binds in a way that enhances the effectiveness of these monomers in supporting movement. A dual physiological role for a-actinin may explain conservation of a-actinin's properties in widely divergent motile systems [40]. Although the ATPase and the filament-forming properties of myosin are clearly separated into different, large fragments by brief tryptic digestion, the F-actin cross-linking and enhancement of contraction activities of a-actinin are not so clearly separated into different fragments by tryptic digestion. Further studies of the tryptic fragments of a-actinin are necessary to determine whether trypsin releases a fragment that binds monovalently to F-actin and increases the effectiveness of F-actin in supporting movement. ACKNOWLEDGEMENTS We are grateful to Joan Andersen for skillful and devoted assistance with preparation of this manuscript. We also thank Joanne Temple, Jean Fatka, Jackie Harvey, Darlene Markley, Mary Bremner, and Diane Rath for indefatigable and expert technical assistance. This work was supported in part by Grants AM-12654 and HL-15679 from the National Institutes of Health, and by grants from the Muscular Dystrophy Association of America, from the American Heart Association (No. 71-679), and from the Iowa Heart Association. During part of this work, Dr. Atsushi Suzuki was recipient of a Visiting Professorship from Training Grant No. FD-00005-05, Food and Drug Association, U.S. Public Health Service. REFERENCES 1 Ebashi, S., Ebashi, F. and Maruyama, K. (1964) Nature 203, 645-646 2 Ebashi, S. and Ebashi, F. (1965) J. Biochem. 58, 7-12
456 3 Seraydarian, K., Briskey, E. J. and Mommaerts, W. F. H. M. (1967) Biochim. Biophys. Acta 133,399-411 4 Maruyama, K. and Ebashi, S. (1965) J. Biochem. 58, 13-19 5 Briskey, E. J., Seraydarian, K. and Mommaerts, W. F. H. M. (1967) Biochim. Biophys. Acta 133,424-434 6 Goll, D. E., Mommaerts, W. F. H. M. and Seraydarian, K. (1967) Federation Proc. 26, 499 7 Goll, D. E., Mommaerts, W. F. H. M., Reedy, M. K. and Seraydarian, K. (1969) Biochim. Biophys. Acta 175, 174-194 8 Masaki, T., Endo, M. and Ebashi, S. (1967) J. Biochem. 62, 630-632 9 Schollmeyer, J. E., Goll, D. E., Robson, R. M. and Stromer, M. H. (1973) J. Cell Biol. 59, 306a I0 Schollmeyer, J. E., Stromer, M. H. and Goll, D. E. (1972) Biophys. J. 12, 280a 11 Schollmeyer, J. V., Goll, D. E., Stromer, M. H., Dayton, W., Singh, I. and Robson, R. M. (1974) J. Cell Biol. 63,303a 12 Suzuki, A., Goll, D. E., Stromer, M. H., Singh, I. and Temple, J. (1973) Biochim. Biophys. Acta 295, 188-207 13 Robson, R. M. and Zeece, M. G. (1973) Biochim. Biophys. Acta 295, 208-224 14 Robson, R. M., Goll, D. E., Arakawa, N. and Stromer, M. H. (1970) Biochim. Biophys. Acta 2130, 296--318 15 Robson, R. M., Huiatt, T. W., Zeece, M. G., Goll, D. E. and Stromer, M. H. (1976) submitted 16 Schollmeyer, J. V., Goll, D. E., Tilney, L., Mooseker, M., Robson, R. and Stromer, M. (1974) J. Cell Biol. 63,304a 17 Arakawa, N., Robson, R. M. and Goll, D. E. (1970) Biochim. Biophys. Acta 200, 284-295 18 Goll, D. E., Suzuki, A., Temple, J. and Holmes, G. R. (1972) J. Mol. Biol. 67, 469-488 19 Suzuki, A., Goll, D. E., Singh, I., Allen, R. E., Robson, R. M. and Stromer, M. H. (1976) J. Biol. Chem., in the press 20 Maruyama, K. and Kimura, S. (1971) J. Biochem. 69, 983-985 21 Kawamura, M., Masaki, T., Nonomura, Y. and Maruyama, K. (1970) J. Biochem. 68, 577-580 22 von Hippel, P. H. and Harrington, W. F. (1959) Biochim. Biophys. Acta 36, 427-447 23 Harrington, W. F., von Hippel, P. H. and Mihalyi, E. (1959) Biochim. Biophys. Acta 32, 303-304 24 Halsey, J. F. and Harrington, W. F. (1973) Biochemistry 12, 693-701 25 Burke, M., Himmelfarb, S. and Harrington, W. F. (1973) Biochemistry 12, 701-710 26 Lowey, S., Slayter, H. S., Weeds, A. G. and Baker, H. (1969) J. Mol. Biol. 42, 1-29 27 Margossian, S. and Lowey, S. (1973) J. Mol. Biol. 74, 301-311 28 Margossian, S. and Lowey, S. (1973) J. Mol. Biol. 74, 313-330 29 Mihalyi, E. and Harrington, W. F. (1959) Biochim. Biophys. Acta 36, 447-466 30 Slayter, H. S. and Lowey, S. (1967) Proc. Natl. Acad. Sci. 58, 1611-1618 31 Weber, K. and Osborn, M. (1969) J. Biol. Chem. 244, 440~4412 32 Starr, R. and Offer, G. (1971) FEBS Lett. 15, 40-44 33 Offer, G., Moos, C. and Starr, R. (1973) J. Mol. Biol. 74, 653-676 34 Gornall, A. G., Bardawill, C. J. and David, M. M. (1949) J. Biol. Chem. 177, 751-766 35 Robson, R. M., Goll, D. E. and Temple, M. J. (1968) Anal. Biochem. 24, 339-341 36 Worthington Biochemical Corporation, Freehold, New Jersey, 1969 Catalog 37 Holmes, G. R., Goll, D. E. and Suzuki, A. (1971) Biochim. Biophys. Acta 253,240-253 38 Gergely, J. (1950) Fed. Proc. 9, 176 39 Stromer, M. H. and Goll, D. E. (1972) J. Mol. Biol. 67, 489-494 40 Singh, I., Goll, D. E., Robson, R. M. and Stromer, M. H. (1976) submitted